Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Gas Dehydration: Part of The Petroleum Processing Technology Series

Download as pdf or txt
Download as pdf or txt
You are on page 1of 714

POL

Petroleum Open Learning

Gas Dehydration
Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY
Petroleum Open Learning

Designed, Produced and Published by OPITO Ltd., Petroleum Open Learning, Minerva House, Bruntland Road, Portlethen, Aberdeen AB12 4QL

Printed by Astute Print & Design, 44-46 Brechin Road, Forfar, Angus DD8 3JX www.astute.uk.com

© OPITO 1993 (rev.2002) ISBN 1 872041 85 X

All rights reserved. No part of this publication may be reproduced, stored in a retrieval or information storage system, transmitted in any form or by any means, mechanical,
photocopying, recording or otherwise without the prior permission in writing of the publishers.
Petroleum Open Learning

Gas Dehydration Petroleum Open Learning

(Part of the Petroleum Processing Technology Series)

Contents Page Visual Cues

* Training Targets 2 training targets for you to


achieve by the end of the unit
* Introduction 3
test yourself questions
* Section 1 - Water in Natural Gas 4 to see how much you
understand
Quantity of Water in Gas
Problems of Water in Gas
Hydrate Prevention
check yourself answers to
let you see if you have been
* Section 2 - Auto Refrigeration 25
thinking along the right lines
The Joules/Thompson Effect
The Low Temperature Separation (LTS) System
The Low Temperature Extraction (LTX) System activities for you to apply
your new knowledge
* Section 3 - Solid Desiccant Dehydration 40
Adsorption
Solid Desiccant Dehydration Plant summaries for you to recap
on the major steps in your
* Section 4 - Liquid Desiccant Dehydration 46 progress

Liquid Desiccants
Glycol Dehydration Plant
1
Petroleum Open Learning

Training Targets
The aim of this unit is to help you understand :

the capacity of natural gas to hold water


the problems which result from the presence of water in gas
the methods used to reduce the water content of natural gas

Upon completion of the unit you should be able to:

• Quantify the amount of water in saturated natural gas under given conditions. q
• List the problems associated with water in gas. q
• Define the conditions that contribute toward hydrate formation. q
• Describe Joules/Thompson Effect. q
• Explain the Auto Refrigeration process. q
• Define Adsorption and Absorbtion. q
• Detail a simple two-tower desiccant dehydration process. q
• Describe a basic glycol dehydration plant. q
• List and explain operational variables in the glycol dehydration process. q

Tick the box when you have met each target.

2
Gas Dehydration Petroleum Open Learning

Introduction
Natural gas can be referred to as Associated Gas or Non-associated Gas. The former is produced together
with liquids from an oil reservoir and is liberated from the liquids at the surface. The latter is produced
independently of an oil accumulation, from what is commonly called a gas reservoir.

Irrespective of whether the gas is associated or non-associated, it invariably contains water in the form of a vapour
or a liquid.

Surface equipment is used to remove the water from the gas. This process is called dehydration, which is the
subject of the present unit.

The unit comprises 4 Sections :

• Section 1, Water in Natural Gas, looks at the amount of water which can be held in gas and discusses the
problems that the water creates. In this section we will also look at options preventing hydrate
formation.

Section 1 will be followed by 3 further sections which deal with process systems used to remove water from gas.

• Section 2, Auto Refrigeration, describes how water is removed by reducing the gas temperature.

• In Section 3, Solid Desiccant Dehydration, you will look at theory of adsorption, and how it is
applied to water removal.

• Finally, in Section 4, Liquid Desiccant Dehydration, we will look at how liquid desiccants work and see
how glycol is used in a typical dehydration plant.

You should be aware that the water removal processes described in Sections 2,3 & 4, are applicable to both
associated and non-associated gas treatment facilities. The actual process chosen for a particular application
depends on a number of factors. These include, location of plant, gas characteristics and so on.

3
Gas Dehydration Petroleum Open Learning

Section 1 - Water in Natural Gas


Natural gas contains water, but how much? In this • Saturated. This is a state in which the gas The main constituents are the following hydrocarbon
section we will look at how we can specify contains the maximum amount of water it can gases:
amount of water contained the gas. We will hold in vapour form.
also review the problems that water in gas can • Methane
cause, and look at inhibiting one of the problems, • Over-Saturated. In an over-saturated state
that of hydrate formation. the gas contains water in excess of the • Ethane
amount it can in its saturated state. The
Quantity of Water in Gas excess water will exist as free liquid. • Propane

Hydrocarbons contained in oil and gas reservoirs • Butane


usually are, or have been, in close contact with
formation water. Any gas is normally, therefore, There are a number of factors that can affect • Pentanes
very wet, at the reservoir temperature and pressure amount of water vapour that may be present in gas.
conditions. Up to a certain point, the water be
held in the gas in the form of vapour. Beyond that These include: The list is longer, but the amounts of other
point the water will appear as liquid. However, hydrocarbons present are usually small.
terms such as very wet are not very scientific, nor • gas composition and gravity
do they provide us with any indication of the actual
water content • temperature The gases are listed above, starting the lightest
at the top. They get heavier, or denser, as you move
• pressure down the list.
In order to rectify this, I have listed below three
terms which are used to describe the three main • the amount of water with which the gas
states of wet gas : been in contact. Density gas is usually expressed as the weight in
pounds per cubic foot at standard conditions of
• Unsaturated. In this state the gas is in a temperature and pressure.
condition where it is able to hold additional
water in the form of vapour.
The composition of natural gas varies, because the
proportions of its constituents will vary from to
field.

4
Petroleum Open Learning

Specific gravity is the more commonly used For the purposes analysing water content, it is
measure of “density”. It is the ratio of a gas density safe to consider natural gas as having a fixed
to the density of air at the same conditions of composition. To further lighten our immediate task,
temperature and pressure. we will initially be looking at the amount of water
needed to render the gas saturated.

As I said a little earlier, the percentage composition


of different natural gases varies. Methane is usually The actual amount of water required to saturate
the most abundant component, and is the principal gas will depend on the pressure and temperature
source of energy in our mains gas supply. As you the gas. This can be represented in the form of a
move down the list you undoubtedly recognise simple graph as shown in Figure 1. Let’s see how
propane and butane, which commonly appear as this figure can be used in practice. Take a look at
bottled liquid gas. Figure 1.

Natural gas also contains impurities such as


hydrogen sulphide (H2S), carbon dioxide (C02) ,
non-combustible gases and water vapour. However
we are going to be concentrating on the water
content.

5
Petroleum Open Learning

6
Petroleum Open Learning

The curve illustrated on the graph is the saturation Answer:


curve for natural gas at 1 000 psi. For varying gas
temperatures in degrees Fahrenheit, it illustrates You can see from the point where the horizontal line
how much water in pounds per standard intersects the vertical axis, that 24 lbs of water are
cubic feet (mmscf) is needed to saturate the gas at required to saturate one cubic feet of gas at
1 000 psi. 70 deg F and 1 000 psi. Or, in abbreviated form 24
Ibs of water per mmscf.
To use the graph, select your temperature on the
horizontal axis, move vertically to intersect the
curve, and then horizontally to find the amount of
water vapour needed for saturation.
Figure 1 gave us the saturation curve for just one
Take an example: pressure. If the curves for other pressures are
illustrated in a similar fashion, the graph as shown in
Question: Figure 2 will be the result.

What is the water content of 1 000 psi natural gas at


saturation, assuming a temperature of 70 deg F?
To find the answer follow the steps listed below.

• find 70 deg F on lower axis

• follow this line up until it intersects the curve

• now move horizontally and read off the figure


on the vertical axis.

7
Petroleum Open Learning

8
Petroleum Open Learning

As you can see, the different pressure curves are This means that, if the temperature of the gas in the
almost parallel, which makes it fairly easy to above example were lowered from 150 deg F to
estimate for intermediate pressures. 120 deg F at a constant pressure, 46 Ibs (94 - 48)
water per mmscf would condense and appear as
free water.
Let’s use Figure 2 :
From the above, a useful fact emerges - one that
Take a reservoir pressure of 3 000 psi and a we will remember and use throughout this unit:
temperature of 150 deg F. Using the method
described earlier, you can see that gas under these • As the temperature drops, the water
conditions will have a saturated water content of vapour required to saturate a given
94 Ibs per mmscf of gas. volume of gas decreases

As I said earlier, up to the saturation point, the water Or, in other words:
in the gas will be in the form of vapour. For the
purposes of this unit, we can consider the water • When the temperature of water-saturated
vapour as being similar in behaviour to a gas. gas is lowered, water vapour condenses
to produce free water.
Now take the same reservoir pressure of 3 000 psi
as in the previous example, but a temperature of Now have a go at the following Test Yourself
120 deg F. Look at the graph in Figure 2 again. question. It should help you to understand what we
This time you will see that only 48lbs of water, in have covered up to now.
the form of water vapour, is required to saturate
every mmscf of gas at these conditions.

From the two examples I have just given, you can


see that gas at the higher temperature of 150 deg F
is capable of holding 46 lbs per mrnsct more water
vapour than if its temperature was 120 deg F.

9
Petroleum Open Learning

Problems with Water in Gas

As gas flows through the reservoir and into the well


Test Yourself 1 bore, it usually becomes saturated with water. In
addition, it picks up free water along the way.

Assume that gas from the reservoir indicated above (3 000 psi, 120 deg F) is produced up a well This is a very important fact regarding natural gas.
to the surface. At the surface the pressure has dropped to 1 500 psi and the temperature to Let me just repeat it :
100 deg F. Now answer the following:
Gas produced to the surface is, in most cases,
saturated with water vapour and is likely to be
a) What water content in lbs/mmscf is required to achieve saturation of the gas at the surface? transporting free water.

As you may gather from the heading, water in gas


is, for most of the time, bad news! Water in gas
b) Will the change in conditions from reservoir to surface result in the gas being unsaturated at gives rise to various problems. Let me now list and
that point, or free water be present? briefly discuss the more important ones.

These are:

• liquid accumulation in the wellbore

• corrosion

• pipeline efficiency
If free water is present, how many pounds will there be for every mmscf of gas?
• gas quality

• hydrate formation

let’s look at each of these problems in turn.

You will find the answers in Check Yourself 1 on page 59

10
Petroleum Open Learning

Liquid Accumulation in the Wellbore


This usually occurs in lower pressure, low flow rate gas
wells. Liquids, of which free water is one, start to build up
in the bottom of the wellbore when the flowing velocity is
too low to lift the liquid to the surface. The wellbore acts
as a separator and the gas bubbles through the liquid.

The liquid column building up in the well causes an


increasing back-pressure to be exerted on the reservoir.
This further reduces the flow rate and thus the velocity of
the well fluids. The process can continue until eventually
the well dies ( ceases to flow) or only flows intermittently.
The deteriorating situation is illustrated in the left hand
sketch in Figure 3.

The normal method for avoiding this problem is to ensure


that the flow velocity is maintained at a level high enough
to prevent fall-back of liquid droplets. The simplest way of
accomplishing this is to increase the flow rate. This is not
always possible, particularly if the well has naturally
declined in performance. Another way of increasing flow
velocity is to install smaller bore production tubing closer
to the perforations. This means that the same amount of
gas has to flow through a smaller cross-sectional area of
tubing. To do this, it must flow more quickly.

This is illustrated in the right hand sketch of Figure 3

There are other techniques which can be used to remove


liquids from the wellbore. However, these are beyond the
scope of this programme and we will not discuss them
here.

11
Petroleum Open Learning

Corrosion Hydrogen sulphide has a similar action, causing Free water can occupy quite a lot of the pipeline volume.
metal loss and pitting when free water is available. This will reduce the amount of pipeline
Corrosion is the next problem on our list. In natural This particularly applies if carbon dioxide-related cross sectional area available for gas flow, resulting
gas systems, corrosion carbon steel occurs when corrosion is also present - they appear to in increased gas flow velocity. The severity of this
free water forms in the presence of carbon dioxide encourage each other! effect will depend on a number of factors:
(C02) or hydrogen sulphide (H2S). You will
remember that these two substances are impurities Protection against carbon dioxide and hydrogen • length of pipeline
which may be found in natural gas. sulphide corrosion attack is provided by:
• flow velocity
Carbon dioxide and water together to form • choice of corrosion-resistant materials (such
carbonic acid which then reacts with an exposed as stainless steel) • undulations in the line
steel surface. The reaction causes chemical
substances to form which are called corrosion • use of protective coatings • the volume of liquid
products.
• application of corrosion inhibitors (chemicals Figure 4a illustrates the effect.
(The most common corrosion product in everyday with special protective properties)
life is, of course, rust, which forms on iron or steel Build up of liquid may continue until the critical point
exposed to the air). at which liquid slugs are formed.
These options may be used singly or in
combination. The choice will be based on both This is shown in Figure 4b.
These corrosion products are removed by the technical and economic factors.
force of the flowing gas stream, exposing fresh
metal for further attack. This action results in metal Pipeline Efficiency
loss and, therefore, corrosion pitting. The rate of
metal loss, called the corrosion rate, depends on Natural gas is usually transported by pipelines, and
many factors, but principally on the amount of water in gas pipelines causes our next problem.
carbon dioxide and free water present. Corrosion
rate also increases dramatically with increase in The presence of free water in a gas pipeline can
temperature. give rise to the complication of two-phase flow. By
two-phase flow we mean that gas and liquid (say,
water) are flowing in the line together. (Gas is one
phase, and liquid is the other).

12
Petroleum Open Learning

Figure 4 a: Liquid Build Up in a Gas Pipeline

Liquid levels can build up in a pipeline, particularly at low spots. This continue until a critical point is
reached. At this point the available flow area is insufficient for the gas flow rate. This results in an
intermittent plug or slug flow which will break the continuity of gas supply at the pipeline destination.

Figure 4 b: Illustration of a Slug Flow in a Pipeline

Usually this is not considered a desirable situation. For example, if the gas is feeding a gas compressor,
this can be seriously damaged by water slugs.

13
Petroleum Open Learning

Gas Quality Hydrates


The influence on quality is the most self evident, As you will see, we have left the most important effect
and undesirable, effect of water in gas. Gas is until last!
usually burnt as a fuel and water is used to put fires
out - hardly compatible qualities! Hydrates are solids that form as snow like crystals.
They are created by a chemical reaction between
For the end-user of gas, and for those who transport natural gas and free water. Once formed, hydrate
it, the quality is strictly specified, especially as crystals can pack together in gas processing plant,
regards water content. partially or completely blocking flow lines or
accessories such as valves. The blockages will
The water content specification is usually called the tend to occur at turbulent regions such as pipe
water dewpoint (in order to distinguish it from the bends or changes of diameter.
hydrocarbon dewpoint, which is a separate aspect
of the gas quality specification). Let us have a look One particular danger of hydrate deposits arises
at what this means. when they form a blockage downstream of a
pipework pressure rating change, for example in
As we have seen earlier, when the temperature of flare or vent pipework. This may subject the lower
saturated gas is decreased, some of the water rated pipework to dangerous over pressures.
vapour condenses and appears as free water. Put
another way - the lower the temperature the less
water vapour it takes to saturate a given volume of
gas. Bearing this in mind, the water dewpoint of Figure 5 shows the general effect.
gas is defined as :

The temperature at which natural gas at any


specified pressure is saturated by the water
vapour it contains.

The quality specification for a natural gas will define


a water dewpoint so that:

• water vapour will not condense as free water


under any foreseeable conditions.

14
Petroleum Open Learning

Now a short exercise:

Test Yourself 2

The hydrate temperature must be below, the


same as, but never above the dewpoint
temperature. Why is this?

Figure 5 : Shows a possible effect of a hydrate blockage

Hydrates can occur at temperatures considerably


above the freezing point of water. At a given
pressure and in the presence of free water, hydrates
will form when the temperature of the gas is at or
below a certain level. Understandably, this is called
the hydrate temperature.

You will find the answer in Check Yourself 2


on page 59

15
Petroleum Open Learning

Take a look at Figure 6. This graph shows the


pressure and temperature conditions needed for
hydrate formation, when a typical natural gas is in
contact with free water.

The example given on the graph shows how


hydrates form in gas 400 psi when the
temperature drops to 50 deg F.

Hydrate formation conditions can be shown


graphically in a slightly different way. Have a look at
Figure 7. You will probably recognise this graph as
being very similar to Figure 2. However, this time,
superimposed on the various pressure curves is a
hydrate temperature line. For each pressure, it
indicates a temperature below which hydrates will
form in the presence of free water.

16
Petroleum Open Learning

17
Petroleum Open Learning

As I said earlier, hydrates are undesirable in gas


processing as they can, in certain circumstances,
disrupt the normal degree of control that we should
have over a process.
Test Yourself 3 As the problem of hydrates is so important, let me
list for you once more the conditions which could
lead to their formation :
Using Figure 7, read off the temperature below which hydrates will form in natural gas at
1 500 psi, in the presence of free water. • gas, with free water present

• temperature and pressure conditions


within the hydrate formation region

Have a go at Test Yourself 4.

Check your answer in Check Yourself 3 on page 59

18
Petroleum Open Learning

Now you can work through an example :

Test Yourself 4

a. We have looked at some problems caused water in natural gas. Which of these will
affect our attempts at processing the gas.

b. By reference Figure 7, indicate whether you think that gases at the following conditions
are in the hydrate formation regions (Yes) or not (No).

Yes No

i) 1 500 psi, 40 deg F.

ii) 300 psi, 50 deg F.

iii) 1 500 psi, 70 deg F.

Iv) 500 psi, 60 deg F.

v) 3000 psi, 70 deg

Check your answer in Check Yourself 4 on page 60

19
Petroleum Open Learning

Hydrate Prevention
From what we have covered thus far, you will
remember that hydrate formation was mentioned as
probably being the most troublesome problem
Activity
associated with water in natural gas.

So, how can we prevent these hydrates from Jot down three things that you could do to a gas stream, in order to discourage hydrate formation.
forming?

20
Petroleum Open Learning

You have probably written down something along 1. Lowering the pressure. 3. Heating.
the following lines:
This is not always possible. The reservoir is at a It is certainly possible to discourage hydrate
1. Reduce the pressure certain pressure and we have no control over this. formation by heating the gas. However this is not
2. Remove the water always a practical solution.
3. Raise the temperature

At this point we are going to cheat a little! - by 2. Removal of water. Consider, for example, a long undersea gas
pipeline. This will lose heat to the surrounding
application of lateral thinking. Hydrate formation is a water. It would not be possible to raise the initial
process which may be compared with ice formation This is, of course, what this unit is all about.
Dropping free water out of the gas whenever temperature of the gas to a point which guarantees
in water. It is possible to prevent ice formation by that the temperature at any point in the line, always
lowering the freezing point of water. This can be possible will reduce the likelihood of hydrate
formation. remained above the hydrate formation temperature.
done by adding a chemical to the water. (Think
about the effect of adding salt to icy roads). If heat is the answer, maximum use is made of heat
However, the pressure and temperature changes
involved in the dehydration process will, in conservation within the process by using heat
In a similar way, we can add a chemical to a gas exchangers. For example. the relatively high
stream to prevent hydrate formation. most cases, give rise to the condition for
hydrate formation before enough water can be temperature of gas at the wellhead may be used to
removed to inhibit such formation. warm up the cold, processed gas, as we shall see
This gives us a fourth method to add to our list: later.

4. Lower the hydrate formation temperature A real chicken and egg situation!

When a chemical is added to the gas to prevent We are, therefore, left with the last two - heating 4. Chemical inhibition.
hydrate formation, it is often known as chemical and chemical inhibition - as the most convenient
inhibition. methods of hydrate prevention. I want to concentrate on chemical injection as a
method used to prevent hydrate formation.
Let’s consider each of the four preventative The decision on the type of inhibition is invariably Ammonia, brines, glycol and methanol have all been
measures we have just listed. made on an economic basis. Usually, a used to lower the freezing point of water in gas.
combination of heating and chemical inhibition is the Methanol and glycol are the inhibitors most widely
result. used. These are fed into the gas by low volume
injection pumps. The injection point is usually just
upstream of the point where hydrate formation
conditions are expected.

21
Petroleum Open Learning

One such point is illustrated in Figure 8 at the Xmas


tree. The inhibitor can be used here to prevent
hydrates forming on the valves of the tree during a
shut-down. This injection point can also be used to
inject inhibitor into the flowing gas stream to prevent
hydrate formation in pipework and processes
immediately downstream.

Methanol is mainly used where only occasional


inhibition is needed, for example plant start-up or
when working on the well. The reason is that,
although methanol is fairly cheap, its recovery is
difficult and costly. It is, therefore, invariably lost. In
addition, methanol is hazardous to store and handle;
it has a fairly low flash point.

Where continuous inhibition is needed, ethylene


glycol (EG) is commonly used. Although it is more
expensive than methanol, its regeneration is a
reasonably straightforward process. There are
three main forms of glycol used in gas processing,
but it is ethylene glycol that is usually used for
hydrate inhibition. We will talk about the other forms
of glycol and their uses later.

The injected glycol mixes with any free water that is


present in the gas and lowers the hydrate formation
temperature, in much the same way as the addition
of anti-freeze to a car engine cooling system lowers
the freezing point of the cooling water. The resulting
glycol-water mixture can be processed to enable
recovery and re-use of the glycol. This regeneration
will be discussed later in this unit, as the process is
identical for regeneration of any of the glycols.

22
Petroleum Open Learning

Test Yourself 5

a. It was mentioned in the text that. in gas processing, the maximum use is made of heat
conservation within the process. Why do you think that this is done?


b. Indicate whether you would use methanol or glycol for the following inhibition requirements: Methanol Glycol

i) For initially starting a new oil well which has a high gas content?

ii) For continuous injection into an offshore pipeline feeding an onshore gas plant?

iii) For long term storage offshore for use in gas well servicing jobs?

23
Petroleum Open Learning

Summary of Section 1

In this first section the origins of water in natural gas have been explained and you know that it can
take the form of vapour or liquid.

You should now be able to find out how much water is required to saturate a given gas, if you
know its temperature and pressure (by reference to the graphs). It is important to remember that
when the temperature of natural gas is lowered, water vapour condenses to produce free water.

We have discussed the problems created by the presence of water in gas, which are :

liquid accumulations in the wellbore


corrosion
lower pipeline efficiency
poor gas quality
hydrate formation.

Finally, we looked at preventing the formation of hydrates in gas, in particular by chemical inhibition
with methanol or glycol. We will be applying our knowledge of hydrate prevention, both by heat
and chemical inhibition, later on in this unit.

You now have the necessary background knowledge for working through the following sections
and understanding the processes described.

24
Gas Dehydration Petroleum Open Learning

Section 2 - Auto Refrigeration


In this section we are going to look at one of the
characteristics of natural gas that assists in the
separation of water from the gas - the Joules/
Thompson Effect.

We will also look at a typical process plant which


uses this principle in practice.

The Joules/Thompson Effect


We have seen that, as the temperature is reduced,
water vapour condenses into free water, which is
fairly easy to separate from the gas. There are
problems involving hydrates, of course, but we can
deal with those by inhibition or process design, as
we will see later on.

The main hurdle we face is how to reduce the


temperature at an acceptable cost. Fortunately,
nature takes a hand. Gas has a property which can
assist us to reduce the temperature fairly easily. Let
me describe this property :

If a natural gas is rapidly expanded by reducing


the pressure, Its temperature will drop.

This temperature drop associated with gas


expansion is known as the JouleslThompson
Effect. The greater the pressure drop, the greater
the temperature reduction. You can see this effect
illustrated in Figure 9.

25
Petroleum Open Learning

This chart is only by way of an example. It will not • Follow this curve to the left until it intersects
be accurate for all gases. However it serves to the vertical line at 1 000 psi.
illustrate the use of such charts.

Test Yourself 6
• Move horizontally from this point to the left
In order to find a temperature drop associated with a hand vertical axis.
given pressure drop, just follow the steps I have
listed here for you. • Read off the temperature at this point. You
should find that it is 18 deg F. A gas is at 2400 psi and 80 deg F. From
• Find the point on the graph which Figure 9, work out what temperature rise
corresponds to the temperature and pressure The temperature drop therefore is would be needed in this gas so that, after
of the gas before expansion takes place. ( 90 - 18 ) = 72 deg F. expansion to 1 500 psi the final temperature
will be 75 deg F.
• Follow the curve to the left until it intersects Of course, the starting pressure and temperature
the vertical line which corresponds to the will not always coincide exactly with one of these
pressure of the gas after expansion. cooling curves. In such a case, a curve parallel to
the nearest printed curve needs to be drawn or
• Read off from the left hand vertical axis of the imagined.
graph, the temperature at this point.
For example, gas at 2 800 psi and 130 deg F is
Let’s do that with some actual figures. expanded to 1 400 psi. With a little imagination, you
will see that the new temperature will be
86 deg F - a drop of 44 deg F.
Take the following example:
Now have a go at Test Yourself 6. In this example
A natural gas at 3 000 psi and 90 deg F is expanded you will have to visualise your own curve from the
to 1 000 psi. What will be the temperature drop? figures given and work from that.

• Find the point on the graph which


corresponds to 3000 psi and 90 deg F. (You
will find that this point lies on the fourth curve
from the bottom of the graph).
Check your answer in Check Yourself 6 on
page 60

26
Petroleum Open Learning

The graph (Figure 9) also gives an indication of


whether we can expect hydrates to form. This is
shown by the broken line. Any point below this line
provides conditions suitable for hydrate formation in
the presence of free water. For instance, the
example we used of 3 000 psi gas at 90 deg F being
expanded to 1000 psi will place it firmly in the
hydrate formation range.

You may have wondered how the expansion of the


gas is brought about, in order to utilise the Joules /
Thompson effect. In fact, there are a number of
ways of doing this. The most common one is to
expand the gas across a choke valve.

A choke valve is a type of valve designed to control


gas or liquid flow. In its simplest form it consists of
a cone and seat arrangement, both of which are
hardened to resist the erosive effects of the flow.
The closer the cone is to the seat, the more the flow
is reduced, or choked. Such a device is illustrated
in Figure 10.

27
Petroleum Open Learning

Before we go on to look at processes using


the Joules / Thompson effect. practice using
Figure 9 further.

Test Yourself 7

Complete the following chart

Expanded to
Starting Starting Final Final Hydrates
pressure temperature pressure temperature expected
(psi) (deg F) (psi) (deg F) Yes/No

4000 102 2000

3800 154 1000

2800 90 1800

2000 86 1200

1800 123 600

Check your answer in Check Yourself 7 on page 61

28
Petroleum Open Learning

The Joules/Thompson Effect forms the basis The effectiveness of this process depends on the
of a common method used to extract water initial pressure being high enough to allow an
from gas. This method is called low adequate pressure drop. Often dehydration can be
temperature separation. achieved with a pressure drop as little as 1 000 psi.
The downstream pressure is usually determined by
Low temperature separation makes use of the the pressure of the pipeline being used to export or
natural characteristics of gas expansion. This, deliver the gas.
together with efficient heat exchange within
the plant design, leads to a very cost effective By dropping the temperature, we may move into the
process. hydrate formation region. This potential problem is
dealt with in one of two ways:
This is illustrated below.
1. Inhibition
A low temperature separation process may be
described as follows : or

The inlet gas is passed through a choke 2. Melting


valve and cooled by the resulting pressure
reduction and expansion. This causes Let’s look at two typical process plants which use
water and liquid hydrocarbons to either inhibition or melting to deal with hydrates.
condense. Dry gas, condensate and free
water can then be separated from each First, inhibition :
other.
Low Temperature Separation with Hydrate
We will be looking at this in more detail Inhibition (LTS)
shortly.
This process is illustrated as a simple flow diagram
in Figure 11.

29
Petroleum Open Learning

30
Petroleum Open Learning

Follow the path of the well fluids as they flow At this point we have all the conditions necessary for The hot glycol-can now be used again. It is first
through the plant. hydrate formation. It means that the glycol injection passed through heat exchanger HE-2, where it is
rate upstream of here needs to be carefully used to warm the incoming glycol/water mix from
The well stream first flows to a high pressure controlled. It must be sufficient to prevent hydrate the cold separator. The regenerated glycol itself is
knockout vessel that separates any free liquid. formation, in the heat exchanger, pipework or the cooled down here, prior to reinjection into the
The free liquids are removed from the vessel and separator. incoming well stream to act as a hydrate inhibitor
passed on to another part of the plant for disposal. again.
In the cold separator, glycol, water and condensate
Ethylene glycol (EG) is then injected into the are separated from the gas and the condensate is Why don’t you go through the process once again at
process gas stream immediately upstream of a heat recovered for further processing and sale. this point, and then have a go at Test Yourself 8.
exchanger HE-1. (You will remember from
Section 1 that EG is a chemical used to inhibit We want to be able to use the glycol again. In order
hydrate torrnation.) to do this, the water glycol mix is further processed
in a glycol regeneration system.
The gas, with glycol added, is then cooled in the We will be looking at this regeneration system in
heat exchanger. This exchanger is known as a gas/ much more detail when we come to Section 4 of this
gas exchanger. It means that cold sales gas is the unit.
cooling medium which cools the incoming gas
stream. (Sales gas is a term often used to describe Briefly, however:
gas treated to meet a laid down specification).
The glycol may have absorbed some hydrocarbons
A further temperature drop occurs as the gas as it mixes with the gas in the process. These must
expands during the pressure drop across the choke be removed. The glycol and water are routed from
valve. the separator, via another heat exchanger (HE-2)
where the mixture is heated up, to a flash tank. In
This cooling causes further condensate and water to this vessel, hydrocarbon vapours are removed from
condense from the gas stream as it enters the cold the warmed mixture.
separator.
In the final part of the process the water is removed
from the glycol. The glycol water mixture passes to
a regeneration package where the mix is heated and
the water is boiled off as steam.

31
Petroleum Open Learning

In this process, the main pieces of equipment, apart


from the regeneration package which we will look at
later, are the separator and the heat exchangers.
Let’s take a look at these pieces of plant in a little
Test Yourself 8 more detail.

The cold separator used in our example is a


The following statements refer to a series of steps in a low temperature separation process Horizontal Three Phase Separator. This type of
with hydrate inhibition (LTS). separator is illustrated in Figure 12.

The steps are out of order. Without looking at the flow diagram (Figure 11), rearrange the steps
in their correct sequence.

1. injection of glycol

2. separation of free liquids in the high pressure knockout vessel

3. condensation of water and condensate in the cold separator

4. cooling of the well stream in the gas/gas heat exchanger

5. expansion of gas across the choke valve

6. separation of water/glycol and condensate in the cold separator

Check your answer in Check Yourself 8 on page 62

32
Petroleum Open Learning

33
Petroleum Open Learning

The three phases are gas, condensate and water in


this case. The condensate, being lighter, sits on top
of the water and flows over a Weir into a separate
compartment. From here the outlet valve is
actuated by a level controller, to maintain a steady
condensate level in this compartment. A second
level controller maintains a constant water
condensate interface.

The gas section is fitted with a series of baffles that


encourage the separation of condensed liquid
droplets from the gas. This ensures that the gas
leaving the separator is liquid-free.

Let us now look at heat exchangers. We repeatedly


refer to heat exchangers in this Unit, and you will
meet up with them in most hydrocarbon processes.

Figure 13 shows a typical heat exchanger.

Figure 13 : Heat Exchanger

One medium flows via the coils and the other in the outer body. Either medium can
be liquid or gas. Due to the temperature difference, the colder medium heats up, and
the hotter one cools down.

34
Petroleum Open Learning

Now we will consider the other process which uses Low Temperature Separation.

This Figure shows a simple flow diagram of a typical process of this type. It is called a Low Temperature Extraction Process (LTX)

35
Petroleum Open Learning

The basic differences between this process and the Cold gas, which has now had its water and liquid
previous one are the routing of the warm well hydrocarbons removed, is taken from the separator
stream, the separator design and the absence of and used as the cooling medium in the heat
glycol injection or regeneration. exchanger. (A by-pass line round the exchanger
incorporating a 3-way valve maintains the correct
Once again, follow the flow paths through the temperature in the process). The gas, now at the
process, using Figure 14. correct specification, can be sold.

This time, the warm incoming stream is first routed The low temperature separator is again a 3 phase
through coils in the bottom of the low temperature vessel, but of a different design. I have included a
separator. This starts to cool the gas, which is then simple drawing of one (Figure 15) so that you can
further cooled as it passes through a heat compare the two.
exchanger.

The effect of cooling is to condense some


hydrocarbon liquids and water from the gas stream.
These free liquids are separated from the gas in
the liquid knockout drum and fed into the liquid
section of the low temperature separator.

After leaving the liquid knockout drum, the gas


passes to the choke at the separator inlet. Here the
gas is expanded to a lower pressure. Again, rapid
expansion of the gas causes a drop in temperature.

At this point, hydrates tend to form because


conditions have now changed to values which
encourage hydrate formation. The hydrates fall into
the liquid section of the separator where they are
melted by the warm coils.

36
Petroleum Open Learning

From here, the condensate and water are removed


under controlled conditions. The condensate is sold
and the water led off to disposal.

The LTX process is obviously more cost effective


than LTS, as glycol inhibition and regeneration are
eliminated. Glycol or methanol injection may be
necessary, however, for start up purposes, when the
well stream will be cold.

Now that you have worked through this section,


have a go at the following Test Yourself question.

37
Petroleum Open Learning

Test Yourself 9
The following statements apply to the LTS process or the LTX process or both. Indicate, by
ticking the box, which one is applicable.

LTS LTX BOTH

1 The incoming well stream passes through coils in the separator.

2 The choke is at the separator inlet.

3 Inhibitor is normally injected into the well stream.

4 The glycol is regenerated.

5 A heat exchanger is used to cool the gas.

6 Hydrates form in the separator.

7 The process makes use of the Joules/Thompson effect.

8 The inhibitor injection point is before the heat exchanger.

Check your answer in Check Yourself 9 on page 62

38
Petroleum Open Learning

Summary of Section 2
This section opened with a description of the Joules/Thompson Effect, where gas expansion
results In a temperature drop. You have seen how this characteristic of gas is used to good effect
in low temperature separation, by condensing liquids from the gas phase.

Two main processes were described:

a. LTS, where hydrate formation is inhibited by the injection of glycol.

b. LTX, where hydrates are melted by passing the incoming, warm, gas stream through coils in
the bottom of the separator.

The LTX was seen to be more efficient as it utilised heat exchange within the process, thus
eliminating the need for expensive glycol inhibition or regeneration.

You now understand the principles of the first method of gas dehydration, using low temperature
separation which can be described as an Auto Refrigeration process.

We will now move on to look at a system which uses the principles of adsorption for gas
dehydration - the Solid Desiccant Dehydration process.

39
Gas Dehydration Petroleum Open Learning

Section 3 - Solid Desiccant Dehydration


In the last section, we examined how the principle of A gas dehydration process which uses this Let’s now look at a Solid Desiccant Dehydration
low temperature separation could be used to phenomenon is the Solid Desiccant Dehydration Plant.
reduce the water content of a gas. Plant.

In this section we will look at the first of two types of In such a plant the gas is dehydrated by passing it
process which utilise a substance called a through a bed of solid desiccant which removes the
desiccant to remove water from gas. The water vapour.
desiccant can either be a solid or a liquid. We will
start with a solid desiccant, which uses the principle This desiccant consists of solid granular materials
of adsorption to achieve this. having an extremely large surface area per unit
weight. This is because the granules have a
Adsorption multitude of microscopic pores and capillary
openings. Common desiccants used in the plant
Adsorption is a process in which a solid selectively are:
removes a particular component from a fluid (liquid
or gas) mixture and holds this component on its
surface. silica gel
sorbead
This solid is known as an adsorbent material. activated alumina
molecular sieves
In our case” the mixture consists of gas and water
vapour, and it is the water which is removed. After a time the desiccant will itself become
The adsorbent material, which removes the water, is saturated with water. This reduces its capacity for
called a solid desiccant. further adsorption and, in order to use it again, it
must be regenerated. In other words, we must get
An everyday example of the adsorption process is rid of the adsorbed water.
the use of sachets of silica gel, packed along with
sensitive photographic or electrical equipment. The This is usually achieved by heating with hot gas,
silica gel is a solid desiccant which prevents which vaporises the water from the desiccant. For
moisture from damaging this equipment. this reason, a dry bed dehydrator usually has at
least two beds of desiccant - one being used to dry
the gas, while the other is being regenerated.

40
Petroleum Open Learning

Figure 16 shows a simple flow diagram of such a plant.

41
Petroleum Open Learning

The diagram shows a two tower dehydration layout. Effective regeneration is the secret of this process.
Take a look at the diagram and follow the process The bed must be thoroughly regenerated or its
through. The solid lines represent the gas being capacity will be reduced. Effective regeneration
processed, and the broken lines the regeneration relies on quantity and temperature of the
cycle gas flow. One tower is on the adsorption cycle regeneration gas. The higher the temperature, the
while the other is being regenerated. less gas is required, but too high a temperature can
ruin the desiccant, and drastically reduce its
Gas being treated flows in at the top of the tower adsorptive properties.
and adsorption takes place from top to bottom.
Water saturation of the desiccant also, therefore, A typical cycle time is 8 hours of adsorption and
starts at the top of the tower. Lighter hydrocarbon 8 hours regeneration. Figure 17 gives a graphical
components are also adsorbed in the lower layers of representation of the regeneration cycle time and
the desiccant bed. As the lower layers temperature.
progressively become water saturated, the
hydrocarbon components are displaced. The
adsorption cycle must stop before the desiccant
bed is totally saturated.

Regeneration takes place in the reverse direction,


that is, bottom to top. The regeneration gas is
heated and fed in to the bottom of the tower. It
passes through the desiccant and out through the
top of the tower. The hot regeneration gas drives the
water from the desiccant as steam. This wet, hot
gas is then cooled and passed through a separator
to remove the liquids. After the bed has been
heated and the water driven off, it has to be cooled
before it is switched to adsorption again.

42
Petroleum Open Learning

Use the figure to have a go at the following Test


Yourself question.

43
Petroleum Open Learning

Solid desiccant dehydration can produce Virtually


dry gas for processes sensitive to feed gas quality,
such as cryogenic-type gas processing plants.
(Cryogenic processing involves extremely low
Test Yourself 10 temperatures, much lower than the LTX and LTS
plants we have been looking at). The effectiveness
of the unit, however, depends upon the incoming
gas being free of liquids, entrained mist and solids.
a. Referring to curve 2 in Figure 17, describe what is happening from the start to the end of Liquids may destroy or damage the desiccant bed,
the regeneration cycle. and solids could plug it.

b. Why do you think that the regeneration gas is fed in at the bottom of the tower?

c. Why is the regeneration gas cooled after leaving the tower, before entering the separator?

Check your answer in Check Yourself 10 on page 63

44
Petroleum Open Learning

Summary of Section 3
You have been introduced to the removal of water vapour from gas using the mechanism of
adsorption.

You saw that adsorption is a process in Which a solid desiccant selectively removes one
component from a fluid mixture. The solid desiccant has the capacity to attract and hold the
component on its surface.

We examined the workings of a dry bed dehydration unit, and concluded that such a unit relies
mainly on effective regeneration of the desiccant. Such plants can dry gas very thoroughly.

We are now going to look at a second process which utilises a desiccant, one in which a liquid,
rather than a solid, is used to dehydrate the gas - the Liquid Desiccant Dehydration process.

45
Gas Dehydration Petroleum Open Learning

Section 4- Liquid Desiccant Dehydration


In the last section we looked at a process Which • Diethylene glycol (DEG); a cost effective Glycol Dehydration Plant
removes water vapour from gas using a solid liquid desiccant where moderate dehydration
desiccant. We now turn to the use of a liquid, rather is required and where DEG has also been In a typical glycol dehydration plant the wet gas is
than a solid desiccant. This is known as an used earlier in the process as a hydrate brought into intimate contact with TEG. The glycol
Absorption process. inhibitor. absorbs water from the gas, which then leaves the
plant as dry gas. The glycol now contains water
Absorption • Trlethylene glycol (TEG); the most which reduces its absorbing properties. In order
expensive but the most effective liquid that the glycol can be used again, it must be
Absorption of water by liquid desiccant is a process desiccant in the glycol family. It possesses regenerated. This is done by heating it and driving
in which the water is taken into the body of the superior dewpoint depression qualities. the water off as vapour (steam).
desiccant. This is in contrast to adsorption, where
the water is held on the surface of the (solid) A balance must be struck between the degree of Let’s follow the flow of gas and glycol through a
desiccant. effectiveness and the cost. Nowadays, therefore, typical dehydration plant.
TEG is the preferred liquid desiccant. TEG has been
Water vapour is removed by bubbling the gas used successfully to dehydrate gases over the
through a hygroscopic liquid, that is, a liquid with following operating ranges and conditions:
an affinity for water.

This hygroscopic liquid is our liquid desiccant. • Dewpoint depression 40 - 140 deg F

• Gas pressure 25 - 2 500 psi


Liquid Desiccants
• Gas temperature 40 - 160 deg F
The liquid desiccant used is almost always one of
the glycols. Glycols which have the necessary
attraction for water are listed below :
We have our liquid desiccant. All we need now is a
• Monoethylene glycol (MEG or EG); rarely plant in which to use it.
used nowadays as a desiccant due to high
evaporation and chemical degradation losses. For the rest of this unit we will be looking at the
If you think back to Section 1 you will equipment used in, and the layout and operation of,
remember that MEG is used as a hydrate a Glycol Dehydration Plant.
Inhibitor.

46
Petroleum Open Learning

Figure 18 shows a simple flow diagram for a typical unit. Look at the figure and trace the flow paths with
the aid of the following brief description.

The gas containing water vapour ( wet gas) enters


the contactor, or absorber tower at the bottom. It
passes up the contactor, through a series of trays
down through which the TEG is flowing. The trays
are so designed that the gas is forced to mix
intimately with the glycol.

The water from the gas is absorbed by the glycol


and dry gas leaves the tower at the top.

47
Petroleum Open Learning

The glycol, now containing water and called rich This simplified description will help you gain a broad
glycol, leaves the contactor tower at the bottom and understanding of the process. Let us now study it in
flows to the regeneration section of the plant. more detail and see how the equipment works.

The TEG also absorbs some light hydrocarbon Gas Flow


vapour, and so the stream is directed to a flash
tank. This reduces the pressure and therefore Let’s start with the flow of gas through the system.
allows most of the hydrocarbon vapours to escape.
The heart of the process is the contactor tower. It
Next, the lEG flows through a filter which removes is in this unit that the dehydration takes place. The
any tarry solids which may have formed in the tower is called either the contactor or the absorber
process. tower. For the rest of this section, however, I will
use the term contactor.
The glycol is then pre-heated in a heat exchanger
and passed to the reboiler where the water is boiled Look at Figure 19(a) which shows a contactor tower
off and the TEG reconcentrated. with a scrubber section and an absorber section.
The glycol is now capable of being used again. It is
known as lean glycol at this point.

From the reboiler, the lean glycol flows to a surge


tank. It is pumped from here to the heat exchanger,
where it is cooled. From there it flows back to the
top of the contactor tower. (Note how the hot glycol
from the reboiler is used as the heating medium in
the heat exchanger).

The process you have just followed is a continuous


process and could be called a Regenerative Glycol
Dehydration Process.

48
Petroleum Open Learning

49
Petroleum Open Learning

Before the gas enters the absorber section of the Finally, before leaving the contactor the gas passes The flash tank is simply a three phase separator
tower, it passes through a scrubber to separate any through a mist extractor. This device extracts any which is capable of separating glycol, gas and
free liquids. The scrubber may be a separate vessel droplets of liquid glycol which may have been picked hydrocarbon liquids from each other.
or, as is the case here, may form part of the tower up by the gas. It helps to reduce glycol losses.
itself. You can probably imagine that, as the glycol flows
The gas leaving the contactor should now be free through the contactor, it can pick up small amounts
It is essential that the contact between the gas and from water vapour and meet the required dewpoint of gas, and liquids which have condensed from the
the TEG is intimate. This is achieved by bubbling specification. gas. These are removed in the flash tank. Any gas
the gas through the TEG via bubble caps which is liberated is led away to be used as fuel or is
positioned on a series of trays within the absorber Glycol Flow disposed of by flaring. Liquid hydrocarbons are
section of the contactor. removed from the tank and are collected for sale or
Let’s now look at what happens to the glycol. otherwise disposed of. What remains is rich glycol
The contactor trays with the bubble caps, are shown which passes on to the next stage in the process.
in Figure 19(b). The TEG, cooled in the coils (at the top of the
tower), passes down through the tower from tray to Within the glycol a certain amount of solid material
The trays incorporate a weir which maintains a fixed tray, dehydrating the gas. This diluted (or rich) may accumulate. This can take the form of dirt,
level of glycol on each tray. As the gas flows up glycol solution collects at the bottom of the absorber scale, rust or tarry reaction substances. A filter is
through the centre of a bubble cap it is forced back section of the contactor tower. used to remove these. It is usually of the type which
down through the glycol and round the outside of the contains a cartridge which can be removed and
cap. This ensures that all the gas must flow through Before the glycol can be used to dehydrate more replaced while the plant is in operation.
liquid glycol as it passes each tray. gas, it must be regenerated.
I said earlier that the glycol must be heated in the
The flow path of the gas through a bubble cap is Before we go on to look at the regeneration process regeneration process. In order to save energy, the
shown in Figure 19(c). itself, we should consider three other pieces of glycol is pre-heated in a heat exchanger before it
equipment. These are: goes to the regenerator. The exchanger uses hot
After the top tray, the gas passes around glycol glycol from the regenerator itself as the heating
cooling coils. These act as heat exchangers to • flash tank medium.
reduce the temperature of the incoming lean glycol.
(The temperature of the lean TEG entering the top • filter Let’s move now to the actual process of glycol
of the tower should be as close as possible to the regeneration.
gas exit temperature. This helps to prevent the • heat exchanger
glycol from foaming, which might occur if there was
too great a temperature difference between lean Look again at Figure 18 and check the location of
glycol and gas). this equipment.

50
Petroleum Open Learning

Simply boiling the dilute glycol would remove the


water as steam. but would also result in
considerable loss of glycol vapour. As the boiling
points of water and TEG are so far apart (212 and
549 deg F respectively). A process called fractional
distillation is used to regenerate the glycol.

The regenerator consists of two parts, the Reboiler


and the Stripper column. Take a look at Figure 20
which shows such a unit.

51
Petroleum Open Learning

You will see that the cool, rich glycol passes first of As long as the temperature at the top of the column
all through a coil in the top of the stripper column. ranges between 210 and 212 deg F, glycol losses
This is called a reflux coil. I will explain its function are minimised. This process is aided by the reflux
shortly. coil which I mentioned earlier. The cool glycol
passing through the coils assists any glycol vapour Test Yourself 11
After pre-heating in the heat exchanger, the glycol is to condense and fall back into the reboiler.
then fed to the stripper column near the top. It
descends through packing in the stripper and mixes Having passed down the stripper column, the glycol Following the flow path of glycol, place the
with a rising stream of water rich, hot vapours. enters the reboiler and is heated further. This jumbled list of equipment given below into a
These vapours are created by heating the glycol in creates the hot, water-rich vapours required for logical process order, starting with the
the reboiler to a high temperature. This temperature stripping. Reboiler;
should be above that of boiling water but below the
boiling point of the glycol itself. From the reboiler the glycol passes to a surge tank
which acts as a storage vessel. From there it is Reboiler
Within the stripper column, two things are pumped via the heat exchanger back to the
happening: contactor to continue the dehydration process. Contactor tower

• the cooler, rich glycol liquid causes the hot So, you should by now have a good idea of how a Heat exchanger
glycol vapour in the rising vapour stream to liquid desiccant dehydration process works. Trace
condense out as a liquid and fall back. the flow path of gas and glycol again and then Filter
attempt the following Test Yourself.
• the hot, water rich, vapour stream strips out Surge tank
the liquid water from the glycol stream as
vapour and carries it to the top of the column, Flash tank
from where the water vapour is vented to
atmosphere. Pump

All being well, at the top of the column the vapour


will be virtually pure water.

Check your answer in Check Yourself 11 on


page 64

52
Petroleum Open Learning

Plant operations The concentration of the glycol fed to the contactor


tower determines the dewpoint depression
A glycol dehydration plant will only work efficiently if achievable for a given contact temperature. The
contact temperature is the temperature at which the
Test Yourself 12
certain process variables are maintained at constant
values. gas leaves the top of the contactor tower. The
temperature of the lean TEG entering the top of the
The more important process variables are : tower should be as close as possible to this contact
temperature. Think about the process plant. What piece of
• glycol concentration equipment is used to try to keep the
temperature of the TEG entering the
• glycol circulation rate contactor as close as possible to that of the
gas leaving it?
• glycol solution condition

Let’s look at each of these in turn.

Glycol Concentration
Glycol concentration refers to the amount of pure
glycol in solution and is measured as a percentage
by weight:

For instance, the rich glycol leaving the contactor


tower is a water rich solution whose glycol content is
less than 95% by weight.

However, after the glycol has been through the


regeneration section of the plant it is
reconcentrated. Now its concentration can vary
from 95% to 99% by weight, although we try to
achieve as near to 100% as possible.

Check your answer in Check Yourself 12 on


page 64

53
Petroleum Open Learning

We can use a graph again to see how glycol


concentration and contact temperature can affect
the dewpoint of the gas.

Figure 21 shows the relationship between


dewpoints, contact temperature and TEG
concentration.

If, for example, we need to achieve a reduction


in water dewpoint of the gas from 55 deg F to
35 deg F at a contact temperature of 120 deg F,
you can see that the TEG concentration must be
increased from 96.0% to 98.0% by weight.

This increased TEG concentration can be achieved


by increasing the reboiler temperature. This can be
shown in another graph, Figure 22, which appears
on the next page.

54
Petroleum Open Learning

Take a look at Figure 22. In order to increase the Let’s work through an example to determine TEG
TEG concentration from 96% to 98%, the reboiler concentration and reboiler temperature for a
temperature would have to be increased from particular dehydration problem. For this example,
322 deg F to 380 deg F. let us assume that dehydrated gas must contain not
more than 5lbs of water per mmscf when delivered
There is a limit, however, to the temperature at at 1 000 psi and 100 deg F.
which we can operate the reboiler. TEG begins to
degrade at temperatures around 450 deg F, which Looking way back at Figure 2, which gives us the
clearly represents the upper limit. water content of natural gas at saturation, you can
see that the required dewpoint is 24 deg F.

If the gas leaves the contactor at 100 deg F it can


be seen from Figure 21 that a 24 deg F dewpoint
requires a TEG concentration of around 97.6%.
(You will have to estimate between the curves to
determine this).

Figure 22 shows that 97.6% TEG concentration


requires a reboiling temperature of 364 deg F (at
sea level).

Again some estimation or interpolation is required.

If you are happy with the above example, try the


following Test Yourself. You will see that you have
to think back to some of the previous sections to
complete it.

Figure 22 : Graph of Triethylene Glycol Reboiler Temperature versus TEG Concentration

55
Petroleum Open Learning

Another method of achieving higher TEG


concentrations is to use stripping gas for final
water removal. Stripping gas is dry, low pressure
gas introduced to remove additional water from the
Test Yourself 13 glycol in the reboiler.

Look again at Figure 20 which shows the glycol


Gas at the wellhead is at 3000 psi and 120 deg F. After separation, its pressure is reduced regenerator. You will see an inlet pipe at the bottom
across a choke to 1 000 psi, after which it is routed to enter the cold separator. The gas enters of the reboiler through which gas is entering the
an absorption tower for dehydration. The required water content should not be more than TEG. This is the stripping gas being used during
4lbs per mmscf. The gas leaves the contactor at 90 deg F. reboiling.

The effect of stripping gas being in contact with lean


i) Is hydrate inhibition (or melting) required? TEG is shown in Figure 23.

ii) What volume of free water is knocked out by the expansion (per mmscf)?

iii) What concentration of TEG is required to achieve the gas delivery specification?

iv) What reboiler temperature is necessary (at sea level) to achieve the TEG concentration?

Check your answer in Check Yourself 13 on page 65

56
Petroleum Open Learning

Now to the next process variable :

Glycol Circulation Rate


It is important to maintain an optimum TEG
circulation rate for effective dehydration.

Each plant will be designed for specific conditions.


As a rule of thumb, however, about 2 gallons of glycol
must be circulated for every 1 lb of water removed at
the 55 deg F dewpoint depression.
These figures are based on operations over a normal
pressure range and a glycol solution of 95%.

Sometimes, greater dewpoint depressions can be


obtained by increasing the circulation rate. However,
an upper limit can be reached where increasing the
circulation rate actually reduces the dewpoint
depression.

And finally:

Glycol Solution Condition


For effective dehydration, the glycol solution must
be kept in good condition. In other words, it must be
free from impurities. Solids must be filtered from the
circulating stream and filters must be kept clean.

57
Petroleum Open Learning

Hydrocarbons in the TEG solution which have not


been removed in the flash tank can cause additional
foaming problems in the contactor. It is therefore
essential that the efficiency of the flash tank is
maintained. Summary of Section 4
In Section 4, I explained the difference between Adsorption and Absorption. However, we
concentrated on the Absorption process.

You saw that Triethylene Glycol (TEG) is most commonly used as a liquid desiccant to absorb
water from gas.

You followed the operation of a glycol dehydration plant, and you will have noted that this consists
of two main units:
Now that you have completed Section 4, have
another look at the process flow diagram and satisfy • the contactor, in which gas dehydration is accomplished
yourself that you understand how the plant
operates. • the glycol regeneration system, in which water is removed from the wet glycol so that it
can be used again

Finally in this section, you looked at some of the process variables which must be maintained at a
constant value for efficient dehydration.

Now that you have completed the whole unit, you should have a basic understanding of the theory
and practice of gas dehydration.

Go back to the training targets on Page 2 of the unit, and check that you are able to meet those
targets.

58
Petroleum Open Learning

Check Yourself - Answers

Check Yourself 1 Check Yourself 2 Check Yourself 3


A water content of 45 Ibs per mmscf will be required to Because above the dewpoint temperature, one of 70 deg F.
achieve saturation of the gas at surface conditions. the conditions for hydrate formation would not exist,
ie free water.
As 48 Ibs of water per mmscf were the
saturation conditions at the reservoir, ie, more
water vapour than required for saturation at the
surface, the gas at the wellhead will be
saturated. As there is a decrease in the
amount of water vapour required for saturation
between reservoir and surface, free water will
exist in liquid form.

The amount will be 48 - 45 = 3lbs per mmscf.

59
Petroleum Open Learning

Check Yourself 4 Check Yourself 5 Check Yourself 6


a. The problems associated with water in a. It saves money to use any naturally available 25 deg F.
gas that could affect gas processing, in sources of heat before considering paying for
the short term, are: additional, external energy. The starting point has to be the final conditions of
1 500 psi and 75 deg F. Follow an imaginary
gas quality b. i) Methanol, as a glycol regeneration cooling curve parallel to the nearest printed curve
pipeline efficiency unit would not be available. until the 2 400 psi line is intersected. This gives
hydrate formation us the temperature to which the gas requires
ii) Glycol, as long as the onshore gas heating. Temperature rise = 105 - 80 = 25 deg F.
Corrosion is not included as it is unlikely to plant has a glycol regeneration facility.
affect actual processing in the short term.
iii) Glycol, because it is safer to store for
b. i) Yes long periods.
ii) No
iii) Yes*
iv) No
v) Yes

* Remember that hydrates can form at or


below the hydrate temperature.

60
Petroleum Open Learning

Check Yourself 7
Expanded to
Starting Starting Final Final Hydrates
pressure temperature pressure temperature expected
(psi) (deg F) (psi) (deg F) Yes/No

4000 102 2000 (63) (Yes)

3800 154 1000 (74) (No)

2800 90 1800 (60) (Yes)

2000 86 1200 (52) (Yes)

1800 123 600 (65) (No)

61
Petroleum Open Learning

Check Yourself 8 Check Yourself 9


The correct sequence of steps should be :
LTS LTX BOTH
2-1-4-5-3-6
1 The incoming well stream passes through coils in the separator. a
a

2 The choke is at the separator inlet. a a

3 Inhibitor is normally injected into the well stream. a

4 The glycol is regenerated. a


a

5 A heat exchanger is used to cool the gas. a


a

6 Hydrates form in the separator. a


a

7 The process makes use of the Joules/Thompson effect. a


a

8 The inhibitor injection point is before the heat exchanger.


a
a

62
Petroleum Open Learning

Check Yourself 10
a. At first, hot gas warms up the tower and the contents. At 240 deg F water will begin to boil and
vaporise. The bed continues to heat up, but more slowly, as water is being driven out of the bed.
After the water, any heavier hydrocarbons will be driven off at a high temperature, and the bed
will become fully regenerated. The bed is cooled for a couple of hours by unheated gas flowing
through it.

b. The regeneration gas is flowed bottom to top due to the lower layers of the desiccant being less
wet. Flowing from top to bottom would result in time being wasted in saturating the drier lower
layers with the wet regeneration gas flow.

c. The regeneration gas is cooled before the separator in order to condense the water removed
from the regenerated tower, which can then be taken out at the scrubber.

63
Petroleum Open Learning

Check Yourself 11 Check Yourself 12


Reboiler The cooling coil at the top of the contactor tower (reflux coil)
Surge tank reduces the temperature of the glycol to near that of the gas.
Pump
Heat exchanger
Contractor tower
Flash tank
Filter

64
Petroleum Open Learning

Check Yourself 13
i) Yes, as the expansion process takes the gas into the hydrate formation range (Figure 9).

ii) From Figure 2, water content at 3 000 psi and 120 deg F = 48 lbs per mmscf.

From Figure 9, expansion to 1 000 psi drops the temperature to 53 deg F.

From Figure 2, water content at 1 000 psi and 53 deg F = 14 Ibs per mmscf.

Water knocked out by expansion = 48 - 14 = 34 Ibs per rnmscf.

iii) Figure 2, shows that for 4lbs water content a dewpoint of 19 deg F is required at 1 000 psi.
From Figure 21, a TEG concentration of 97.5% is necessary to achieve sales specification.

iv) From Figure 22, the reboiler temperature will need to be set at about 360 deg F.

65
POL
Petroleum Open Learning

Gas Flow Measurement


Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY
POL
Petroleum Open Learning

Gas Flow Measurement


Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY
Petroleum Open Learning

Designed, Produced and Published by OPITO Ltd., Petroleum Open Learning, Minerva House, Bruntland Road, Portlethen, Aberdeen AB12 4QL

Printed by Astute Print & Design, 44-46 Brechin Road, Forfar, Angus DD8 3JX www.astute.uk.com

© OPITO 1993 (rev.2002) ISBN 1 872041 85 X

All rights reserved. No part of this publication may be reproduced, stored in a retrieval or information storage system, transmitted in any form or by any
means, mechanical, photocopying, recording or otherwise without the prior permission in writing of the publishers.
Gas Flow Measurements Petroleum Open Learning

(Part of the Petroleum Processing Technology Series)

As a large part of this subject deals with calculations, you will require to be comfortable with
maths up to about standard grade level. Visual Cues
Although some of the equations are fairly complex, all relevant data and information is
provided to assist you to solve the problems. training targets for you to
All formulae required for calculations in your examination will be provided for achieve by the end of the unit
you. However, it is necessary that you are able to recognise the symbols
in formulae and allocate the correct units of measurement to each
symbol in your calculations
You will also find that a scientific calculator will be useful for this programme. test yourself questions to see
how much you understand

Contents Page
check yourself answers to
let you see if you have been
thinking along the right lines
* Training Targets 4

* Introduction 5
activities for you to apply your
* Section 1 - Gas Flow Measurement Applications and Gas Physics 6 new knowledge

The Need for Accurate Measurement


Units of Measurements
summaries for you to recap on
the major steps in your progress

1
Petroleum Open Learning
Petroleum Open Learning

Contents (cont’d) Page Visual Cues


training targets for you to
* Section 2 - Fluid Flow Principles 26 achieve by the end of the
unit
Fluid Flow Principles
Fluid Flow Properties
test yourself questions
to see how much you
* Section 3 - Measurement Devices and Methods 34 understand

Types and Applications


Orifice Plate Principles
check yourself answers to
Orifice Plate Flow Calculations
let you see if you have been
thinking along the right lines

* Section 4 - Orifice Plate Metering Equipment 46


activities for you to apply
Types of Plate
your new knowledge
Sensing Devices
Metering Stations
Safety Implications
summaries for you to recap
* Check Yourself - Answers 54 on the major steps in your
progress

3
Petroleum Open Learning

Training Targets

When you have completed this unit on Gas Flow Measurement, you will be able to:

• State the gas laws and perform relevant calculations using the appropriate units of measurements.
• Define molecular mass, gas density and specific gravity, and perform calculations when given appropriate formulae.
• Define Reynold’s number and, given Reynold’s equation, define the terms in it and use it in a calculation.
• Describe Bernoulli’s principle and state the types of pressure in a flowing fluid.
• Describe the main types of gas flow measuring devices.
• State the relationship between differential pressure and flowrate.
• Describe the main features of orifice plate meters.
• Perform a flowrate calculation in which all necessary formulae and data are given.
• Draw a schematic diagram of a typical multi-stream system which complies with fiscal standards.
• Describe the main safety implications associated with metering systems.

Tick the box when you have met each target

4
Oil and
Gas Gas Separation
Flow Measurement Systems Petroleum Open Learning

Introduction
Introduction
For most of this century there has been a necessity to measure gas flow accurately, both in commercial applications and scientific investigations.

The measurement of gas flow is more complex than that of liquid since gas is more sensitive to physical factors, such as pressure, temperature, composition,
etc. It has therefore been subjected to considerable research effort, which has led to a high degree of accuracy now being possible.

In this book we will be mainly concerned with natural gas flow measurement, but the principles are generally applicable to all gas phase matter.

This book comprises four sections :

Section 1, Gas Flow Measurement Applications and Gas Physics, outlines areas in which the ability to accurately measure gas flow is essential. It then
covers the basic physics of gas behaviour, which are essential to an understanding of the measurement and flow calculation methods.

Section 2, Fluid Flow Principles, presents the concepts of laminar and turbulent fluid flow, develops the Bernoulli and continuity principles to produce a
simple flowrate / differential pressure relationship.

Section 3, Measurement Devices and Methods, describes various fluid flow measurement devices which are applicable to gases. It then gives a more
detailed treatment of the orifice plate method, since this is the one most widely used. It ends with the ISO 5167 formula, and an explanation of its terms.

Section 4, Orifice Plate Metering Equipment, describes : various orifice plate designs, differential pressure and gas density sensing and measuring
equipment and a typical gas metering station. It ends by drawing attention to the safety aspects of gas metering systems.

5
Gas Flow Measurement Petroleum Open Learning

Section 1 - Applications and Gas Physics


The Need for Accurate Measurement Gas Sales Contracts
We will start by looking at some examples of All natural gas sales contracts are based on
activities that require accurate measurement of accurately measured volumetric flow rates with
natural gas flow rates. specific reference pressure and temperature
conditions.
Oil Field Evaluation
Hydrocarbon Taxation
A typical field evaluation exercise involves flowing
reservoir fluid from an exploration well to a test In most oil and gas producing countries, governments
separator, where the liquid and gas phases are impose various types of revenue on the production
separated. Accurate measurement of the gas and companies. Most of these revenues are applied on
liquid flow rates from the separator is essential a volumetric basis, so flow rates must be measured
to the achievement of a reliable appraisal of the to a high degree of accuracy.
reservoir performance.
Fiscal Standards
Process Control & Optimisation
In the last two examples, in addition to the need
The control and optimisation of gas processes, for accuracy, there is also the implication of
in both offshore and onshore operations, often complex legal considerations. This has led to
requires gas flow rate monitoring as the process the establishment of a set of fiscal standards,
variable in automatic control systems. the purpose of which is to achieve consistent
levels of high accuracy and reliability in fluid flow
An example of this is using the gas flow rate through a measurement.
centrifugal gas compressor as a measured variable
to provide automatic flow control / recycling for anti-
surge protection.

A comprehensive explanation of surge protection


is provided in the Petroleum Gas Compression
programme which forms a part of this Petroleum
Processing Technology Series.

6
Petroleum Open Learning

Units of Measurement
It is not possible to fully appreciate the methods
and procedures of gas flow metering without a
basic understanding of the physical behaviour
of gases, in particular the relationship between
pressure, volume and temperature.

Boyle’s Law
Figure 1 illustrates the principle behind Boyle’s
Law, which describes the relationship between
the volume occupied by a given mass, or number
of molecules, of gas and its pressure, while the
temperature remains constant.
The weight (W) balances the force due to the If we double this force to give us four times
Figure 1 depicts a piston in a cylinder which pressure (P), and the gas occupies a volume (V), the original force, Figure 1 (c) shows that
contains a fixed mass of gas. The highly energetic as shown in Figure 1 (a). a gas pressure of 4P is produced and the
gas molecules collide with each other and with the
volume is reduced to 1/4 of V1.
cylinder walls and the piston face, resulting in a Now consider what happens if we double the force
force being exerted. The property we describe as on the piston, a condition we achieve by applying The relationship between the pressure and volume
pressure is defined as the magnitude of that force a second weight of the same mass as the first one is now clear. Doubling the pressure halves the
divided by the area over which it acts. (we assume that the piston itself is weightless), to volume; quadrupling the pressure reduces the
exert 2W. volume to a quarter of its original value. Boyle’s
The force acting on the bottom face of the piston
Law expresses this formally with the statement that:
is therefore the pressure multiplied by the cross To balance this force the gas must now exert At constant temperature, the absolute pressure
sectional area of the piston. To prevent the piston twice the pressure, i.e. 2P. If we measure the new of a given mass of an ideal gas is inversely
being driven out of the cylinder, a force of the volume we find it to be 1/2 of V1, as shown in proportional to the volume.
same magnitude must be applied downwards, in Figure 1 (b).
this case by a weight (W).

7
Petroleum Open Learning

We need to define some of the terms in this statement. Charles’ Law


Thus : P = kv
• Absolute pressure units must be used so k = PV which is the mathematical way of stating that the Charles’ Law describes the relationship between the
instead of gauge units. Most field pressure pressure multiplied by the volume gives a constant value.
volume and temperature of an ideal gas, while the
measurement devices read in gauge units,
which means that atmospheric pressure Referring to Figure 1 (a) and (b), we can write : pressure is kept constant.
must be added to the indicated value to
obtain the absolute pressure. P1 V1 = k As in the case of Boyle’s Law, we can use a cylinder
and P2 V2 = k /piston arrangement to demonstrate the principle behind
NOTE : HENCE P1 V1 = P2 V2 Charles’ Law, as shown in Figure 2.
Unless otherwise stated, all pressure references in this
programme will be given as bar, this will infer that the which is the mathematical expression of Boyle’s Law. This time, however, we keep the pressure constant by
pressure is bar absolute (bara). leaving the force on the piston unchanged, and heat the
This equation can be used to calculate a new pressure or gas in the cylinder. Not surprisingly, we find that the gas
• The behaviour of the gas must be “Ideal”. volume, where the original pressure and volume and one of volume increases.
The kinetic model that was designed to the new conditions are given, at the same temperature.
enable the prediction of gas behaviour was
based on two main assumptions that EXAMPLE
pertain to an ideal gas
5 Ltr of an ideal gas is contained in a cylinder at 2 bar.
1. The gas molecules are small spheres, the A piston then compresses the gas until the volume is
volumes of which are negligible in reduced to 3 Ltr.
comparison to the volume that the gas
occupies. What will the new pressure be, once the temperature has
stabilised to its initial value ?
2. There are no attractive or repulsive
intermolecular forces, and the behaviour We will use the left side of the equation to represent the
of the molecules when colliding is similar to initial conditions, and the right the final ones.
that of billiard balls, the collisions being elastic.
Thus :
The Boyle’s Law statement that gas pressure (P) is inversely 2 X 5 = P2 X 3
proportional to its volume (V) can be written
P2 = 2 X 5 = 3.33 bar
as P is proportional to 1v 3

Which means that the pressure is equal to a constant (k)


divided by the volume.
8
Petroleum Open Learning

If the Ifvolume
the volume is measured at various temperatures and the
is measured at various temperatures
results are plotted on a graph, we obtain a diagram like
and the results are plotted on a graph, we obtain
Figure 3 when we use the Celsius temperature scale
a diagram like Figure 3 when we use the Celsius
temperature scale.

Figure 3 : Volume / Temperature Graph Figure 4 : Volume / Temperature Graph (Celsius and Kelvin Scale)
(Celsius Scale Only)

We see that there is a simple linear relationship Figure 4 is similar to Figure 3, but with the graph line extrapolated to intersect the Temperature axis.
between the volume and temperature (the graph This point is taken as 0 for our other temperature scale, and we see that it corresponds to -273.15°C.
is a straight line). However calculations involving Absolute Zero is the term that is commonly applied to this temperature, since it is physically impossible
temperatures below 0°C are slightly inconvenient to achieve lower temperatures than it. Absolute zero has been approached experimentally, but has never
due to the presence of negative numbers. This been quite achieved; and is therefore a theoretical value rather than a practical one. We see in Figure 4
problem is solved by employing a different that the gas would occupy no volume at that point; a futile observation, since no substance would be in
temperature scale which only has positive values. the gas phase at such a low temperature.

9
Petroleum Open Learning

Absolute zero is the lowest point on the absolute To derive a mathematical expression of Charles'
temperature scale, which is measured in units Law we can employ a similar argument to the one
called Kelvin (K) in the SI system. In the Imperial we used for Boyle's Law.
system the units are known as Rankine (˚R).
The statement that the volume of gas is
As evident in Figure 4, the unit step sizes for the proportional to its temperature can be written as :
Kelvin and Celsius scales are the same. So a
one degree Celsius temperature change is also V = cT
one Kelvin. (A convention, which is by no means
universally applied, is to omit the term degree where c is a constant
when using absolute temperature units). O°C,
then, is 273.15 K, and 100°C is 373.15 K, so to so c=V
convert from °C to K we simply add 273.15 to the T
°C value.
We see in Figure 5 that if the absolute Referring to Figure 5 (a) and (b) we see that
Absolute zero on the Rankine scale is equal to temperature is doubled, the gas volume will also
-459.67°F, and a degree on the Rankine scale is be doubled. V1 = c and V2 =c
the same size as a degree on the Fahrenheit T1 T2
scale. Charles' Law, then, states that: at constant
pressure, the volume occupied by a given hence : V1 = V2
In most practical situations sufficient accuracy is mass of gas is proportional to its absolute T1 T2
achieved by using 273 as the conversion factor temperature.
between Celsius and Kelvin, and 460 between
Fahrenheit and Rankine. However, where high
This equation can be used to evaluate the new
accuracy is required, such as in fiscal gas flow
volume or temperature of an ideal gas for a
measurement, the more exact values should be
change in which the pressure stays constant.
used.

10
Petroleum Open Learning

Combined Ideal Gas Law


Boyle's and Charles' laws combine to give the
equation:
Test Yourself 1.1
P1V1 = P2V2
T T 2 Ltr of an ideal gas at 10˚C and 2 bar is compressed to a
1 2 volume of 0.5 Ltr. Given that the heat of compression raises its
We will apply it in the following examples: temperature to 25˚C, what will its pressure be?

A 5 Ltr sample of gas at 15˚C and 3 bar a is heated to


45˚C and has its pressure reduced to 1.5 bar. Assuming
that it behaves ideally, what will its new volume be?

The first step is to ensure that the pressures and You will find the answer in Check Yourself 1.1 on page 54.
temperatures are in absolute units. The pressures
are quoted in bar a, which means that the values are
absolute. However we will have to add 273 to the
temperatures to convert them from ˚C to K.

Using P1V1 = P2V2


T
1 T2 Molecular Mass
we will ascribe the initial conditions to left side, and the
changed ones to the right. Molecular mass is a physical property of all substances. A comprehensive description of it can be found in
elementary chemistry text books and training manuals, but for our purposes a simple description is sufficient.
Thus we need to find V2.
3 x 5 1.5xV2 All matter consists of atoms. In many substances two or more atoms combine to form molecules. As we wish to
(15 + 273) (45 + 273) keep this description simple, we will accept the atomic mass units (a.m.u.) given in the following text.

V2 = 3 x 5 x 318 Let us consider methane, the lightest alkane hydrocarbon and the main component of natural gas. It is a
(1.5 x 288)
molecule comprising one carbon atom bonded to four hydrogen atoms. Carbon has an atomic mass of 12
atomic mass units (a.m.u.). Hydrogen has an atomic mass of 1 a.m.u. The molecular mass of methane is
V2 = 11.04 Ltr
the sum of the masses of its constituent atoms, which is therefore 12 + (4 x 1) = 16 a.m.u.
11
Petroleum Open Learning

Although it may not be scientifically correct, the An appropriate unit of measurement for molecular Having seen how the molecular mass of pure
terms molecular mass and molecular weight weight (Mw) is obtained by using the units in the substances like methane and ethane are
(Mw) are often used interchangeably. calculated, we will now determine the molecular
n= m equation. Hence: mass of a mixture of components, such as
The very small size of atoms and molecules Mw natural gas. We have already seen that methane
makes calculations using their individual masses is the main constituent of natural gas, but it
inconvenient. A more practical approach is to n(kg-mol) = m(kg) also contains smaller quantities of heavier
consider the mass of a large number of them, and Mw hydrocarbons such as ethane, propane and
this involves the concept of the mole. Mw = m(kg) butane. The relative amounts of these can vary
n(kg-mol) considerably between samples of gas, depending
The mole, usually written as mol, or gramme- on factors such as the reservoir conditions,
mole (g-mol) is defined as the atomic or processing methods, etc. These variations can
molecular mass of a substance expressed in So, in this case, the unit of molecular mass is have very significant effects on the behaviour of
grammes. kg/kg-mol. It may also be written as kg kg-mol-1 gas during handling and measurement of its flow
It will have the same value when expressed in rates.
The number of moles of substance is usually units of g/g-mol, so molecular mass is one of the
assigned the symbol n, and is easily calculated few physical quantities for which it is acceptable We will use a simple example of a
for a given mass (m) of material using the to omit its unit of measurement. two-component mixture of methane and ethane.
relationship:
To perform the calculation, we will obviously
n= m need to know the relative quantities of each
Mw
Test Yourself 1.2 component. These are expressed as mole-
fractions, which simply means the relative
EXAMPLE number of molecules of each constituent. Let
Calculate the molecular mass of ethane, us assume that our mixture has mole-fractions
Calculate the number of kg-mol in 40 kg of which is a molecule comprising two carbon of 80% and 20% for methane and ethane
methane. and six hydrogen atoms. respectively. In other words, 80 of every 100
molecules of the mixture are methane, and 20 are
We have already seen that the molecular mass of ethane.
methane is 16.
You will find the answer in Check Yourself
n = m = 40 = 2.5kg-mol 1.2 on page 54.
Mw 16 12
Petroleum Open Learning

The procedure is shown in the following table, and involves adding up the The physical behaviour of the heavier mixture will be considerably different
results of multiplying the mole-fraction of each component by its molecular from that of the lighter one.
mass.
The procedure for calculating the molecular mass of mixtures of more than
Component Mol. Mass Mol. Fraction Mol. mass x two components is exactly the same as for two.
Mol.Fraction

Methane 16 0.80 12.80


Gas Constants
Ethane 30 0.20 6.00 The form of the ideal gas equation we looked at earlier :
Mixture Mol. Mass 18.80
P1V1 = P2V2
The mixture in the table could be considered as consisting of molecules with T1 T2
an average mass of 18.80 kg / kg-mol. In the following exercise you will see
the effect changing the concentrations of the components has on the mixture
molecular mass. implies that

PV = a constant
T

Test Yourself 1.3 In other words, for a given type and mass of an ideal gas, the absolute
pressure multiplied by the volume and divided by the absolute temperature
Calculate the molecular mass of a mixture comprising 60% will always produce the same answer.
methane, and 40% ethane.
If we call the constant C, we can rewrite the equation as:

PV = CT

A disadvantage of the equation, as it stands, is that C will only be constant for


You will find the answer in Check Yourself 1.3 on page 54. a gas with a given molecular mass. It will have different values for methane
and ethane, for example. What we need is a constant that will have the same
value regardless of the type of gas under consideration.

13
Petroleum Open Learning

This is achieved by using the number of moles EXAMPLE


(n) of gas, instead of the mass, in the equation.
Thus: What is the volume of 1.5 kg of air at 1.5 bar and 25°C? (Take the molecular
PV = nRT weight of air as 29, and assume ideal behaviour).

Because it applies to all ideally behaving gases, We first find the number of moles of air using:
R is termed the universal gas constant. Its
value will depend on the units by which the other n = m = 1.5 = 0.0517 kg-mol
terms in the equation are measured; the following Mw 29
table shows values of R for various combinations
of units. and use this value in : PV = nRT, having selected the appropriate value of R
from the table as 0.0831. Note that this will make our volume units m3. As these
are the most commonly used units for gas measurement calculations, we will use
0.0831 for all our calculations where a value of R is required

Inserting these values gives:

P V T n R 1.5 x V = 0.0517 x 0.0831 x 298


kPa m3 K kg-mol 8.130
V = 1.280 = 0.853 m3
bar m3 K kg-mol 0.0831 1.5
Ltr
bar K g-mol 0.0831
Test Yourself 1.4
bar cm3 K g-mol 83.1
psia ft3 oR Ib-mol 10.73
How many kg of methane will, when behaving
ideally, occupy a volume of 1 m3 at 2 bar and
20°C?

You will find the answer in Check Yourself 1.4


on page 54.

14
Petroleum Open Learning

Reference Pressures and There are various conventions for expressing EXAMPLE
reference volumetric and volumetric flow rate
Temperatures What is the actual volume of 5m3 (st) of an ideal
units. For example the standard cubic metre
may be written as sm3. The oil industry often gas at 3 bar and 25°C.
Gas flow rates are often quoted in volumetric units,
such as m3 / minute and ft3 / minute. Having expresses gas flow rates in millions of volume We will use
studied the very significant interdependence of units per day, which would be 106 sm3 / d or 106
pressure, temperature and volume, you will now sft3 / d. However, the oil industry would normally P1V1 = P2V2
be aware that it is meaningless to express a write these as MMSCMD or MMSCFD, although T1 T2
volume of gas without stating the pressure and you may also come across ksm3/hr (1000m3 (st)
temperature at which it is measured. This has led / hour in place of MMSCMD. Before performing and assign the standard conditions to the left side.
to the establishment of reference conditions for any calculations, you must always ensure that you
gas volumetric measurements. know which units the quantities you are using are
Hence:
expressed in, and the units that will apply to the
The values of reference pressures and temperatures result.
may vary between countries and contracts, but the P1 = 1.013 bar
most common ones are Gas volumes or volumetric flow rates measured at T1 = 15°C = 288 K
pressures and temperatures other than reference V1 = 5 m3
1. 1.013 bar ; 15°C ones are sometimes called actual volume, or P2 = 3 bar
2. 1.013 bar ; 0°C actual volumetric flow rate. 1 m3 at 10 bar and T2 = 298 K
40°C is an actual cubic metre at that pressure and
3. 14.73 psia ; 60°F temperature, and would require conversion to be and we are required to find V2.
expressed in terms of reference conditions, as the
The first one is usually termed Standard following example shows.
Pressure and Temperature, and the second 1.013 x 5 = 3 X V2
Normal Pressure and Temperature. However 288 298
you should be aware that some textbooks use
Standard Pressure and Temperature (STP) with 0.01007 X V2 = 0.01759
a reference temperature of 0°C. Clearly, the first
two conditions apply to the metric system, and the V2 = 0.01759 = 1.747 m3
third to the imperial system. 0.01007

A standard cubic metre, then, is the quantity of gas


that has a volume of 1m3 at 15°C and 1.013 bar.
15
Petroleum Open Learning

Non-Ideal Gas Behaviour

In our work so far, we have repeatedly used the


term ideal to describe the characteristic
behaviour of gas. Real gases can deviate
significantly from ideal behaviour, since the
simple mathematical model from which ideal
gas behaviour was predicted is not adequate to
describe all gases under all possible conditions.
The model has been modified to make prediction
more accurate over a wider range of conditions,
but it is beyond the scope of this book to present
the explanation that these modifications would
require.

Although precise gas flow measurement


techniques would require use of fairly complex
formulae, we will be able to gain an appreciation
of how real gases deviate from ideal behaviour,
and even to make reasonably accurate
calculations, by introducing a term called the
compressibility factor.

The availability of this factor is due to the large


amount of empirical data that has been gleaned
from extensive experimental work over many
years, particularly in the natural gas industry. The This provides curves that represent values of Z plotted against pressure for various temperatures;
compressibility factor is usually given the symbol so to find the appropriate factor, the point that represents the relevant pressure and temperature is
Z, and is a function of the type of gas, pressure identified and the corresponding value is read from the vertical axis. In Figure 6, for example, we see
and temperature. that if the gas pressure and temperature is P1 and T1 respectively, the compressibility factor is Z1.
Note that if the stated temperature lies between two curves, interpolation is necessary.
It is customary to present Z in the form of charts,
the general form of which is shown in Figure 6. 16
Petroleum Open Learning

When Z has been determined, it is inserted in the


ideal gas equation thus : The general characteristic, then, is for curves
of temperatures below about 300°C to give
PV = ZnRT decreasing values of Z (increasing deviation
from ideal behaviour) as the pressure increases.
The first point to note is that if Z=1, the gas is However, at a certain pressure, a minimum
ideal ( multiplying by 1 has no effect ). We see Z value is reached and further pressure
in Figure 6 that Z = 1 at low pressure, which increment causes Z to increase until it reaches
includes atmospheric and standard pressure. 1, where the gas is again ideal. Increasing the
So, in most cases, ideal gas behaviour can be pressure beyond that point causes Z to become
assumed at these pressures. Points on the chart progressively greater than 1, which is again an
furthest away from the Z=1 line denote the largest increasing deviation from ideal behaviour, but in
deviations from ideal behaviour, the Z1 in Figure 6 the opposite sense from values of Z that are less
being an example. than 1. This should become clearer when we look
at some examples.
In Figure 6, T2 represents a higher temperature
than T1, and we see that T2 curve is generally
closer to Z = 1 than the T1 curve. This
demonstrates the characteristic that higher
gas temperatures tend to produce behaviour
that is closer to ideal than low ones. However,
temperatures above 300 - 400°C will have curves
that show increasing deviation above the Z = 1
line with increasing pressure.

17
Petroleum Open Learning

We will now look at the compressibility factor


chart for methane, depicted in Figure 7.

The chart covers ranges of pressure and


temperature that encompass most processing
conditions. The main features of the chart are:

1. At pressures of 10 bar or less the gas


virtually behaves ideally.

2. At temperatures below -20oC methane


becomes rapidly non-ideal as the
pressure rises from 10 to 100 bar.

3. The maximum deviation from ideal


behaviour occurs at pressures between
100 and 200 bar, for temperatures below
100oC.

4. Methane virtually behaves ideally between


about 100 and 200oC, at pressures below
250 bar.

18
Petroleum Open Learning

EXAMPLE

Calculate the volume occupied by 1 kg-mol of b) Accounting for non-ideal gas behaviour, we find
Activity methane at 70 bar and —35°C, in Figure 7 that Z = 0.68 (interpolation between
the T = —30°C and T = —40°C curves was
a) ignoring non-ideal behaviour required). So we use this value in :
Use Figure 7 to find values of Z for each
of the following sets of conditions :
b) accounting for non-ideal behaviour. PV = ZnRT
1. 70 bar ; 20oC
a) Assuming ideal behaviour, we simply apply V = ZRT
P
2. 100 bar ; -70oC
PV = nRT
= 0.68 x 0.0831 x 238
3. 150 bar ; 0oC
Since n = 1 kg-mol, we can ignore it. So 70

V = RT = 13.449
P 70

V = 0.0831 x (-35 + 273) V = 0.192 m3


70
Your answers should be :
= 19.78
1. 0.88
70
2. 0.36
We see that, in this case, failure to consider the
3. 0.74
V = 0.283 m3 non-ideality would have led to an error of

(0.283 - 0.192) x 100 = 47.4%. In other words,


The following example will enable you to 0.192
appreciate the magnitude of the effect that non-
ideal gas behaviour can have. we would have overestimated the actual volume of
the gas by almost 50%.

19
Petroleum Open Learning

EXAMPLE

Find the volume occupied by 3m3 of methane at


Test Yourself 1.5 65 bar and 20°C, after it has been compressed to
150 bar and chilled to -30°C.
Using Figure 7 to find the compressibility factor, calculate the Using:
volume of 32 kg of methane at 100 bar and -50°C. Estimate the P1V1 = P2V2
percentage error you would have incurred by assuming the gas Z1T1 Z2T2
to behave ideally.
and assigning the initial conditions to the left side,
P1 = 65 bar, V1 = 3 m3, T1 = 293 K and, from
Figure 7, Z1 = 0.88. The final conditions are: P2 =
You will find the answer in Check Yourself 1.5 on page 54. 150 bar, T2 = 243 K, at which Z2 = 0.61 ; and V2
is the volume we need to calculate.

We have already seen that, to perform gas We have also seen that the compressibility factor 65 x 3 = 150 x V2
calculations involving changes in conditions from varies with pressure and temperature, so 0.88 x 293 0.61 x 243
P1, V1 and T1 to P2, V2 and T2 , we use: different values of Z are likely to be required for
each set of conditions; ie. Z1 for P1, V1 and T1, 1.012 V2 = 0.756
P1V1 = P2V2 and Z2 for P2, V2 and T2.
T1 T2 V2 = 0.756 = 0.747 m3
The complete equation will be : 1.012
You will now be aware that, unless the gas is
ideal at both sets of conditions, corrections will P1V1 = P2V2 Until now, we have used Z values which were
have to be applied. Z1T1 Z2T2 less than 1, which caused the volume occupied
by a given mass of gas to be less than that
predicted by the ideal gas equation.

Examining PV = ZnRT, from which P = ZnRT


V
we see that the pressure of a given mass of gas
at a given volume and temperature would also be
less than if it were ideal.
20
Petroleum Open Learning

It should now be clear that, if Z is greater than 1, Figures 8 and 9 are examples of such charts, and apply to mixtures of molecular mass 18.85 and
the volume at a given pressure, or the pressure at a 23.2 kg / kg-mol respectively. Charts are also available for lower, intermediate and higher molecular
given volume, will be greater than that predicted by masses. When applying factors to mixtures with molecular masses that are between values for which
the ideal law. You will also observe, when studying charts are available, reasonable accuracy can be achieved by interpolation.
Figure 7, that Z is only significantly greater than 1
at pressures and temperatures considerably higher
than those we normally encounter.

As stated earlier, natural gas is a mixture of


hydrocarbons, so it is worth taking a brief look at
the selection of compressibility factors for such
mixtures.

When we looked at mixtures earlier, we saw that,


in addition to the type of components present, the
molecular mass of the mixture is determined by
the concentration, or relative amount of each
component. A mixture with a high concentration
of methane, the lightest hydrocarbon, produces a
lower molecular weight mixture than one with a
lower methane concentration.

The only difference between finding compressibility


factors for gas mixtures and for pure gases, is in
the selection of the appropriate chart. Instead of
the name of the gas to which the chart applies,
a gas mixture chart is identified by the average
molecular mass of the mixture.

21
Petroleum Open Learning

Comparing Figures 8 and 9, it is clear that,


except at temperatures well above 100°C, the
deviation from ideal gas behaviour is considerably
greater for the heavier gas. For example, at
100 bar (10,000 kPa) and 5°C, we see that the
Z values are 0.47 and 0.68 for the heavier and
lighter gases respectively. This is consistent with
the fact that its behaviour is closer to ideal when
a gas is relatively light.

We have spent some time looking at the gas laws


and the implications of non-ideal behaviour. We
did so because they are a very important aspect
of gas flow measurement.

We will finish this topic with an exercise in which


you will calculate volumetric flow rates instead
of simply volumes. This should not present
problems, since volumetric flow rate is just
volume divided by time and we substitute the
symbol Q for V in the equation
PV =ZnRT (PQ =ZnRT). This exercise is slightly
longer than the ones you have done so far, and
it is worth ensuring that you understand how the
answers are worked out, especially if you do not
get them right first time.

22
Petroleum Open Learning

Density Example

Test Yourself 1.6 In fluid flow measurement, and particularly when


the fluid is gas, density is an important physical
Find the density of methane at standard pressure
and temperature (1.013 bar and 15°C).
property.
ρ = PMw = 1.013 x 16
1. For two types of natural gas, with
Density is defined as mass divided by volume, ZRT 0.0831 x 288
molecular masses of 19 kg / kg-mol
and is usually given the Greek symbol ρ.
and 23 kg / kg-mol, calculate their
ρ = 0.678 kg / m3
actual volumetric flow rates in m3 /
So ρ=m
minute, given the following data:
V Note that we ignored Z since the gas is ideal at
standard conditions.
Mass flow rate = 50 kg / minute
Line pressure = 130 bar
Having seen how sensitive gas volume is to
Line temperature = 10oC
pressure and temperature, you will appreciate that
density will be similarly affected.
Test Yourself 1.7
2. Find the standard (15oC, 1.013 bar)
We have already seen that the number of moles,
volumetric flow rates for both your
answers to part 1. Find the density, at standard pressure and
n = m , and that this can be substituted for temperature, of the following natural gas
Mw components:
in PV = ZnRT, so that PV =ZRTm Ethane (Mw = 30)
You will find the answers in Check
Mw Propane (Mw = 44)
Yourself 1.6 on page 55.
This is rearranged to : Butane (Mw = 58)
m = PMw = ρ
V ZRT You will find the answers in Check Yourself
1.7 on page 56.
which is the form that is often used in gas flow
calculations.
23
Petroleum Open Learning

Specific Gravity So, in addition to being equal to the density of


gas divided by the density of air, specific gravity is
Another method of expressing the density of a equal to the molecular mass of gas divided by
material is to use the measurement specific the molecular mass of air. i.e.:
gravity (s.g.). Specific gravity is sometimes
referred to as relative density, which is an apt sgg = Mw g
term since it is defined as the density of the Mw a
substance being evaluated divided by the
density of a reference substance. .˙. sgmethane = 16 = 0.55
29
In the case of liquids, the reference material is
water; for gases it is air. Note that sg is a dimensionless number
(it has no units).
So for gases we can write:

s.g. = ρg
ρ
a
where the subscripts g and a denote the gas
being evaluated and air respectively.

To avoid having to account for non-ideal gas


behaviour, measurements are usually referred to
standard pressure and temperature.

Consider the specific gravity of methane:

sgmethane = ρ methane
ρ a

As the density (ρ) of both methane and gas are


calculated at standard pressure and temperature,
we need only use the molecular mass (Mw) for
gas and air in the equation. 24
Petroleum Open Learning

Summary of Section 1

Applications in which accurate gas flow measurement is required are :

• Oil Field Evaluation


• Process Control and Optimisation
• Gas Sales Contracts
• Hydrocarbon Taxation

Boyle’s Law and Charles’ Law combine to express the relationship between the pressure, volume and temperature of gases when they behave
ideally, with the proviso that absolute temperature and pressure units must be used in the calculations.

The number of moles of a substance is found by dividing its mass by its molecular mass, a procedure that can be applied to mixtures as well as
pure substances.

The number of moles (n) can then be used in PV = nRT, where R is defined as the universal gas constant.

The interdependence of these properties demands the use of reference pressure and temperature at which gas volumes are calculated.

Most gases only obey the ideal gas laws at certain pressures and temperatures, so the compressibility factor (Z) is introduced to compensate for
non ideal behaviour.

Gas density can be evaluated from the ideal gas equation, and gas specific gravity is defined as its density divided by that of air, both values being
referred to standard conditions. Gas specific gravity is also found by dividing the molecular mass of the gas by that of air.

25
Gas Flow Measurement Petroleum Open Learning

Section 2 - Fluid Flow Principles


We will start this section with a general outline of fluid flow principles. Figure 10 represents fluid flowing through a pipe, in a condition known as
You should note that the term fluid applies to gases as well as liquids. laminar flow. This is characteristic of very gentle flow, in which we see from the
velocity profile that the fluid velocity is zero at the pipe wall, and progressively
Fluid Flow Characteristics increases to a maximum at a point midway across the pipe.

For calculation purposes, the mean or average velocity is the important value;
The Streamline Concept in a laminar flow situation it would typically be about half the maximum velocity.

The construction of streamlines in a diagram is designed to enable Consider now what happens if the flow rate is increased. The laminar profile is
a visual image of fluid flow characteristics to be achieved, In simple maintained until a certain fluid velocity is reached, at which point eddy currents
terms, streamlines are drawn such that adjacent lines represent start to appear, indicating a breakdown of the laminar pattern as the layers start
different fluid flow speeds. to mix, and the onset of turbulent flow.

Turbulence commences near the centre of the pipe, where the velocity is
Flow Types : Laminar and Turbulent
greatest, and spreads towards the pipe wall as the flow rate increases. At the
pipe wall a thin layer of laminar flow will survive unless very severe turbulence
occurs. A flow pattern exists between the turbulent and laminar regions which is
known as the boundary layer or transition layer, as shown in Figure 11.

26
Petroleum Open Learning

The velocity profile is much flatter in a turbulently Reynold’s Number Viscosity can be considered simply as an
flowing fluid, as we see in Figure 12. indication of a fluid’s resistance to flow. Treacle
An indication of whether fluid flow is likely to be at temperatures below 10°C, for example, has a
laminar or turbulent, or between them, can be much higher viscosity than water. Gases generally
obtained by calculating a value called Reynold’s have considerably lower viscosities than liquids,
Number (Re), using the following formula: but this is partially compensated for in Reynold’s
number calculations by their densities also being
Re = ρDVavg lower.
µ
Reynolds number is an important factor in flow
Where: calculations, and is often incorporated in a quantity
called the discharge coefficient, as we will see
D = Internal Pipe Diameter (m) later.
vavg = Average Fluid Velocity (m / s)
ρ = Fluid Density (kg / m3)
= Fluid Viscosity (kg / m s)
Test Yourself 2.1
µ

Note that Re is dimensionless, since the units on


the right side of the formula cancel each other.
Find Reynold's number for a process gas,
with a viscosity of 1.2 x 10-5 kg / m s and a
density of 20 kg / m3, which flows through
a 125 mm internal diameter pipeline at an
At the boundary layer there is considerable friction average velocity of 2 m / s.
between the moving fluid and the static fluid layer, As an approximate guide, values of Re less than
giving rise to the term boundary layer drag. A 2 000 indicate laminar flow; while values greater Predict whether or not the flow is likely to
Telsa pump employs this principle in its design, than 30 000 indicate turbulence. For intermediate be turbulent.
which is essentially a disc without blades that values, the flow would be partially turbulent. Note,
rotates at a very high speed. The boundary layer however, that this prediction applies to straight You will find the answers in Check Yourself
provides the friction which allows the disc to sections of pipelines; at elbows, for example, 2.1 on page 57.
impart centrifugal acceleration to the liquid being turbulence will occur at lower Reynold’s numbers.
pumped.
27
Petroleum Open Learning

Fluid Flow Properties Pressure Energy can be considered as a form of


potential energy in terms of the ability of the fluid
Since we will only be concerned with points in a
flow line immediately upstream and downstream
to do work, such as driving a piston or impeller. of flow measuring devices and thus relatively
Bernoulli’s Principle This energy is expressed as the pressure close to each other, we can simplify the
multiplied by the volume (PV). However, we have expression as follows:
Daniel Bernoulli was responsible for considerable seen that density (ρ) is mass (m) divided by
advancement of fluid flow theory by developing a volume (V), i.e. The fluid temperature will be constant, so the
principle based on the conservation of energy. internal energy will not change and the term Um
The basis for this principle is that the total energy þ = m, so V = m and the pressure energy can be discarded.
of the fluid remains constant at all points through V ρ
which it flows. The main assumptions in the There will be no significant height difference
development of the theory are that: the fluid is therefore equals Pm / ρ. between the points, so the mgh term can be
incompressible, frictionless and adiabatic (no ignored.
heat energy enters or leaves it). Kinetic Energy is due to the fluid’s motion,
and can be considered as the energy that will Removing these terms and dividing by the mass
The total energy of a flowing fluid is the sum of be converted to another form, or forms, when it m to get the energy per unit mass (Em) gives:
the following components: stops moving. It is a function of its mass (m) and
average velocity (v) and is calculated from the
Internal Energy (U) can be considered for our term mw2 which, along with the potential Em = P + v2
purposes, without describing its thermodynamic 2 ρ 2
definition, simply as a function of the fluid energy expression mgh, you will recognise if you
temperature. If U is defined as the internal energy have studied elementary physics. Multiplying energy per unit mass by density gives
per unit mass, the total internal energy of the fluid pressure.
is Um. Adding these terms to express the total fluid
energy (E) gives: ρ Em = P+ þv2
Potential Energy is the energy the fluid has by 2
virtue of its position above some reference E = Um + mgh + mP + mv2
level. If it is a height h above this reference, its ρ 2
potential energy is mgh, where m is its mass and
g is the acceleration due to gravity. Conversely, it
is the energy required to propel it to a height h.

28
Petroleum Open Learning

Here we see that the pressure of a flowing fluid, The Continuity Equation Alternatively, or additionally, you could consider a
subjected to the constraints I have described, particle which is travelling at the average velocity v
consists of two components : P is referred to as of the flowstream. If, for example, the velocity is 2
the m / s, the particle will move 2 m along the pipe in
1 second and the volume of fluid displaced will be
2 x A (the volume of a cylinder is calculated by
static pressure, and ρv2 is called the dynamic multiplying its cross-sectional area by its length;
2 so the volume of fluid moving along the pipe
pressure, because it is associated with the fluid in one second is, in effect, that of a cylinder of
velocity. Clearly, if the fluid is stationary, the total cross-sectional area A and length 2 m).
pressure would be P. Using the symbol PT to
represent the total pressure: The mass flowrate Qm is derived from ρ = M so
V
ρ Em = PT = P + ρv2 m = ρV, and, replacing V with Qv,
2
Qm =ρ Qv = ρ v A
We now have an equation with only three physical Figure 13: Velocity / Cross-sectional Area
quantities : pressure, density and velocity. If the Relationship We now have two simple formulae relating
fluid density is known, and we measure the total volumetric and mass flow rates to pipe dimensions
and static pressures, we can calculate the velocity. Figure 13 represents fluid flowing, at an average and fluid velocities.
This is the principle used by the Pitot Tube, which velocity v, through a pipe of cross-sectional area A.
I will describe later.
The volumetric flowrate Qv is given by:

Qv =Av

You will find that the easiest way to confirm this


is to consider the units involved. Using the SI
system, A is in m2, and v is in m / s; so multiplying
the units gives:

m2 x m/s = m3/s, which is volume per unit time.


29
Petroleum Open Learning

Now consider a situation in which the pipeline


diameter changes, as in Figure 14.

Test Yourself 2.2


Water flows through a 200 mm internal diameter
pipeline at an average velocity of 3 m / s. Taking
the density of water as 1 000 kg / m3, find the
volumetric and mass flow rates.

You will find the answer in Check Yourself 2.2


on page 57.

Figure 14 : Pipeline Diameter Reduction

Here we see fluid flowing at average velocity V1, from the section of pipe with cross-section
area A, to the section with area a where its average velocity is v2. To account for compressible fluids
which might experience a change in density, we note that the densities are ρ1 and ρ2 in the wide and
narrow sections respectively.

The continuity equation, as its name might suggest, is based on the principle that the mass flowrate
must be constant through all cross-sections of a flowstream. So the mass flowrate in the wide section
(ρ1 A v1) is equal to the mass flowrate in the narrow section (ρ2 a v2) :

ρ1 A v1 = ρ2 a v2 =Qm

30
Petroleum Open Learning

EXAMPLE

A liquid flows through a pipeline the diameter of


which changes from 150 mm to 75 mm. If the
average velocity in the wide section is 0.5 m / s,
what will its velocity in the narrow section be ?

Let A and a be the cross-sectional areas of the


wide and narrow sections respectively, with v1 the
velocity through A and v2 the velocity through a.

Note: We are using π/ 4 x D2 to calculate cross-


sectional area, although you may be more familiar
with π r2

Av =av
1 2
Figure 15 : Pipeline Diameter Reduction
π x 0.1502 x 0.5
v2 = A v1 = 4
a π x 0.0752
4 Figure 15 is silimar to Figure 14, but with the addition of two pressure gauges, P1 and P2, which
measure the static pressures of the wide and narrow pipe sections respectively.
The π s cancel to give:
Consistent with the assumption that energy losses due to friction are negligible, the total pressure PT
4
remains constant; thus we can write:
v2 = 0.01125 = 2m/s
0.005625
P1 + ρv12 = P2 + ρv22
2 2
We can also apply the Bernoulli pressure
relationship, in which the total pressure is the sum
Clearly the pressure P2 must be less than P1 to compensate for v2 being greater than v1 and to obey
of the static and dynamic elements, to a situation
this equation.
involving a changing flowstream diameter.
31
Petroleum Open Learning

Fluid Flow Equation Applying it to the varying area flow depicted in While it is possible to measure fluid flow by
Figure 15, but assuming that the density does not applying this equation to a changing cross-
From the continuity principle we can express change ( ρ1 = ρ2 = ρ ), we get: sectional area pipeline configuration as shown
mass and volumetric flow rates in terms of the in Figures 14 and 15, a high degree of accuracy
flowstream dimensions and the fluid densities
and velocities. We have been able to calculate

( )
ρ1+ρ Qm 2 = ρ2+ρ
2 ρA 2 ( )
Qm
ρa
2 would not be achieved. In practice, there would
be considerable pressure energy loss due to
turbulence and friction. As stated when describing
flowrates in the preceding example and exercise,
but only because we were given the average Reynold’s number, the discharge coefficient term
velocity of the fluid. In practice, accurate Now we have an expression relating the mass will be introduced to compensate for this.
measurement of this quantity is difficult due to the flowrate to the change in static pressure,
susceptibility of the measuring devices to fouling, flowstream cross-sectional area, and fluid Gas flow measurement would be considerably
and other problems; so it is preferable to avoid density; no longer requiring velocities. Cross- inaccurate from a calculation using the equation
measuring it directly. sectional areas are known, pressures are easily as it stands, since compressibility is not
measured and densities can be measured or accounted for. Again we will see that the equation
Taking Bernoulli’s equation: calculated. This is the principle behind most of the will be modified by incorporation of a factor to
gas measurement devices that will be described correct this.
PT = P + ρv2 in this book.
2 These corrections and other modifications to the
It is customary to refer to the static pressure equation will be described in the next section. You
and the mass flowrate expression : change across a measurement device as the will be relieved to know that you are not expected
differential pressure, and it is often called delta to remember the fluid flow equation, either in this
Qm = ρv A p, which is written as ∆p. So ∆p = P1 - P2. form or when modified. However, you should be
able to describe the terms in it, and understand
which we rearrange in terms of v : When ∆p is substituted for P1 - P2 and the the terms and concepts of the Bernoulli and
equation is rearranged and simplified, we get: Continuity principles from which it was developed.
v = Qm One important relationship that you should keep
ρA in mind is that the flowrate is proportional to the

Substituting this for v in the Bernoulli expression


gives:



Qm =
� 2∆pρ( A2a2
A2 _ a2
) square root of the differential pressure.

Because of the absence of the correction factors, I


will not include a calculation exercise at this point;


PT = P+ρ
2 ( )
Qm 2
þA
instead, the following exercise will invite you to
test your knowledge of the terms and concepts.
32
Petroleum Open Learning

Test Yourself 2.3 Summary of Section 2

1. Which two components does the total pressure Fluid flow can be categorised as laminar or turbulent, and
of a flowing fluid consist of ? Reynold's number can be used to predict which category
applies.
2. Which quantity is always constant through all
cross-sections of a fluid flowstream ? Bernoulli's principle of fluid energy conservation can
be simplified and expressed in pressure terms, the
3. State which quantities are denoted by the total pressure being the sum of the static and dynamic
following terms and give their SI units components.

a. A and a Bernoulli's principle, and continuity principle of mass flow


b. v (note that this is lower case) conservation, combine to produce a fluid flow formula
c. ρ which does no account for frictional losses.
d. Qv
e. Qm
f. ∆p

4. If the mass flowrate and the fluid density are


known how is the volumetric flowrate calculated ?

5. What is the relationship between the flowrate


and the differential pressure ?

You will find the answers in Check Yourself 2.3 on


page 57.

33
Gas Flow Measurement Petroleum Open Learning

Section 3 - Measurement Devices and Methods


Types and Application Clearly, the dynamic pressure is found by
subtracting the reading on the static pressure
The gas flow measurement devices I will discuss gauge from that on the gauge attached to the
can be described as static, in that they have no nozzle. If we call the dynamic pressure Pd , we can
moving parts. There is a device which is used for find the velocity from :
gas flow measurement called a turbine meter. As
its name suggests, it contains a small turbine or Pd = ρv2
fanlike wheel which is rotated by the flowstream, 2
the speed of rotation being a function of the so
v= 2Pd
flowrate. The rotational speed of the turbine ρ
shaft is converted to an electrical signal which is
processed to give an indication of the gas flowrate.

Note that v is the velocity at the nozzle, not the
average velocity of the stream. Before the mass or
Some models work quite satisfactorily in some volumetric flowrate can be evaluated it is essential
situations, but the device has not been universally to know the relationship between v and the
accepted in applications where high accuracy and average velocity. We see that, in the example
precision are required. For this reason there will be represented by Figure 16, the nozzle is in the
no further description of it in this book. Figure 16 Pitot Tube
centre of the stream, where it measures the
maximum velocity. As stated when describing flow
The static flow measurement devices can be patterns, this could be about twice the average
categorised into two groups: velocity head and velocity if the flow were completely laminar.
differential pressure, the names of which give an A diagrammatic sketch of a Pitot tube is shown in
This, then, is one disadvantage of the Pitot tube;
indication of their operating principle. I will only Figure 16. The main feature is a nozzle (1)
another is that the nozzle is prone to blockage
describe one velocity head device: the Pitot tube. pointing in the direction of the flow source and
by foreign matter, as would be the case in the
connected to a pressure gauge. A second pressure
majority of natural gas applications. However
Pitot Tube tapping (2) is subjected to the pressure at the pipe
it should be stated that the device is widely
wall, perpendicular to the flow direction, and is
employed with success where the gas is clean and
I referred to the Pitot tube in the preceding section, thus measuring the static pressure. The pressure
the flow characteristics of the system have been
when I described pressure as consisting of a static at the nozzle is the total pressure: the static
comprehensively determined.
and a dynamic element. This is the device that is pressure
normally used to measure the static and dynamic An averaging Pitot tube device is also used in
pressure of a flowing gas, from which the velocity plus the dynamic pressure, P +ρv2
some cases. This has several nozzles spread over
is then calculated. 2
a larger area of the flow stream, thereby providing
increased accuracy. 34
Petroleum Open Learning

We will now look at differential pressure flow


measurement devices; their principle being to
use the static pressure differential created when
the flowstream diameter is reduced.
Test Yourself 3.1

The pressure gauge attached to a pilot Nozzles


tube nozzle reads 10 m bar g, while the
static pressure gauge reads 9 m bar g, Nozzles, an example of which is shown in
and gas density is 3 kg /m3. What is Figure 17, are used in high velocity applications;
the gas velocity at the location of the especially in hostile environments where erosion
nozzle? or corrosion would damage devices such as
orifice plates. They produce lower differential
Hint: Remember to convert the pressure pressure at a given flow rate than most other
to Pa (1 m bar - 100 Pa), to get the velocity devices.
in m / s. Strictly seaking, you should also
use absolute pressure valves, but, since
the calculation will involve the difference
between two pressures, the same result
will be obtained if the gauge values are
used.

You will find the answer in Check Yourself


3.1 on page 58

35
Petroleum Open Learning

Venturi Meter Dall Tube


In the preceding section it was mentioned that energy losses due to friction occur, and will cause a The dall tube has a similar aerodynamic design
reduction in the total fluid pressure. One objective in the design of a flow measuring device should be to the venturi meter, but is shorter and hence less
to enable the maximum recovery of pressure energy after the fluid has left the measurement location. energy efficient. Its smaller size often makes it a
preferred option to the venturi meter. Figure 19
The venturi meter is designed to produce a smooth flow, with the minimum turbulence, into and out of shows a typical model.
the narrow diameter section where the velocity is increased. These features are evident in Figure 18.

The advantage of low energy loss is often outweighed by their high cost and space required
for installation.

Figure 19 Dall Tube


Figure 18 Venturi Meter 36
Petroleum Open Learning

Orifice Plate

The simplest, and cheapest, method of creating


a restriction in a pipeline is to insert a disc with a
hole in it, so that the fluid has to flow through the Test Yourself 3.2
hole. The concept could hardly be more simple,
but it belies the deeper understanding of fluid flow What is the main difference between
principles that the achievement of accurate and the Pitot tube and the other four ∆p flow
precise flowrate measurements requires. measurement devices mentioned?

An orifice plate is mounted between flanged ends in


a pipeline, and it is this relative ease of installation You will find the answer in
and subsequent maintenance that has made orifice Check Yourself 3.2 on page 58.
plates the most popular gas flow measurement
device in commercial applications, especially in the
case of natural gas.

It should be noted, however, that the simplicity


and low cost are at the expense of the smaller line
pressure energy losses enjoyed by devices like the
venturi meter. As we would expect, the presence
of an orifice plate will create considerably more
turbulence than an aerodynamically designed
restriction.

We will consider orifice plate design requirements


in the next section.

37
Petroleum Open Learning

Orifice Plate Principles

Configuration and Pressure Profile


Figure 20 is an orifice plate configuration which
also shows the behaviour of the fluid in terms of
streamlines, as it flows through the measurement
region.

The principle, as we have seen, is that the


presence of the orifice reduces the flowstream
diameter, and the resulting increase in velocity
causes a decrease in static pressure from which
the flowrate is calculated.

Note, however, that the maximum velocity is not


exactly at the orifice, but is at a distance equal to
approximately half the pipe diameter downstream
of it, called the vena contracta. This effect is
mainly due to the inertia of the fluid causing it to
continue converging after it passes through the
orifice. So the static pressure is slightly lower
at the vena contracta than at the orifice, which Figure 20 Orifice Plate Meter
means that the ideal location for the downstream
pressure tapping is at the vena contracta rather
than at the orifice plate.

This brings us to a description of the location of


pressure tappings. Figure 21 on the next page,
shows how the pressure varies at locations on
the flowpath.

38
Petroleum Open Learning

Obviously the best measurement resolution would be


obtained from the highest differential pressure(∆p) between
the upstream and downstream regions. In Figure 21, we
see that this is between point 1, the corner between the
upstream surface of the plate and the pipe wall, and point
2, the vena contracta.

However practical considerations must again prevail; it


is easier and cheaper to bore and fit pressure tappings
to the flanges instead of the pipe wall, so most industrial
installations are fitted with flange tappings. Corner
tappings such as would be required to measure the
pressure at point 1 are also less convenient than flange
ones.

In applications where pipe tappings are used it, is


customary to use the D and D configuration. D is the
2
pipe internal diameter, and the upstream tapping is
located at a distance D from the upstream surface of the
plate, while the downstream one is D from the
2
downstream surface : at the vena contracta.

The differential pressure, then, measured in the Figure 21


diagram is that between points 3 and 4. We also see that
the static component of the line pressure has recovered
at point 5, but not to its upstream value. This reflects the
pressure energy loss due to turbulence, which it somewhat
exaggerated in the diagram. You should also note that a
continuous slight pressure reduction is shown to represent
the frictional losses that occur in all pipelines.

39
Petroleum Open Learning

Meter Run Specifications Beta Ratio

In addition to the orifice plate assembly, the The beta ratio (ß) is defined as d, where d is the Test Yourself 3.3
design and configuration of the line in the vicinity D
of the meter is subjected to certain specifications, diameter of the orifice and D is the internal
particularly in fiscal and other contract situations. diameter of the pipe. ß is an important factor in 1. What is the name of the region where
The plate assembly and pipeline lengths upstream orifice calculations and it is recommended that it the minimum flowstream diameter
and downstream of it are referred to as the meter should always be greater than 0.2 and less than occurs in an orifice meter, and what is
run. 0.7 in natural gas applications. its approximate location?

The objective of these specifications are, as we 2. On Figure 21, between which points
would expect, to achieve as smooth and would the maximum ∆p, and hence
symmetrical a flow of fluid through the meter as best measurement resolution, be
possible. The presence of bends, valves or other obtained?
devices within a certain distance of the meter could
cause measurement inaccuracy, so specific 3. What is the most common location for
minimum lengths of straight pipe, both upstream pressure tappings?
and downstream of the plate, are stipulated.
4. Why does Figure 21 show a slight but
The minimum lengths are quoted as multiples of the continuous line pressure drop in the
pipe diameter, and depend on the types of fitting direction of flow, outside the region of
such as bends, valves, reducers and expanders the plate?
on the upstream side, and on the diameter of the
orifice relative to the pipe diameter. Tables of values
5. What is the ß ratio of a meter in which
are available in international standards publications
the orifice diameter is 130 mm, and
such as ISO 5167; these range from 5 to 80 times
the pipe internal diameter is 250 mm ?
the pipe diameter on the upstream side, and from 4
to 8 on the downstream side.

In some applications where the physical layout of You will find the answers in Check Yourself
the plant makes the minimum straight length 3.3 on page 58.
unattainable, it is possible to install straightening
vanes upstream of the meter, which help to smooth
out flow disturbances. 40
Petroleum Open Learning

Orifice Plate Flow Calculations C is the factor that compensates for the energy losses Again, you are not expected to remember this
due to turbulence and friction that were mentioned formula for ℇ, but you should note that its value
The ISO 5167 Formula earlier, and it is interesting to note that a typical value increases (approaches 1) as the line pressure
for the venturi tube is 0.98, thus reflecting its more P increases; which confirms that gases become
aerodynamically efficient design.
At the end of Section 2 we had the following less compressible as their pressure and density
equation for mass flowrate: E is called the velocity of approach factor. It has
increases. You should also be careful to avoid
confusing ℇ with Z, the non-ideal gas behavioural
2∆pþ A2a2


Qm =
(
A2 - a2 ) enabled us to eliminate the A and a pipe and orifice
cross-sectional area terms since it employs the 13ß factor which was described in an earlier section.
factor in the following formula:
We saw that this equation required to be modified Qv, the volumetric flowrate, is found by dividing
before it could be used for accurate flow calculations, the mass flowrate Qm by the gas density at the
E= 1 - ß4
and you will recognise similarities between it and reference pressure and temperature:
the following flow equation from ISO 5167: ℇ, termed the expansion factor, is important in gas
measurement since it accounts for the compressibility, Qv = Qm
Qm = CEℇ π d2 2∆p ρ
and hence density change, of gases when their pressure ρ
4
changes as they flow through the meter. Liquids, being
You will be relieved to learn that it is beyond the essentially incompressible, have an ℇ factor of 1, which We saw in Section 1 that:
scope of this book to show how this equation is means that it can be ignored. For gases, it is obtained
derived from the preceding one. It looks more from the following formula: þ = PMw
simple, but that is because the modifications have
been mainly incorporated in the first three terms, ℇ = 1 - ∆p (0.41 +0.35 ß4) ZRT
which we will now look at. PY So, if the volumetric flowrate is required to be
referred to standard conditions, P = 1.013 bar, T =
C is defined as the discharge coefficient, and is
The only term that may be unfamiliar to you is Y, 288 K and Z = 1.
a function of Reynold’s number Re and ß. It can
which is the specific heat ratio of gases. It is given by:
be calculated from a formula, or obtained from
This expression can also be used to obtain a value
tables. ISO 5167 presents discharge coefficient Y = Cp
Cv for ρ, the fluid density upstream of the orifice, in
tables for various pressure tapping locations and
the ISO equation, so non-ideal gas behaviour is
pipe internal diameters, but if high accuracy is not
where Cp is the specific heat at constant pressure, accounted for here. Clearly, the P, T, Z and Mw
critical, a value of 0.605 can be used in typical
and Cv is the specific heat at constant volume. These values would be the prevailing ones upstream of
applications.
values vary between gases, and typical values of Y are the meter. It is fairly common nowadays, however,
1.4 for air and 1.3 for methane. to measure the gas density directly.
41
Petroleum Open Learning

At the end of Section 2, I asked you to take Figure 22 demonstrates the problem associated
particular note of the relationship between the with this square root relationship. Were the
flowrate Qm and ∆p, in that Qm is proportional relationship linear, the scale on the left would
to the square root of ∆p. This relationship, of apply and the resolution to which it is read is
course is maintained in the ISO 5167 equation, constant over the complete range. With the
and has an important implication with regard to square root scale, however, we see that high
the resolution to which flow measurement can be resolution is available on the upper region, but
made. it deteriorates lower down and is extremely
poor near the bottom. For this reason a
range switching facility is recommended, and
stipulated in fiscal systems, so that low flowrate
measurements can be measured to greater
accuracy.

Orifice Plate Size Selection


Reliable flow measurement requires an adequate
∆p value. If the flowrate is reduced, the differential
pressure across the meter will also be reduced,
and, if it falls below a certain value, the reliability
of the flowrates calculated from it will decrease.
If the flowrate is anticipated to remain near this
value for a considerable time, the plate should be
replaced with one with a smaller diameter, which
will increase the differential pressure.

Conversely, excessively high flow rates will lead


to unreliability, as well as creating undue line
pressure loss due to excessive friction. In such
cases, a larger diameter plate would be fitted.

42
Petroleum Open Learning

EXAMPLE Automated Calculation Methods


An orifice plate meter, with an internal pipe diameter Having seen the large number of data necessary to calculate flowrates, you will appreciate
of 200 mm, has been operating satisfactorily with that longhand methods of performing the calculations are rather tedious and time consuming.
a ß ratio of 0.65. The gas flow through the line is Fortunately, recent advances in electronic technology have alleviated this problem.
to be reduced to a rate which will not produce an Differential pressure, line pressure, temperature and density measurement signals can be digitised
adequate ∆p signal. Calculations predict that ∆p and processed electronically to yield flowrate readouts directly.
would be satisfactory if ß is reduced to about 0.45.
What is the diameter of the plate which is currently We will look at the general layout of such systems in the next section.
fitted, and what is the diameter of the one that will
replace it ?
d
-
ß =D
Test Yourself 3.4
d = ß D = 0.65 x 200 = 130 mm 5. It has been decided to increase the
gas throughput in a 250 mm internal
1. Which two factors are used to obtain
To get a beta ratio of 0.45: diameter pipeline which contains an orfice
the discharge co-efficient C?
meter operating with a beta ratio of 0.44.
The increased rate would create a higher
d = 0.45 x 200 = 90 mm 2. What is the factor E called.
than necessary differential pressure
Given that E = √1-ß4, what is its value
across the plate in addition to causing
You should note that the numbers in this example for a meter in a 250 mm internal
excessive energy loss due to friction.
happen to give round figures for the plate diameter. diameter pipe, with an orfice diameter
It is predicted that increasing the beta
of 110 mm ?
Orifice plates are only available in certain sizes, so ration to about 0.65 will alleviate the
it may not always be possible to achieve a desired problem. What is the diameter of
3. Why is it ℇ, the expansion factor,
ß value exactly; hence my use of about 0.45. the plate to be removed and of the plate
important when measuring gas
which will replace it, given that plates are
flowrates ?
only available with diameters in multiples
of 10 mm?
4. Why is it desirable to have a range
switching facility on instruments that
read gas flow measurements You will find the answers in Check Yourself 3.4
obtained from orifice meters ? on Page 59.

43
Petroleum Open Learning

The following exercise might look a bit formidable,


but is actually not too difficult since you are given
most of the formulae and the numbers you will Test Yourself 3.5
need, so it is really a question of inserting them
as appropriate. Given the following: a. Evaluate E and p, then calculate Qm, the
mass flowrate of the gas in kg/s.
Qm = CEℇ π d2 2∆p ρ
4 b. Find Qv, the volumetric flowrate in
standard (1.013 bar, 15°C) cubic metres
per hour.
E= 1 - ß4

þ = PMw
ZRT

d = 125mm
D = 257.4 mm
∆p = 100mbar (10000 Pa)
Line pressure P = 24 bar g
Line temperature T = 59°C
Mw = 22.7 kg / kg-mol You will find the answers in Check Yourself 3.5
Z = 0.937 on page 60.
C = 0.605
ℇ = 0.9987

44
Petroleum Open Learning

Summary of Section 3
The main features of the following devices were described:

• Pitot tube
• flow nozzles
• venturi
• dall tube
• orifice plate

The Pitot tube is the only velocity head device, the others being dependent on differential
pressure measurement.

The orifice plate was identified as the most popluar gas flow measurement device, and the rest
of the section was devoted to describing its effects on flowing fluid and how these effects could
be used to measure flowrates. The description showed that:

• The maximum fluid velocity, and hence minimum static pressure, is at the vena contracta.

• Flange-mounted pressure tappings are the most popular type, although they do not tap
into the regions of the meter at which the maximum differential pressure occurs.

• The ISO 5167 formula is used for orifice and contains terms which correct for fluid frictional
energy losses and gas compressibility.

45
Gas Flow Measurement Petroleum Open Learning

Section 4 - Orifice Plate Metering Equipment


Types of Plate In addition to variations to the location of the orifice
on the plate, there are different designs of orifice
edge, two of which are shown in Figure 24.
Plate Geometry Variations
The most common type is the square edge with
Orifice plates are manufactured with various bevel, and you should note that the plate must be
geometrical designs, examples of which are installed with the bevelled edge facing
shown in Figure 23, to suit different applications. downstream. The only exception to this rule is a
plate designed for special applications, which is
known as a conical entrance plate.

The square edge profile is used in applications


where the facility to measure flow in either
direction is required. Clearly, its shape will make
it slightly less aerodynamically efficient than the
bevelled edge plate.

As you would expect, the location of the orifices


and the design of their edges will affect the flow
characteristics; so the appropriate factors, mainly
C and ℇ, have to be adjusted accordingly.
Concentric is the most common type, because
the central location of the orifice allows a It is very important that plates which been
symmetrical flow pattern. damaged are replaced, even if the damage is only
slight. Entrained high velocity solid, or even liquid,
The segmental type is mainly for slurries, and particles will wear the sharpness off the leading
the eccentric plate is for gas streams which may edge; the permissible wear is less than that
contain liquid. The hole, being at the bottom of which is visible to the naked eye. Other damage
the plate, allows liquid to flow through instead which can occur is buckling due to excessive
of being trapped, as it would be by a concentric temperatures, or high pressure surges.
plate. A variation on this is a concentric plate with
a small hole near the bottom.

46
Petroleum Open Learning

Senior Orifice Plate Fitting Figure 25 is a sketch of the basic design which comprises an orifice plate
carrier which is moved by a handle operated rack and pinion arrangement,
We have seen that orifice plates are normally fitted between flanges, so a and a chamber situated above the plate.
plate changout demands the inconvenience of depressurising and purging the
meter run. To alleviate this, the Senior fitting assembly was produced. The following description is intended as an explanation of the operating
principle of the system, and it is important that you do not regard it as an
operating procedure. If your work involves the operation of Senior fittings,
then you must follow the procedures pertinent to your installation.

This explanation applies to a unit fitted to a hydrocarbon gas system.

1. The chamber is purged with inert gas and pressurised to the pipeline
pressure.

2. The carrier and plate are retracted into the chamber, which is then sealed
from the process stream, depressured and purged.

3. The plate is removed from the carrier and chamber, and the replacement
plate is inserted.

4. The chamber is sealed and purged with inert gas and pressurised to the
pipeline pressure.

5. The plate and carrier assembly is lowered into the meter run.

6. The chamber is safely vented and purged.

In theory, this method can be followed without the gas flow being stopped. In
practice, however, many plant operations managers doubt that it is as safe as
a proper line isolation, depressurisation and purge procedure. Nonetheless,
even if the meter section is isolated, depressurised and purged, the Senior
device saves plate changeout time by eliminating the work and problems that
can be associated with flanges.
47
Petroleum Open Learning

Sensing Devices The main feature is a cell containing a


liquid-filled twin-walled diaphragm
Line pressure and temperature sensing devices capsule, which is subjected to the
are widely used throughout industry, and the upstream pressure on one side and the
types used in gas flow metering systems are the downstream pressure on the other.
same as those for other applications. We will
therefore confine our attention to the devices that The position of the centre of the capsule
measure the orifice plate differential pressure and depends on the pressure difference
the gas density. between its sides; an increase in flowrate,
for example, will reduce the pressure on
Electronic measuring and signal processing the downstream side thus causing the
devices are becoming increasingly predominant. diaphragm centre to be displaced to the
Advancement in the design of intrinsically right.
safe circuitry has allowed them to replace the
pneumatic instrumentation that was traditionally A force bar, which is linked to the capsule,
the only viable means of measurement and transfers the movement to a flapper/
control in hazardous environments. However, this nozzle arrangement in the signal
conversion process is by no means universal, and conditioning system, which is shown
pneumatic systems are still in use, particularly on simply as a block diagram. The movement
older plant installations. is converted to a pneumatic output signal
that can be measured directly or, more
Differential Pressure commonly nowadays, converted to an
electrical signal first.
I will describe two differential pressure sensors,
one pneumatic and one electrical. The electrical device is the capacitive
differential pressure transmitter, a sketch
The pneumatic one is the torque balance of which is shown in Figure 27, on the
transmitter, and a simplified sketch is shown in next page.
Figure 26.

48
Petroleum Open Learning

Gas Density
A device which is often used to measure gas
density directly uses the vibrating cylinder
principle. The process gas is passed over the
inner and outer surfaces of a thin metal cylinder,
which, like all solid objects, has a natural or
resonant vibration frequency. However, the vibration
frequency of the cylinder is affected by the gas
molecules which interact with its surface and vibrate
with it. The significant property, here is the mass of
gas, the relationship being that increasing the mass
will decrease the vibration frequency. This means
that, if the frequency is measured, it can be used to
evaluate the gas density.

Figure 28, on the next page, is a schematic


diagram of the system. The cylinder requires an
activating signal to make it vibrate; this is achieved
by passing current through the cylinder activating
coil. The vibration frequency is picked up as an
alternating current by the sensing coil, and this
signal is amplified and processed to produce a
square wave output. This type of signal is ideal for
transmission and can be readily converted to an
analogue or digital form.
The upstream pressure is transmitted to the isolating diaphragm on one side of the cell, and the The frequency range of a typical sensing system is:
downstream pressure to the other side. Movement of the isolating diaphragms is transferred to the 4900 Hz at a gas density of 0 kg / m3 (no gas
sensing diaphragm by silicone oil, and any change in the position of the sensing diaphragm changes present) to 3900 Hz at 60 kg / m3. Units are also
the electrical capacitance between it and the capacitor plates. The resultant electrical signal is available which are capable of measuring densities
converted to a current reading in the 4 - 20 mA DC range, which can then be processed to produce the up to 400 kg / m3.
required analogue or digital output.
49
Petroleum Open Learning

50
Petroleum Open Learning

Metering Stations

Typical Multi-Stream System


Metering stations to which fiscal standards apply,
and others in applications where maintenance
of production rates is essential, comprise two
or more meter runs; thus allowing fluid flow to
continue while a meter run is shut down for
maintenance.

Figure 29 is a schematic diagram showing the


main components of a two stream metering
station with a computer based electronic signal
processing system.

We see that each stream has its own computer,


and they are connected to the main or station
computer. The operator interface with the system
is the input / output terminal, which displays live or
recorded data and through which entries such as
changes to meter factors are made.

The temperature, pressure, density and differential


pressure transducers transmit their data to their
stream computer where the signals are processed
and the computations are done and sent to the
station computer.

51
Petroleum Open Learning

An additional feature shown in Figure 29 is a facility 1. The gas will be measured in either volume 5. Totalisers on individual meter runs and on
to measure the gas specific gravity, upstream of or mass units, depending on the agreement station summators will have sufficient digits
the meter runs. This is an option that is sometimes between the interested parties. Volumes will to prevent cycling occurring more frequently
used so that a more accurate measurement of the be measured in cubic metres and mass in than once every two months.
density is obtained. The sensing element in the tonnes. Volume measurement will be
specific gravity transducer is usually a vibrating referred to the metric standard conditions of Safety Implications
cylinder, as in the case of the density sensor, but 15°Celsius and 1.01325 bar.
with the facility to incorporate a reference gas. Most of the safety precautions and procedures that
2. While in most cases gas density will be apply to other items of process equipment handling
In addition to mass and volumetric flowrates, measured directly using a density combustible gases are pertinent to metering
measurements of line pressure, temperature and transducer, in some instances it may be equipment, and includes:
density can also be displayed. A fiscal requirement calculated, by an agreed method, from a
is that flow totalisers are used so that the total knowledge of the composition of the gas • isolation, depressurisation, purging and
mass or volume of gas for a period of time can be together with the measured operating pressurisation
measured. pressure and temperature.
• checking for leaking fittings
This description has only outlined the basic 3. Enough meter runs will be provided to
facilities of an automated metering station. Many ensure that at least one standby meter will • monitoring for hazardous gases
installations have additional features such as the be available, at the design production rate.
ability to transmit data to central monitoring sites. Isolation valves will be provided so that You should also be aware that many metering
individual meters can be removed from systems are installed in high pressure applications.
The following is an example of some of the main service without shutting down the metering
metering station design requirements that would system. I will repeat the statement I made when describing
likely be agreed between the relevant parties, the Senior orifice plate assembly, which is that you
which would include the UK Department of Energy, 4. All computing functions will be done by a must always follow the safety precautions and
in a fiscally controlled contract. You should note digital microprocessor based flow computer, procedures applying to the installation you are
that this does not include specifications applicable one of which will be allocated to each meter working on. Nothing in this book is a substitute for
to hardware such as the orifice plates, the meter run. them.
pipes and the installation of measuring elements.
Please regard this as an outline of some of the All the constants and factors which are
main clauses in a typical agreement in the UK, and held in the flow computer will be accessible
not as a specific contract. If you work on a fiscal for inspection in a general display register
metering system you should try to learn as much as and it will be possible to modify these
possible about the terms of the contract. values, with authorisation, after overriding 52
some form of security lock.
Petroleum Open Learning

Summary of Section 4

We looked at some variations of orifice pate design, and at the Senior fitting
which is intended as a means of simplifying plate changeouts. The pneumatic
torque balance is ofter used to measure differential pressure where electric
instrumentation is not employed. Electrical measurement of differential
pressure often uses capacitance changes in a sensing cell as the output
signal.

Gas density can be measured by the principle that the mass of gas in contact
with a thin-walled metal cylinder affects its vibrational frequency.

The main features of two-stream gas metering station were described, along
with some of the fiscal standards it complies with.

I finished by drawing your attention to the fact that metering equipment is


subjected to specific seafety procedures along with the plant in which it is
installed; and you must apply them if you are working on that system.

53
Petroleum Open Learning

Check Yourself - Answers

Check Yourself 1.1 Check Yourself 1.3 Check Yourself 1.5


Using P1V1 = P2V2 Molecular mass of mixture Find n from n = m = 32 = 2 kg-mol
T
T Mw 16
1 2
= (0.60 x 16) + (0.40 x 30) = 21.6
From Figure 7, at 100 bar and -50°C, Z =0.50
we will assign the initial conditions to the left
side, so we need to calculate P2 : Check Yourself 1.4 PV = ZnRT
2x2 = P2 x 0.5 Start by finding the number of kg-moles of V = ZnRT
(10 + 273) (25 + 273) methane using PV = nRT P
so: 2 x 1 = n x 0.0831 x 293 V = 0.50 x 2 x 0.0831 x 223
P2 = 2 x 2 x 298
100
283 x 0.5 n= 2 = 0.08214 kg-mol
24.348 V = 0.185 m3
P2 = 8.42 bar a
n= m If the compressibility factor had not been used,
Check Yourself 1.2 Mw
V = (2 x 0.0831 x 223) = 0.370 m3
0.08214 = m 100
Molecular mass of ethane 16
which is twice the correct value;
= (2 x 12) + (6 x 1) = 30 m = 0.08214 x 16 = 1.31 kg ie. an error of 100% !

54
Petroleum Open Learning

For the heavier gas:


Q = 0.51 x 2.174 x 0.0831 x 283
130
Check Yourself 1.6 3
Q = 0.201 m / minute

To replace these actual flow rates at line conditions to standard ones,


For both gases we will use PV = ZnRT, but the first point to note is that, we use:
since we were given a mass flow rate, we will calculate a molar P1V1 = P2V2
T1 T2
flow rate using n = m , where m is the mass flow rate.
Mw and replace V1 and V2 with Q1 and Q2 respectively. We will assign
For the lighter gas, n = 50 = 2.632 kg-mol / minute standard conditions to the left side, so we need to find Q1 for both gas
19 types.
For the heavier gas, n = 50 = 2.174 kg-mol / minute
For the lighter gas:
23
1.013Q1 = 130 x 0.314
This means that V will be solved as a volumetric flow rate, so we will 228 283
replace it with the symbol Q. Hence : PQ = ZnRT
Q = ZnRT 0.003517Q1 = 0.1442
P Q1 = 0.1442 = 41.00 m3 (st) / minute
From Figure 8, Z for the lighter gas is 0.66, and from 0.003517
Figure 9, Z for the heavier gas is 0.51.
For the heavier gas:
So, for the lighter gas: 1.013Q1 = 130 x 0.201
Q = 0.66 x 2.632 x 0.0831 x 283 228 283
130
Q1 = 0.0923 = 26.25 m3 (st) / minute
Q = 0.314 m3 / minute 0.003517

55
Petroleum Open Learning

Check Yourself 1.7


ρ = PM
w
ZRT

1. ρ = (1.013 x 30) = 1.270 kg / m3


(0.0831 x 288)

2. ρ = (1.013 x 44) = 1.862 kg / m3


(0.0831 x 288)

3. ρ = (1.013 x 58) = 2.455 kg / m3


(0.0831 x 288)

56
Petroleum Open Learning

Check Yourself 2.1 Check Yourself 2.3


Re = ρDVavg 1. Static pressure and dynamic pressure.
µ
2. Mass flowrate

3.a. A and a : Flowstream cross-sectional area (m3)


ρ = 20 kg / m3 b. v (note that this is lower case) : Average fluid velocity (m / s)
D = 0.125 m c. ρ: Fluid density (kg / m3)
vavg = 2 m / s d. Qv : Volumetric flowrate (m3/ s)
µ = 1.2 X 10-5 e. Qm : Mass flowrate (kg / s)
f. ∆p: Differential pressure i.e. change in static pressure. the SI unit
Re = 20 x 0.125 x 2 is Pa, but mbar is often used.
1.2 X 10-5
Re = 416,667; indicating turbulent flow.
4.
Divide the mass flowrate by the density Qm
ρ ( )
Check Yourself 2.2 5.

The flow rate is proportional to the square root of the differential
pressure.

Qv = A v = ( 2 X 3 = 0.0942 m3 / s
π
4
)x (0.2)

Qm = ρ Qv = 1000 x 0.0942 = 94.2 kg / s

57
Petroleum Open Learning

Check Yourself 3.1 Check Yourself 3.2 Check Yourself 3.3


1. The vena contracta. It is located at a
The total pressure, PT = P + þ v2 The pitot measures the ∆p between the static distance of approximately half the pipe
2 and total pressures. The others measure the diameter downstream of the plate.
static ∆p across the restriction.
PT = 10 mbar 2. 1 and 2.
P = 9 mbar
3. On the meter flanges.
ρ v2 = 10 - 9 = 1 mbar = 100 Pa
2 4. Fluid friction causes pressure energy
loss.
v2 = 2 x 100 = 200
ρ 3 5. π = d = 130 = 0.52.
D 250
v = 200 = 8.16 m / s
3 It has no units since the length units cancel
each other

58
Petroleum Open Learning

Check Yourself 3.4


1. Re and ß.

2. E is called the velocity of approach factor.


ß= 110 = 0.44
250
E = 1- ß4 = 1 - 0.444 = 0.981

3. ℇ accounts for gases being compressible and thus experiencing a change in density when
flowing through the meter.

4. Because the resolution is poor at the lower end of a square root scale.

5. ß=d
D

The diameter of the plate to be removed:


d = ß D = 0.44 x 250 = 110 mm

The diameter of the plate to be installed:


d = 0.65 x 250 = 162.5 mm

So the nearest available size is 160 mm

59
Petroleum Open Learning

Check Yourself 3.5


b. Qv = Qm Note that we need the volumetric flowrate at standard
a. E= 1- ß4 ρ
conditions, so we calculate þ using P = 1.013 bar and T = 288 K, Z = 1
ß = d = 125 = 0.486
D 257.4 þ = PMw = 1.013 x 22.7 = 0.961 kg / m3
ZRT 0.0831 x 288
E= 1 - 0.4864 = 0.972
Q = 4.77 = 4.964 sm3 / s = 17869 sm3 / hr ( 17.87 ksm3 / hr)
ρ = PMw = 25 x 22.7 = 21.9 kg / m3 v
0.961
ZRT 0.937 x 0.0831 x 332

Qm = CEℇ π d2 2∆pρ
4

Qm= 0.605 x 0.972 x 0.9987 x π x 0.01563 2 x 10000 x 21.9


4

Qm = 4.77 kg / s

60
POL
Petroleum Open Learning

Injection Water Treatment


Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY
POL
Petroleum Open Learning

Injection Water Treatment


Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY
Petroleum Open Learning

Designed, Produced and Published by OPITO Ltd., Petroleum Open Learning, Minerva House, Bruntland Road, Portlethen, Aberdeen AB12 4QL

Printed by Astute Print & Design, 44-46 Brechin Road, Forfar, Angus DD8 3JX www.astute.uk.com

© OPITO 1993 (rev.2002) ISBN 1 872041 85 X

All rights reserved. No part of this publication may be reproduced, stored in a retrieval or information storage system, transmitted in any form or by any means, mechanical,
photocopying, recording or otherwise without the prior permission in writing of the publishers.
Injection Water Treatment Petroleum Open Learning

(Part of the Petroleum Processing Technology Series)

Contents Page Visual Cues


* Training Targets 4
training targets for you to
achieve by the end of the unit
* Introduction 5

test yourself questions


* Section 1 - The Reasons for Water Injection 6 to see how much you
understand
Reservoirs and Rock Properties
Reservoir Drive Mechanisms
Oil Recovery
check yourself answers to
Pressure Maintenance by Water Injection
let you see if you have been
thinking along the right lines

* Section 2 - Reasons for Treating Injection Water 15


activities for you to apply
Bacteria your new knowledge
Suspended Solids
Dissolved Gases
Dissolved Solids
Other Chemical Treatments
summaries for you to recap
on the major steps in your
progress

1
Petroleum Open Learning

2
Petroleum Open Learning

Contents (cont'd) Page Visual Cues


* Section 3 - Sea Water Treatment Equipment 19
training targets for you to
Bacteria Control achieve by the end of the unit
Sodium Hypochlorite Generation
Filtration Equipment
Basket Filters test yourself questions
Sand Filters to see how much you
Dual Media Filters understand
Cartridge Filters
Oxygen Removal
Gas Stripping
check yourself answers to
Vacuum Deaeration
let you see if you have been
thinking along the right lines

* Section 4 - A Typical Injection Water Treatment System 37


activities for you to apply
Sea Water Intake and Coarse Filters
your new knowledge
Chlorination Facilities
Sea Water Reservoir
Fine Filtration
Deaeration
Water Injection Pumps summaries for you to recap
Water Injection Wells on the major steps in your
progress
* Check Yourself - Answers 54

3
Petroleum Open Learning

Training Targets

When you have completed this unit on Injection Water Treatment you will be able to :

• Define the terms porosity and permeability as applied to reservoir rock

• Explain three basic reservoir drive mechanisms

• Explain how reservoir pressure is maintained by water injection

• Explain the reasons for treating injection water

• Describe the function, construction and operation of sea treatment equipment,


including filters, sodium hypochlorite generators, and deaerators

• Describe a typical injection water treatment

• List and describe common chemicals used in the treatment of injection water

Tick the box when you have met each target.

4
Oil and Gas
Injection Water
Separation
Treatment
Systems Petroleum Open Learning

Introduction
The production efficiency of most oil reservoirs can be improved by injecting water into the layer of water which
underlies the oil in the reservoir rock. But why should this be so ? Where do we get the water from and what
treatment does the water require before it is injected ? In this unit we are going to look at these questions and
try to answer them.

The unit is divided into four sections.

In Section 1, we will look at the reasons for injecting water into a reservoir in oilfield operations. We will concentrate
on the nature of a typical reservoir and the properties which influence the way in which oil is produced. You will
see that the most common type of water used for injection purposes is sea water.

Section 2 will focus on the problems which could result from injecting untreated sea water into a reservoir rock,
and the methods used to overcome these problems.

In Section 3, I will explain the construction and operation of equipment used in the treatment of injection water.

Finally, in Section 4, we will look at a typical sea water injection system. I will take you through this hypothetical
system step by step. This will give you an overall picture of the operation and control of such a process.

5
Injection Water Treatment Petroleum Open Learning

Section 1 - The Reasons for Water Injection


Before we can go on to consider the equipment This type of reservoir is known as a dome
used and procedures involved in water injection, structure. It looks rather like a number of
it is necessary to look at the reasons for injecting upturned saucers lying on top of each other.
water. This will involve a little bit of reservoir Each saucer represents a layer of rock.
engineering. Don't let the term put you off. We
will tackle it in a rather simple way. The rock layer which contains oil, water and gas
is the reservoir rock. On top of the reservoir
lies a layer of rock called the cap rock. The cap
rock forms the upper boundary of the reservoir
Let’s start by having a look at a typical petroleum and stops any fluids from migrating upwards.
reservoir and the way in which the oil is The whole structure is often called a reservoir
produced: trap.
Reservoirs and Rock Properties In order for reservoir fluids to be brought to the
surface, wells are drilled into the reservoir rock.
Look at Figure 1, which shows a
However, in order for these fluids to enter the
simplified cross section through a
wells, they must be capable of flowing through
typical reservoir structure.
the rock.
A reservoir rock, therefore, must be capable of
holding fluids within itself and allowing the fluids
to move through it. The two properties of rock
which govern this are:
• porosity
• permeability
Porosity is the property of the rock which
enables it to hold fluids within itself. The oil, gas
and water are contained in tiny holes in the rock
called pores.
Figure 1

6
Petroleum Open Learning

Sandstone is a common reservoir rock. It is made Figure 2 is an illustration of a porous and permeable Before we move on, complete the following Test
up of grains of sand which are cemented together rock. Yourself question.
at the points where they touch. Between the sand
grains are void spaces - the pores.

The ratio of the volume of the pores to total rock


volume expressed as a percentage is the rock

Test Yourself 1
porosity.

This means that, if you have a sandstone reservoir


with a porosity of 25% , for every 4 m3 of reservoir a) What is the porosity of a rock whose
rock, 1 m3 consists of holes and 3 m3 solid sand pore spaces occupy one fifth of its total
grains. volume?

Another common reservoir rock is limestone. This b) What is the likely permeability of a cap
is a rather brittle rock which contains lots of tiny rock?
cracks and fissures. These tiny cracks give the
limestone its porosity. c) Which of the following materials is likely
to be porous and permeable?
Permeability is a measure of the ability of a fluid The following example may help you to visualise
to flow through the rock from one pore to another. porosity and permeability in a reservoir. i) sponge
In order for it to be able to do this, the pores must ii) glass
be interconnected. We have all seen an ordinary building brick, but iii) sandstone
have you ever examined one carefully ? Take a iv) slate
Permeability is measured in d’arcys, named after close look at a brick under a magnifying glass and
a French engineer who studied the flow of liquids you will see that the surface is a mass of tiny holes.
through filters. He found that the flow increased in A building brick is both porous and permeable.
proportion to the pressure increase. However he If you place one in a bowl of water you will see
also discovered that the flow was affected by the small bubbles rising from the surface. This occurs
thickness, or viscosity, of the fluid. as water flows into brick and displaces the air. In You will find the answers to Test Yourself
many ways an oil reservior rock is similar, in terms 1 on page 54.
Generally there is a wide spread of permeability of porosity and permeability, to a building brick.
values in reservoir rocks.
7
Petroleum Open Learning

When wells are drilled into a reservoir, the fact that


the reservoir is porous and permeable enables the Gas / oil ratio - This is a term which relates
reservoir fluids to flow towards the wellbores. But the volume of gas produced from a well to the
what causes this fluid flow ? volume of oil production. It is usually abbreviated
to G.O.R. and has often been expressed in units
The presence or absence of pressure within the of cubic feet of gas per barrel of oil (cu.ft. / bbl).
reservoir has a great deal to do with its ability to It is now more common to use the SI Units
produce fluids. It was recognised in the early days of of cubic metres of gas per cubic metre of oil
the petroleum industry that, when pressure was high, (m3/m3).
oil was easily produced. When pressure dropped,
however, assistance was needed to help the wells Oil in place - is an estimate of the total amount
produce. of oil in the reservoir.

The fact that pressure is required to drive oil to the Recoverable oil - is an estimate of the amount of
wellbore is important but it doesn’t fully explain how oil oil which may be produced from the reservoir and
is produced. A complex set of circumstances causes recovered for sale.
this flow through the pores of the reservoir rock.
Recovered oil - means the actual amount of oil
Most of the oil is driven to the wells by one of three which has been removed from the reservoir.
basic mechanisms which utilise existing energy forces
within the reservoir. They are referred to as drive
mechanisms. We will look at these mechanisms
shortly. Before we do this, however, have a look at the
list of terms highlighted opposite. They are common
terms used when discussing oil reservoirs. When you are familiar with the terms above we
can move on to the drive mechanisms.

8
Petroleum Open Learning

Reservoir Drive Mechanisms


The three basic mechanisms are:

• dissolved gas drive

• gas cap drive

• water drive

They are considered to be natural drive mechanisms.


No outside assistance is introduced.

Let’s consider each of them in turn.

Dissolved Gas Drive


Figure 3 is an illustration of a reservoir with a
dissolved gas drive.

In this type of reservoir, the energy to drive the oil to Characteristics Trends
the wellbores comes from gas dissolved within the oil.
This gas is liberated from the liquid as bubbles when Reservoir Pressure Declines rapidly and continuously
the pressure in the reservoir declines. Being highly
expansive, the bubbles of gas provide the energy G.O.R. Low initially then rises rapidly to a maximum and finally falls
to push the oil towards the wells as the pressure
continues to decline.
Water production Very little
Reservoirs behave characteristically during their
producing lives. The trends of these characteristics for Recoverable oil 5 to 30 percent of original oil in place
a dissolved gas drive reservoir have been included in
the table opposite. Table 1 - Dissolved gas drive reservoirs

9
Petroleum Open Learning

Gas Cap Drive


Let's move on to gas cap drives. Figure 4 is an
illustration of such a reservoir.

In many situations, oil has accumulated where there


are considerable volumes of gas present. In such
situations, all of the gas may not dissolve in the oil
at the temperature and pressure of the reservoir. The
undissolved gas will then migrate to the top of the
reservoir, forming a layer of free gas above the oil.
This layer of gas is called a gas cap.

The gas cap is usually in a compressed state. As such


it becomes a source of energy to drive the oil to the
well bore and lift it to the surface.

The characteristic trends of a gas drive reservoir are


given in the table opposite. Characteristics Trends

Reservoir Pressure Falls slowly and continuously

G.O.R. Rises continuously in wells higher in the reservoir

Water production Low

Recoverable oil 20 to 40 percent of original oil in place

Table 2 - Gas cap drive reservoirs

10
Petroleum Open Learning

Water Drive
Finally let's look at water drive mechanisms, as
illustrated in Figure 5.

The greatest natural source of energy available to


produce oil is the water which underlines the oil in
certain reservoirs. In these reservoirs, the porous
rock occupies tremendous volumes. However, the oil
bearing part of the rock will be only a relatively small
proportion of the total. The water occupies a very large
volume compared with the oil.

We generally consider water to be incompressible. It


can be compressed to a very small extent, however,
and when such tremendous volumes are involved, this
compressed water can have quite an influence on the Characteristics Trends
behaviour of the oil reservoir.
Reservoir Pressure Remains high
As oil is produced, the pressure in the reservoir
tends to decline. This decline is counteracted by the
expansion of the water, as it moves to replace the G.O.R. Remains low
produced oil.
Water production Increases gradually to significant amounts
These types of reservoir are usually the most efficient.
Once again I have included a table which shows the Recoverable oil 35 to 65 percent of original oil in place
characteristics of water drive reservoirs.
Table 3 - Water drive reservoirs
Of course, in the natural world, reservoirs are seldom found which fit exactly into one of the classifications
we have just been looking at. Combinations must occur which further complicate the picture.

Before moving on, have a go at the Test Yourself question on the following page.
11
Petroleum Open Learning

Oil Recovery
Look again at the expected oil recovery from the three types of drive mechanism.
Test Yourself 2
• dissolved gas drive expected recovery — 5 to 30 % of original oil in place
Identify the type of drive mechanism present of the
following reservoirs : • gas cap drive expected recovery — 20 to 40 % of original oil in place

1. There is very little water production and we • water drive expected recovery — 35 to 65 % of original oil in place
expect to recover up to 35 percent of the
original oil in place.
Even the most efficient natural drive is likely to leave
2. The G.O.R. was 0.5 m3 of gas per m3 of oil more than 35% of original oil in place in the
when oil production began. It rose to 3 m3 per reservoir ! It would make sense to try to improve the
m3 after 10 years and is now 1 m3 per m3. recovery of oil in these circumstances.

3. The reservoir pressure has declined So, how could we get more oil from a reservoir than is
continuously, there is little water production possible when relying on natural drive
and we expect to recover only 18 percent of mechanisms ?
the original oil in place.
We have seen that, with high reservoir pressure,
4. The expected oil recovery is only 55 percent of oil production is improved. If the reservoir pressure
the original oil in place. could be maintained we could expect an increase in
recoverable oil.
5. Water production is low and the G.O.R. has
been rising especially in wells which are This could be done by one of two ways.
completed higher in the reservoir.
• inject gas into a gas cap
You will find the answers to Test Yourself 2 on page 54.
• inject water into the water layer

12
Petroleum Open Learning

We have also seen that the most efficient natural drive


mechanism is the water drive. It would seem reasonable
to assume, therefore, that pressure maintenance by
water injection would be beneficial.

In days gone by, water was injected later on in the


life of a reservoir after maximum recovery had been
obtained by natural drive mechanisms. It was referred
to as secondary recovery. Nowadays water injection
is planned to start as soon as the field comes into
production and is an integral part of many production
facilities.

Pressure Maintenance By
Water Injection
Figure 6 is an illustration of an artificial water drive
being used on an oil reservoir.

As the oil is removed from the reservoir via the oil


production wells, water is injected into the water layer
beneath the oil. The water layer is often called the
aquifer. The injection of water into the aquifer helps to
maintain the reservoir pressure, many of the problems
which decrease the efficiency of the reservoir are
avoided.

It is usually necessary to inject more water than the


volume of oil being removed - in some cases as much Some reservoirs produce oil at over 16,000 m3 per day and water may be injected at a rate of over
as 2 m3 injected per m3 produced. 30,000 m3 per day. Imagine. This amount of water being forced through the reservoir rock, every day for
the life of the reservoir.

13
Petroleum Open Learning

Summary of Section 1
In this section we have looked at the nature of oil and You saw that relying on these drives can be inefficient
gas reservoirs. We started by considering the shape of in terms of oil recovery. Even the best drive is likely to
a basic dome type reservoir. From there we saw that the leave behind more than 35% of the original oil in place.
two rock properties which influence the way in which oil It would make sense to try to improve the ultimate
and gas are produced are: recovery. One way of improving this recovery would be
to maintain the pressure in the reservoir for as long as
porosity possible. This could be done by injecting gas into the
gas cap or water into the water layer.
permeability

I pointed out that the most efficient drive mechanism is


We then considered reservoir drive mechanisms. Here water drive. Here the reservoir pressure tends to remain
you saw there are three basic mechanisms which utilise high and oil recovery is maximised.
natural forces to drive fluids through a reservoir towards
a well. These are:
It would appear, then, that to simulate a natural water
• dissolved gas drive drive by injecting water into the aquifer could prove
beneficial.
• gas cap drive

• water drive
In the next section we will see where this water comes
from and why it is necessary to treat it before injection.

14
Injection Water Treatment Petroleum Open Learning

Section 2 - Reasons for Treating Injection Water


In Section 1 you saw that the injection of water and what it is. It is basically everyday water which Aerobic bacteria require oxygen in order to
into an aquifer could increase the recoverable contains the following survive. They are responsible for producing the
oil from a reservoir. It may be necessary to inject green slimes which we see on sea walls, weirs
considerable amounts of water. I gave an example • bacteria
and other highly oxygenated areas of water. This
of over 30,000 m3 of water being injected daily. • suspended solids type of bacteria does not cause serious problems
The actual amounts required will depend on the in itself. However, it consumes oxygen and creates
size of the field and the extent of the natural water • dissolved gases anaerobic conditions which can lead to the growth
drive. But in any case we will still be looking at a of anaerobic bacteria.
• dissolved solids
very large amount being injected continuously.
Each of the above substances poses its own Anaerobic bacteria thrive in conditions where
Where do we get all this water ? In some land there is no oxygen. They are responsible for
particular problem in water injection systems. Let’s
locations water supply can be a problem. Surface producing the black slimes which we see in
consider each in turn and think of ways in which
supplies such as lakes or rivers can be used if sewers and other dark and dank places where the
the problem may be overcome.
available. If not, water wells may have to be drilled oxygen content may be low or zero.
or produced water used.
Bacteria A special type of anaerobic bacteria is called
Offshore, there is no problem of supply. Oceans sulphate reducing bacteria. These bacteria
Sea water contains tiny micro-organisms called
full of sea water are available. In fact, sea reduce the sulphates present in the water. (The
bacteria. They are very simple life-forms which are
water may be pumped to land locations for term ‘reduce’ means to remove oxygen or add
split into two categories. They are
injection purposes. As this is the most common hydrogen.) In the case of sulphate reduction, the
type of water injected into oil reservoirs, • aerobic bacteria result is the production of hydrogen sulphide. This
we will concentrate on sea water treatment gas, which has the formula H2S, is very corrosive
throughout the rest of this unit. • anaerobic bacteria and extremely toxic. It is essential to try to limit the
production of H2S in oilfield operations.
If we were to take raw sea water, however, and
inject it directly into the reservoir, we would Certain bacteria are slime forming as you saw
very quickly find ourselves in trouble. Sea water earlier. These slimes can cause fouling in pipework
requires a great deal of treatment before we can and, if injected, cause plugging in the reservoir.
inject it and this is what we are going to look
at in this section. We will look at the problems In view of this, I think you will agree that it is
associated with the injection of sea water and the necessary to maintain relatively low numbers of
reasons for treating it. bacteria. In particular, it is necessary to control the
growth of sulphate reducing bacteria.
So let’s first look at the make-up of sea water,
15
Petroleum Open Learning

In addition to bacteria, other marine organisms The combination of sea water and oxygen is,
can enter the system. Hard shell creatures such however, very corrosive. It will cause corrosion
as barnacles and mussels enter as embryo. These problems in the pipes and flowlines used to inject
can accumulate and grow on pipes, restricting flow.
Dead shell debris can be carried to the reservoir Test Yourself 3 water into the reservoir. Corroded equipment
will require expensive replacement or repair. In
once again to plug the reservoir pores. addition, the products of corrosion, i.e. rust, may
enter the reservoir as small particles and block the
The only way of preventing the problems associated Explain what the consequences would be pores in the reservoir rock.
with the growth of bacteria and other organisms of injecting water containing sand or clay
is to kill them with some form of chemical. The particles into a reservoir rock. It is necessary then to remove the oxygen from
chemical which is used to do this is called a the sea water. This can be done by mechanically
biocide. deaerating the water, injecting oxygen
scavenging chemicals, or both.
You will find the answer to Test Yourself
Suspended Solids 3 on Page 54.
If we look closely at sea water we can see with the
naked eye a whole variety of suspended solids.
In areas of clear water we may see small fishes,
shrimps, jelly fish and other life forms. In areas To prevent the problem which you saw in the
of muddy water we can see very fine particles of
sand and clay. All of these items are classed as
answer to Test Yourself 3, the suspended solids
must be removed. This is done by filtering the sea Test Yourself 4
suspended solids. water to ensure that all but the very tiniest particles
of suspended solids are taken out.
We have said that it is necessary to
Bearing in mind what we have already said about
remove oxygen from injection water. This
the properties of reservoir rocks I’m sure you can Dissolved Gases
in itself could create further problems.
visualise what would happen if we tried to inject
What problems are these?
water containing suspended solids into the rock. Sea water contains dissolved gases, the main
one of which is air. The air is made up mainly of
oxygen and nitrogen. It is the oxygen from the air
You will find the answer to Test Yourself
dissolved in the sea water which provides fish and
4 on Page 55.
other living organisms with the means to live.

16
Petroleum Open Learning

Dissolved Solids Other Chemical Treatment You will see where each of these chemicals is
injected into the water when we look at a typical
A large number of chemical compounds are In addition to the biocide and scale inhibitor system in Section 4.
dissolved in sea water. We only have to taste sea which may be added to the sea water, it may be
water to decide that one of the dissolved solids necessary to inject other chemicals. We will look at
is common salt. It is this salt, sodium chloride, some of these in more detail in a later section but
which gives sea water its distinctive taste. let me just mention one or two at this point.

Elements such as potassium, calcium and Oxygen scavengers — I mentioned earlier


magnesium are also present as compounds which that the removal of oxygen can be done by
are dissolved in the water. These compounds are mechanically deaerating the water or by injecting
called dissolved solids. oxygen scavenging chemicals. These scavengers
are chemicals which will react with oxygen to
If we allow sea water to evaporate, the dissolved remove it. In theory, all the oxygen could be
solids will come out of solution as small crystals. removed in this way but it would be a very
As the crystals leave the liquid, they are said to expensive operation. Mechanical deaeration by
have been precipitated from the water. itself usually leaves a small amount of dissolved
oxygen in the water and the oxygen scavenger is
In a sea water injection system the precipitated used to remove this residual amount.
crystals will form a deposit known as scale. This
scale could block pipework and, if it formed in the Corrosion inhibitors — Chemicals which will
reservoir, could block the pores in the reservoir retard the effects of various types of corrosion may
rock. be injected into the water.

To prevent the formation of scale, a chemical Defoamers — The operation of mechanical


known as a scale inhibitor is injected into the deaeration equipment may be improved by injecting
sea water. chemicals to prevent foaming of the water.

Polyelectrolytes — These chemicals are


coagulants. They cause fine particles in the water
to stick together, forming larger particles. These
are then more easily filtered from the water.

17
Petroleum Open Learning

In the next section we will be looking at


the equipment used to treat sea water.

Summary of Section 2 Before you move on, have a go at the


following Test Yourself question.

In this short section we have looked at In order to overcome these problems you
the make up of sea water which renders it
unsuitable for injection in its raw state.
saw that:
Test Yourself 5
You saw that: • Bacteria and other marine organisms
are controlled by dosing the water
• Bacteria can cause slimes to form with a biocide Which of the problems associated
which can foul the system with the use of sea water as injection
• Suspended solids are removed by water is tackled by each of the following
• Bacteria can cause the formation of filtering the water treatments?
hydrogen sulphide, a corrosive and
toxic gas • Dissolved solids are treated by 1. Passing the water through filters.
injecting a scale inhibiting chemical
2. Injecting oxygen scavengers.
• Suspended solids can plug the pores
of the reservoir • Dissolved gases are removed by 3. Passing the water through
deaeration or oxygen scavenging mechanical deaerators.
• Dissolved solids can create scale
You also saw that other chemicals may be 4. Injecting scale inhibitors.
• Dissolved gases, particularly oxygen, injected into the water to prevent foaming,
5. Injecting polyelectrolytes.
can cause corrosion inhibit corrosion, and assist in filtration.
6. Injecting biocides.
You will find the answer to Test Yourself
5 on Page 55.

18
Injection Water Treatment Petroleum Open Learning

Section 3 - Sea Water Treatment Equipment


We have seen that we need to do a number of Bacteria Control
things to the sea water before we can inject it into
the reservoir. They are: I stated earlier that we use a biocide to kill the
bacteria present in the sea water.
• kill off all the bacteria and other marine
organisms so that they cannot create slimes A simple domestic bleach was advertised in the
or corrosive products 1980’s as being capable of killing all known
germs dead. It sounds ideal for our purpose.
• remove the suspended solids so that the Take the time to look at the small print on a bottle
pores of the reservoir rock do not get of strong domestic bleach. You will find that it
blocked contains sodium hypochlorite. This substance is
• remove the oxygen so that the sea water a biocide — a chemical which is capable of killing
does not corrode the pipework and bacteria. Sodium hypochlorite is a very powerful
equipment biocide and is the most common one used in sea
water treatment.
• ensure that the sea water does not produce
scale The component of the bleach which does the job
is chlorine. This is a gas, however, which would
• treat the water with other chemicals to be difficult to handle. Hypochlorite is a convenient
prevent foaming, inhibit corrosion and so on way of storing chlorine in liquid form.
In this section, we are going to look at some of Occasionally the sodium hypochlorite may be
the equipment used to do these jobs. We will look purchased and delivered to the injection water
at bacteria control first, as this is usually the first treatment plant as bulk chemical. However, it can
treatment which the sea water receives. be made from sea water on-site and this is often
much more convenient.

Let us take a look now at how this is done.

19
Petroleum Open Learning

solution. It can be used as a biocide in this form.


Sodium hypochlorite generators may contain as
few as two chlorination cells in the smaller units to
over twenty cells in the case of large units.

One of the disadvantages of the chlorination cell


is the fact that hydrogen is produced. This is an
inevitable by-product of this process. Hydrogen
gas is of course very explosive, so it must be
safely removed.

This is done by routing the sea water plus the


products of electrolysis to a degassing tank. In the
tank, the hydrogen is liberated and excess air is
introduced to dilute it. The hydrogen can then be
safely vented.

We will look at where the biocide is used in the


next section. Let’s now move on to look at the
equipment used to remove the suspended solids.

Filtration Equipment
Figure 7 shows a simplified illustration of a Sea water flows through the space between the The next treatment which the sea water has to
sodium hypochlorite generator, where the process pipe and the tube. The chemical reaction takes undergo is that of removing the suspended solids.
takes place. It consists of a length of pipe with place as a high voltage direct current flows from This is done using a filtration system.
a titanium coated steel tube suspended in the the pipe walls, through the sea water, to the central
centre. Electrical connections are made to the pipe tube and back to the power source. Filtration is the process whereby solids are
and tube. The combination of outer pipe and inner removed from liquids by means of a permeable
length of tube is called a chlorination cell. As the electricity passes through the sea water barrier which will allow the passage of liquid but
electrolysis takes place. The sodium hypochlorite will strain out solid particles. The equipment used
which is generated remains in the sea water as a dilute to do this are called filters.

20
Petroleum Open Learning

The efficiency of a filter is measured by the


diameter of the largest solid particle which will
pass through it.

This is usually expressed in microns. A micron is


a measurement of length equal to one millionth of
a metre.

Most sea water treatment plants are fitted with


filter systems which are designed to remove:

• all suspended solids which have a diameter


of more than 5 microns

• 95% of all suspended solids which have a


diameter of 5 microns or less

There are a range of filter types in common use.


We will look at some of those which you may find
in water injection systems.

Before we take a look at the different types of filter,


I want to point out two features of these pieces of
equipment. If we pour the mixture across the mesh screen once If we then pass the water through the screen for
it will filter out the largest particles. The medium a third time the medium sized particles assist the
First take a look at Figure 8 which shows three and small particles will fall through the screen. screen in removing most of the tiny particles. If we
illustrations of a glass of water and a funnel kept passing the water through the screen we would
shaped mesh screen. But look what happens if we pour the medium and eventually be able to remove nearly all of them.
small particles across the screen for a second
The water contains solid particles of three different time. On the second pass through the screen the Before moving on, remember that the particles
sizes. The mesh screen will only filter out the larger particles assist the screen to remove the which have already been filtered increase the
largest size of particle. medium sized particles. efficiency of the filter in removing other particles.

21
Petroleum Open Learning

The second point to remember is that, as filtration


progresses, the filter itself will eventually become
plugged with solids. When this happens the filter
must be cleaned. As you will see, a common way of
cleaning a filter is to flush away the filtered particles
with water. This process is called backwashing.

Let’s now look at the four main types of filter used for
filtering sea water and see how they work. They are:

• basket filters
• sand filters
• dual media filters
• cartridge filters

Basket Filters
Basket filters use a sheet of fine wire mesh to
remove the solids from the sea water. They are most
often used for coarse filtration, and are usually found
at the very start of the filtration process.

The sea water is pumped through the filter and the


suspended solids are trapped on the front face of
the mesh. As we have already seen, the solids build
up on the surface of the mesh and create a more
efficient filtering medium than the mesh itself.

Take time to study the illustration in Figure 9, which


shows a basket filter in normal operation.

22
Petroleum Open Learning

In this mode, the water enters the outer casing


through the water inlet on the inside of the filter
basket. As the water flows across the basket the
solids are deposited on the inside of the mesh screen
and removed from the water. The filtered water
leaves via the filtered water outlet. Whilst filtering,
the backwash valve is in the closed position.

The pressure differential transmitter (PDT), measures


the differential pressure between the water inlet and
the filtered water outlet. As the solids build up on
the wire mesh screen of the basket the differential
pressure will rise. When this pressure reaches a pre-
set level, the PDT will activate a filter backwash
through a backwash controller.

Figure 10 shows the flow of water through a basket


filter when it is in the filter backwash mode.

23
Petroleum Open Learning

When the differential pressure switch activates a In our example:


filter backwash, the backwash valve opens and the
cleaning head motor starts to rotate the cleaning • the support plate is a stainless steel plate
head. which is drilled with holes or slots which are
smaller than the support material
With the backwash valve open, filtered water from
the outside of the filter basket flows: • the support material consists of layers of
ceramic balls, graded according to their
• back through the basket diameter
• into the cleaning head • the top layer is garnet sand
• through the backwash valve to drain
Garnet sand is a special type of sand which will not
The filtered solids are washed off the front face of the chip or flake easily.
wire mesh screen into the cleaning head. When the
cleaning head has rotated through 360˚ the whole of Once again, as suspended solids are filtered from
the filtering surface will have been cleaned. the water and collect on the sand, the differential
pressure across the filter will increase. When
The main advantage of this type of filter is that this pressure reaches a predetermined value, a
it can be backwashed whilst it is still on line and backwash sequence is activated. Water is pumped
filtering water. in the reverse direction through the filter to remove
the filtered solid patricles.
Sand Filters The major drawback of this type of filter is that only
the top few inches of the filter bed are effectively
As the name implies, this type of filter uses grains
used. This means that the rate of filtration is quite
of sand as the filtering medium.
low. Therefore many filters are required for the
large volumes of water used in water injection
Figure 11 shows a simplified cross section through
systems.
a sand filter.

The water to be filtered enters the filter at the top


and filtered water leaves from the bottom.
24
Petroleum Open Learning

Dual Media Filters The larger grains of anthracite have larger gaps Backwashing is carried out as for the sand filter.
between them and filter out large particles from the
Figure 12 shows a dual media filter which uses sea water. The smaller particles will pass through In Figure 13, I have illustrated a dual media filter
anthracite and garnet sand as the filter media. the grains of anthracite to the layer of garnet sand. with some of the ancillary valves and equipment.
The sand will then filter out the smaller particles.
This technique increases the efficiency of the filter
and reduces the frequency of backwashing.

In the illustration you can see that the layer of


anthracite lies on top of the layer of garnet sand.
The particles of anthracite are larger than the
grains of sand but, because they are lighter, they
lie on top of the sand.

By having a layer of large particles on top of a


layer of small particles we create two different
levels of filtration within the same filter.
25
Petroleum Open Learning

Study the illustration carefully and identify: Pressure Differential Transmitter • the paper bag in a vacuum cleaner. When it
is full, the cleaner bag (cartridge) is removed
• Turbidity Analyser (TA) The pressure differential transmitter (PDT) and replaced with a new one.
measures the pressure difference between the
• Pressure Differential Transmitter (PDT) inlet and outlet of the filter. As the filter becomes In our examples, the cartridge is removed and
blocked up with filtered particles, the differential replaced with a new one. This results in an
• Backwash Sequence Controller pressure will rise. ongoing cost and prevents the cartridge filter being
used more widely.
Turbidity Analyser When this differential pressure reaches a pre-set
limit, the PDT sends a signal to the backwash Because of this lack of economy, cartridge filters
The turbidity analyser (TA) is an instrument which sequence controller. are normally used where:
measures the turbidity of the water. Turbidity is a
measure of clearness. By measuring the level of The backwash sequence controller will activate • a back-up is required to the normal filtering
turbidity we have an indication of whether or not a backwash cycle, which is a series of timed method, e.g. if an upstream filter fails, the
the filter is performing properly. events designed to clean the filter. We will look at cartridge filter will remove the particles to the
the backwash cycle in a dual media filter in Section required standard
Most turbidity analysers shine a light through a 4 of this unit.
stream of water, which is picked up by a receiver. • a polishing stage of filtration is required,
The level of light transmitted is measured. The e.g. the cartridge filter is used to remove
Cartridge Filters small amounts of very tiny particles to polish
level of light received is also measured. The
difference between the two is a measure of the water which has already been filtered by
The cartridge filter is one of the easiest filters to
turbidity of the water. other means
install, operate and maintain. It is also one of the
most popular filters in everyday life.
It should be noted that turbidity is only an indication
that a filter is operating correctly. Two or three
Two types of cartridge filter found in common
large particles per m3 (large enough to clog the
usage are:
reservoir) may register the same level of turbidity
as 150 to 200 very fine particles per m3 (small
• the filter in the engine lube-oil system of your
enough not to affect the reservoir).
car. When the filter becomes blocked, the
paper cartridge is removed and replaced
with a new one.

26
Petroleum Open Learning

Cartridge filters are reliable because they:

• guarantee a specified level of filtration

• very seldom fail to filter to the required


standard

• are easy to install and monitor

Figure 14 shows a cartridge filter arrangement. We


can see that normal flow through the filter is from
top to bottom. The water to be filtered:

• passes through valve V1

• passes through the filter cartridges

• exits the filter via valve V2

Study the illustration carefully and identify the


cover plate, and the cartridges.

27
Petroleum Open Learning

The cover plate allows the operator to Oxygen Removal


gain access to the filter, to remove the old
cartridges and replace them with new ones. In Section 2 we established that sea
Because the cartridges are renewed, there Test Yourself 6 water contained dissolved oxygen and
is no backwash requirement on a cartridge we determined that, if the oxygen is not
filter. removed:
Are the following statements true or false ?
The cartridges used for water filtration are • it will cause corrosion problems in the
True False
most often a series of cotton or synthetic pipes and flowlines used to inject the
material layers wrapped around a slotted a) A non-permeable barrier is used to o o sea water into the reservoir
stainless steel tube. The type of material, remove suspended solids in a filter.
and the number of layers, will decide the • it will support bacteria which
particle size which the filter will remove. b) A seven micron particle will pass o o produces slimes, causing blocking of
Cartridges may be installed, which will filter through a ten micron filter. the small pores in the reservoir rock and
out all particles greater than, say, 5 microns in rendering the rock impermeable
diameter, regardless of the state of the water c) A common way to clean a filter is o o
entering the filter. to backwash it.
The process of removing oxygen from the
d) A basket filter uses garnet sand o o water is called deaeration. This can be
as the filter medium. accomplished by:

So much for filters for the time being. Before e) In a sand filter the flow is : through the
o o • increasing the temperature of the
moving on to the oxygen removal part of support plate, through the support water
this section, have a go at the following Test material and then through the sand.
Yourself question. • decreasing the concentration of oxygen
f) A cartridge filter has a removable o o in the gas mixture in contact with the
cover plate which allows access to water
the unit to replace filter elements.
• reducing the total pressure in the
You will find the answers to Test Yourself 6 on system
Page 55.

28
Petroleum Open Learning

You will see in Section 4 that the injection water is


used as cooling water in various process systems.
This means that the water itself is heated, thus
assisting in the oxygen removal.

Decreasing the concentration of oxygen in the gas


mixture in contact with the water, can be achieved
by a process known as gas stripping.

Let’s have a look at this now.

Gas Stripping
The process of stripping dissolved oxygen from
water is conducted in large stripping towers.
Inside the tower the water is spread out to increase
the surface area, and natural gas is passed
over the surface of the water. This reduces the
concentration of oxygen which is in the atmosphere
in contact with the water.

Figure 15 shows a simplified view of a gas stripper


tower.

29
Petroleum Open Learning

The flow of liquid and gas through the gas stripper The packed section contains packing rings. These
is counter current. As the liquid falls downwards are designed to increase the surface area of the
through the column, the stripping gas flows water. A popular type of packing ring used in
upwards. stripping columns is the pall ring which may be
made from plastic or stainless steel. One of these
In the illustration, the pre-heated sea water: rings is shown alongside the stripping tower in
Figure 15.
• enters the stripper at the top of the column
Other features of the stripper tower which you can
• passes through a water distribution pipe see in Figure 15 include:
which is often called a sparge pipe
• the distribution plate which ensures that the
• flows downwards through a packed section sea water is evenly distributed across the
packing
• accumulates at the bottom of the stripper
• the demister pad which ensures that the gas
• leaves the stripper from the bottom of the leaving the stripper is free from liquid droplets
column and mists

• the oxygen scavenger inlet which is used


The stripping gas is often hydrocarbon gas from the to introduce oxygen scavenging chemical to
oil and gas process. The flow through the tower is: the tower

• into the stripper through the stripping gas • the vortex breaker on the deaerated water
inlet pipe outlet which prevents gas from being sucked
into the pump with the water
• through a gas distribution sparge pipe

• upwards through the packed section Look again at Figure 15 and try to visualise what
is going on in the gas stripper. When you have
• out through an gas outlet located at the top done that we can have a look at the other type of
of the tower deaeration, i.e. vacuum deaeration.

30
Petroleum Open Learning

Vacuum Deaeration
You should note that the term vacuum is used to
denote any pressure below atmospheric pressure.
We also speak of a deep vacuum, which is
a vacuum well below atmospheric pressure. A
shallow vacuum on the other hand, is a vacuum
which is only slightly below atmospheric pressure.

Figure 16 is a simple illustration of a two stage


vacuum deaerator. You will probably notice that it
looks similar to the gas stripping deaerator.
However, it does have a number of different
features. The first thing we should note is that there
are two separate sections of packing. The upper
section operates under a shallow vacuum. The
lower section operates under a deeper vacuum.
This, then, is a two stage vacuum deaerator. Single
stage and three stage deaerators have been used
but two stages are the most common.

In this vessel, the sea water from the process heat


exchangers enters the deaerator through a sparge
pipe and flows downwards over the upper section
of packing.

Oxygen and water vapour are sucked out of the


vessel via an overhead vapour line, creating a
shallow vacuum in the vessel.
Figure 16
31
Petroleum Open Learning

When the water reaches the bottom of the upper Figure 17 is a simplified drawing of a vacuum pump.
section of packing, it flows through a set of seal (An actual pump would look very different from this.
chimneys. These operate in a similar manner to the However, the illustration is intended to explain the
‘U' bend under a sink. The water has to flow upwards principles of operation.)
and over a weir before it can reach the lower section
of packing. The pressure difference between the two If you look at the first part of Figure
sections is maintained by the height of the chimney 17 you will see that I have drawn a
weirs. shaft, fitted with four vanes, rotating
inside an empty casing. I have
A simplified sketch of a seal chimney is shown positioned the shaft and vanes
alongside the deaerator in Figure 16. central to the casing.
After the sea water leaves the seal chimneys it is In the second drawing of Figure 17, I
re-distributed and flows downwards over the lower have shown a water inlet and outlet
section of packing. Oxygen and water vapour are which introduces service water into
again sucked out of the vessel via a vapour line which the casing at the near end and
goes to the deep vacuum section. removes it from the far end. The
spinning action of the shaft and the
The deaerated water leaves the lower section of
vanes imparts a centrifugal force
packing, falls into the bottom of the column, and
to the water. The water is thrown
leaves via the deaerated water outlet.
against the inside wall of the casing
Vacuum deaerators are very efficient. Most two due to this force.
stage vacuum deaerators can achieve an oxygen
The water forced against the casing
concentration as low as 0.1 to 0.15 ppm (parts per
in this way is the liquid ring which
million) in the deaerated sea water outlet.
gives the compressor its name. The
We have seen how a vacuum deaerator works. Now central part of the casing is filled
let’s take at look at two items of equipment which may with air.
be used to create the vacuum in the deaerator.
The most common method of creating this vacuum is by
the use of a liquid ring compressor. These are also
called vacuum pumps, which is the term I shall use.

32
Petroleum Open Learning

Now take a look at Figure 18 and see if you can You will notice that I have drawn the shaft and
detect the difference in the position of the shaft vanes offset from the centre of the casing. Service The differing sizes of these void spaces means
and vanes. water is again introduced into the casing and again that the pressure in each will change as the shaft
forced outwards to form the liquid ring. and vanes rotate. The pressure will:

You should notice that: • decrease when the void space increases

• although the shaft and vanes are • increase when the void space decreases
offset the water still forms a uniform
liquid ring against the casing If we control the flow of air into the unit so that it
enters a void space at the low pressure side and
• the offset shaft and the uniform leaves at the high pressure side, we have created
liquid ring combine to form different a liquid ring compressor. The compressor will
sized void spaces in fact suck the air into the low pressure side and
create a vacuum.

In real vacuum pumps, slide valves control the air


inlet and outlet.

Vacuum pumps are only capable of creating a


limited amount of vacuum. Therefore, in order
to create the deeper vacuum required in the
second stage of deaeration, the vacuum pump is
augmented by an ejector. This piece of equipment
is also called an eductor or venturi.

Figure 19, on the next page, shows how such a


unit works.

33
Petroleum Open Learning

Test Yourself 7
Do the following statements apply to : gas stripping
deaerators, vacuum deaerators, both types of
deaerator or neither of them?

a) Water enters the column at the top and is


distributed through a sparge pipe.
b) A vortex breaker ensures that gas is not drawn
out of the vessel with the deaerated water.
c) The upper packing support plate accommodates
seal chimneys.
d) The concentration of the oxygen in the gas
mixture in contact with the water is reduced.
e) Pall rings may be used as packaging.
f) Renewable cartridges help to maintain the
The discharge side of the ejector is connected water quality.
to the suction of the vacuum pump. A line
g) A liquid ring compressor is connected to the
connects the nozzle of the unit to the deep
upper outlet.
vacuum section of the deaerator. You should by now have a good idea of how
oxygen is removed from the injection water. h) Water flows through a distribution plate located
As the vacuum pump draws atmospheric Check your understanding now by having a above the packing.
air through the nozzle of the ejector the air go at Test Yourself 7. You will find the answers to Test Yourself 7 on
speeds up. This creates an area of very low Page 56.
pressure at the nozzle which in turn pulls a
deep vacuum on the deaerator.
34
Petroleum Open Learning

Summary of Section 3
In this section we have been looking at the equipment and processes which are used to treat sea water.
This equipment ensures that the water is rendered suitable for injection into a reservoir.
We started by considering the equipment used You saw that injection water filters are typically The temperature is increased because the
to make sodium hypochlorite from sea water. designed to remove all suspended solids which water is used as a cooling medium in heat
You saw that the process is one of electrolysis have a diameter of more than 5 microns. In exchangers.
using a chlorination cell. addition they will remove 95% of suspended
solids which have a diameter of 5 microns or The remaining two methods use a deaeration
From there we went on to look at the different less. I pointed out that equipment is provided to tower.
types of filter which could be used to reduce the backwash the filters when they become plugged
suspended solids content of the injection water. with solids. In the first method, gas stripping reduces
The types which we looked at were: the concentration of oxygen. In the second, a
In the final part of the section we had a look at vacuum is created in the tower by a vacuum
• basket the equipment used to remove dissolved oxygen pump augmented by an ejector.
from the sea water. I said that the process of
• sand removing oxygen can be accomplished by:

• dual media • increasing the temperature of the water In the next section we will see how this equipment
is used in a typical sea water injection system.
• cartridge • decreasing the concentration of the Before you move on to that, however, try the
oxygen in the gas mixture which is in following Test Yourself question to check your
contact with the water understanding of Section 3.

• reducing the total pressure in the system

35
Petroleum Open Learning

Test Yourself 8
To which piece of injection water treatment equipment or system do the following
components belong ?

If the piece of equipment is a filter, state which type. You should indicate if the
components belong to more than one piece of equipment or system.

sparge pipe ...................................................... backwash controller..................................................

degassing tank.................................................. cleaning head motor..................................................

support material ............................................... demister pad.............................................................

vortex breaker................................................... vacuum pump............................................................

ejector................................................................

seal chimney.....................................................

cartridges...........................................................

You will find the answers to Test Yourself 8 on Page 56.

36
Injection Water Treatment Petroleum Open Learning

Section 4 - A Typical Injection WaterTreatment System


In this, the final section of our study of injection Look first at Figure 20. This is a simple block Familiarise yourself with the basic process before
water treatment, we are going to look at a typical diagram which shows the flow through our system. moving on.
system which you might find on an offshore oil From the diagram you can see that the sea water
production platform. is pumped:
• from the sea, via the sea water transfer
However, you should remember that each water
pumps, through the coarse filters, to the sea
treatment system is unique to itself. Each one
water reservoir
will differ, in types of equipment used, process
layout and so on. The sea water treatment system • from the sea water reservoir, via the sea
which I will describe is not meant to represent any water supply pumps, through the fine filters,
particular facility. It is simply a hypothetical system crude oil coolers and the vacuum deaerator
which will give you a feel for the full sea water to the surge vessel
treatment process. • from the surge vessel, via the booster
pumps and injection pumps, to the injection
wells and then into the reservoir

Figure 20
37
Petroleum Open Learning

We are going to follow the flow through the system in


more detail now. In order to make the task simpler, I
will guide you through it section by section.

Sea Water Intake and Coarse Filters


Take a look at Figure 21 which shows the sea water
transfer pumps and the coarse filter system. You
will see that I have only shown one pump and one
filter. In practice there would probably be three of
each but I have eliminated the rest for simplicity.

The sea water transfer pumps are positioned


inside caissons. These are long pipes which are
open to the sea and extend below the platform. The
pumps are selected and started manually by the
operator.

At each pump suction there is fitted a simple filter


to prevent larger marine organisms from entering
from NaOCI generator
the system. It is also at this point that the first
treatment, the introduction of sodium hypochlorite,
takes place.

The sea water leaving the sea water transfer


pumps is analysed by an instrument called an
analyzer indicator controller (AQIC). This monitors
the sea water to ensure that there is a sufficient
concentration of sodium hypochlorite (NaOCI) to kill
all of the bacteria. If the NaOCI concentration is too
low the controller opens valve AV1 further to allow
more NaOCI to flow into the suction of the pumps.

38
Petroleum Open Learning

The sea water then flows to the coarse


filters where the first stage of filtration takes
place. The coarse filters are basket filters
which can be backwashed whilst they are on
line. A pressure differential switch high
(PDSH1) monitors the differential pressure
across the filters.

If the differential pressure exceeds a pre-set


limit, for example 1 barg differential pressure,
PDSH1 sends a signal to the backwash
sequence controller. The controller starts
a backwash cycle which backwashes each
filter in turn. We will look at a backwash
sequence controller in more detail when we
come to the fine filters.

The filtered water now flows to the sea water


reservoir. You will notice however that a
portion of it is taken as a side stream to be
used as feed for the NaOCI generator. Let’s
leave the main flow for a time and see how
the NaOCI generator fits into the overall
system.

Chlorination Facilities
Figure 22 shows the main pipelines
and controls to be found on the NaOCI
generator.

39
Petroleum Open Learning

Follow the line which enters the Figure at the top The NaOCI, sea water and hydrogen mixture A dosing pump increases the pressure of the
left. You will see that the feed from the coarse leaving the chlorination cells enters a hydrogen NaOCI and sea water so that it can be injected into
filters enters the NaOCI generator via: removal tank. You will remember from Section 3 a higher pressure area.
that the hydrogen removal tank is kept constantly
• a flow switch low-low (FSLL1) purged with air from an air blower. You should note from Figure 21 that the injection
point to the main sea water line is downstream
• an emergency valve (XV1)
The hydrogen is diluted with the air which is then vented of the take-off point to the NaOCI generator.
• a pressure control valve (PCV1) from the hydrogen removal tank to the atmosphere. To If the injection point was upstream of this take
minimise the danger of an explosion, a flame arrestor off, then NaOCI would be re-circulated back to
The flow switch is required to prevent the NaOCI is fitted in the vent line to prevent the backward the generator resulting in high and dangerous
generator from being damaged due to a low flow movement of a flame should the vent ever ignite. concentrations of NaOCI.
of sea water through the system.
If the supply of air from the blowers was to fail, Another line from the hydrogen removal tank
If this switch is activated, the local control system a dangerous concentration of hydrogen would takes the NaOCI / sea water mix to the transfer
will close the emergency valve, shut off the rapidly build up in the hydrogen removal tank. To pump caissons. This is a simple gravity feed via
electrical supply and send an alarm to the operator guard against this, a pressure switch (PSLL1) is the valve AV1 to the pump suctions. Look back to
to warn him of the problem. installed in the discharge line from the air blowers. Figure 21 and identify this injection point.
The switch will sense a lack of air and activate a
The pressure control valve (PCV1) maintains a shutdown of the chlorination facilities via the control You will note that there is a signal to the chlorination
constant pressure on the chlorination cells during system. control system from a chlorine analyser at the inlet
normal operation. to the sea water reservoir. If there is insufficient
The liquid in the hydrogen removal tank is, of NaOCI at this point, then AIC2 will signal the
The NaOCI generator is a skid mounted unit course, a mixture of NaOCI and sea water. NaOCI generator to increase the amount of
which is contained within its own housing. If there NaOCI being produced and injected into the
is a leak of water within the housing, an operator The liquid level, is maintained by a controller discharge of the coarse filters.
could be in severe danger of electrocution. To (LIC1) operating valve LV1.
prevent this danger from arising, the area under Before moving on to the next part of our system,
the chlorination cells is monitored by a level The sodium hypochlorite and sea water mixture, in try Test Yourself 9, on the next page.
switch. This switch is designated LSHH1 on the the line after LV1, goes either :
drawing. It is activated by a rising water level. If the
level switch is activated, the local control system • into the main sea water line just downstream
will again close XV1 , shut off the electrical supply of the coarse filters
and send an alarm to the operator to warn him of • to the shock dosing connections. ( We will
the problem. be looking at these later ) 40
Petroleum Open Learning

Sea Water Reservoir The pre-service flush line, allows injection water
to be circulated through the fine filters and back to
the reservoir prior to starting the system. You will
The sea water from the coarse filters (which is now see this when you move on to the next part of this
chlorinated) flows towards the sea water reservoir.
Test Yourself 9 section.
Take a look now at Figure 23, on page 42, which However the shock dosing connection requires a
State the function of the following shows this part of the system. The main sea water little more explanation at this point.
equipment in the sea water intake, coarse line is a continuation of the one leaving the coarse
filters and chlorination facilities. filters which you saw in Figure 21. Follow the flow As a general rule the NaOCI is injected at a
through this part of the system starting at the reasonably steady rate during operations. NaOCI
a) The analyser indicator controller at the bottom left hand side of Figure 23. is extremely effective but, after a number of weeks,
discharge from the transfer pumps.
a few of the bacteria may become resistant to the
You can see that the sea water flows towards the effects of NaOCI. When this occurs the regular
b) The differential pressure switch sea water reservoir via a level control valve dosage rate becomes insufficient and a shock
connected across the coarse filters. and an analyser indicator controller. These are dose, i.e. a large amount, of concentrated NaOCI
designated LV2 and AIC2 respectively. is injected into the system over a short period of
c) The low flow switch at the inlet to the
time. This usually kills off any resistant strains of
NaOCI generator. LV2 maintains the correct level in the reservoir and bacteria. This shock dose can be injected at the
is operated by a signal from a level controller connection upstream or, as you can see in the
d) The air blower in the chlorination which is marked LIC2 in the drawing. Figure, downstream of the reservoir.
facilities.
As we have already seen, AIC2 controls the The sea water reservoir has two outlets. They are:
amount of NaOCI injected into the sea water as it
leaves the coarse filters. • an outlet to the sea water supply pumps
Note that, just before entering the reservoir, two • an outlet to the fine filter backwash pumps
lines join the main sea water line. They are:
The outlet to the sea water supply pumps is the
You will find the answers to Test Yourself • a pre-service flush line from the fine filters main outlet from the sea water reservoir. The
9 on Page 57. pumps transfer the water through fine filters.
• a shock dosing connection for NaOCI injection

41
Petroleum Open Learning

NaOCI
shock dosing

The backwash pumps provide water to clean the • a biocide connection Let’s take the time to have a look at why we need
fine filters. We will look at these in more detail these chemicals before moving on to the fine
when we move on to the fine filter system shortly. • a scale Inhibitor connection filter section.

You will notice that there are four connections • a ferric chloride connection
to the discharge line of the sea water supply
pumps. These come from a chemical dosing • a polyelectrolyte connection
skid and are:
42
Petroleum Open Learning

I pointed out earlier that a shock dose of If you look again at Figure 23 you will see that the
concentrated NaOCI usually kills off any resistant main flow line passes from the sea water supply
strains of bacteria. However, if this shock dose of pumps to the fine filters. We are now going to look
NaOCI fails to do the trick, then a dose of another at this system but before we do, trace the lines
biocide is used. The biocide connection is one of again and identify the equipment associated with
the points where such an alternative biocide would the sea water reservoir part of the overall system.
be injected.

In Section 1, I stated that the sea water may start Fine Filtration
to precipitate dissolved solids as scale if it was
not treated. The scale inhibitor is injected at this Figure 24, on the next page, shows one of the
point to ensure that there is no scale precipitation fine filters and its associated pipework. The
later in the process. main flowline from the sea water supply pumps
continues into this drawing at the top left hand
You will remember from Section 2, that side. In our example the fine filters are dual media
polyelectrolytes are coagulants. They could filters. Figure 24 is quite complicated. However, if
also be termed flocculants. The word flocculant you follow the lines carefully, together with the
comes from “flocking” or “gathering together”. explanation, I don’t think you should have any
These chemicals assist in gathering together tiny problems with it. Once again I have only shown
particles of material and converting them into one filter with its associated controls. In fact there
larger particles. These larger particles are easier would probably be a number of fine filters working
to filter out. in parallel.

Ferric chloride does a similar type of job. It


produces small electrically charged particles which
attract other particles to them. The combined
particles are large enough to be removed by the
filters.

Both of these substances, then, are aids to


filtration. They are injected into the water at this
point, which is upstream of the fine filters.

43
Petroleum Open Learning

44
Petroleum Open Learning

Before looking at the valve and control arrangements The main flow is shown entering the system near The filtered sea water leaves the fine filter via valve
in Figure 24, locate the two turbidity analyzers. the top left hand part of the drawing. Follow this bold XV3. This valve operates as part of the backwash
One measures the turbidity in the main sea line now. As you can see, the flow passes through sequence which we will look at shortly. The water
water line before the filter. The other monitors the the filter and continues towards the deaerator. This then joins the water discharged from the other
turbidity of the water leaving the filter. line then is very straightforward. The rest of the filters before being fed to the deaerator via the
lines are part of the filter control system. Let’s crude oil coolers.
Think about the location of the two analysers, then spend some time identifying the lines and controls,
attempt the following Test Yourself question. to see what their function is. Before continuing, go over the last few paragraphs
again, together with the drawing. Make sure that
The total flow of sea water to the fine filters you understand the way in which the total flow is
passes through flow element 2 (FE2) which distributed between the filters.
measures the water flowrate at this point. A signal
from FE2 is fed to flow relay 2 (FY2). Look now at two other devices which are fitted to
Test Yourself 10 the fine filter. They are:
FY2 determines how much water should be flowing
What is the significance of having two through each filter. It then sends a signal to a flow • pressure safety valve (PSV1). This is a
turbidity analysers in the system? indicator controller located on each of the fine valve which will open and release excess
filters. In this case it is designated FIC3. This pressure from the filter should the pressure
signal tells the FIC how much water should be exceed a predetermined maximum value.
flowing through its own filter. It is called a set point
signal. The actual amount of water passing this • pressure differential switch high
filter is measured by the second flow element FE3. (PDSH2). This compares the pressures
A signal from this element is also fed to FIC3. The upstream and downstream of the filter. If
controller compares the actual flow rate with the the differential pressure exceeds a pre-
flow rate determined by the flow relay. If there is a set limit, PDSH2 will automatically activate a
discrepancy between the two it sends a correcting backwash sequence, through a controller.
signal to the flow control valve FV3. The valve
responds by opening or closing to maintain the
correct flow.
You will find the answer to Test Yourself 10
on Page 57.

45
Petroleum Open Learning

We can now see how the backwash is carried This opens lines to dispose of the water in the filter At this point you may be looking for FIC4 on Figure
out. Let’s start from the point where the filter is via a dirty water sewer. It causes the water level in 24. In fact it appears on the previous Figure we
operating normally. the filter to fall to just above the filter bed. were looking at, Figure 23.

The flow of water would then be along the following After five minutes has elapsed the backwash Go back to Figure 23 for a moment and look at
path: sequence controller will: the second line leaving the sea water reservoir.
This line goes to the backwash pump then through
Through valve FV3 - through the filter - through 5. Close valve XV7. a strainer. The strainer ensures that reasonably
flow element FES - through valve XV3 - across the clean water is available for backwashing.
turbidity analyzer - on towards the deaerator. 6. Open valve XV9.
After the strainer, the water flows past a flow
As the water is filtered, the differential pressure, 7. Start the air scour blower. indicator controller (PIC4) and through a flow control
as measured by PDSH2, would start to increase. valve (FV4). This system controls the amount of
When it reaches, say, 1 barg, the backwash The air scour blower blows air through the filter. water to be pumped to the filter as backwash
sequence would be activated. The sequence of This violently agitates the filter bed to knock off water, the actual amounts being determined by the
events from now would be as follows: any tiny particles which may be stuck to the grains backwash sequence controller.
of garnet sand.
The backwash sequence controller will send a Now back to Figure 24 and the sequence of events.
backwash sequence in progress alarm to the After ten minutes has elapsed the backwash
operator, and then: sequence controller will: After twenty minutes has elapsed the backwash
sequence controller will:
1. Send a zero flow signal to FIC3 to ensure that 8. Stop the air scour blower.
FIC3 closes FV3. 13. Reduce the flow signal to FIC4.
9. Close valve XV9.
2. Close valve XV3. 14. Open valve XV5.
10. Start the sea water filter backwash pump.
The fine filter is now isolated from the main sea 15. Close valve XV6.
water line. The backwash sequence controller 11. Open valve XV10 to direct the water back
will now: through the filter and to disposal via XV6. The flow is reduced to allow the filter bed to settle
down as the filter is re-filled. The water will now
3. Open valve XV6. 12. Send a flow signal to FIC4 to regulate the flow back to the sea water reservoir instead of to
amount of water being pumped. the dirty water sewer.

46
Petroleum Open Learning

After a further five minutes, the backwash The fine filter is now back in service. A backwash There is just one line which I haven’t mentioned up to
sequence controller will: sequence completed signal will be sent to the now which appears in Figure 24. That is the spill back
operator. He will then check the outlet turbidity meter line at the top right hand of the drawing. Disregard
16. Open valve XV8. to ensure that the filter is operating efficiently. that for the moment. We will come back to it shortly.

17. Open valve XV4. The backwash sequence which I have just At this point you may want to follow the sequence a
described takes approximately sixty minutes to couple more times to ensure that you can visualise
18. Close valve XV5. complete. what is going on. When you have done that, have a
go at the following Test Yourself question.
19. Close valve XV10.

20. Increase the flow signal to FIC4 to maximum.

The filter is now in a pre-service flush sequence.


The water is flowing through the filter in the normal
Test Yourself 11
way, then back to the sea water reservoir.
The following steps, which are out of order, are the first part of a typical backwash sequence.
Place the steps in the correct order. Start with inlet valve closes and finish with filter is now
After a further fifteen minutes, the backwash
being backwashed.
sequence controller will:
a) inlet valve closes. h) controller closes outlet valve from filter.
21. Close valve XV4.
b) lower valve to dirty water sewer closes. i) valve in line from air scour blower opens.
22. Close valve XV8. c) air scour blower starts. j) valve in line from backwash pump opens.
d) sea water backwash pump starts. k) valve in line from air scour blower closes.
23. Remove the flow signal to FIC4. e) valves open to dirty water sewer. l) filter is now being backwashed.
f) air scour blower stops.
24. Stop the sea water filter backwash pump. g) signal sent to controller on discharge of
backwash pump to regulate flow of
25. Remove the zero flow signal to FIC3, backwash water.
which opens FV3.
You will find the answers to Test Yourself 11 on Page 57.
26. Open valve XV3.

47
Petroleum Open Learning

Deaeration
The deaerator column in our system is a
vacuum deaerator. Figure 25 shows the main
components of the system and the associated
controls. Take a few minutes to study this Figure
before moving on. First of all try to establish the
main flow of sea water through the deaerator
column to the surge vessel and on towards the
booster pumps.

Figure 25

48
Petroleum Open Learning

The sea water leaving the fine filters first flows The spill-back line : After being injected with anti-foam agent, the sea
through a turbidity analyser then a pressure water passes into the dearator column where it
indicator controller (PIC3). The turbidity analyser • protects the sea water supply pumps from gives up its dissolved oxygen. It then accumulates
measures the turbidity of the combined flow of sea damage if the flow to the deaerator column at the bottom of the column, the water level there
water from all of the fine filters. The PIC works is stopped being controlled by LIC3 and LV3.
together with a pressure control valve, PV3, to
prevent excess pressure building up in the line to • allows the fine filters to be brought on line At this point it can be treated with:
the deaerator. without flowing sea water to the deaerator
column • a non-hypochlorite biocide
The sea water is then passed across crude oil
coolers. These coolers are not really part of the • allows the filtered sea water to be diverted • an oxygen scavenger
injection water treatment system. They simply back to the sea water reservoir if the fine
use the cold sea water as a cooling medium in filters ever go off specification You should be able to remember how a vacuum
heat exchangers to reduce the temperature of oil. deaerator works. If you need to refresh your
However the sea water itself is heated up in the Back to the flow through the system memory, however, I suggest that you go back to
exchanger and , as you saw Section 3, heating up the relevant part of Section 3.
the water will assist in the deaeration process. After passing through LV3 the filtered sea water is
injected with an anti-foam chemical. The deaerated water leaves the column and flows
The sea water then flows through a level control to a surge vessel. This is in effect a storage tank
valve (LV3) which is activated by a level indicator The anti-foam chemical is injected at this point from which booster pumps can take their suction.
controller (LIC3). LIC3 controls the level of water because, as the sea water passes across LV3, it A balance line connecting the surge vessel to the
in the bottom of the deaerator column. may be subjected to a large pressure drop. The deaerator column ensures that the water level in
pressure drop would cause a lot of the dissolved each remains the same.
If LV3 closes too far, the pressure on the filter side air to come out of solution.
of LV3 will rise. This will activate PIC3 which will You will notice a line entering the surge vessel at
open PV3. Water from the filters will be spilled The changing conditions may cause the water to the top, from the booster and injection pumps. This
back to the sea water reservoir. This is the line foam. If foam is produced, the deaerator column is part of the pump protection equipment which we
that joins the pre-service flush line from the fine cannot operate correctly. A lot of water will be will look at shortly.
filters, which you saw in Figure 24. lost with the vapour and a lot of oxygen will pass
through the column with the water. The anti-foam Now let's take a look at the remaining instruments
chemical breaks down the bubbles in the foam. and equipment which service the deaerator
column and surge vessel.

49
Petroleum Open Learning

The first thing to notice is that there are two pressure The only lines that we haven't identified on Figure Water Injection Pumps
relief valves on the column. PSV4 is located on 25 are the ones the ejector and vacuum pump.
the top of the column and PSV5 is located on the We can do that now. If we move on to Figure 26, on the next page, we
outlet line at the bottom of the column. Both PSV's can see that the main flow from the surge vessel
are set to open at the maximum working pressure The top line to the vacuum pump is connected to is through the booster pumps and the injection
of the column (say, 1 barg). the deaerator above the top packing. The other pumps to the sea water injection wells via a
line is connected above the lower packing. It leads header.
If the deaerator column ever completely filled with to the ejector and , from there, joins the upper line
water, the pressure at the bottom of the column before the vacuum pump. From the pump the line The booster pumps, as their name suggests,
could exceed its maximum working pressure. goes to an air / water separator. From here, the boost or increase the pressure of the treated sea
air is vented to the atmosphere and the water is water. The pressure is increased from that of
A pressure alarm high (PAH4) is positioned just disposed of via the dirty water sewer. the surge vessel to the pressure required at the
above the top layer of packaging and is set to suction of the injection pumps.
initiate an alarm if the pressure exceeds, typically, Note the location of a hydrocarbon detector
35 millibar. This could occur if: (HD). If the crude oil coolers leak, crude could
enter the sea water and cause severe problems
• the level of water reached more than 0.35 m with the deaerator column and other equipment.
above PAH4 HD will detect the presence of hydrocarbons in the
air leaving the air / water separator and activate
• the level of water was below PAH4 and an alarm to warn the operator that there is a
the internal gas pressure was greater than problem.
35 millibar.
We are almost at the end of the system now. All
PAH4 would activate and alert the operator to the that remains is to have to look at the injection
fact that a pressure problem has occurred in the pumps and injection wells.
deaerator column. In both instances the alarm
would activate before the maximum pressure at
the bottom of the column was reached.

A pressure relief valve also protects the surge


vessel from over pressure. On the drawing it is
designated PSV6.

50
Petroleum Open Learning

The combined discharge of the booster pumps is fitted


with:

• an oxygen analyser (AE Os)

• a chlorine analyser (AE CI;)

• a pH analyser (AE pH)

The oxygen analyser ensures that the oxygen content


of the water is still below the maximum permitted level.
The chlorine analyser ensures that there is a sufficient
amount of Na0CI in the injection water. The pH analyser
measures the acidity or alkalinity of the injection water.

If you remember, we injected the scale inhibitor into the


discharge of the sea water supply pumps just before
the water entered the fine filters. The level of pH is an
indication of the tendency of the sea water to form scale
deposits. As a general rule, if the water has a high pH
(above 7.0) it is more prone to forming scale. The level
of pH is monitored at this point and the rate at which the
scale inhibitor is injected is adjusted accordingly.

The injection pumps will increase the water pressure


to that required for injection into the reservoir. This could
be well over 200 barg in many cases.

The discharge lines of both the booster pumps


and injection pumps are fitted with minimum flow
controllers. They are FIC5 and FIC6 respectively.

51
Petroleum Open Learning

All centrifugal pumps require protection to The treated sea water then flows across:
prevent them from pumping at too low a flow
rate. The minimum flow controllers measure wing valve (XV20)
the flow from the pumps. If this flow falls below
a predetermined level, the controllers open their upper master gate valve (XV21)
respective control valves (FV5 and FV6) and
allow the water to recycle back to the surge lower master gate valve (LMV7)
vessel. When sufficient water is flowing through
the pumps, FV5 and FV6 are fully closed. sub-surface safety valve (XV22)

The treated sea water leaves the injection and down into the reservoir.
pumps and enters a common water injection
header. The flow lines to the individual water The injection well is completed in an almost
injection wells are then taken from this header. identical way to the oil and gas producing
wells. Our open learning packages on Oilwell
Water Injection Wells Drilling Technology and Oilwell Production
Technology explain the construction of these
There may be as few as two or three water wells in great detail.
injection wells in small oil reservoirs, or as
many as twenty or more in larger ones. In
Figure 27, I have shown a schematic drawing
of a typical well.

The flow of treated sea water is measured by You have almost completed this unit on injection
a flow element (FE7) which is upstream of a water treatment but before going through the
choke valve. A choke is a specially designed summary, have a go at the final Test Yourself
valve which allows the flow of high pressure question.
injection water to be controlled. In our example
the choke is controlled manually and would be
adjusted by the operator to give the required
flow rate, as measured by FE7.

52
Petroleum Open Learning

Test Yourself 12 Summary of Section 4


Make a simple block diagram of a water
In this section we have looked at the operation We looked at the operation of the coarse
injection system similar to the one at the start
of a typical sea water treatment system filters and fine filters and saw how they are
of Section 4. In addition, your diagram should
which uses: backwashed.
show:
• sodium hypochlorite as a biocide We saw how the vacuum deaerator is
1. The chemical injection points.
operated and controlled, and how:
• basket filters as coarse filters
2. Where the water is taken from for the
• anti-foam agent is injected into the
filter backwashing.
• dual media filters as fine filters deaerator feed to prevent foaming
3. Where water analysers may be located.
• vacuum deaeration to remove oxygen • oxygen scavenger is injected into the
from the sea water system to remove the last few traces of
oxygen
We considered how the sodium hypochlorite
is generated and injected into the sea water We then looked at the booster pumps and
during treatment. We also saw how a different the injection pumps and saw how they were
biocide may be used to shock dose the operated and controlled.
system to prevent resistant bacteria from
surviving the sodium hypochlorite injection. Finally we looked at a typical water injection
well and located the various valves and
We saw how scale inhibitor is injected into the controls.
sea water and how the efficiency of the scale Now go back to the training targets on page
You will find the answer to Test Yourself 12
inhibitor was measured by a pH Analyser. 4 and make sure that you have met those
on Page 58.
targets.

53
Petroleum Open Learning

Check Yourself 1 Check Yourself 2 Check Yourself 3


a) One fifth is 20%. The particles of sand or clay would block the
So the porosity is 20%. 1. Gas cap drive. pores of the reservoir rock, reducing permeability
2. Dissolved gas drive. and hence productivity.

3. Dissolved gas drive.


b) A cap rock prevents oil migration so its
permeability is likely to be zero. 4. Water drive.

5. Gas cap drive.

c) Sponge

Sandstone

54
Petroleum Open Learning

Check Yourself 4 Check Yourself 5 Check Yourself 6


Removing oxygen creates conditions
which are ideal for the growth of anaerobic 1. Removal of suspended solids. a) FALSE - a permeable barrier is used.
bacteria.
2. Removal of dissolved gases (i.e. oxygen, b) TRUE.
which makes the water corrosive).
c) TRUE.
3. Removal of dissolved gases.
d) FALSE - a basket filter uses a fine mesh
4. Reduces scale deposition caused by screen.
dissolved solids being precipitated.
e) FALSE - the flow is in the opposite
5. Assists in removal of suspended solids.
direction.
6. Preventing growth of bacteria. f) TRUE.

55
Petroleum Open Learning

Check Yourself 7 Check Yourself 8


sparge pipe...................... deaerators
a) Both.
degassing tank................ NaOCI generator
b) Both (air is a gas).
support material............... filters (sand & dual media)
c) Vacuum deaerator.
vortex breaker.................. deaerators
d) Gas stripping deaerator.
ejector.............................. vacuum deaerator
e) Both.
seal chimney................... vacuum deaerator
f) Neither.
cartridges......................... cartridge filter
g) Vacuum deaerator.
backwash controller......... filters (basket, sand, dual media)
h) Both
cleaning head motor........ basket filter

demister pad.................... deaerators

vacuum pump.................. vacuum deaerator

56
Petroleum Open Learning

Check Yourself 9 Check Yourself 10 Check Yourself 11


a) To monitor sea water, in order to ensure By checking the water turbidity upstream and
The correct sequence would be
a sufficient concentration of NaOCI to kill downstream of the filter, the efficiency of the
bacteria. filter can be monitored. (a) — (h) — (e) — (b)
(i) — (c) — (f) — (k)
b) To monitor differential pressure. When
this pressure exceeds a pre-set maximum, (d) — (j) — (g) — (I)
the differential pressure switch initiates
a backwash, through a backwash
sequence controller.

c) To protect the NaOCI generator from


being damaged due to a low flow of sea
water.

d) To purge the hydrogen removal tank with


air.

57
Petroleum Open Learning

Check Yourself 12

58
POL
Petroleum Open Learning

Natural Gas
Liquids Recovery
Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY
POL
Petroleum Open Learning

Natural Gas
Liquids Recovery
Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY
Petroleum Open Learn-

Designed, Produced and Published by OPITO Ltd., Petroleum Open Learning, Minerva House, Bruntland Road, Portlethen, Aberdeen AB12 4QL

Printed by Astute Print & Design, 44-46 Brechin Road, Forfar, Angus DD8 3JX www.astute.uk.com

© OPITO 1993 (rev.2002) ISBN 1 872041 85 X

All rights reserved. No part of this publication may be reproduced, stored in a retrieval or information storage system, transmitted in any form or by any means, mechanical,
photocopying, recording or otherwise without the prior permission in writing of the publishers.
Petroleum
Petroleum
Open
Open
Learning
Learn-
Natural Gas Liquids Recovery
(Part of the Petroleum Processing Technology Series)

Contents Page Visual Cues


BOOK1 training targets for you to
achieve by the end of the unit

* Training Targets 4 test yourself questions


to see how much you
* Introduction 5 understand

* Section 1 - The Theory of NGL Recovery 7


check yourself answers to
Chemistry Fundamentals let you see if you have been
Hydrocarbons and Chemical Bonding thinking along the right lines
The Physical Properties of Hydrocarbons
Boiling Point of Alkanes
Dew Point Curve activities for you to apply
Vapour Pressure your new knowledge
Mulit-Component Mixtures
Absorption

summaries for you to recap


* Section 2 - NGL Recovery Using Compression & Cooling 23 on the major steps in your
progress
A Typical Compression and Cooling System

1
Petroleum
Petroleum
Open
Open
Learning
Learn-

2
Petroleum
Petroleum
Open
Open
Learning
Learn-

Contents (cont'd) Page Visual Cues


* Section 3 - NGL Recovery by Refrigeration 32
training targets for you to
Mechanical Refrigeration achieve by the end of the unit
A Typical Mechanical Refrigeration NGL Recovery System
Auto Refrigeration
test yourself questions
to see how much you
* Section 4 - NGL Recovery Using Absorption Process 47 understand

A Typical Absorption System


Gas Flow check yourself answers to
Absorption Oil Flow let you see if you have been
thinking along the right lines

* Check Yourself - Answers 57


activities for you to apply
your new knowledge

summaries for you to recap


on the major steps in your
progress

3
Petroleum
Petroleum
Open
Open
Learning
Learn-

Training Targets

When you have completed this unit on Natural Gas Liquids Recovery, you will be able to :

• State the reasons for recovering natural gas liquids


• Explain in simple terms the behaviour of hydrocarbons with respect to changes in energy level
• Define the terms bubble point, dew point and vapour pressure
• Describe the construction and operation of equipment used in a typical compression and cooling NGL recovery plant
• Explain the operation of a refrigeration process
• Differentiate between mechanical refrigeration and auto refrigeration
• Explain what is meant by the Joules Thompson effect
• Describe two types of NGL recovery plant which use refrigeration systems
• List the equipment used in an NGL recovery plant using the absorption process
• Outline the flow through an absorption system used to recover NGLs

Tick the box when you have met each target.

4
Petroleum
Petroleum
Open
Open
Learning
Learn-
Oil and Gas Separation
Natural Liquids Recovery
Systems
Introduction
Natural Gas is produced from petroleum A natural gas stream is capable of being partially or Finally you should be aware of the expression
reservoirs as associated gas ( associated with totally liquefied, using various types of processing condensate. This is a hydrocarbon mixture which
an oil accumulation) or non-associated gas ( equipment. We will look at why we might want to exists as a gas in the reservoir but condenses to a
produced independently of oil). This natural gas is do this in a moment. liquid as the pressure and temperature is reduced
a complex mixture of hydrocarbons which, in the when the gas is produced.
main, belong to a family of hydrocarbons called Total liquefaction of natural gas is only possible
the alkane or paraffin series. We will talk about using processing techniques involving extremely In the Petroleum Processing Technology series
this hydrocarbon family in more detail in Section low temperatures. These are called cryogenic of units, we are concerned with the processing
1. techniques. The processes are normally carried of oil and gas in the oilfield or on the production
out in large, shore-based installations, similar to platform. Therefore, in this unit, we will concentrate
The first ten members of this alkane series are oil refineries. The liquid produced is referred to as on the recovery of natural gas liquids or NGLs.
listed below. liquefied natural gas ( LNG)

• Methane Partial liquefaction can be achieved with less But why bother recovering the liquids? There are a
• Ethane complicated processes than those required for number of reasons for this, and I think we should
total liquefaction. These processes are often look at them in some detail before starting the unit.
• Propane incorporated into the facilities on an oil and
gas production platform. In these systems, the • If the associated gas is being re-injected after
• Butane components of the natural gas stream are liquefied separation, valuable components are being
• Pentane with the exception of the methane and most of the lost. There is a strong case for recovering as
ethane. The resulting liquid is called natural gas much as possible from the gas stream before
• Hexane liquid ( NGL ) re-injection, and the NGLs are the easiest
fraction to recover. For example, from a total
• Heptane Whilst I am quoting abbreviations regarding gas of 25 million cu ft of gas produced daily,5
• Octane liquids, it is worth while mentioning another one 000 bbl of liquid might be recovered with a
which you will come across. This is the expression potential value somewhere in the region of
• Nonane LPG which means liquefied petroleum gas. LPG, $100,000.
which consists of liquefied propane and/or butane,
• Decane is extensively used for domestic heating and
cooking.
At normal temperatures and pressures the first
four members of the series exist as gases. The
other components are liquid at these conditions.

.
5
Petroleum
Petroleum
Open
Open
Learning
Learn-

• Gas being marketed has to meet certain Composition can be controlled by using an adsorp-
specifications regarding water content, tion or absorption process.
calorific value (heating capability),
composition etc. Processes designed to In an adsorption process the components are
remove water vapour and recover the gas deposited on the surface of a solid material then
liquids, will ensure that the sales contract regenerated off.
specifications are met.
In an absorption process the gas is contacted with
• Gas is usually transported via a pipeline to liquid. The heavier components of the gas stream
the sales point. If the heavier liquefiable are absorbed into the liquid. These can then be
components are not removed they could recovered from the liquid and the absorbing liquid
condense in the pipeline as free liquids. This used again.
would result in loss of pipeline efficiency.
The liquid accumulation would necessitate The unit is divided into four sections:
frequent pipeline cleaning or pigging.
• Section 1 will concentrate on the theory of
So you can see that it makes sense in most cases NGL recovery.
to recover the gas liquids and in this unit we will
look at how this is done. • In Section 2 we will be looking at NGL
recovery by compression and cooling.
Liquefaction of gas, whether it be partial or total,
always involves control of pressure and • Section 3 will cover NGL recovery by
temperature. refrigeration.

In addition, the composition of the gas and • Finally in Section 4 we will be looking at NGL
recovered liquid streams may be controlled to recovery by absorption techniques.
ensure the correct type and amounts of liquid are
obtained.

Pressure is controlled directly. Temperature is


controlled by adding or removing heat.

6
Natural Gas Liquids Recovery Petroleum
Petroleum
Open
Open
Learning
Learn-

Section 1 - The Theory of NGL Recovery


ELEMENT CHEMICAL SYMBOL
In this section we are going to start by looking Another term with which you should be famiiar at
at the chemical and physical properties of this point is mixture.
hydrocarbons. This will enable us to see how the Aluminium AI
recovery processes work. We will do this in a Mixtures are not pure substances in theselves.
rather simple way so don’t get concerned about Carbon C They are made up from two or more pure
the science involved. Let’s start by having a look substances which are not chemically combined.
at some basic fundamentals of chemistry. These pure substances may be elements or
Hydrogen H compounds which retain their own properties when
forming the mixture.
Chemistry Fundamentals Table 1 : Elements with English names
A combination of iron filings and sand is a
All substances are matter. This can be ELEMENT LATIN NAME CHEMICAL mixture. They could be separated from each
described as anything which occupies space and other quite easily. A magnet would attract
can be seen and felt etc. Chemistry is the science
SYMBOL the iron but not the sand. Both the iron and
which investigates the composition and structure Copper Cuprum Cu sand are pure substance elements. As a
of substances and the changes which substances combination they are a mixture but they retain
may undergo. their own properties. ( I.e. iron is magnetic,
Iron Ferrum Fe sand is not ).
Pure substances can be elements or compounds.
These are basic terms used extensively in Silver Argentum Ag Each element is made up from even simpler
chemistry and we must understand these before building blocks, called atoms.
we carry on. Table 2 : Elements with symbols from Latin names
These are the smallest particles that can be
Elements are the building blocks of all matter. identified as an element.
They are substances which cannot be broken
down into simpler substances. There are about 90 Compounds are pure substances which consist An atom is also the smallest particle of an element
naturally occurring elements, including things like of two or more elements which are chemically that can combine chemically with another atom.
Hydrogen, Oxygen and Carbon. combined to form a new substance.
When atoms combine chemically, they form
Instead of using the full name to describe an The compound may not bear any resemblance to molecules.
element, symbols are used. These are usually the the elements from which it is made. For instance
first one or more letters of the full name, either water is a compound which is normally a liquid. It These are the smallest particles of a
English or foreign ( usually Latin). I have listed is made from the chemical linking of the elements combination of atoms which can exist and still
some common examples of each in the following hydrogen and oxygen which are normally gases. retain the characteristics of the combination.
tables.
7
Petroleum
Petroleum
Open
Open
Learning
Learn-

Atoms of a single element can combine to form molecules


of that element. For instance the gas oxygen is a pure
substance. It is composed solely of atoms of the element
oxygen combined into molecules.

Water on the other hand is also a pure substance. It is a


compound composed of atoms of oxygen and hydrogen
combined to form water molecules.
The following list of substances are either elements or compounds.
A word of warning at this point. I said earlier that chemical
symbols are used to describe elements. This same symbol Indicate which.
is also used to describe one atom of the element. Therefore
the symbol C can mean the element carbon or one atom
nitrogen - (N) ...............................................................
of carbon. This is important when we are looking at the way
atoms combine to form molecules.
sodium chloride - (NaCl) ...............................................................
We will be looking at the way this happens later. Before we
move on however, have a go at the following Test Yourself water - (H2O) ...............................................................
Question on elements and compounds.

helium - (He) ...............................................................

sulphuric acid - (H2SO4) ...............................................................

carbon - (C) ...............................................................

iron - (Fe) ...............................................................

You will find the correct answers in Check Yourself 1.1 on Page 57.

8
Petroleum
Petroleum
Open
Open
Learning
Learn-

In this unit we are concerned with the recovery of Of course, each arm of the carbon atom need not
natural gas liquids. But what is natural gas? In the necessarily grasp the arm of a hydrogen atom.
introduction I defined it as a complex mixture of Two atoms of carbon could ‘hold hands’ leaving
hydrocarbons. the remaining six carbon atom arms grasping
hydrogen arms. This could be represented as
Hydrocarbons are compounds of the elements shown in Figure 1.3.
hydrogen and carbon. These elements have the
chemical symbols, H for hydrogen and C for carbon.
We can now look at the formation of these
compounds.

Hydrocarbons and Chemical


Bonding
Methane
When atoms combine to form molecules they do so in
a special way called chemical bonding. The easiest
way to visualise this bonding is to imagine each atom
Figure 1.1 : Methane Molecule
to be a ball having a number of arms. For chemical
bonding to occur, each arm of an atom must grasp The simple molecule shown in Figure 1.1 is the
the arm of another atom leaving no arm free. The hydrocarbon molecule methane. A simple method
actual number of arms on an atom will depend on the of representing the structure of methane is shown
specific element. in Figure 1.2

For instance, carbon has four arms whilst hydrogen Figure 1.3 : Ethane Molecule
has only one. So, for hydrogen and carbon atoms to
combine into a simple hydrocarbon molecule, each This is the hydrocarbon molecule ethane.
of the four arms of the carbon atom must grasp the
single arm of each of four hydrogen atoms. This is Using small drawings to represent molecules is
shown pictorially in Figure 1.1. somewhat clumsy. So chemists have developed a
system of shorthand, in which a chemical formula
is used to describe the number and type of atoms
which make up a molecule.
Figure 1.2 : Methane Molecule

9
Petroleum
Petroleum
Open
Open
Learning
Learn-

For instance ethane, which has two carbon atoms In the alkane series, which starts with metane,
and six hydrogen atoms, can be written: each subsequent member has one extra cabon
atom and two extra hydrogen atoms.
C 2H 6
Note that when there is only one atom of a
particular element in the molecule, the subscript
Test Yourself 1.2 In the Introduction, I listed the first ten members of
the series:
“1” is left off. The chemical formula for methane • Methane
Using the general formula for alkanes,
therefore is
determine the chemical formula for octane • Ethane
CH4 which has eight carbon atoms and nonane • Propane
As I pointed out in the Introduction, methane and which has nine.
• Butane
ethane are the first two members of a family of
hydrocarbons called the alkanes. In such a family • Pentane
each of the members are related by a general • Hexane
formula. The general formula for the alkane family • Heptane
is:
• Octane
CnH(2n+2)
• Nonane
In the formula, ‘n’ represents the number of carbon • Decane
atoms in the molecule. The greater the number of
carbon atoms, the larger and, therefore heavier, You will find the correct answers in Check You will also remember from the Introduction
the molecule. Yourself 1.2 on Page 57. that, at normal temperatures and pressures, the
first four members of the series exist as gases.
So a hydrocarbon molecule of the alkane family The heavier components are liquid under these
which has 4 carbon atoms would have the conditions.
chemical formula
Scientists do not use the word family’ to describe a Although I have just said that pentane and the
C4H(2x4)+2 i.e. C4H10 heavier components are liquids at normal
related group of compounds such as the alkanes.
This is the fourth member of the alkane family They call them an homologous series. temperatures and pressures, a natural gas stream
-Butane. may contain some of these components in very
This is from the Greek ‘homos’ meaning ‘same’, small quantities. This constituent of the gas stream
and ‘logos’ meaning ‘speech’. In other words is usually referred to as pentanes plus (pentanes
related or similar. + ) or C5+

10
Petroleum
Petroleum
Open
Open
Learning
Learn-

A typical gas stream leaving a separator in a However, other chemicals may also be present in I’m quite sure that you are fully aware of the state
production facility could have the following analysis. the natural gas stream. Compounds of oxygen, of everyday matter. For instance at normal room
nitrogen and sulphur, together with water vapour, temperature and pressure a steel bar is in the solid
can occur in greater or lesser amounts. These state, water is liquid and the air you are breathing
Component mol % substances can cause problems in the production is a gas.
Methane (C1) 83.9% and processing of natural gas. However for
the present we will ignore these impurities and You will note, however, that I said at normal
Ethane (C2) 11.6% concentrate on the hydrocarbons of the alkane temperature and pressure. If the temperature and /
Propane (C3) 3.3% series. or the pressure is changed, the state of the matter
Butane (C4) 1.0% may change.
Each of the components of the hydrocarbon
Pentanes + (C5+) 0.2% mixture has its own unique physical properties. I We would actually be adding or removing energy
think it would be worthwhile having a look at these from the matter by changing its temperature or
now. pressure. It is the energy possessed by the matter
These components are often referred to by which determines what state it is in.
abbreviating the chemical formula as indicated. The Physical Properties of
i.e. Butane C 4H10 is referred to as C4. Hydrocarbons A very familiar example of changing the state of
a substance is that of adding or removing energy
Mol % stands for molar percentage. It is just I said earlier that all substances are referred to as from water.
one way of expressing the proportion of each matter. Matter is all around us, from the air that
substance in a mixture of several substances. It is you breathe to the food that you eat and the liquids Water is a pure substance which is a compound
used extensively in chemical calculations but we that you drink. of hydrogen and oxygen i.e. H2O. At atmospheric
don’t need to go any further into this at this point. pressure and room temperature water is a liquid.
Matter can exist in three different physical states. If however we raise the temperature of the water
In addition to the alkane series, other They are: by adding energy in the form of heat to the water,
hydrocarbons can be produced from a petroleum it will eventually boil and turn to steam - a gas or
reservoir. Components of the cycloparaffln series •Solid - which has a definite shape and vapour.
or the aromatic series can form a small proportion volume. Similarly, if we were to remove heat energy, the
of the reservoir fluids. •Liquid - which has a definite volume water would eventually freeze and turn to ice - a
but no specific shape. solid.
You can see then that the gas stream is a mixture •Gas - which has neither definite shape
of hydrocarbons in varying proportions. This is of nor volume. This process of changing the state of water is
course how I defined natural gas. easily represented on a simple graph and I have
included one on the next page.

11
Petroleum
Petroleum
Open
Open
Learning
Learn-

As we continue to add heat you will notice that the


temperature stops increasing and remains constant
until point C is reached. The heat energy added
between points B and C is called latent heat, and
is used to convert solid ice to liquid water. When
all the ice has turned to water, a further addition of
heat energy causes the temperature to start rising
again until point D is reached. Once again the heat
added between C and D is sensible heat. At point
D the first bubble of steam appears, as the water
starts to boil. The liquid has started to change to a
vapour or gas. Between point D and point E, the
temperature again remains constant as further
latent heat converts liquid water to steam. When
all the water has turned to steam, a further addition
of sensible heat causes the temperature to climb
again towards point F.

Of course this process would work in reverse. If


we started at point F where all the water is steam
and began to remove heat energy we would
follow the graph back to point A, passing through
Figure 1.4 : Changing State of Water
temperatures corresponding with the boiling point
and freezing point of water.

In Figure 1.4 the vertical axis of the graph is the temperature of the water and the horizontal axis is the heat input. We have talked so far about the three states of
Note that I have not indicated any units for either temperature or heat input. You should also note that the whole proc- matter. You will often find these states referred
ess takes place at a constant pressure, i.e. atmospheric pressure. to as the three phases of matter: the solid, liquid
and vapour (gas) phases. I will continue to use the
Take a look at the graph and try to visualise what is happening in this representation. terms vapour and gas to mean the same thing.

At the starting point A, the water is in its solid state of ice. As we add heat energy the temperature of the ice would Before leaving this exampleof changing the state,
start to rise and would continueto rise until point B is reached. The heatwhich is added during this time is called sen- or phase, of matter have a go at Test Yourself 1.3.
sible heat. At point B the ice would start to melt.

12
Petroleum
Petroleum
Open
Open
Learning
Learn-

So both pressure and temperature have an affect


on the way a substance changes its phase. In
Test Yourself 1.3 other words its phase behaviour. This being
so, it is useful to prepare diagrams to illustrate
this effect. These diagrams are called phase
diagrams.
If the substance in Figure 1.4 is water at
atmospheric pressure, what would be the The simplest type of phase diagram is one which
temperature at points B, C, D and E. shows the behaviour of a pure substance. That is,
a substance where all the molecules are identical,
such as pure water or pure ethane. The diagram
is in the form of a graph with the vertical axis
representing pressure and the horizontal axis
representing temperature.
You will find the correct answers in Check
Yourself 1.3 on Page 57.

Figure 1.5 : Pure Substance Phase Diagram

You will remember that, in the example shown in


You can see in the figure that there are three
Figure 1.4, the phase changes were achieved by
regions. A solid region, a liquid region and a
changing the amount of heat energy in the system,
gaseous region. The lines separating these regions
at constant pressure (that is, at a constant level
are called lines of equilibrium or saturation. For
of pressure energy). We can, however, affect
example, along the line B D the substance could
phase behaviour by changing the level of pressure
be all solid, all liquid or a mixture of both. Similarly,
energy, as an alternative to changing the level
along line A B the substance could exist as a solid,
of heat energy. For example, the boiling point of
a gas or a mixture of both. The point B is called
water is 100°C (212°F) at atmospheric pressure.
the triple point. At this point all three phases could
However, if the pressure was decreased, for
exist together. It is difficult to imagine ice, water
example by conducting the experiment on top of
and steam existing together, and ot course this
a mountain, the boiling temperature would also
condition is very seldom seen.
decrease.

13
Petroleum
Petroleum
Open
Open
Learning
Learn-

If changing conditions of temperature and pres-


sure took a substance across line B D, the phase
change would be from a solid to a liquid. Across
line B C there would be a phase change from liquid
to gas. Test Yourself 1.4
However if changing condltions of temperature and
pressure took a substance through line A B, the
phase change would be from a solid directly into If the phase diagram in Figure 1.5 is for water, explain what would happen
a vapour. No liquid would be formed. A common as conditions change along the line from Y to X.
substance which does this is solid carbon dioxide,
or dry ice. It sublimes from a solid to a fog like
vapour and is often used on stage to give the mist
effect you see at pop concerts.

Look at the figure again and try to imagine what is


happening as conditions of temperature and
pressure change. Then have a go at Test Yourself
1.4.

How many phase changes would there be?

You will find the correct answers in Check Yourself 1.4 on Page 57.

14
Petroleum
Petroleum
Open
Open
Learning
Learn-

Of course we are not really interested in the solid phase when we are discussing Natural Gas Liquids Recovery. So Once again I have not included any actual
the most important line as far as we are concerned is the line B C, the line which separates the liquid and vapour values of temperature or pressure so we are not
regions. Along this line the substance would exist as a liquid, a vapour, or a mixture of the two. This line is known specifying which alkane it is.
by a number of names. It is called the bubble point curve, the dew point curve or the vapour pressure line. We
will have a look at the meaning of these terms shortly. In the region to the left of the line B C the
substance is all liquid, whilst to the right it is all
Take a look at the following figure which shows the line B C from a phase diagram for a pure hydrocarbon of the gas.
alkane series.
The point C is called the critical point of the
substance. At this point the properties of the liquid
phase and the vapour phase become identical.
The actual temperature and pressure at this point
for a particular substance are called the critical
temperature and the critical pressure.

For a pure substance the critical temperature is


the temperature above which a vapour cannot be
liquefied no matter how much pressure is applied.

Similarly the critical pressure is the minimum


pressure necessary to liquefy a vapour at its critical
temperature. Don’t worry about these terms. I have
included them for interest only and we will not be
looking at them in any detail.

Fig 1.6 : Pure Hydrocarbon Phase Diagram

15
Petroleum
Petroleum
Open
Open
Learning
Learn-

Boiling Point of Alkanes ALKANE BOILING POINT Vapour Pressure


AT ATMOSPHERIC
If we increased the temperature of the alkane PRESSURE Closely related to boiling point is the property of
along the line X Y the following would happen. Methane -162°C -259°F vapour pressure. It is the pressure which would
When the temperature (T) corresponding to point be exerted by a liquid in a closed container if there
Z is reached, the first bubble of gas would appear Ethane -89°C -128°F is a vapour space above the liquid level. That may
and the liquid will start to boil. You can probably Propane -42°C -43.7°F seem rather complicated, so let me try to explain it.
see now why the line BC is called the bubble
point curve. The actual temperature at this point is Butane -0.5°C 31.1°F When a liquid is contained in a closed vessel,
the boiling point Pentane 36°C 97°F some of the liquid molecules will leave the liquid
of this alkane at the pressure (P) which also and enter the vapour space. This is evaporation.
corresponds to point Z. Note that I said the boiling Hexane 69°C 156°F These molecules strike the sides of the container
point at a specific pressure. If we crossed the line Heptane 99°C 209°F causing the pressure in the container to rise. Some
B C at a different position the liquid would start to of the molecules in the vapour space will strike the
boil at a different pressure and temperature. Octane 126°C 258°F surface of the liquid and re-enter the liquid. This is
the opposite of evaporation, i.e. condensation.
The boiling point is one of the physical properties Table 3: Boiling Points of Alkanes
of hydrocarbons. But to compare one with another As long as more molecules are leaving the liquid
we must use the same pressure. The commonly than entering it the pressure in the vessel will
used standard is atmospheric pressure. So, the You will notice from the table that the boiling point continue to rise. Figure 1.7 shows this situation in
boiling point of our alkane Is the temperature at increases as the alkanes. get heavier. In other a simple way.
which the liquid bolls at atmospheric pressure. words the more carbon atoms, the higher the boil-
ing point.
These boiling points vary quite considerably and I
have listed the first 8 members of the alkane series
in the following table. Dew Point Curve

Now go back to Figure 1.6. If we were to start at


point Y and reduce the temperature the opposite
would happen. When we reached the bubble point
curve the gas would start to condense.
At this point the first dew drop of liquid would ap-
pear. You can now see why this line B C is also Figure 1.7 : Vapour pressure Rising
called the dew point curve.

16
Petroleum
Petroleum
Open
Open
Learning
Learn-

At a certain temperature however, a state of


equilibrium is reached between liquid and vapour.
This occurs when exactly the same number
of molecules are leaving the liquid as are re-
entering it. At this temperature the pressure in
the vessel would remain constant. This pressure
is the vapour pressure of the liquid at that
temperature.

This is shown again pictorially in Figure 1.8.


Figure 1.8 : Phase Equillibrium

If the liquid in the vessel is propane at a


temperature of 38°C (100°F), the pressure would
be 12 bar (175 psi). If, however, heat was added
to the vessel, the temperature would increase.
More molecules wouId start to leave the liquid and
the equilibrium would be lost. The pressure in the
vessel would rise once more. The pressure would
increase until equilibrium was again established
at a new temperature. Figure 1.9 shows this
happening.

So the vapour pressure is dependent on


temperature. In order then that the vapour
pressures of different hydrocarbons may be
compared, they must be quoted at a standard
temperature. This temperature is 38°C or (100°F). Figure 1.9 : Vapour Pressure Rise - Heat Induced

17
Petroleum
Petroleum
Open
Open
Learning
Learn-

In the following table I have listed the vapour pres- Multi-Component Mixtures
sures of the first five alkanes.
The phase behaviour we have been looking at
so far has been for a pure substance (or a single
ALKANE VAPOUR PRESSURE component). The trouble with the gas leaving a
AT 38°C 100°F separator is that it is a mixture of hydrocarbons. In

pressure
other words a multi-component mixture. Each
Methane 344.8 bar 5 000 psi of these components has a different boiling point.
We can no longer represent the behaviour of this
Ethane 5.4 bar 780 psi mixture with a single vapour pressure line. So, the
Propane 12 bar 175 psi phase diagram now becomes a phase envelope.
The dew point curve and bubble point curve are
Butane 2.5 bar 36.9 psi now different lines.
Pentane 0.06 bar 0.9 psi
Look at Figure 1.10. It is a phase diagram for a
multi-component mixture of hydrocarbons with a
Table 4 : Vapour Pressure of Alkanes fixed composition.

In the diagram the line A C is the bubble point temperature


You can probably see now why the line B C in curve and line B C is the dew point curve. At Figure 1.10 : Hydrocarbon Phase Envelope
Figures 1.5 and 1.6 can also be called the vapour conditions to the left of the bubble point curve we
pressure line. would have a mixture which is all liquid. To the
right of the dew point curve we would have all gas.
Within the envelope we would have a mixture of
So by reducing the temperature, increasing the gas and liquid. If the hydrocarbon mixture is all gas at point Z and
pressure or a combination of the two, any of our The dotted lines within the envelope are lines we begin to take away heat energy, the following
hydrocarbons can be liquefied. Some are more of quality. Each one represents a constant will happen:
difficull than others, of course. It is relatively simple percentage of liquid. At point X, we have a mixture
to liquefy butane at atmospheric pressure. We just composed of 50% liquid and 50% gas. The mixture would cool until point Y is reached.
have to cool it below -0.5°C (31.1°F). However liq- At this point the first drop of liquid would appear
uefying methane is much more difficult. At atmos- Let’s see what happens if we remove heat at from the component with the highest boiling
pheric pressure, the temperature here would have constant pressure from a mixture whose phase point. This is the dew point for this mixture at the
to be reduced below minus -162°C (-259°F), a very diagram is represented by Figure 1.10. Follow this corresponding
low temperature indeed. along the line Z V. pressure.

18
Petroleum
Petroleum
Open
Open
Learning
Learn-

When more heat is removed, more and more liquid Compression and cooling, or refrigeration, are
is formed as the temperature is reduced to the by far the most common methods used in the
boiling points of lighter hydrocarbons. When the recovery of gas liquids. There are other methods
temperature corresponding to point W is reached however, and in the introduction I mentioned
the mixture would be all liquid. Point W is of course Test Yourself 1.5 adsorption and absorption. Of these the
the bubble point of the mixture. adsorption process is more commonly used in
Assume that the substance indicated in gas dehydration operations. A separate unit in the
Although the behaviour of a hydrocarbon mixture is petroleum processing technology series covers
Figure 1.10 is a mixture of Heptane hexane
more complicated than that of a single component, this. Its use in natural gas liquids recovery is not
what I said earlier still applies. We can liquefy and pentane. Determine the temperatures at
very widespread so we will disregard it in this unit.
some or all of the components of a natural gas points W and Y if there was a temperature However the absorption process is often used so
stream by reducing the temperature, raising the rise from V to Z at atmospheric pressure. we must consider the theory behind this before
pressure or a combination of both. The actual looking at some of the processes in more detail.
amounts of liquid obtained will depend on the
composition of the stream and the degree to which Absorption
we can cool or compress the gas. NOTE:
Absorption is a process which involves contacting
All of this is a rather roundabout way of showing
The actual shape of the graph will vary the gas with a liquid called lean oil or absorption
how we can liquefy natural gas.
depending on the composition of the hydro- oil. When this is done some of the components
carbon mixture. I am using this general of the gas will dissolve in the oil. The heavier
•If we can create conditions of
shape for the purpose of the Test Yourself. components, the gas liquid components, will
temperature and pressure so that we
dissolve more readily but some of the lighter gas
cross the vapour pressure line from the
components will also be absorbed.
gas to the liquid state, the problem is
solved
It is quite simple to see that gas can in fact
dissolve in a liquid.
Before I move on to another method of recovering
If you take a sealed bottle of soda water and stand
some of the lighter hydrocarbons from a gas
it on a table, you will see a bottle of clear liquid with
stream, have a go at the following Test Yourself
no apparent activity within it.
question.

You will find the correct answers in Check


Yourself 1.5 on Page 58.

19
Petroleum
Petroleum
Open
Open
Learning
Learn-

If you now gently remove the cap of the bottle what The process is made more efficient at lower temperatures. (In general, gases are
happens? You should see the contents start to fizz more soluble in liquids at lower temperatures). Also, if the contact area between gas
and gas escape from the surface of the liquid. The and liquid can be increased the process will again be more efficient.
gas must have been dissolved within the liquid. In
fact the gas is carbon dioxide which is dissolved
in plain water. The process of removing the cap On an offshore oil production platform the
reduced the pressure in the bottle causing the gas recovered gas liquid components may be
to be liberated from the liquid. sent ashore still dissolved in the absorbing
oil. This oil is injected into the main crude
In order for the water to absorb the gas, they oil transportation pipeline. Of course, when
would need to be in contact with each other under it arrives at the reception terminal, further
pressure. You have probably seen this in practice. processing is necessary to recover the
The equipment which is sold for making fizzy natural gas liquids.
drinks at home uses this principle.

The gas ( carbon dioxide) is supplied in small An altemative to this is to remove all the
pressurised cartridges. When the cartridge is NGLs as gases from the absorbing oil at the
mated with a bottle of water and the cartridge processing location. These will then have to
punctured the carbon dioxide is forced into the be liquefied again for onward transportation
water under pressure. or storage.

You can see then that this process can be used The absorption process then is only part
to recover the gas liquids components from a of the story. In order to recover natural gas
gas stream by dissolving them into a liquid under liquids by this method a two stage process is
pressure. involved.

In addition to pressure two other factors influence First the NGL components of the total gas
the efficiency of the process. These are : stream are absorbed into the absorption oil.
Then the components are removed from the
• temperature oil as gas. The simple block diagram Figure
1.11 shows the principle of this.
• contact area between gas and liquid Figure 1.11 : Absorption Process Block Diagram

20
Petroleum
Petroleum
Open
Open
Learning
Learn-

The terminology associated with such a process I


have listed below :

• Rich gas Gas containing absorbable


( NGL ) components. Test Yourself 1.6
• Dry gas Gas having given up its
absorbable components. a) Classify the following substances as either solid, liquid or gases at room temperature.

• Rich oil Oil having absorbed the steel - water - acetylene - copper - oxygen.
NGL gases.

• Lean oil Oil having had the absorbed b) Match the following symbols with the list of element names.
NGL gases removed.

• Absorber Process vessel where CI - a - Sl - He - Fe - C.
absorption takes place.
Carbon - Hydrogen - Iron - Oxygen - Chlorine - Sodium - Silicon - Helium.
• Stripper Process vessel where
removal of NGL gases from
lean oil takes place. c) Using Figure 1.10 match the letters in the multi-component phase diagram with the
descriptions below.
Of course the actual process is somewhat more 1. The 50% liquid line.
complex than the simple example I have just given
2. The gas region.
you.
3. The bubble point curve.
In Section 4 we will look at one such process in 4. The dew point curve.
more detail. 5. The liquid region.

Before moving on to the next section though, have a You will find the correct answers in Check Yourself 1.6 on Page 58.
go at the final Test Yourself in this section.

21
Petroleum
Petroleum
Open
Open
Learning
Learn-

Summary of Section 1
In this section we have been looking at some Using the example of water, you saw that adding or We defined boiling point and vapour presure.
of the basic theory behind natural gas liquids removing energy from a substance can change the We then showed how this idea of phase behaviour
recovery. state of the substance. Ice can be turned to water could be applied to a multi-component mixture.
and then to steam and vice versa. Heat energy and We used a modified phase diagram to illustrate
Having defined natural gas in the introduction, pressure energy are responsible for these changes this, which we called a phase envelope.
we went on to consider the simple chemistry of of state.
hydrocarbons. All this helped to show that, by decreasing the
You saw that the three states of matter, solid, liquid temperature and increasing the pressure of natural
You saw how the elements hydrogen and carbon and gas are referred to as phases. The effects that gases, certain of the components can be liquefied.
can bond together to form the hydrocarbon temperature and pressure have on a substance is The actual amounts of liquid recovered from a gas
compounds of natural gas. called its phase behaviour. of a certain composition will depend on the degree
You also saw that these compounds form the to which these energy levels are altered.
hydrocarbon series called the paraffin or alkane We prepared diagrams to illustrate this effect called
series. Although members of other hydrocarbon phase diagrams. Firstly we concentrated on pure At the end of the section we looked at an
series, together with impurities, can be present in substances or single components. The region alternative method of recovering natural gas liquids
natural gas, in this unit we concentrated on the which we were most interested in was the change by absorption. This is a process which
alkanes. in state from liquid to gas and from gas to liquid. involves contacting the gas with an oil under
You saw that the line on the phase diagram which pressure. The NGL components of the gas are
Having covered the very basic chemistry of separates these regions can be called the bubble absorbed by the oil and then recovered by
hydrocarbons we went on to look at the physical point curve, the dew point curve or the vapour stripping.
properties of these compounds. pressure line.
In the next section we will look at the simplest of
the recovery processes, that of compression and
cooling.

22
Petroleum
Petroleum
Open
Open
Learning
Learn-
Natural Gas Liquids Recovery
Section 2 - NGL Recovery Using Compression and Cooling
As we showed in Section 1, NGL recovery by Cooling the gas also promotes a phase change
compression and cooling is the simplest and of some of the gas components. More liquids are
cheapest method available. In fact, natural gas produced!
liquids are often produced as a by-product during
petroleum gas compression operations. So, in a gas compression facility we are going to
Test Yourself 2.1
get some liquids produced whether we want them Ethane, Propane and Pentane, are leaving a
Imagine an oil production plant where a or not. But these liquids are valuable, so in some
separator at atmospheric pressure and 38°C
considerable amount of associated gas is also instances, it is worth recovering them for sale.
produced. This gas is then available for sale. (100°F). By how much would each have to
Let’s say that the separation facilities on this Unfortunately, this type of process is relatively be cooled in order to liquefy them?
plant involve 2 stages of separation operating at inefficient in recovering the maximum amount of
pressures of 25.9 bar (375 psi) and 5.2 bar (75 NGL. Where large quantities of gas or NGL are in-
psi). Gas from the separators will require to be volved it is more common to increase the recovery
dehydrated and then routed through gas export of gas liquids by reducing the temperature much
compressors into the pipeline for transportation further than is necessary simply to protect the com-
to the point of sale. If we say that a gas export pression equipment. This is done by refrigerating
pressure of 138 bar (2 000 psi) is required, then the gas rather than just cooling it. We will look at
it is obvious that the gas from each stage of some plants which do this in the next section.
separation will have to be compressed in order to
achieve this. However, in this section let’s look at a typical
Compression and Cooling System which is used to
You will remember from Section 1 that recover NGL from the gas stream leaving a crude You will find the correct answers in Check
compressing the gas is likely to promote a phase oil separation plant. Yourself 2.1 on Page 59.
change of some components. So, the very fact
that we have to compress the gas from the Before you move on, have a go at the following
separators will cause NGLs to be produced. Test Yourself question.

The process of gas compression also causes its


temperature to rise. If the gas is being compressed
in a number of stages this increase in temperature
can cause damage to the compression machinery.
(The gas compression programme in the
petroleum processing technology series covers
this in more detail.) To prevent damage to the
compressors the gas must be cooled after each
stage of compression.
23
Petroleum
Petroleum
Open
Open
Learning
Learn-

A Typical Compression and


Cooling System
Take some time to study Figure 2.1
which shows a typical compression and
cooling NGL recovery system. We will
follow the flow through the system.

Figure 2.1 : Compression and Cooling System

24
Petroleum
Petroleum
Open
Open
Learning
Learn-

In our example the gas to be treated is from a two


stage Separation System. The gas is compressed
in two stages by two centrifugal compressors. The
compressors are positioned on a single shaft and
driven by a single motor.

The temperatures and pressures which I will use


are purely hypothetical and are not meant to
represent a specific process plant.

Let’s follow the first part of the process now.

Look at Figure 2.2. It is simply the first part of the


overall system which I have reproduced for greater
clarity. First of all identify where the gas from the
low pressure or second stage separator enters the
system.

The gas leaves the 2nd stage Separator at a


pressure 5.2 bar (75 psi) and temperature of,
say, 80°C (176°F) and enters the 1st stage
compressor suction knockout drum.

This drum is a vessel whose job it is to remove


any droplets of free liquid from the gas. It is an
essential part of any gas compression facility. If
any free liquids reach the compressor, serious
damage could be done to the equipment. You may
also come across the term suction gas scrubber
to describe the knockout drum.

Figure 2.3, on the next page, shows a typical Figure 2.2 : 1st Stage Gas Compression
suction knockout drum.
25
Petroleum
Petroleum
Open
Open
Learning
Learn-

The drum in our system is a vertical steel cylinder which has an


inlet connection approximately three quarters of the way up.

Gas enters the vessel and impinges on a baffle plate which is


positioned opposite the inlet nozzle. This causes a change in
flow direction and reduction of gas velocity. At this point liquid
droplets separate from the gas and fall to the bottom of the
vessel.

These liquid drops will be a mixture of condensed gas liquids


and water.

The liquid accumulates at the bottom and is taken out of the


drum under level control. The recovered liquid is dumped to a
Closed Drain System. From there liquids will be pumped back
to the 2nd stage Separator. Already some small amounts of gas
liquids have been recovered.

The gas rises towards the outlet of the drum and passes
through a mist extractor. As its name suggests this unit is
there to extract any very fine droplets of liquid which may
remain in the gas in the form of mist. It consists of a number of
knitted wire mesh pads through which the gas must flow. As the
gas flows through, any mist droplets impinge on the wire mesh
and stick to it. Further small droplets coalesce until they are
large enough to fall down to the bottom of the vessel.

The gas, now completely free of liquid, leaves the 1st stage
Compressor Suction K.O. Drum through the dry gas outlet, at
the top of the vessel.

Figure 2.3 : Suction Knockout Drum

26
Petroleum
Petroleum
Open
Open
Learning
Learn-

The liquid level in the knockout drum is controlled Normally however the gas flows forward to the 1st stage Compressor. This raises the pressure from 5.2 bar (75
via a level controller (LC) and control valve. A sight psi) to 25.9 bar (375 psi). During compression the temperature of the gas increases from 80°C (176°F) to 127°C
glass or level gauge (LG) is also incorporated in (260°F). If the gas is not cooled at this point the efficiency of the next compressor would be affected.
the level control instrumentation.
So this hot, medium pressure gas, is cooled by passing it through a heat exchanger. In our system, the heat
Further instrumentation protects the system exchanger is of the shell and tube type. Figure 2.4 shows an exchanger of this type.
against loss of level control. A level switch low (
LSL ) activates an alarm if the level should go too
low. Similarly a level switch high ( LSH) activates a
high level alarm.

If, despite the activation of a high level alarm


the level of liquid in the vessel continues to rise,
a dangerous situation could occur. Too high
a level could result in carry over of liquid into
the compressor with possible serious damage
being done. This situation is prevented by
having a shutdown incorporated into the vessels
instrumentation.

A level switch high high ( LSHH ) activates a


signal which shuts down the system, preventing
potential hazardous situations arising.

A pressure controller on the top of the drum


controls the pressure of gas in the vessel. If the
pressure is too high and the compressor is not
pulling the pressure down for whatever reason,
then the pressure controller opens the valve and
allows excess gas to escape to the flare. Figure 2.4 : Shell and tube Heat Exchanger

27
Petroleum
Petroleum
Open
Open
Learning
Learn-

The hot gas passes through the inside of the


tubes where it is cooled by sea water passing
through the shell.
Test Yourself 2.2
The temperature of the gas leaving the heat
exchanger is controlled by a temperature controller
(see Figure 2.2). The temperature controller opens
and closes a valve to control the flow of cooling insert the correct word or words from the list provided into the blanks in the following sentences.
water through the unit. If the gas temperature
leaving the exchanger is too high the temperature
controller will open the valve and allow more sea 1. The process of gas compression causes the gas ................................................... to rise.
water to flow through. If the gas temperature is too
low the temperature controller will close the valve
and allow less sea water to flow through the heat
exchanger. 2. The suction gas scrubber removes ................................................... from the gas.

Go back over the process so far and satisfy


yourself that you understand it before having a go 3. Within a suction K.O. drum wire mesh pads form a .....................................................
at the following Test Yourself question.

4. In a .................................and tube heat exchanger, gas flows through the tubes and water
flows through the shell.

List of words.

wier - mist extractor - solids - temperature - liquids - shell - filter

You will find the correct answers in Check Yourself 2.2 on Page 59.

28
Petroleum
Petroleum
Open
Open
Learning
Learn-

Medium pressure gas leaves the 1st stage


Separator at a pressure 25.9 bar (375 psi) and
a temperature of, say, 82°C (180°F). The two
streams, now mixed together, enter the 2nd stage
We have now increased the Compressor Suction K.O. Drum.
pressure and cooled the gas.
More gas liquids will have This drum removes any liquid which is in the gas
formed. At this point the mixture coming from the 1st stage separator, plus any
is joined by the gas which is liquids which have been formed during the
leaving the 1st stage Separator. compression and cooling of the original gas from
the 2nd stage separation.

The 2nd stage Compressor Suction K.O. Drum


operates at a higher pressure than the 2nd stage
Separator. (Remember, it operates at 5.2 bar (75
psi)) The recovered liquid can therefore be piped
straight back to the 2nd stage Separator.

The gas, again completely free of liquid, leaves the


2nd stage Compressor Suction K.O. Drum
where a pressure controller controls the pressure
of gas in the vessel. It operates in a similar manner
to the one previously described.

The gas then enters the 2nd stage Compressor.

The 2nd stage Compressor compresses the gas


to a pressure 50 bar (725 psi) and the temperature
Figure 2.5 shows this part of the process. will rise to, say, 138°C (280°F).
You can see that it is almost identical to
part 1.

Let's quickly go through this part now.

Figure 2.5 : 2nd Stage Gas Compression


29
Petroleum
Petroleum
Open
Open
Learning
Learn-

The hot, high pressure, gas is then cooled by The gas and liquid mixture enters the NGL
passing it across a shell and tube heat exchanger Knockout Drum which separates the two phases.
similar to the one used to cool the 1st stage This vessel is yet another separator which works
discharge gas. in the same way as the Compressor Suction K.O.
Drum. In this case however the vessel is a horizon-
The temperature of the gas leaving this heat tal cylinder.
exchanger is again controlled by a temperature
controller. The gas leaving the NGL Knockout Drum is routed
to dehydration. The liquids accumulate at the bot-
The pressure of the gas is maintained at 50 bar tom of the vessel where the level is controlled by a
(725 psi) as it passes the heat exchanger but the level control system.
temperature of the gas is lowered to, say, 29°C
(85°F). This combination of high pressure and In our case, the NGLs will be transported from the
relatively low temperature causes the heavier plant via the main oil pipeline operating at, say,
hydrocarbons to turn into a liquid. All that remains 69 bar (1 000 psi). This injection of NGLs into the
now is to separate these liquids from the gas crude is often referred to as spiking the crude with
before the gas is dehydrated, further compressed NGLs. The Knockout Drum operates at a lower
and exported. pressure than the main oil pipeline. In order to
get the NGLs into the line, the pressure must be
There will be no NGL knockout after the export boosted by an NGL Booster Pump.
compressor discharge gas is cooled, as the
pressure will be above the point at which any The level controller on the NGL Knockout Drum
liquid could exist. is different to the previous ones we looked at. It
works in conjunction with the booster pump. The
control valve is positioned downstream of the
pump. If the level in the vessel goes down the
This brings us to the final part of this simple controller opens the control valve and allows liquid
system which is shown in Figure 2.6. to circulate back to the drum. If the level goes too
high all the NGLs are pumped into the crude line.

Figure 2.6 : NGL Knockout Drum


30
Petroleum
Petroleum
Open
Open
Learning
Learn-

Summary of Section 2
In this short section we have looked at a simple process used to recover liquids from
natural gas.

In the section we looked at a typical Compression and Cooling System and saw:

• what equipment was used

• how the NGL was recovered

• how the process was controlled

• what happened to the NGL after it had been recovered

As I pointed out in this section the simple compression and cooling process recovers
a limited amount of liquids. To make the process more efficient, the degree of
cooling must be much greater. We need to refrigerate the gas. In the next section
we will look at some methods of doing this.

31
Petroleum
Petroleum
Open
Open
Learning
Learn-
Natural Gas Liquids Recovery
Section 3 - NGL Recovery by Refrigeration
In this section we are going to take the recovery When a substance is liquefied by compression, heat energy is liberated. In a refrigerator this heat is dissipated
process one step further. More NGL recovery through a condenser which resembles a car radiator. The liquid then passes through an expansion valve where
can be achieved if the temperature is reduced it vapourises. The necessary heat for vapourisation comes from the surroundings (in this case, the refrigerator
to a greater degree than that obtained by simply contents) which are thereby cooled. The refrigerant is then recycled.
cooling. In order to reduce the temperature of
the gas even more, it is necessary to refrigerate Figure 3.1 is a block diagram which shows this process very simply.
it. There are two methods commonly used to do
this. The first method can be called mechanical
refrigeration and the second, auto refrigeration.
We can look at these two processes in turn . Let’s
start with mechanical refrigeration.

Mechanical Refrigeration
Mechanical refrigeration systems use a similar
process to that found in a domestic refrigerator.

Before we look at a refrigeration process to recover


NGL let’s take a look at an ordinary domestic
refrigerator, and see how it works.

We have already seen that many substances


that are gases at normal temperatures can be
liquefied by increasing their pressure. The liquids
can then be vapourised again by increasing the
temperature. In a refrigerator a substance which is
liquefied then vapourised is called a refrigerant.

Fig 3.1 : Mechanical refrigeration Block Diagram

32
Petroleum
Petroleum
Open
Open
Learning
Learn-

In the next figure Figure 3.2, the main On the outside of the refrigerator we have
components of an ordinary domestic
refrigerator are shown. Take a look at •An electric power supply which drives an electric
the figure and identify the following motor.
components.
•A compressor which is driven by the electric motor.

•A compressor suction line which takes refrigerant


gas into the compressor.

•A compressor discharge line with a check valve


which stops anything flowing back through the
compressor.

•A set of condensing coils where the heat in the


refrigerant is dissipated and the refrigerant turns
from a gas into a liquid.

•An accumulator which stores a small amount of


liquid refrigerant.

Inside the refrigerator, in the freezer section, we have:

•An expansion valve where the liquid refrigerant


‘flashes off’, changing into a liquid/gas mixture.

•A set of chilling coils where the refrigerant is boiled


off, completing the change from a liquid to a gas.

Fig 3.2 : Domestic Fridge Schematic Diagram


33
Petroleum
Petroleum
Open
Open
Learning
Learn-

Inside the refrigerator, in the cooler section, there is a


temperature switch which will switch off the electric power
to the electric motor when the temperature falls to a pre-
set level, and, switch on the electric power to the electric
motor when the temperature rises to a pre-set level.

Now that you have identified the refrigerator components


we can now look at what is going on inside the unit.

Starting from the discharge of the compressor Figure 3.3


shows what is happening as follows:

The refrigerant leaving the compressor is a hot, high


pressure, gas. As the refrigerant gas passes through
the condensing coils it loses heat and turns into a liquid.
It arrives in the accumulator as a relatively cool, high
pressure liquid.

From the accumulator it flows through the expansion


valve where the pressure is reduced. This expansion
causes the refrigerant liquid to start to vapourise and
become a cold, low pressure, gas/liquid mixture.

This mixture now enters the chiller coils. The remaining


liquid is boiled off by heat entering the chiller coils from
the air in the freezer section of the refrigerator. This
completes the vapourisation of the liquid refrigerant. The
refrigerant gas is now at low pressure and temperature.

Fig 3.3 : Domestic Fridge


Schematic Diagram

34
Petroleum
Petroleum
Open
Open
Learning
Learn-

This gas enters the compressor. The


compressor raises the pressure and the
temperature of the gas and the process starts all
over again.

The system will keep running, circulating a


Test Yourself 3.1
limited amount of refrigerant around the system,
until the temperature falls to the point where the
Starting with ‘Refrigerant leaves the compressor’, put the following steps of a refrigeration
temperature switch operates and stops the electric
motor.
process in the correct order.

It will stay switched off until the temperature in the 1. Refrigerant leaves the compressor. ..............................
cooler section rises and starts the electric motor
again. 2. Refrigerant flows through accumulator. ..............................

Think about the operation of a fridge and have a go 3. Refrigerant enters condensing coils. ..............................
at the following Test Yourself question.
4. Refrigerant enters compressor. ..............................

5. Refrigerant starts to vapourise and becomes cold. ..............................

6. Refrigerant loses heat and liquefies. ..............................

7. Refrigerant completely vapourises. ..............................

8. Refrigerant mixture enters chiller coils. ..............................

9. Refrigerant flows through expansion valve. ..............................

Now let’s see how we can apply the theory of


operation of a domestic refrigerator by comparing You will find the correct answers in CheckYourself 3.1 on Page 60.
it with a typical natural gas liquids recovery plant.

35
Petroleum
Petroleum
Open
Open
Learning
Learn-

A Typical Mechanical Refrigeration


NGL Recovery System

Figure 3.4 shows a typical Mechanical Refrigeration


System which uses Propane as the refrigerant.

Figure 3.4 : Mechincal Refrigeration NGL Recovery System

36
Petroleum
Petroleum
Open
Open
Learning
Learn-

The boiling point of Propane is -42°C (-43.7°F). It The first point of contact is the chiller. It is in this
is often used as a refrigerant in gas processing. vessel that the low temperature created by the re-
It is not used in domestic refrigerators because frigeration system is used to refrigerate the natural
it is flammable. If leaked, it could cause a fire or
explosion.
Activity gas.

The second one is the refrigerant condenser


Within this system, there are actually two where the heat generated during refrigerant com-
interconnected sub-systems. They are: Take a close look at Figure 3.4. See if you pression is dissipated.
can determine where these two points of
• the refrigeration system which reduces contact are.
the temperature of the refrigerant
Let’s now work our way through the two systems
• the natural gas system where the starting with the refrigeration system. We can go
reduction in temperature of the through this rather quickly as it is very similar to the
refrigerant is used to recover the NGL domestic refrigerator we looked at earlier.

There are two points of contact between the two


systems.
Take a look at Figure 3.5 on the next page. It is the
refrigeration system from Figure 3.4 which I have
isolated from the rest of the drawing.

37
Petroleum
Petroleum
Open
Open
Learning
Learn-

Some of the liquid refrigerant leaves the


acumulator and flows across the expansion valve,
where partial vapourisation takes place. This gas/
liquid mixture now flows into the chiller.

Warm natural gas is passing through the coils in


the chiller, as we will see shortly. This is the gas
which needs to be refrigerated in order to form
NGL. The heat content of this gas is transferred to
the refrigerant in the chiller. The net effect of this is
to vapourise some of the liquid refrigerant, whilst
reducing the temperature of the natural gas. The
amount of natural gas flowing through the chiller
will vary. Therefore, the amount of liquid refrigerant
vapourised in the chiller will vary, affecting the
liquid level. This level is maintained by a level
controller operating the expansion valve.

The refrigerant leaves the chiller as a cold, low


pressure gas and enters the compressor suction
knockout drum.

This drum removes any entrained liquids which


may be carried over by the refrigerant gas from the
chiller.
Figure 3.5 : Refrigeration System
The refrigerant gas leaves the compressor suction
We will start at the same point as we did with the As it passes over the refrigerant condenser the knockout drum and enters the compressor as a
domestic refrigerator, that is, at the discharge of refrigerant exchanges heat with cold gas coming cold, low pressure gas.
the compressor. from the gas/gas heat exchanger.
After compression the refrigerant leaves the
The refrigerant leaving the compressor is a high The refrigerant condenses to high pressure, compressor as a hot, high pressure gas and the
pressure, high temperature gas. relatively cool liquid as it enters the refrigerant process starts again.
accumulater,

38
Petroleum
Petroleum
Open
Open
Learning
Learn-

Now let’s go through the actual gas liquids


recovery part of the system. Figure 3.6 is once
again the relevant part of Figure 3.4 which has
been isolated from the rest of the drawing.

The natural gas enters the NGL recovery system


after it has been dehydrated. This is to remove any
chance of hydrate formation as the gas passes
through the refrigeration process. (A separate unit
in the Petroleum Processing Technology Series
covers dehydration).

The incoming natural gas stream enters the gas


/ gas heat exchanger. It is called a gas / gas
exchanger because both the medium to be cooled
and the cooling medium are gas.

Figure 3.6 : NGL Recovery System

39
Petroleum
Petroleum
Open
Open
Learning
Learn-

As the warm incoming natural gas enters the The NGLs accumulate at the bottom of the NGLs
system it passes through the exchanger tubes gas separator, where a liquid level is maintained by
and is cooled by the cold natural gas leaving the a level controller.
system through the exchanger shell. This pre-cools
the incoming gas and pre-warms the outgoing gas. From there they flow to an NGL flash separator.
In this vessel any small amounts of light
The pre-cooled gas leaves the gas/gas exchanger hydrocarbon gases which may still be entrained in
and enters the refrigerant chiller. As it flows the NGLs are separated. These gases are usually
through the tubes in the chiller, it is refrigerated to disposed of via a flare system.
a temperature of say -15°C (-59°F).
From there, the NGLs are pumped via a level
As the natural gas is refrigerated, the heavier control valve to a transport system for sale.
hydrocarbons turn into natural gas liquids. This
mixture of residual cold gas and liquids then The system we have just been looking at is a
enters the NGL/gas separator where the NGL is very common system which you could find on an
separated from the gas. offshore oil and gas production platform. However
there are other systems. I mentioned one earlier
The cold gas leaves the separator from the top where refrigeration is achieved using what is called
of the vessel and flows through the gas/gas heat an auto refrigeration system. We will look at
exchanger where, as we have seen, ~ cools down one of these systems shortly. Before you move
the incoming warm gas. on to this however have a go at the following Test
Yourself question on mechanical refrigeration.
Next the gas flows through the refrigerant
condenser. Here it condenses the refrigerant and
is itself warmed up further to a level where it can
be put into a pipeline as a sales gas.

Note : Depending on operating pressures,


this gas may require to be further
compressed prior to being exported as
sales gas.

40
Petroleum
Petroleum
Open
Open
Learning
Learn-

Auto Refrigeration
In this part of Section 3 we are going to look at an
alternative method of reducing the temperature of
Test Yourself 3.2 gas in order to recover NGLs. It is a method which
uses one of the properties of the gas itself. This
property can be stated as follows :
Are the following statements true or false ?
TRUE FALSE If a natural gas is rapidly expanded by reducing
1. Heat generated during refrigerant compression is its pressure, its temperature will drop.
dissipated in the refrigerant condenser.
This temperature drop associated with rapid gas
expansion is known as the Joules/Thompson
2. . The chiller is used to cool down the refrigerant. Effect. The greater the pressure drop, the greater
the temperature reduction. You can see this effect
illustrated on next page in Figure 3.7.

3. As liquid refrigerant flows across the expansion


valve partial condensation takes place.

4. Incoming natural gas flows through a heat exchanger


which uses liquid refrigerant as the cooling medium.

5. In the NGL flash separator, small amounts of entrained light


hydrocarbon gases are removed from the NGL stream.

You will find the correct answers in Test Yourself 3.2 on Page 60.

41
Petroleum
Petroleum
Open
Open
Learning
Learn-

This chart is only by way of an example, it will not


be accurate for all gases. Also, temperature and
pressure are shown in imperial units, however it
serves to illustrate the use of such charts. You
can use this chart for SI Units by converting them
to imperial units, then use the graph to find your
answer, and then convert your answer back into SI
Units.

In order to find a temperature drop associated


with a given pressure drop, just follow the steps I
have listed here for you.

• Find the point on the horizontal axis of


the graph which corresponds to the initial
pressure before expansion takes place.

• From this point move vertically upwards to


where this pressure intersects the pressure
drop curve. (Note that on the graph the
pressure drop is indicated as A P i.e. initial
pressure - final pressure.)

• From the intersection point, move horizontally


to the left hand vertical axis of the graph.

• Read off the temperature at this point.

Let's do that with some actual figures.


Figure 3.7 : Typical Natural Gas Expansion - Temperature Reduction Curves

Note: take care to note that pressure and temperature on this graph are given in psig and fahrenheit only.

42
Petroleum
Petroleum
Open
Open
Learning
Learn-

Work through with me, the following example,


where I will start of with SI Units and convert them
to imperial to use the graph and then convert my Test Yourself 3.3 Test Yourself 3.4
answer back to SI Units. :
If gas at an initial pressure of 207 bar (3000 psi) is
expanded to a final pressure of 69 bar (1 000 psi), If a gas at an initial pressure of 310 bar (4 What could be the advantage in using the
what will be the temperature drop? 500 psi) is expanded 10 34.5 bar (500 psi), turbine ?
what will be the drop in temperature?
• First calculate the pressure drop.
The pressure drop will be (3000-1 000) =
2 000 psi. (Use Figure 3.7)

• Find the point on the horizontal axis of the


graph of an initial pressure of 3 000 psi. Check your answer in Check Yourself
3.3 on Page 60. Check your answer in Check Yourself
• Move vertically upwards to intersect the 3.4 on Page 60.
AP = 2 000 curve.

• From the intersection point move horizontally


We have seen that a pressure drop will reduce the We will now take a look at an Auto-Refrigeration
to the left to read off the temperature drop on
temperature of a gas. But how do we create the system which uses both the valve and the turbine
the left hand vertical axis.
drop in pressure? Basically there are two methods in the system.
in common use.
• You should have found that the temperature
drop is 70°F, which you can convert to 21°C.
• expand the gas across a valve
Of course, the actual pressure drop will not always
• expand the gas across a turbine
coincide exactly with one of the curves. In such a
case a curve parallel to the nearest printed curve
In fact expanding the gas across a turbine makes
needs to be drawn or imagined. Now have a go at
the gas do work as it expands and the cooling
Test Yourself 3.3.
effect is greater. Think about using the turbine to
create the pressure drop and have a go at the
following
Test Yourself.

43
Petroleum
Petroleum
Open
Open
Learning
Learn-

First of all, you should appreciate that a gas must


be at a high enough pressure initially in order that
the required pressure drop can take place. It is
seldom worth compressing the gas and raising
its pressure just to drop it again as part of an
Auto-Refrigeration system. The cost involved in
purchasing and running compressors for such a
system is far too high under normal circumstances.

Because of this, auto-refrigeration systems are


less common than mechanical refrigeration
systems.

Figure 3.8 is an illustration of an auto refrigeration


system which uses both forms of expansion to
recover NGL.

Take a look at the drawing and identify the


following components:

• liquid / gas separator Fig 3.8 : Auto Refrigeration


NGL Recovery System
• Joules Thompson (JT) valve

• turbo-expander driving a compressor

• recovered NGL knockout drum

• make up compressor

44
Petroleum
Petroleum
Open
Open
Learning
Learn-

Now let’s follow the flow through the system and The gas leaving the recovered NGL knockout drum
see how the process works. flows to the suction of the compressor connected
to the turbo-expander. As the gas is compressed it
In this case the gas entering the system comes is heated due to the compression process and is
from a high pressure pipeline, although in other warm enough, depending on operating pressures,
cases it may come from gas dehydration. to leave the system as a sales gas.
It first of all enters a liquid / gas separator. Here,
any liquids which may have condensed in the If the turbo-expander is out of commission then
pipeline are separated. These liquids which may the JT valve acts as a secondary expansion valve.
consist of water and NGLs are dumped to the NGL This system will not recover as much NGL as the
Knockout Drum. Turbo expander but it will allow the process to keep
running and will recover some of the NGL.
If there is any water in the gas at this stage,
methanol will require to be injected upstream of the When the JT valve is in operation, the make-up
expansion process to prevent hydrate formation compressor will be required to boost the pressure
caused by the resultant chilling effect. of the gas leaving the recovered NGL knockout
drum to sales gas pressure.
The gas leaves the separator and, normally enters
the turbo-expander. As it passes across the turbo- This is a typical example of an Auto-Refrigeration
expander it is made to do work by driving the system. You may find that some systems have an
compressor which is attached by a shaft. export compressor to compress the gas up to gas
pipeline pressure.
As the gas is expanded across the turbine its
temperature is reduced and NGLs are condensed.

The NGL is separated from the gas in the


recovered NGL knockout drum and, under level
control, is despatched to NGL Sales. Any water
which might be present, is also separated and sent
to disposal.

45
Petroleum
Petroleum
Open
Open
Learning
Learn-

Summary of Section 3
In this section we have looked at what is probably I pointed out that there are two points of contact We performed some simple calculations to show
the most common natural gas liquids recovery between the two systems. The first one is the this effect using a graph of gas expansion /
system found in offshore petroleum producing chiller. It is in this vessel that the low temperature temperature reduction curves.
operations. i.e. NGL recovery by refrigeration. created by the refrigeration system is used to chill
the natural gas. A system which uses this effect to achieve
As you worked through the section you first of all temperature reduction is known as an auto
saw how a simple domestic refrigerator worked. The second one is the refrigerant condenser refrigeration system.
You saw that when a substance is liquefied where the heat generated during refrigerant
by compression, heat energy is liberated. In compression is dissipated by the chilled gas. You saw that there are two methods used to
a refrigerator this heat is dissipated through a achieve this auto refrigeration.
condensor. The liquid then passes through an We then worked through the plant using a typical
expansion valve where it vapourises. The flow diagram. • expand the gas across a valve
necessary heat for vapourisation comes from
the surroundings which are thereby cooled. The We then looked at an alternative method of • expand the gas across a turbine
refrigerant is then recycled. reducing the temperature of gas in order to recover
NGLs. A method which uses one of the properties Finally we followed the flow diagram of a simple
We then applied this knowledge to a mechanical of the gas itself. This property can be stated as plant which uses the two types of expansion to
refrigeration system used to recover NGLs. follows: achieve refrigeration and hence
NGL recovery.
You saw that within this system, there are two If a natural gas is rapidly expanded by reducing
interconnected sub systems which are: it’s pressure, it's temperature will drop. In the next section we will look at one more method
of recovering the natural gas liquids. This uses an
• the refrigeration system which reduces This temperature drop associated with gas absorption process.
the temperature of the refrigerant expansion is known as the Joules/Thompson
Effect. The greater the pressure drop, the greater
• the natural gas system where the reduction the temperature reduction.
in temperature of the refrigerant is used to
recover the NGLs

46
Natural Gas Liquids Recovery Petroleum
Petroleum
Open
Open
Learning
Learn-

Section 4 - NGL Recovery Using an Absorption Process

In Section 1 you saw that contacting gas with a


lean oil causes the heavier gas components to be
absorbed into the oil. These can then be recovered
and transported as NGLs. Test Yourself 4.1
In this section we are going to look at a typical
system which uses this type of process. As usual
we will follow a drawing of a system and see how Figure 4.1 is the same figure that you saw in Section 1.
the plant is constructed and how it works. Where I described a simple absorption system.

Before we start however, see if you can remember You will see that there are no labels
the simple process which I described in Section 1.
Complete the figure with the
correct labels.

Now have a go at the following Test Yourself


question.

You will find the correct answer in


Check Yourself 4.1 on page 6.1 Figure 4.1 : Absorption system

47
Petroleum
Petroleum
Open
Open
Learning
Learn-

A Typical Absorption System

Figure 4.2 is a line drawing of a typical


system. Study the drawing for a while and
identify the various components.

Figure 4.2: Absorption System

48
Petroleum
Petroleum
Open
Open
Learning
Learn-

You will probably have noticed that the system is, in essence,
the same as the very simple system which you labelled in your
answer to Test Yourself 4.1. At the heart of the system are the
two towers, the absorber tower and the stripper tower.

You will notice that there is a continuous circulation of oil round


the system. This is the oil which absorbs the NGLs as the gas
is contacted with it. We will refer to it as Absorption Oil. It is a
hydrocarbon liquid and in some cases it can be dead crude oil.

In our example the absorption oil is classed as lean oil or rich


oil as it flows around the circuit. The lean oil is oil which has
been stripped free of NGL gases. The rich oil is oil which has
absorbed the NGL gases.

Before we start to follow the process flow, let’s take a look at


the absorber tower which as I said earlier is at the heart of the
system.

Look at Figure 4.3 which shows a typical absorber tower. The


figure is a simplified cutaway view which shows the internal
features.

As you can see, the absorption oil enters the tower through the
inlet pipe near the top. It is spread out through spray bars and
a distribution plate before flowing down the tower. It flows down
through two sections packed with devices which increase the
surface area of the liquid. (You will remember from Section 1
that if the contact area between gas and absorption oil can be
increased the process will be more efficient).

Figure 4.3 : Absorption Tower

49
Petroleum
Petroleum
Open
Open
Learning
Learn-

The selection of the packing devices for any type of contactor is dependent upon: The reason for having two separate sections is to ensure that
the liquid and gas flows can be evenly re-distributed. If all the
• the gas/liquid to be contacted liquid went down one side of the tower and all the gas went up
• the flow rate requirements the other side there would be very little contact between the
• the size of the column two.
• the materials of manufacture for each type of packing
You will notice that halfway down the tower there is a re-
Figure 4.4 is an illustration of a few of the different kinds of packing devices. When the liquid distribution plate. After the oil has passed the upper section of
flows over them it forms a thin film and increases its surface area. packing it accumulates on the plate. It then flows through small
caps which spread the oil out over the next section of packing.

The lower section of packing is supported on a plate located


above the gas inlet. When the oil has percolated through all the
packing and the support plate, it accumulates in the bottom of
the tower. A liquid level is maintained and the oil exits the tower
from the outlet under level control.

The gas enters the absorber tower just beneath the lower
packed section. It flows upwards through the two packed
sections and finally leaves the tower through the gas outlet at
the top of the vessel.

During its passage up the tower the gas is forced to intimately


mix with the downcoming oil. It is this mixing under the right
conditions which cause the NGLs to be absorbed into the oil.

Let’s now go back to Figure 4.2 and work our way through the
process. We will start with the gas flow through the system then
follow the flowpath of the absorption oil.

Figure 4.4 : Absorption Tower Packing Devices

50
Petroleum
Petroleum
Open
Open
Learning
Learn-

Gas Flow The first feature you should note is that the absorber can be
by-passed. This may be required when the absorption system
Once again I have isolated the gas flow section from the is out of action for repair or maintenance.
rest of Figure 4.2 in order to simplify the explanation.
Under normal operating conditions the gas enters the tower at
the bottom and flows upwards. As the gas rises through the
absorber tower, the heavier gases (the NGLs) are absorbed
into the absorber oil which is flowing down the tower. The
gas, stripped of NGL, leaves the tower at the top. If there is
sufficient gas, it may be transported for sale. If the amounts of
gas are too small to justify collection for sale, it may be used
as fuel with the surplus being flared.

Under normal operating conditions there is a small differential


pressure from top to bottom across the packed sections. If the
packing becomes dirty, the differential pressure could rise and
the gas flow could cause the packing to be dislodged. This
may cause channelling through the packing resulting in a loss
of efficiency. To guard against this the differential pressure is
monitored by a differential pressure transmitter PDT 01.
This will give an alarm before the packing begins to lift.

Figure 4.5 : Absorption System Gas Flow

51
Petroleum
Petroleum
Open
Open
Learning
Learn-

Before we continue, try Test Yourself 4.2.


From the answer to the Test Yourself you can see
that it is necessary to maintain a pressure in the
tower. This is done by a back pressure controller
PC 01 opening and closing pressure control
Test Yourself 4.2 valve PCV 01 in the gas outlet line.

We said that increasing the contact area The flow of gas is reasonably straightforward.
between gas and liquid increases the We can now go on to look at the flowpath of the
efficiency of the absorption process. State absorption oil as it circulates round the system.
two other factors which influence the
efficiency.

You will find the correct answers in


Check Yourself 4.2 on Page 61.

52
Petroleum
Petroleum
Open
Open
Learning
Learn-

Absorption Oil Flow


Look at Figure 4.6 which shows the route taken by
the absorption oil flowing round the system.

We are going to start at the oil circulation pumps


so locate these on the drawing. At this
point the oil is lean. You will remember that this
means that the absorption oil has had all the NGLs
stripped from it.

The pumps are used to increase the pressure of


the lean oil which enables it to enter the high
pressure absorber tower.

Downstream of the pumps a flow element FE 01


monitors the flow of lean oil to the absorption tower
and switches on a stand-by pump in the event of a
flow failure.

It is important to ensure that a constant ratio of


gas to lean oil flow rates is maintained within the
absorber tower. Instrumentation is provided to
do this. If the gas flow increases the absorption
oil circulation rate will increase. Conversely if the
gas flow decreases the absorption oil circulation
rate will decrease. Note that I have omitted this
instrumentation from the drawing for the sake of
simplicity.

Figure 4.6 : Absorption System Absorption Oil Flow

53
Petroleum
Petroleum
Open
Open
Learning
Learn-

You saw in Section 1 and the answer to Test The rich oil flash drum is filted with pressure In fact this is what happens on some offshore oil
Yourself 4.2 that the colder the absorption liquid controller PC 02 which controls the pressure by producing platforms. The rich oil is injected into
the more gas it can absorb. So, before passing into opening and closing pressure control valve PV the main oil transport line going ashore. Further
the absorber tower, the lean oil flows through a 02. processing plant at the shore terminal is used to
lean oil cooler. remove the NGLs which are then further refined
The level in the rich oil flash drum is controlled by into pure products of propane, butane etc.
The cooler is a heat exchanger which uses sea level controller LC 02 in conjunction with LCV 02.
water to cool the lean oil to as Iow a temperature This control valve is positioned downstream of the But we are following a complete plant so let’s go
as possible. next equipment in the system, the rich oil filters. back to the system which we were
following.
The cooled lean absorption oil enters the Absorber As the oil passes across LCV 02, the pressure
Tower near to the top. It flows downwards and falls. Because of this reduction in pressure, some The flow of oil plus flashed gas is now routed to the
spreads out over the packing contacting the gas of the gases which are absorbed into the oil will Stripping Tower. This tower is almost identical to
flowing upwards through the tower. begin to separate. This is often called flashing off. the absorption tower in size and design. It is used
If the valve were situated before the filters, there to remove the rest of the absorbed components
The oil accumulates at the bottom of the tower. It would be a mixture of liquid and free gas flowing from the absorption oil.
is now rich oil and has absorbed as much NGL as through them. This would affect the efficiency of
it can. the filters. Placing the control valve downstream, The gas/oil mixture enters the Stripper Tower
helps prevent separation taking place before we through a set of spray bars located above the
The liquid level in the tower is controlled by a level top of the packed section. The spray bars do two
controller LC 01 through a level control valve want it to.
things.
LCV 01 located in the liquid outlet line from the So, the total flow of rich oil from the rich oil flash
tower. drum is passed through a set of filters. These filters They:

From the absorber tower the rich oil flows to a remove any fine solids which may have entered
rich oil flash drum. Small amounts of the light the system with the gas.
• immediately Increase the surface area of the
hydrocarbon gases may have been absorbed into Stop for a moment here and think about the rich oil to assist in removing the gas.
the oil. The rich oil flash drum acts as a separator situation in our NGL recovery plant.We have
to remove the gas which is mainly methane. This reached a point where we have removed the NGLs • evenly distribute the oil across the top of the
gas joins the gas leaving the absorber tower and is from the gas stream and they are dissolved within packing.
sold, used as fuel gas or sent to the flare. the absorption oil.
The rich oil falls down the tower. and the large
It would be possible to transport the NGLs to surface area generated over the packing plus the
another location still dissolved in the oil and drop in pressure allows the absorbed gas to be
recover them in a separate processing plant. released.

54
Petroleum
Petroleum
Open
Open
Learning
Learn-

In order to assist in the stripping process heat is A feature of most absorption systems is that the
often added to the tower. Also a small amount of interplay of :
dry lean gas may be introduced at the bottom.
This flows up and helps to vapourise the absorbed • gas flow rates
components. Test Yourself 4.3
• tower level control
The gas which is released comprises the NGLs in
gaseous form which have been recovered from the • flash drum level control Place the following items in the correct order
original gas stream. This gas flows upwards and in the absorption oil circulation path starting
leaves the tower at the top. The pressure in the • absorption oil flow rates with absorber tower.
tower is controlled by pressure controller PC 03
in conjunction with pressure control valve PCV these can all add up to result in a slight slugging of a) absorber tower .................
03 on the gas outlet line. flow. The lean oil surge drum evens out this slugging
effect by providing a buffer in the system. b) filters .................
The absorption oil accumulates at the bottom of
the tower where it can now be called lean oil. The The drum is fitted with: c) stripper tower .................
level of the oil is controlled by level controller LC
03 operating level control valve LV 03. • level switch high-high (LSHH 01)
d) lean oil cooler .................
The next piece of equipment in the system is • level switch low-low (LSLL 01).
the lean oil surge drum. This keeps a sufficient e) rich oil flash drum .................
amount of oil in the system to ensure that a These switches will operate to shutdown the system
continuous flow of absorption oil is maintained if the slugging becomes too severe. f) lean oil surge drum .................
around the circuit
This completes our brief section on gas liquids g) oil circulation pumps. .................
recovery using an absorption system. Before I
summarise the section have a go at the final Test
Yourself in this unit.

You will find the correct answers in


Check Yourself 4.3 on Page 61.

55
Petroleum
Petroleum
Open
Open
Learning
Learn-

Summary of Section 4
In this section we looked at a system which uses Next we followed the path of the natural gas You have now completed this unit on natural gas
the principle of contacting gas with a lean oil. through the absorber tower. You saw that the gas liquids recovery. Before you move on to another
This causes the heavier gas components to be flowing upwards is contacted with a downcoming unit in the Petroleum Processing Technology
absorbed into the oil which can then be recovered stream of absorbtion oil. As the two streams co- Series I would like to make some final comments.
and transported as NGLs. mingle the oil absorbs the NGLs and the residual
gas flows from the tower to sales or flare. The unit you have just completed relates to
You saw that at the heart of the system are the NGL recovery in general. It is not meant to
two towers, the absorber tower and the stripper Within the oil circulation system we looked at the apply to any specific plant or process. If you
tower. following pieces of equipment. are involved with process plant operation you
must be completely familiar with the specific
There is a continuous circulation of oil round the • oil circulation pumps plant under your control. You must always
system. This is the oil which absorbs the NGLs as follow laid down procedures and operational
• lean oil cooler
the gas is contacted with it. guidelines and adopt safe working practices at
We referred to it as Absorption oil. It is a • absorber tower all times.
hydrocarbon liquid and in some cases it can be
dead crude oil. The oil is classed as lean oil or • rich oil flash drum
rich oil as it flows around the circuit. The lean oil is
• filters
oil which has been stripped free of NGL gases. The
rich oil is oil which has absorbed the NGL gases. • stripping tower

We looked at the construction of a typical packed • lean oil surge drum


column type of absorber tower and saw how it
worked I explained the function of each of these pieces
of equipment and pointed out how they are
controlled.

56
Petroleum
Petroleum
Open
Open
Learning
Learn-

Check Yourself 1.1 Check Yourself Check Yourself 1.4


nitrogen - (N) - element 1.2 At point Y the substance would exist as steam.
If heat energy is removed at constant pressure
sodium chloride - (NaCl) - compound the temperature of the substance would fall
a) C8 H (2 x 8) +2 i.e.C8H18 towards point X. When the line BC is reached
water - (H20) - compound
the steam would start to condense and the
helium - (He) - element b) C9 H (2 x 9) +2 i.e. C9H20 first drop of liquid would form. At this point the
temperature would remain stable until all the
sulphuric acid - (H2SO4) steam had condensed to liquid water.
- compound

carbon - (C) - element Check Yourself The temperature would then fall again
until line BD is reached. Once again the
iron - (Fe) - element 1.3 temperature would remain constant until all
the liquid freezes into solid ice. When the
substance is solid, the temperature again falls
B 0ºC (32ºF) to point X.

C 0ºC (32ºF) So, there have been 2 phase changes. From


gas ( steam) to liquid and from liquid to solid.
D 100ºC (212ºF)

E 100ºC (212ºF)

57
Petroleum
Petroleum
Open
Open
Learning
Learn-

Check Yourself 1.5 Check Yourself 1.6


W 36°C (97°F). i.e. the temperature at which a) Solid - liquid - gas - solid - gas.
pentane ( having the lowest boiling point)
starts to boil.
b) CI- Chlorine, O - Oxygen, Si - Silicon,
He - Helium, Fe - Iron, C - Carbon,
Y 99°C (209°F). I.e. the temperature at which
Heptane ( having the highest boiling point)
boils. c) 1 - X, 2 - Z, 3 - W, 4 - Y, 5 - V,

58
Petroleum
Petroleum
Open
Open
Learning
Learn-

Check Yourself 2.1 Check Yourself 2.2


Ethane 228ºF (100 - (-128)) 1. The process of gas compression causes the gas temperature to rise.

Propane 143.7°F (100 - (- 43.7)) 2. The suction gas scrubber removes liquids from the gas.

Pentane 3°F (100 - 97) 3. Within a suction K.O. drum wire mesh pads form a mist extractor.

4. In a shell and tube heat exchanger, gas flows through the tubes and
water flows through the shell.

59
Petroleum
Petroleum
Open
Open
Learning
Learn-

Check Yourself 3.1 Check Yourself 3.3


The correct order is :
110°F (43.3°C)
1-3-6-2-9-5-8-7-4

Check Yourself 3.2 Check Yourself 3.4

1. True If the gas is being used to power a turbine as it expands, the


turbine could be used to drive some other machinery. The
2. False - It is used to refrigerate the natural gas. most common would be a compressor. However, a pump or
generator could be driven by the turbine.
3. False - Partial vapourisation takes place.

4. False - The cooling medium is cold gas from


the NGL / gas separator

5. True

60
Petroleum
Petroleum
Open
Open
Learning
Learn-

Check Yourself 4.1 Check Yourself 4.2


a) Pressure Contact must be made under
pressure.

b) Temperature The process is made more efficient


at lower temperatures.

We discussed this in Section 1. Go back and refresh your


memory of this if you had any difficulty.

Check Yourself 4.3


The correct order is :

a) - e) - b) - c) - f) - g) - d)

61
Petroleum
Petroleum
Open
Open
Learning
Learn-

62
POL
Petroleum Open Learning

Oil and Gas


Separation
Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY

2
POL
Petroleum Open Learning

Oil and Gas


Separation
Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY

2
Petroleum Open Learning

Designed, Produced and Published by OPITO Ltd., Petroleum Open Learning, Minerva House, Bruntland Road, Portlethen, Aberdeen AB12 4QL

Printed by Astute Print & Design, 44-46 Brechin Road, Forfar, Angus DD8 3JX www.astute.uk.com

© OPITO 1993 (rev.2002) ISBN 1 872041 85 X

All rights reserved. No part of this publication may be reproduced, stored in a retrieval or information storage system, transmitted in any form or by any means, mechanical,
photocopying, recording or otherwise without the prior permission in writing of the publishers.
Oil and Gas Separation Systems - Workbook 2 Petroleum Open Learning

(Part of the Petroleum Processing Technology Series)

Contents Page Visual Cues


* Training Targets 2 training targets for you to
achieve by the end of the unit
* Section 4 -Control of Separators 3
The Control Loop test yourself questions
Separator Level Control to see how much you
Separator Pressure Control understand

* Section 5 - Separator Safety Systems 21


check yourself answers to
Level Control and Systems let you see if you have been
Pressure Control and Safety thinking along the right lines
Emergency Shut-down Valves

* Section 6 - Operations of Separators 34 activities for you to apply


your new knowledge
Routine Operation Checks
Start-up Procedure
Shut-down Procedure
Blow-down Procedure summaries for you to recap
on the major steps in your
progress

1
Petroleum Open Learning

Training Targets

When you have completed Workbook 2 of this unit you will be able to :

• Explain the basic principals of process control

• Describe the equipment used in separator level control

• Describe a simple separator safety system

• Explain the routine operational checks on a separator

• Describe in simple terms a separator start up procedure

Tick the box when you have met each target.

2
Oil and Gas Separation Systems Petroleum Open Learning

Introduction
Section 4 - Control of Separators
In the last part of Section 3 of this programme you In any continuous process such as oil and gas The Control Loop
looked at the external features of separators. Two separation there are a number of factors which
of the items you saw were pressure gauges and must be kept within certain limits. These are called There are 4 main elements in a typical control loop
level sight glasses. These pieces of equipment are the process variables. The four most common of and these are :
used to enable the operator to check pressure and these are.
liquid levels inside the vessels. * The Process Variable
* Liquid Level * The Measuring Unit
Pressure and liquid level are features of the * Pressure * The Controller
process which can vary. Each can increase or * Temperature * The Correcting Unit
decrease with variations in separator throughput. * Fluid Flow
However, in order to obtain optimum separation, Figure 26 shows a simple block diagram of a
the pressure and liquid level must be maintained As I pointed out earlier, we are going to concentrate
control loop.
at a constant value. on Liquid Level and Pressure, but the basic method
of achieving control applies to all four. It relies on
So, apart from the equipment used to check the having built into the system a suitable control
liquid level and pressure, separators have two loop.
major controls.

* Liquid Level Control Correcting Measuring



* Pressure Control
Unit Process Variable Unit

In this section we will look at each of these control


systems and see how they work.
desired
Please note that we will be covering the value
subject of control at a very basic level. Other
programmes in this series will delve much measured
more deeply into Instrumentation and Control. correcting value
Before we proceed, however, let’s consider the
signal
fundamentals of process control.
Controller

Figure 26 : The Control Loop

3
Petroleum Open Learning

We can look at the four elements in turn. The controller may work using air (pneumatic In this figure the process variable is the level of
operation), liquid (hydraulic operation) or water in the tank. The measuring unit is the level
The Process Variable electronics. indicator. The correcting unit is the valve in the
water outlet line. Finally, the controller in this case
This is that part of the process which has to be The Correcting Unit is the plant operator. Let’s call him Joe.
controlled within certain limits, i.e. Level, Pressure,
etc. The actual value of the process variable This part of the control loop is usually a valve. On Imagine in this simple example that water is
which the operator wishes to maintain is called the receipt of the signals from the controller it opens or entering the tank through the inlet line at the top of
desired value. We need not say any more about closes to alter the process variable. The measured the tank. Joe has opened the valve on the outlet
the process variable at this point. value is then returned to the one indicated by the line so that exactly the same amount of water is
desired value. leaving the tank as is entering. As long as this
The Measuring Unit situation remains steady, the water level in the
The following example should help you see how a tank will stay constant. This level, as shown on
This unit measures the actual value of the variable. simple control loop works. the level indicator, is the measured value of the
It could be a Pressure Measuring Instrument, a
Look at Figure 27. process variable. Also, because that is the level
Flow Measuring device and so on. The measuring
which Joe wishes to maintain in the tank, it is the
unit obtains the measured value.
desired value. There is no difference between the
measured value and the desired value.
The Controller
Supposing that the flow of water entering the tank
It is the job of the controller to compare the
is somehow increased. What will happen to the
measured value of the process variable with the
water level? Of course, if nothing is done to the
desired value. If it senses a deviation between
outlet valve, the level will start to rise.
the two it then sends a correcting signal to the final
element in the loop, the correcting unit. Figure 28 on the next page, shows this.
For instance, supposing you wanted to maintain
the pressure in a separator at 250 psi, but the
pressure had increased to 275 psi. The desired
value is 250 psi and the measured value is 275 psi.
There is obviously a deviation. A mechanism within
Figure 27
the controller would sense this and instruct the unit
to send an appropriate correcting signal.

4
Petroleum Open Learning

Figure 28 Figure 29

But Joe is keeping his eye on the level indicator Joe, the controller, has maintained a constant level Of course having an operator like Joe standing
and sees the change in level. He compares the in the tank by : by the tank all day long would be a waste of an
new measured value with the desired value and operator’s time, and pretty boring for Joe. It would
notes that the level has increased. In order to • Comparing the measured value of the level in be much more sensible to have an instrument to do
reduce the level again the valve in the outlet line the tank with the desired value. this simple control job.
must be opened more. Joe does this manually and
• Noting the difference between measured and Let’s see how this basic principle of automatic
increases the outlet flow until the measured value
desired values. control is applied to level and pressure control in
of the liquid level once again matches the desired
value. a separator.
• Sending an appropriate signal to the correcting
element in the loop. (In this case an instruction
This is shown in Figure 29.
to his hands to open the valve).

5
Petroleum Open Learning

Separator Level Control


Level control in a separator is very similar to the level

Test Yourself 7
control system in the simple example I just gave.

Before we look at the control loop used, let’s consider the


reasons for maintaining the liquid level in a separator at a
Fill in the missing words from the following paragraph. constant height. They may seem fairly obvious but I think
it is worthwhile listing them here.

Liquid Level Control is required :


That part of a continuous process which an operator wishes to control within certain limits is
• To prevent liquids being carried out with the
called the ................ ................, and its target value is called the ................ value. gas. (Known as carry-over.)

A measuring unit obtains a measured value from the process and feeds it to a • To prevent gas from leaving the separator
through the liquid outlet. (Blow round.)
.................................... whose job it is to compare the two values. If a deviation exists
• To help maintain the pressure on the vessel.
between the two values, the ........................ sends a correcting signal to the final element (Fluctuating levels affect the pressure.)

• In a 3 phase separator, to prevent oil from


in the loop - a correcting unit which is usually a ................ ................ .
leaving through the water outlet or vice
versa.

• To ensure optimum retention time.

Think first of all of an oil level control system on a 2 phase


horizontal separator.

Figure 30 shows the system in its most basic form.


You will find the answers in Test Yourself 7 on page 42

6
Petroleum Open Learning

The hardware associated with the system consists of:

• The displacer mechanism - which is the


measuring unit.

• The level controller - in this case we are


looking at a pneumatic controller.

• The level control valve - the correcting unit.

Let me describe each of these units and briefly


explain how they work.

Displacer Mechanism
This piece of equipment measures liquid level by a method
which is based on the principle of Archimedes. The
principle states that, if a body is immersed in a liquid, it
will apparently lose weight equal to the amount of liquid it
displaces.

So, a cylindrical weight partially submerged in a liquid will


have a certain apparent weight. (Less than its actual weight
because of the buoyancy effect of the liquid.)

If the liquid level rises or falls, then more or less of


the cylinder will be submerged. Its apparent weight will
therefore vary. The apparent weight can be measured to
give an indication of the level of the liquid.

Figure 31 on the next page, shows how the apparent


weight varies with the liquid level.

Figure 30 : Oil Level Control System

7
Petroleum Open Learning

The loss or gain of apparent weight has to be transmitted


to the controller as a signal which is proportional to the
increase or decrease in liquid level. This is done by
means of a torque tube mechanism.

Figure 32 shows the torque tube assembly.

Figure 31a Figure 31b

Much of the cylinder submerged - buoyant effect means Less of the cylinder submerged - higher
low apparent weight. apparent weight.

8
Petroleum Open Learning

A hollow rod called the torque tube (C) is connected


via a torque tube plate (D), to the supported end
of the float rod. The opposite end of the torque
tube is welded to the outer flange of the assembly
(indicated E).

A small diameter shaft, the torque tube rod (F), fits


inside the torque tube and is welded at the torque
tube plate. This shaft protrudes through the outer
flange where it is free to rotate.

Look carefully at Figure 32 again and try to


visualise what will happen to the assembly as the
liquid level moves up or down.

Imagine the level going down in the vessel. As it


does so the apparent weight of the displacer will
increase. The increased weight hanging on the
float rod will give a twisting motion to the torque
tube. (Remember that the torque tube is fixed at the
flange end but only supported on the knife edge at
the free end).

Because the torque tube rod is welded to the torque


tube, it will rotate as the tube is twisted. The rotation
of this rod at its free end will be in proportion to the
Figure 32 : Torque Tube Assembly
increase or decrease in liquid level. The rotation is
transmitted via linkage to the next element in the
The displacer cylinder may be contained in a chamber which is connected to the separator but mounted outside
control loop, the controller.
it. The liquid level in the chamber is obviously the same as the level in the separator. Having the displacer placed
outside the separator like this means that it is unaffected by any turbulence in the vessel.

The displacer is suspended from one end of a float rod, marked (A) on the figure. The other end of the float rod
is supported on a knife edged bearing (B).

9
Petroleum Open Learning

Level Controller
We have seen already that the job of a controller is to
compare 2 signals, the measured value signal and the
desired value signal. If a deviation exists between the
two, the controller then has to send a correcting signal
to a control valve.
to recorder
There are various types of controller, but in this section CONTROLLER
we are going to look at the basic principle of operation
for one type of pneumatic controller.
measured

DIFFERENTIAL
In order to perform its job the controller has 4 separate,

MECHANISM
but interconnected, units. They are : value
FLAPPER/NOZZLE 20 SUPPLY AIR
• The Differential Mechanism
Desired UNIT
• The Flapper/Nozzle Assembly
value PILOT
• The Feedback Unit VALVE TO CORRECTING
FEED BACK
ELEMENT
UNIT
• The Pilot Relay

Figure 33 shows the relationship of the four units in


the form of a block diagram

Figure 33

10
Petroleum Open Learning

We can see how each of the units works, and how One end of the desired value link acts as a pivot
it interacts with the others. point for the differential arm. The position of the
pivot point can be altered by means of an adjusting
Differential Mechanism knob at the other end of the link (not shown).

It is the job of the differential mechanism to The measured value link is also connected to the
compare the desired and measured values of the differential arm. It is connected at the other end to
process variable. If a deviation exists between the the measuring unit, or element (not shown).
two, the unit feeds this information to the next unit
in the controller, the flapper/nozzle assembly. From the centre of the differential arm a deviation
link transmits any motion of the arm to the flapper/
There are two types of differential mechanism nozzle assembly.
which can measure the deviation, and these are
known as : Look again at Figure 34. You can see that, providing
there is no movement of the M.V. link relative to the
• The Motion Balance Mechanism D.V. link, there will be no movement of the deviation
link. However, if the measuring element causes the
• The Force Balance Mechanism M.V. link to move, there will be a movement of the
deviation link.
Motion balance mechanisms use two mechanical
linkages to compare the measured and desired
values (abbreviated as M.V. and D.V.).
Force balance mechanisms use pressure applied
to bellows to compare desired and measured
values.

Figure 35 on the next page, shows this -again, in


a very simple form.

Figure 34 shows this in simple form


Figure 34 : A Motion Balance Mechanism

11
Petroleum Open Learning

Supposing the measuring element output gave an increased pressure signal. The M.V. bellows
would expand against the D.V. bellows. This movement would cause the free end of the force bar to
move. The movement could then be fed to the flapper/nozzle assembly as a deviation.

Force balance is the mechanism most commonly used in Pneumatic Controllers.

I have mentioned several times the flapper/nozzle unit. Let’s look now at this piece of equipment.

Flapper/Nozzle Assembly

The flapper/nozzle unit consists of three items, the flapper, the nozzle and the restrictor.

Figure 36 is a simplified drawing of the unit.

Figure 35 : Force Balance Mechanism

The output from the measuring element is fed as a pressure to the


measured value bellows.

The movement of the bellows is opposed by a second set of bellows, the


desired value bellows. These are pressurised by a signal which relates to
the desired value of the process variable.

Sandwiched between the two sets of bellows is one end of a bar known as
the force bar. The bar is pivoted using a fulcrum and the other end of the
bar is free to move.

I’m sure that you can visualise what happens when a deviation between
desired value and measured value occurs.

12
Petroleum Open Learning

It is the job of the assembly to send a correcting


pressure signal from the controller to the final
element in the control loop, the control valve.

Here is a basic explanation of how that is achieved.

An air supply is fed to the line upstream of the


restrictor, typically at a pressure of 1.3 bar.

This air can pass through the restrictor to the nozzle


outlet and also the variable back pressure outlet.

If the flapper is positioned away from the nozzle it


is possible for the air to pass through the restrictor
and out through the nozzle. Because the diameter of
the restrictor is small compared to that of the nozzle,
there will be very little pressure build-up in the space
between the restrictor and the nozzle. This means Figure 37
that there will be no build-up of pressure in the back
pressure line.

However, if the flapper is moved towards the nozzle, You can see from the graph, that the section of the With a supply pressure of 1.3 bar, the operating range
the area of nozzle through which the air can pass is curve between points 1 and 2 is almost a straight chosen is from 0.2 to 1 bar. So, nearly all pneumatic
reduced. This means that the nozzle back pressure line. The nozzle back pressure varies from 0.2 bar at instruments will record, transmit and control within
will increase. point 1, to 1.0 bar at point 2. Between the 2 points, this 0.2 to 1 bar range.
the back pressure obtained will be proportional to the
If the nozzle is completely covered then the back distance that the flapper is away from the nozzle. You can probably imagine now, how the controller
pressure would build up to the supply pressure of is able to send a correcting signal to the correcting
1.3 bar. Pneumatic instruments are usually designed to unit.
operate over a standard pressure range. This range
Figure 37 shows a graph of the back pressure must lie on the straight line portion of the graph If the movement from the differential unit is linked
obtained from a flapper/nozzle system against the which you saw in Figure 37. to the flapper, a varying back pressure signal will
position of the flapper relative to the nozzle. be obtained from the nozzle. This signal will be
proportional to the deviation between measured and
desired values of the process variable.

13
Petroleum Open Learning

nozzle
In an actual controller, the movement of the flapper
is extremely small. The total movement required to
change the signal from 0.2 to 1 bar is only a little
more than 0.1 mm. Such a small movement means
that a direct linkage from the differential unit to the
flapper is impracticable. The deviation signal must
be adjusted to compensate for this.

This brings us to the next unit in the controller, the


feedback unit.

Feedback Unit

Feedback simply means feeding back the output


signal of the controller to the input of the same
unit. This allows relatively large movements of the
differential unit to cause very small movements of
the flapper. We will see how this is done shortly.
Feedback is also used to introduce more complex
control actions to the loop. These control actions
are beyond the scope of this programme. However, Figure 38 : Feedback Unit
you will come across them if you follow other
programmes covering Instrumentation and Control,
in the Petroleum Processing Technology Series. The pivot point of the flapper is attached to the The increased bellows pressure will move the pivot
We can see now how a simple feedback unit movable end of a set of bellows. The movement of point of the flapper against the spring until the spring
works. these bellows is opposed by a spring. and bellows forces are balanced.

The mechanism is shown in Figure 38. The nozzle back pressure is fed to these bellows in As this happens, the movement of the pivot point will
addition to being the controller output signal to the lift the flapper away from the nozzle.
correcting unit.
This sequence of events, movement of the flapper
When a deviation occurs, the flapper moves towards then away from the nozzle, will continue
towards the nozzle and causes an increase in back until a steady state is reached.
pressure.

14
Petroleum Open Learning

control signal
When this state is reached, the flapper will have moved very slightly nearer to the
nozzle. Just enough to increase the output to the correcting unit.

Pilot Relay

The final unit in our pneumatic controller is the pilot relay. This is a device which is diaphragm
connected to the nozzle back pressure line.

Its function is to act as a signal amplifier. actuator spring

The controller output has to operate an actuator on the correcting element which is
usually a control valve. This actuator, as you will see, requires a relatively large volume
of air. Because of the restriction in the supply line to the nozzle, only a limited volume of
air can be supplied to the actuator. The pilot relay can boost this air supply, for proper
operation of the valve. valve stem

I think that this explanation of what the pilot relay does is sufficient at this stage.

The Level Control Valve


valve plug valve body
The function of a control valve is to throttle, or regulate, the rate of flow of a fluid.

You will remember from our simple example earlier, that Joe had to open and close the
valve in the outlet line from the tank. He was operating a level control valve manually. In valve seat
an automatic control system, however, the control valve is operated by the signal from
the controller. In the case of a pneumatic controller the signal is air pressure, which
varies between 0.2 and 1 bar.

There are many different types-of control valve in use. Figure 39 is a simplified drawing
of a typical valve which could be used in a level control application.

Figure 39 : Level Control Valve

15
Petroleum Open Learning

The valve is called a diaphragm motor valve and If there is a complete loss of control pressure, the Test Yourself 8 will help you draw together all
consists of the following items. actuator spring will cause the valve to move into the these aspects of control loop operations.
fully open position. This type of valve is sometimes
• diaphragm called a fail open valve. It is possible to change the
action of the valve or the controller so that a loss
• actuator spring of control signal would cause the valve to close. In
this case the valve would be referred to as a fail
• valve stem closed valve.

• valve body You will notice that the valve illustrated in Figure
39 has two valve plugs and seats. It is known as a
• valve plug double ported valve. Although single ported valves
are sometimes used, the double valve is preferred
• valve seat for level and pressure control duties on separators.

The actuator spring is attempting to hold the valve Take another look at Figure 39 and satisfy yourself
in the open position by pushing up the diaphragm that you understand how the control valve works,
and lifting the valve stem. This is called a normally and how it fits into the complete level control loop.
open valve.

The control signal is applied to the top of the


diaphragm. Increasing pressure of the control
signal overcomes the resistance of the actuator
spring and gradually closes the valve.

The valve is arranged so that a signal pressure of


3 psi will just start to close the valve. With a signal
pressure of 15 psi, the valve will have moved to its
fully closed position.

16
Petroleum Open Learning

We have just been looking at separator level control. You


will remember that the other major separator control is that

Test Yourself 8
of pressure. Let's finish off this section by having a look at
separator pressure control.

Separator Pressure Control


The following items from a level control loop are part of the measuring unit, the level controller
or the level control valve. Indicate by a tick in the box provided, to which part of the system each As with level control, the basis of a pressure control loop
belongs. consists of :

• a process variable (in this case, the pressure


Measuring Unit Level Controller Level Control Valve
in the separator)
pilot relay
• a measuring unit (some form of pressure
differential mechanism measuring device)

torque tube rod • a controller (again, a pneumatic controller in


this example)
actuator spring

valve plug • a correcting unit (once more a control valve)

flapper nozzle assembly


The controller and control valve work in the same way as
diaphragm the units used in level control. The measuring unit, however,
is obviously going to be different.
feedback unit

float rod

valve stem

You will find the answer to Test Yourself 8 on page 43

17
Petroleum Open Learning

Think for a moment about what the measuring unit


has to do. It must transmit the measured valve to

Activity
the controller as a signal. In level control this is done
by the torque mechanism which transmits rotation
of the tube via a linkage. Somehow the pressure
measuring unit must perform a similar function.
Think about the way in which a pressure gauge works. How could the measured pressure
We have already looked at a pressure measuring be transmitted to the flapper of a controller ?
instrument in Section 3, the bourdon tube type
pressure gauge. This mechanism could be used to
transmit a measured valve signal to the controller.

18
Petroleum Open Learning

In a most simple way, the transmission could be as is shown in Figure 40. As the valve opens, more gas is allowed to leave
the separator and the pressure is reduced.

Obviously, if the pressure in the separator goes


down, the movement of the bourdon tube pushes
the flapper towards the nozzle. This will cause the
output signal pressure to rise, causing the control
valve to close. If less gas now leaves the separator,
the separator pressure will then increase.

Test Yourself 9

In the pressure control loop I have just


described, is the control valve in the fail
open or fail closed mode?
Figure 40 : A Pressure Control System
You can see that the free end of the bourdon tube
is connected via a linkage to the flapper/nozzle
assembly of the controller. In the set up shown, if
the pressure in the separator rises, the bourdon
tube tries to straighten and the movement pushes
the flapper away from the nozzle. This causes the
You will find the answer to Test Yourself
output signal from the controller to fall which in turn
9 on page 43
causes the control valve to open.

19
Petroleum Open Learning

Summary of Section 4
Before going on to summarise Section 4, I should emphasise once again that the subject of process measurement and control is very complex. We have only
scraped the surface of the subject in this section but you should now know how a simple control system works.

We started the section by considering the four With the aid of a simple example involving an After a detailed look at the principle of operation of
most common process variables : operator called Joe, we saw how the elements of a level control loop we finished off the section by
a control loop work together to maintain the liquid working through the basics of separator pressure
• Liquid Level
level in a tank at a constant value. control. Here you saw that the basis of a pressure
• Pressure
control loop is the same as that for level control.
• Temperature From there we moved on to consider a separator
However, the measuring element is obviously
level control system. We saw that the hardware
and a different unit to the measuring element of a
associated with a typical system consists of:
level control system. We used a bourdon tube
• Fluid Flow • The displacer mechanism type pressure measuring element in the example
Of these, we concentrated on the control of liquid • The level controller we considered, but you should remember that
level and pressure. • The level control valve many other types of measuring instruments are
We looked at each of these elements of the system available.
You saw that a suitable control loop is required
to achieve control, and such a loop has four main in some detail and saw how they are constructed and In the next Section we will be looking at additional
elements : how they work. We paid particular attention to the equipment and instrumentation which may be
controller with its four separate, but interconnected, fitted to a separator to ensure safe operation of the
• The process variable units which are known as : vessel or train of vessels.
• The measuring element
• The controller • The differential mechanism
• The correcting unit • The flapper/nozzle assembly
• The feedback unit
• The pilot relay

20
Oil and Gas Separation Systems Petroleum Open Learning

Section 5 - Separator Safety Systems


In the last section we looked at control of the two
main process variables in a separator, i.e. Level
and Pressure. These two control systems usually
operate with relatively few problems. However
there is always the possibility that, for some reason, Activity
they fail to maintain control. This may happen, for
example, because of instrument malfunction in
the control loop. If this should occur a potentially Jot down in the space below your ideas of what might result from the following separator
hazardous situation will arise. malfunctions.

Think for a moment about the possible • oil level goes too high
consequences of losing either level or pressure
control in a separator. • oil level goes too low

• pressure continues to increase

• pressure continues to decrease

• water level goes too high

• water level goes too low

21
Petroleum Open Learning

Here are a few ideas of mine. How do they compare We have already looked at similar equipment in Section
with what you have written down? 1. There, you saw that flow lines can be fitted with
pressure switches to warn the operator of high or low
• If the oil level goes too high, a situation will be flow line pressures. You also saw that these pressure
reached where oil gets carried over with the gas, switches are tied into the Emergency Shutdown (ESD)
causing problems downstream system.

• Should the oil level go too low, there is a danger of Let us now consider a typical 3 phase separator, together
gas leaving the separator through the oil outlet with its protective devices.

• If the pressure increases too much, there is a risk of Level Control and Safety
exceeding the safe working pressure of the
separator Let’s start by having a look at the oil level in a separator.
We can see what would be the sequence of events if
• In the situation where the pressure falls too much, the oil level started to rise, and continued to rise.
there will be insufficient pressure to push the liquids
from the separator First of all you will remember from the previous section
that an increase in level will cause the level control
• If the water level rises above the weir, water will valve to open.
contaminate the oil leaving the vessel LCV 01
Take a look at Figure 41 which shows the oil
• Should the water level go too low, oil will flow from accumulation and outlet side of a 3 phase horizontal
the separator through the water outlet separator. Figure 41

In this section, we are going to look at the equipment


designed to prevent such situations arising in a typical
separation system.

The section will concentrate on three variables (oil


level, water level, and pressure) and the degrees of
protection afforded by this equipment.

22
Petroleum Open Learning

You will see that this is a simple illustration of a level to rise or fall, then the separator must be protected HIGH and LEVEL SWITCH LOW- LOW.
control loop. automatically.
If the level reaches the setting of either of these
LC 01 is the level controller and LCV 01 is the level This is done by having a second degree of protection switches, a signal is sent to the ESD system which
control valve. So in this case an increase in oil level using level switches connected to ESD valves. These automatically isolates the vessel and makes it safe, by
causes LC 01 to open LCV 01. switches are connected to the separator independently activating the appropriate ESD valves.
of the level controller.
However, the level might continue to rise (due to We will look at the location and operation of ESD valves
equipment malfunction, etc.), and start to approach a Figure 42 shows this. shortly.
hazardous situation. In order that the operator can be
warned of the situation, alarm signals are generated It is quite common for all the instrumentation relating to
by the controller. separator level control and safety, to be located externally
to the vessel. In such a case, the instrumentation can
If you look again at Figure 41, you will see LAH 01 be mounted on pipework sometimes referred to as an
connected to the controller LC 01. The letters stand for instrument bridle.
LEVEL ALARM HIGH
Figure 43 on page 24, shows a typical set up
If the level should reach the setting of LAH 01 an audio/ incorporating an instrument bridle.
visual alarm would be generated.

This is usually a noise (rapid high pitched beeper) and


a flashing light which would indicate the alarm condition
at a location which is normally manned, e.g. a central
control room.

The alarm would alert the operator, who could then try
to rectify the situation before the actual hazardous
situation is reached.

If a falling level is the problem, you can imagine that a Figure 42


similar alarm is generated by LAL 01. You will notice that the switches are designated LSHH
If the situation is not rectified and the level continues and LSLL These stand for LEVEL SWITCH HIGH-

23
Petroleum Open Learning

You can see that the bridle is connected via valves to the
top and bottom of the separator.

Take a look at Figure 43 and identify the bridle which


incorporates the level controller, the level alarms, the level
switches and the sight glasses. You will remember from
Section 3 that the sight glasses give a visual indication of
the actual level inside the separator. Note that there are
two sight glasses which overlap each other. The normal
operating level would lie within the overlap, enabling the
level to be checked through both sight glasses.

We have just looked at the oil side of the separator. In a 3


phase vessel, the water level must also be controlled.

Of course, in this section of the vessel, the water is


covered by a layer of oil. So there is an interface between
the water and oil. It is this interface which is measured by
the controller.

Apart from that, the control and safety of the water end of
the separator works in a similar manner to the oil end.

Figure 43 : Instrument Bridle

24
Petroleum Open Learning

Pressure Control and Safety

You saw in the introduction to this section that an increase or decrease in pressure in
Test Yourself 10 the separator is also potentially hazardous. Let's look now at this problem.

Make a simple sketch of the water outlet end of a separator. I am sure that by now you will have realised that there are several degrees of pressure
Your sketch should show an instrument bridle with the protection on a separator.
relevant instrumentation.
Look at Figure 44. This shows a simple pressure control loop.

You will find the answer to Test Yourself 10 on page 44 Figure 44

25
Petroleum Open Learning

PC 01 (the pressure controller) activates PCV 01 (the This operating pressure is well below the maximum Pressure Safety Valves
pressure control valve). which the separator is capable of holding.
To prevent the separator system from becoming over-
In addition PC 01 generates alarm signals (PAH 01 and However, the vessel could be ruptured if the pressure pressured, every separator is fitted with Pressure Safety
PAL 01) if the pressure goes too high or too low. went high enough. Valves (PSVs). (These valves may also be called
Pressure Relief Valves, or PRVs).
Separate pressure switches PSHH ( Pressure switch Because of this, a further level of protection against
high-high) 01 and PSLL (Pressure switch low-low) 01 excessive pressure is fitted to separators, in common The set pressures of the PSVs are determined by the
are connected to the ESD system. The switches are with other pressure vessels. maximum operating pressure of the separator. The sizing
shown in Figure 45. of the PSVs is determined by the maximum amount of
These are called Pressure Safety Valves (PSVs). gas which may be required to be vented.

PSVs are special valves fitted to the top, or gas section-


of the separator. At a pre-determined pressure, which is
higher than the set pressure of PSHH 01, the valves will
open and gas from the separator is vented. Usually the
vented gas goes to a flare.

Let us take a more detailed look at pressure safety


valves.

Figure 45
If either of these switches are activated, the separator is
made safe by isolating it via ESD valves.
Separators are designed to operate at a certain pressure.

26
Petroleum Open Learning

Figure 46 show a typical PSV arrangement. There are The PSVs are normally situated on the top of the vessel
two PSVs fitted to the separator - PSV ‘A’ and PSV ‘B'. upstream of any demister pad. This is to ensure that,
should the demister pad become blocked, then the PSVs
will still function properly.

Separators are fitted with two PSVs, either one of which


will cope with the full pressure relief requirements of the
vessel.

The usual method of operation is that one PSV is ‘on-line’


whilst the other PSV is isolated and on ‘stand-by’.
PSV A
The isolating valves are interlocked so that:

* both PSVs may be on line,

* PSV ‘A’ may be on line, or,

* PSV ‘B’ may be on line.

The interlock system ensures that it is never possible for


both PSVs to be isolated from the separator at the same
time.

From the shading of the valves you can see that, in the
illustration, PSV ‘A’ is on-line and PSV ‘B” is on standby.

Figure 46

27
Petroleum Open Learning

Pressure Safety Valves can vary in design and construction but the most
common type is one which relies on a spring to hold a valve closed. When the pressure beneath the valve seat reaches a pre-set
value, the valve lifts against the spring tension and allows
Figure 47 is an illustration of this type of valve. gas to escape. Figure 47 shows the valve in the relief
position.

In circumstances where the separator may be handling


corrosive gases such as carbon dioxide (CO2) or hydrogen
sulphide (H2S), then a rupture disc may be fitted between
the separator and the PSV.

Rupture Discs

A rupture disc is a disc of malleable metal which is designed


to burst at a pre-set pressure. Figure 48 is an illustration of
a rupture disc and how it is installed.

The rupture disc is fitted to protect the seats of the PSV from
corrosion and is normally set to fail at a pressure just below
the set point of the PSV. To assist the disc to spread out as
it fails, a disc cutter is normally positioned just above the
rupture disc.

This is shown in Figure 48 on the next page.


Figure 47

28
Petroleum Open Learning

Activity
A pressure switch is often fitted in the space
rupture disc above the rupture disc and below the PSV.
It is indicated as P.S. in Figure 48. If the
switch is activated it sends an alarm to the
operator in the control room. Can you work
out why the pressure switch is fitted?

disc cutter

rupture disc

Figure 48 : A Rupture Disc Installation

29
Petroleum Open Learning

The set pressure of the switch is below the lifting


pressure of the PSV. If the rupture disc bursts or
leaks, then the pressure switch will be activated.

The pressure switch tells the operator that the rupture


disc is no longer operable and requires changing.

Emergency Shutdown Valves

We have said that the High-High and Low-Low


switches will be tied into the ESD system to stop liquids
or gases from entering or leaving the separator. This
is achieved by the use of Emergency Shutdown
Valves (ESDVs).

ESDVs are valves which are operated as part of the


ESD system.

They are normally air or hydraulically actuated


valves. The valves are spring loaded to fail to their
‘safe’ position in the event of an air or hydraulic
failure.

The ESDVs ensure that the process is isolated


in a safe condition in the event of an Emergency
Shutdown.

Figure 49 shows a separator with its ESDVs.


LCV-02 ESDV3 ESDV2 LCV-01

Figure 49

30
Petroleum Open Learning

The ESDV’s in Figure 49 are situated as follows :

• ESDV 1 - on the main fluid inlet line

• ESDV 2 - on the oil outlet line


Activity
• ESDV 3 - on the water outlet line Look at Figure 49 and locate the four different controllers. Write down what each controller
does and what it uses to achieve the control required.
• ESDV 4 - on the gas outlet line
YES. There are FOUR controllers, its not a printing error!
Also indicated in the drawing are the four main
controllers associated with a 3 phase separator.

31
Petroleum Open Learning

The controllers are as follows : Before we finish this section, let us look, once again, at the overall sequence of events which could occur
if control of one of the process variables is lost.
• Level Controller LC-01 controls the oil level
on the downstream side of the weir by opening Use Figure 50 to follow the sequence.
and closing LCV-01.

• Level Controller LC-02 controls the oil/water


interface level on the upstream side of the weir
by opening and closing LCV-02.

• Pressure Controller PC-01 controls the pressure


in the separator by opening and closing PCV-01.

• The overall level control for the left hand side


of the separator is achieved by a weir. The weir is
the simplest type of level control ever invented.

You will see from the drawing that there is one further
ESDV - ESDV 5.

This is situated on the top of the separator and is known


as a 'blow-down' or 'depressurising' valve.

If the separator has been isolated via the ESD system,


it could still remain fully pressurised, and it may be
necessary to depressurise the vessel as an extra safety
precaution. If so, this is done through the blow-down
valve. If ESDV 5 operates, it will vent all the gas from the
separator to the flare, thus dropping the pressure.

32
Petroleum Open Learning

Let us assume that the problem is a rising oil level. The


following will happen :

• LC-01 will open LCV-01

If the level continues to rise Summary of Section 5


• LAH-01 will generate an alarm
In this section on separator safety systems we have covered a very complex subject in a rather simplistic
If the level still rises way. Once again I must emphasise that the situations I described are not taken from any particular
• LSHH-01 will send a signal to the ESD process or separation system. Each process is unique. If you are involved with process operations,
system which will then close ESDV 1 you should make sure that you are completely familiar with the equipment on your own particular
installation.
When ESDV 1 closes, no more reservoir fluids will enter
the separator. Remember, safety is your concern.

Of course, the sequence I have just described is just one At the start of this section we considered the consequences of losing either level or pressure control in a
small part of the whole ESD system. Other things may separator. You saw that loss of control could result in a potentially hazardous situation.
happen. Depending on the hazardous situation which has We looked first at the problem of level control. You saw that an increasing or decreasing level can :
been detected, other ESD valves may open or close.
• generate alarms
For instance, in addition to closing ESDV 1, the system • cause an emergency shutdown
may also close ESDVs 2, 3 and 4. This would completely
isolate the vessel. The vessel may then be depressurised Moving to pressure control, you saw that alarms and shutdowns are also initiated by the pressure going
by opening ESDV 5 and venting the gas to a flare too high or too low.
system.
You also saw that additional pressure safety features such as Pressure Safety Valves and Rupture Discs
A complete ESD system is very complex. In essence may be fitted to a separator.
however it can be described as a system of sensors,
actuators and valves which are capable of automatically We finished the section by having a brief look at ESD Systems and valves, and you saw an example of
shutting down a process or part of a process. This a simple sequence of events which could occur on loss of level control.
renders the plant safe in the event of a hazardous Now that you have completed Section 5, you can move on to the final section in this unit, where we will
situation arising. look at separator operations.

33
Oil and Gas Separation Systems Petroleum Open Learning

Section 6 - Operation of Separators


In this, the last section of the programme, we are going
to look at some routine separator operations. You must
remember however that each system is unique and will have
its own operating procedures. Here we will consider general
operations of a hypothetical separator train.

Routine Operational Checks

The operator of a separation system will often have other


systems under his control. This is because, normally,
separators are smooth running items of equipment with very
few operational upsets.

The operator will usually check the following, on a regular


basis, during the course of a shift:

• Levels : All sight glasses and level controllers.


The operator should ensure that the sight
glasses are easily readable and that the levels
which they indicate correspond to the levels
indicated by the level controller. This may
involve draining the sight glass and blowing it
clear with gas.

Figure 51 illustrates the type of valve found at the


top and bottom of most sight glasses. You have
seen this type of valve before. It is similar to that
shown in Figure 25 on Page 35 in Workbook 1.
You will remember that we referred to it as a ball
check valve.
Figure 51 : Ball Check Valve

34
Petroleum Open Learning

The valve shown in Figure 51, however, Emulsions However, in some cases, the emulsion can be
is capable of shutting off any of the three treated in the separator itself. This involves the
outlets. It seals by the movement of a free A common operating problem is that caused by the injection of a chemical into the well fluids. This
floating ball. water and oil forming an emulsion. chemical, which is called a demulsifier, helps to
break down the emulsion and allows the separator
This is a mixture of two immiscible liquids where
Its operating position is indicated in the to do its job.
one of the liquids is dispersed throughout the other
drawing, with the valve stem fully retracted.
in the form of very small droplets. In the oilfield, the
Foaming
dispersed liquid is usually the water.
If the sight glass breaks, whilst the valve is in
Another problem which may be found in separators
this position, the flow of oil and gas, as they An emulsion may be classed as ‘tight’ or loose’.
is that of foaming.
escape, will cause the ball to move and seal
off the leaking glass. Milk is a tight emulsion. It is a mixture of butter fats
This is caused when the oil fails to release the gas
and water and it cannot be easily broken.
quickly enough as it passes through the vessel, and
* Pressures : All pressure gauges and a layer of oily bubbles forms on top of the liquid
Salad dressing is a loose emulsion. It is a mixture
pressure controllers. The operator should surface.
of oil and vinegar. When you shake the bottle an
ensure that all the vessels are working at the
emulsion forms and the small globules of oil and
desired pressure and the readings on the The level control displacer on the oil side of the weir
vinegar can be seen with the naked eye. If you let
pressure gauges correspond with the is designed to operate in a liquid. It cannot float in
the bottle stand for a few minutes the emulsion will
readings being given by the pressure foam.
break down and the oil will begin to float on the top
controllers. If differences in these readings
of the vinegar. When the float sinks in the foam it indicates a false,
are discovered, the cause should be
low level to the level controller and the oil outlet
investigated and the fault remedied. If emulsions are found in a separation process they
valve will close. This can result in the carry-over of
may be tight or loose. The type will depend, for
liquids with the gas stream and a possible shutdown
* Leakages : All vessels, interconnecting example, on the nature of the oil being produced
of the gas facilities downstream.
pipework and instruments. The operator and the amount of water present.
should ensure that there are no escapes of To stop this happening, anti-foam agents are often
liquids or gases from any of this equipment. One of the functions of a separator is, of course,
injected into the inlet stream to prevent foaming.
to remove the water from the oil. The presence
We will now take a look at some of the problems of an emulsion could obviously make this more
which can occur during normal operations. difficult. In fact, in extreme cases, water removal
from an emulsion may have to be done in a special
treatment plant.

35
Petroleum Open Learning

Slugging As the vessel is empty and depressurised, the


Start-up Procedure following switches will have been activated :
A third, but less common, problem in separators is
that of slugging. Before a start-up procedure is initiated, a number • LSLL-01. This will have closed ESDV 1, ESDV 2,
of checks have to be made to ensure that the ESDV 3 and ESDV 4 through the ESD system
Slugging occurs when, for some reason or another, separation train is ready.
there is an intermittent, rather than a constant, flow • LSLL-02. This will have closed ESDV 3
of well fluids into the separator. In some instances We will assume that the separators are empty, but
the flow may cease altogether for a few seconds are in a condition to receive hydrocarbons. • PSLL-01. This will have closed ESDV 1 and
and then a slug will arrive. ESDV 4
We will first check that:
This intermittent flow can cause rapid fluctuations The fact that ESDV 1 is closed means that we are
in separator levels and pressures. The controllers • all the valves, in the inlet manifold upstream not able to get fluids into the separator through this
react to these changes by rapidly opening and and the pipework downstream, are in the route.
closing their respective valves in an attempt to bring correct open/closed position for start-up.
the situation under control. In severe cases the We will need to inhibit the output signal from LSLL-
control system may become unstable resulting in a • well fluids will be available when we require 01 before we can open ESDV 1. We will then have
shutdown. them. to keep it inhibited until we have established a level
of oil, over the weir, higher than LSLL-01. When this
These are the three most common problems • all the sight glass bridles, sight glasses, occurs LSLL-01 will automatically re-set and will
associated with the operation of separators. level controllers, pressure gauges, pressure cease to have an effect on ESDV 1 and the other
controllers etc. on the separator are on-line ESDVs.
We will now have a look at a simple start-up and able to function properly.
procedure and a simple shut-down procedure for LSLL-02 only closed ESDV 3 so, although it will be
our separator train. As you work through the next few paragraphs, refer activated, we will not require to inhibit it in order to
to Figure 52 on the next page to remind yourself of open ESDV1.
the valve numbering.
PSLL-01 will also be activated. This switch also will
Let us consider the status of the vessel before we prevent us from opening ESDV 1. However, this is
introduce the well fluids. not a problem. If you refer to Figure 52, you will see
that there is a small by-pass line around ESDV 1.

36
Petroleum Open Learning

37
Petroleum Open Learning

So, having completed our checks, and with When the oil level reaches the set point of LSLL- Shut-down Procedure
LSLL-01 by-passed, we are ready to start-up 01, we can re-activate it. ESDV 1 is now fully
our separator. under the control of the ESD system. ESDV 2 If we slowly reduce the flow of fluids into the
will also be opened and LC-01 will take over separator, levels and pressure will fall and the
control of the oil level. controllers will close the relevant control valves.
We first of all open the small by-pass around
ESDV 1. This will allow a small flow of well fluids When ESDV 1 is open, and the system is If none of the valves leak, the pressure and
into the separator. The pressure will begin to operating automatically, we must always levels will be maintained at that point.
rise and the liquid level at the back of the weir remember to close the by-pass valve. If we
will also rise. don’t do this, and ESDV 1 activates, there Unfortunately this seldom happens. Control
will still be flow into the separator via the by- valves often do leak and levels and pressure will
When the pressure passes the set-point of pass, thus defeating the objective of the ESD continue to fall and, eventually, will activate the
PSLL-01, ESDV 1 will be opened. (Remember system. ESD system.
that LSLL-01 is still inhibited.)
When the water level reaches the set-point of In order to maintain levels and pressures, and
ESDV 4 will also be opened and, as the pressure LSLL-02, ESDV 3 will be opened. With the water thus facilitate easy start-up, the ESD valves are
in the separator rises to the desired value, PC- level at the desired value, LC-02 will maintain closed before these levels and pressures fall
01 will take over control. control. too low.

As the pressure in the separator is increasing, The ESD system on the separator is now fully If the separator is shut down in this condition,
so will the liquid level upstream of the weir. commissioned and will operate if we have a then it can be brought back on stream much
problem. more quickly.
By using the sight glasses we can keep a check
on the level of the well fluids as they fill the At this stage we have flow of fluids into the
space at the upstream side of the weir. separator and all controllers on-line in automatic
control.
The oil will now start to spill over the weir into the
oil end of the separator. We can then observe When we are satisfied that everything is
the level of the oil building on the downstream functioning normally, we can begin to increase
side of the weir. the flow of fluids into the separator up to the
operating rate.

38
Petroleum Open Learning

Blow-down Procedure One other feature on Figure 52 should be mentioned.


The restriction orifice (RO), downstream of ESDV
One procedure we have not mentioned is blow- 5, is a flat plate with a precision hole bored through its
down. centre. The RO serves the following functions :

As we have already explained, ESDV 5, which is • it reduces the pressure drop across ESDV 5,
shown in Figure 52, is a blow-down valve. It is fitted thus reducing wear on the valve
so that we may depressurise the separator in a rapid,
but controlled, manner. • it can be accurately sized so that the
flare system is not overloaded by too much
The control logic of a blow-down valve on a separator gas
is often designed so that it will not open if any of the
other ESDV’s on the vessel are open. Now that you have completed this section, try the
following Test Yourself.
The blow-down valves may be activated :

• automatically, by the ESD system, in the event


of a hazardous situation arising.

• by a manual signal from the Main Control Room,


or a local depressuring panel.

The blow-down valve would not normally be used as a


depressuring valve for maintenance or during normal
operations.

39
Petroleum Open Learning

Test Yourself 11
The following statements refer to :

a operational checks,
b separator problems,
or
c separator start - up procedures

Indicate to which of the three areas, a, b or c, each statement belongs.

1 Inhibit output signal from the level switch low low.


2 Increase flow of fluids into the separator.
3 Ensure level in the sight glass is readable.
4 Inject demulsifying chemical.
5 Bypass the ESDV on the separator inlet
6 Ensure that the vessel pressure is at normal operating value.
7 Inspect the connection between vessel and instrument bridle for leakage.
8 Liquid carry over occurs because of foaming.
9 Observe the level building on the upstream side of the weir.
10 Check that the inlet manifold valves are open.

You will find the answer to Test Yourself 11 on page 44

40
Petroleum Open Learning

Summary of Section 6
In Section 6, the final section in this unit on Oil and Gas Separation Systems, we have had a brief look at the operation of separators.

We started by considering the routine operational checks that an operator may have to make on a regular basis during the course of his shift or tour of duty. You saw that the
operational variables of level and pressure are constantly monitored. You also noted that the operator keeps a close look out for possible leakages which could give rise to
potentially hazardous situations.

Although separator operations are usually trouble free, there are some problems which Before you leave this unit and move on to another unit in the Petroleum Processing
may be encountered. You saw that potential problems included : Technology Series, I must make some final comments regarding the operation of
process plant:
• emulsions
• The Unit that you have just completed relates to separation
• foaming in general. It is not meant to describe any particular plant or
process
• slugging
• If you are involved in the operation of processing facilities,
Finally we looked at three basic procedures, including start-up, shut-down and blow- you should remember that each plant is different. You must
down. We went through a step-by-step procedure to be followed when starting a be completely familiar with the specific plant and equipment
separation system. The procedure was a hypothetical one, based on the separators under your control
described in previous sections of the unit. We then looked at shut-down and blow-
down in more general terms. • Laid down procedures and operational guide-lines must be
followed, and safe working practices adopted at all times

41
Petroleum Open Learning

Check Yourself 7
Fill in the missing words from the following paragraph.

That part of a continuous process which an operator wishes to control within certain limits is

called the PROCESS VARIABLE, and its target value is called the DESIRED value. A

measuring unit obtains a measured value from the process and feeds it to a CONTROLLER

whose job it is to compare the two values. If a deviation exists between the two values, the

CONTROLLER sends a correcting signal to the final element in the loop � a correcting unit

which is usually a CONTROL VALVE.

42
Petroleum Open Learning

Check Yourself 8 Check Yourself 9


The following items from a level control loop are part of the measuring unit, the level controller If signal pressure is required to close the
or the level control valve. Indicate by a tick in the box provided, to which part of the system each valve a loss of signal pressure causes the
belongs. valve to open. It is therefore in a fail open
mode.
Measuring Unit Level Controller Level Control Valve
pilot relay 3

differential mechanism 3
torque tube rod 3


actuator spring 3

valve plug 3

flapper nozzle assembly 3

diaphragm 3

feedback unit 3
float rod 3


valve stem 3

43
Petroleum Open Learning

Check Yourself 10 Check Yourself 11


Indicate, to which of the three areas, a, b or c, each
statement belongs.

1 Inhibit output signal from the level


switch low low. (c)
2 Increase flow of fluids into the separator, (c)
3 Ensure level in the sight glass is readable, (a)
4 Inject demulsifying chemical. (b)
5 By-pass the ESDV on the separator inlet, (c)
6 Ensure that the vessel pressure is at
normal operating value. (a)
7 Inspect the connection between vessel
Figure 53 and instrument bridle for leakage. (a)
8 Liquid carry over occurs because
Figure 53 shows the water end of a 3 phase This instrumentation consists of: level controller
of foaming. (b)
separator. An instrument bridle is connected to incorporating level alarms, high-high and low-
the separator with the top connection in the oil low level switches and, of course, one or more 9 Observe the level building on the upstream
and the bottom connection in the water. The sight glasses. (Only one is shown in Figure 53 side of the weir. (c)
instrumentation on the bridle is identical to the for simplicity).
10 Check that the inlet manifold valves
instrumentation on the bridle at the oil end of are open. (c)
the separator.

44
POL
Petroleum Open Learning

Oil Pumping
and Metering
Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY
POL
Petroleum Open Learning

Oil Pumping
and Metering
Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY
Petroleum Open Learning

Designed, Produced and Published by OPITO Ltd., Petroleum Open Learning, Minerva House, Bruntland Road, Portlethen, Aberdeen AB12 4QL

Printed by Astute Print & Design, 44-46 Brechin Road, Forfar, Angus DD8 3JX www.astute.uk.com

© OPITO 1993 (rev.2002) ISBN 1 872041 85 X

All rights reserved. No part of this publication may be reproduced, stored in a retrieval or information storage system, transmitted in any form or by any means, mechanical,
photocopying, recording or otherwise without the prior permission in writing of the publishers.
Oil Pumping and Metering Petroleum Open Learning

(Part of the Petroleum Processing Technology Series)

Contents Page Visual Cues


* Training Targets 4
training targets for you to
achieve by the end of the unit
* Introduction 5

test yourself questions


* Section 1 - Centrifugal Pumps: Terms and Concepts 6 to see how much you
understand
Liquids, Gases and Fluids
Mass, Force and Weight
Density and Secific Gravity
Centrifugal Force check yourself answers to
Kinetic Energy and Pressure Energy let you see if you have been
Pressure
Head Pressure thinking along the right lines
Net Positive Suction Head (NPSH)
Cavitation
Flow v Differential Pressure activities for you to apply
your new knowledge

* Section 2 - Construction and Operation of Centrifugal Pumps 18


Impellers and impeller Speed summaries for you to recap
Pump Casings
Bearings on the major steps in your
Pump Configurations progress
Centrifugal Pump Performance Curves
A Centrifugal Pump Arrangement
Minimum Flow System
1
Petroleum Open Learning

2
Petroleum Open Learning

Contents (cont'd) Page Visual Cues


* Section 3 - Oil Metering and Sampling 33
training targets for you to
Differential Pressure Metering achieve by the end of the unit
Turbine Meters
Metering Systems
Meter Proving
Sampling Systems test yourself questions
to see how much you
understand
* Section 4 - Pig Launching Facilities 42
Types
Pig Launchers check yourself answers to
Pig Launching Problems let you see if you have been
Basic Rules for Pig Launching thinking along the right lines
Safety Systems

* Section 5 - A Typical Oil Pumping and Metering System 49 activities for you to apply
your new knowledge
Booster Pumps
Sampling System
Metering System
Oil Pipeling Pumps
Pig Launching summaries for you to recap
on the major steps in your
progress
* Check Yourself - Answers 65

3
Petroleum Open Learning

Training Targets

When you have completed this unit on Oil Pumping and Metering, you will be able to:

• Explain some of the technical terms and concepts which lie behind the operation of a centrifugal pump

• List the component parts of a centrifugal pump.

• Explain the operating principles of a centrifugal pump.

• Describe the construction and operation of turbine and differential pressure meters.

• Explain the function and operation of a typical meter run.

• Describe the procedure for proving a meter

• List the essential elements of an oil sampling system

• Detail the main features of a pig launching system, and its method of operation

• Describe a typical layout for the oil handling (or oil pumping and metering) section of a production facility

Tick the box when you have met each target

4
Oil Pumping
and Gas Separation
and Metering
Systems Petroleum Open Learning

Introduction
In this unit, we will be looking at the oil handling section of a production facility. The system is often referred to as the oil
pumping and metering system.
The equipment needed for this system will usually be situated between the final stage of separation
(into oil, gas and water streams) and the point where crude oil leaves the production facility for The unit is divided into five sections :
transfer to a pipeline, oil tanker or terminal. The layout is illustrated in Figure 1.

in Section 1, we will look at some basic terms


and concepts relating to centrifugal pumps.


Section 2 concentrates on the construction
and operation of centrifugal pumps.

Section 3 gives you an overview of the


metering and sampling part of the system.

In Section 4, pigs and pig launching facilities


are described.

Finally, in Section 5, we will go through a


typical oil pumping and metering system.

Figure 1: Oil Pumping, Sampling and Metering

5
Oil Pumping and Metering Petroleum Open Learning

Section 1 - Centrifugal Pumps : Terms and Concepts


In this first section, we will look briefly at a number Weight is a measure of this force acting on the
of concepts which relate to the operation of object. Therefore, a one pound mass will have
centrifugal pumps. I will also explain some of the a force of one pound weight acting on it, due to
terms often used when we are discussing how
these concepts can be applied in practice.
gravity.
Test Yourself 1
The two terms “mass” and “weight” cause a lot
5 litres of water has a mass of 5 kg
Throughout this unit we will be concentrating on of confusion. Very often they are used as if they
centrifugal pumps because these are the most mean the same thing. In many cases, however,
5 litres of crude has a mass of 4.25 kg
common ones used in oil pumping and metering this is not all that important and I think that the
services. brief explanation given above should be sufficient
5 litres of salt water brine has a mass of
to guide you through the remainder of this unit
5.5 kg
Liquids, Gases and Fluids without any undue problems.
What are the specific gravities of gas oil and
Both liquids and gases are called fluids because Density and Specific Gravity brine?
each has the ability to flow.
The density of a substance is defined as the mass
In this unit we will use the term fluid when per unit volume of that substance. For the same You will find the answers to Test Yourself
describing something which can happen to a gas material, density can be expressed in a variety of 1 on page 65.
or a liquid. When we need to make a distinction, I units. For example, the density of water is :
will use the specific term liquid or gas.
1 gram per cubic centimeter -1 gm/cm3
Mass, Force and Weight
Specific gravity (s.g.) compares the mass of a
For solids and liquids, the reference material used
certain volume of a material with the mass of an
The mass of an object is a measure of the quantity is usually water. For gases, the reference is often
equal volume of a reference substance. In other
of matter present. This object may have various to air.
words :
forces acting on it, the most important of which
is likely to be the force of gravity. You can easily
demonstrate that there is a force acting on the mass of a certain volume of material
specific gravity (s.g.)
object. Hold it out and release it - the force of mass of an equal volume of reference substance
gravity will pull it towards the earth.

6
Petroleum Open Learning

Centrifugal Force It shows a spinning disk. If we let a drop of water


fall onto the face of the disk, near to the centre
Have a look at Figure 2. spot, it will follow the type of path shown. This is
because the drop is affected by two forces during
its travel:

centrifugal force, which tends to throw


the droplet outwards, causing it to speed
up as it approaches the edge of the disk

friction, which will cause the disk to


attempt to drag the droplet round with it as
it rotates

The relative size of these two forces will determine


the angle at which the droplet leaves the disk
edge.

This angle is important, as you will see when


we come to the section on Construction and
Operation of Centrifugal Pumps (Section 2). The
design features of the pump encourage a flow
path for the liquid being pumped, which is very
similar to the droplet trajectory in Figure 2.

This ensures that the pump imparts the maximum


amount of energy to the liquid. In this case, energy
of motion, or kinetic energy is transferred.

7
Petroleum Open Learning

Kinetic Energy
We have seen that kinetic energy is energy of
motion, or movement. Test Yourself 2
The amount of kinetic energy possessed by any A small car has a mass of 1 000 kilograms,
moving object depends upon: and is travelling at 180 kilometres per hour.
• its mass (“weight”)
A large truck has a mass of 20 000 kilograms,
• its velocity (“speed”) and is travelling at 30 kilometres per hour.
In mathematical terms, kinetic energy (KE) can be
calculated by using a formula: Which one has the greater kinetic energy ?
KE = 1/2 mass x velocity2

If the mass is expressed in kilograms and the You will find the answer to Test Yourself 2
velocity in metres per second, the kinetic energy on Page 65.
will be in joules.

To confirm your understanding of this relationship,


try the following Test Yourself.

8
Petroleum Open Learning

Kinetic Energy and Pressure Energy Let us go into a little more detail on these pressure
and velocity changes by considering six separate
Figure 3 illustrates the flow of a fluid across a restriction, and how the fluid velocity and points in the process:
pressure vary during this process.
point A : fluid is flowing along the pipe at a steady
speed and (almost) constant pressure. You will
remember that the kinetic energy of this fluid can
be calculated by the equation:

KE = 1/2 mass x velocity2

The velocity of the fluid at point A is constant. In


addition, the mass of fluid passing each point in
the pipeline per unit of time (mass flow rate) is
also constant. This means that the kinetic energy
content of the fluid at that point is also constant.

point B : the fluid starts to enter the restriction.


The mass flow rate remains constant but, because
the pipe diameter is smaller, the fluid velocity must
increase.

Looking again at the kinetic energy equation, you


will see that the kinetic energy of the fluid will start
to increase at this point as the fluid speeds up.

9
Petroleum Open Learning

Now let me introduce you to another principle of Pressure This cubic metre of water weighs 1000 kg. In other
science - Conservation of Energy. words, due to the effects of gravity, it is applying a
Pressure expresses the relationship between downward force of 1000 kg, spread over its base.
This tells us that the total energy content of a force (or weight) and area, as follows:
system will always remain constant The pressure on the base of the cube is therefore:
kilogram force (or weight)
If the kinetic energy content of our system increases 1000 kg/m2
pressure = area
then, to compensate for this, some other form of
Like density, pressure can be measured in a variety
energy possessed by the system must decrease. However, we have just seen that pressure is usually
of units. The most common are pounds per square
This other form of energy is pressure energy.
inch (psi), or kilograms per square centimetre (kg/ expressed in bar.
Figure 3 shows that, as the velocity (kinetic
cm2). We normally use the SI term bar, as 1 bar is
energy) increases, the pressure (pressure energy) As you will see from Figure 4, the base of this water
almost the same as 1 kg/cm2. (1.019 kg/cm2 to
decreases. cube has an area of one square metre, or:
be exact).
point C : this is a new steady state. The fluid has
Picture a metre cube of water: 100 cm x 100 cm = 10,000 cm2)
a higher velocity and a lower pressure but both
of them are steady as the fluid passes across the
So, on each square centimetre of the base a
restriction.
downward force of 1000 kg is applied. 1000 kg
points D and E reverse the changes which is applied. 10,000
occurred at points A and B.

It is worth noting that, across the process overall, The pressure on the base can, therefore, also be
a small reduction of pressure has occurred. Due to expressed as:
turbulence in the system, some pressure energy
has been converted into heat energy. You will no 1000 = 0.1 kg/cm 2, or 0.1 bar.
doubt appreciate that, under conditions of high 10,000
flow rates, high turbulence, or extended restrictions
(say, a long pipeline run), pressure losses will be
greater.

We will look into the effects of pressure loss and


flow a little later on in this section.
10
Petroleum Open Learning

Head Pressure We have already shown that centrifugal force can


impart kinetic energy to a substance as a result of
The term head pressure or head is often used in a spinning action. We have also seen that kinetic
the oil and gas industry, especially when referring
to pumps. It is the pressure developed by a head,
Test Yourself 3 energy can be converted into pressure energy.

or column height, of liquid. Centrifugal pumps are dynamic pumps which,


The specific gravity of crude oil is 0.85, and
primarily, impart kinetic energy to the fluid being
that for a particular salt water is 1.1.
In the paragraph entitled “Pressure”, we saw that pumped. They do not create pressure directly.
the head pressure applied by 1 metre depth of Pressure results from the liquid slowing down, and
What will be the head pressure developed
water will be 0.1 bar. For 3 metres of water the the kinetic energy converting to pressure energy.
by 3 metres of crude oil, and 4.5 metres of
head pressure would be 0.3 bar; for 30 metres, 3
this brine?
bar, and so on. The pressure developed will depend on the
density of fluid being pumped.
Now let us combine what we know about specific
gravity and head pressure. Try the following Test A centrifugal pump, working at a fixed flowrate,
The answers to this Test Yourself are on
Yourself to combine these two factors together: will generate the same height of head, but will
page 65.
generate a lower head pressure, when pumping
crude oil, than when water is being pumped,
because water is heavier than crude oil.

The different categories of head pressure referred


to in pumping operations are shown in Figure 5
on the next page.

11
Petroleum Open Learning

The suction head represents the head pressure


present at the pump suction.

The discharge head represents the head


pressure delivered by the pump.

The total head (which is the difference between


suction and discharge heads) represents the
additional pressure imparted to the liquid by the
pump.

12
Petroleum Open Learning

Net Positive Suction Head (NPSH) If these effects are observed in water and lemonade, The accompanying drop in pressure may cause
it is reasonable to assume that they will happen in gas or vapour to be released for either of the
I would like you to think about two common situations other liquids as well. So let us now visualise how these reasons described above. It is important that we
in which you have seen bubbles coming out of a effects can influence the operation of a pump. prevent this happening, for reasons that I will
liquid. explain a little later under “Cavitation”.
We already know that fluids can only flow from areas
1. If you heat up a pan of water, two things happen: of high pressure to areas of low pressure. Suppose We must therefore always have sufficient pressure
that the liquid being pumped enters an area of low at the pump suction to prevent gas or vapour
• long before the water boils, bubbles are seen pressure. Then: release for whatever reason. The minimum
rising through the liquid as dissolved air comes pressure necessary to do this is called the net
out of solution when the temperature begins to • if the liquid was near its boiling point, positive suction head (NPSH).
increase the pressure drop may cause the
liquid to boil and thus release gas or A further pressure reading which is relevant to
• at the boiling point, the liquid bubbles vapour the suction end of the pump is called the static
vigorously as the water is rapidly converted suction line pressure. As the name implies,
into steam • if the liquid was near to the pressure this is the measured pressure at the pump suction
at which dissolved gases are released, when pumping has stopped.
Water at sea level boils at 100°C (212°F). I am sure the pressure drop may cause these
you will have heard, however, that the boiling point gases to come out of solution We now have three pressure values which relate to
of water (or any other liquid) falls as you climb from the pump suction:
sea level, so that it can be difficult to cook an egg In either case we can predict that, if the
properly on top of a high mountain. This is because pressure is increased again, the released a. the pressure at which gas or vapour
atmospheric pressure falls the higher up we get. gases will go back into the liquid, either is released
because boiling stops or the released gases
2. If you open a bottle of fizzy lemonade, bubbles re-dissolve. b. the static suction line pressure
are seen rising through the liquid as dissolved gas
comes out of solution when the pressure is released c. the NPSH
(reduced). When a centrifugal pump is running, a low
pressure area is created at the suction. This
encourages liquid further upstream to flow into
the pump suction.

13
Petroleum Open Learning

To be safe, most pumps will be operated just As these tiny cavities created in the liquid
above their NPSH. An adequate safety margin collapse, the liquid tends to rush in from all
for most applications would be 1 metre, or angles to fill the cavity. The cavity is said to
Test Yourself 4 10%, head of water pressure above the NPSH
specified by the manufacturer (whichever is
implode.

the larger). This inrushing liquid can transmit very large


As a check on whether you have understood forces. When the bubbles are near a metallic
what I have ben saying about pressures at surface, these forces are applied directly to the
In general, the industry standard is to work
the suction end of the pump, list these three solid surface. When a pump is cavitating, this
in terms of “head of water”. This is because
pressure valves: process is being repeated many thousands
everyone knows the density of water and
pumps can easily be tested to make sure that of times each second and the effect results in
a. the pressure at which gas or vapour noise, vibration and eventual erosion of metal
they produce the level of head specified.
is released from the surfaces. In very severe cases, for

b. the static suction line pressure Cavitation example where the pump is handling liquids
carrying small solid particles, the impeller can
be eroded in a relatively short space of time.
c. the NPSH We have discussed at some length the
importance of NPSH and other factors in
An equally important factor is that severe
in order of decreasing pressure, and preventing the release of gas or vapour bubbles
cavitation can result in a failure of the pump to
see if you can explain the reasons for your in the suction of the pump. We will now look at
deliver flow at the expected head.
answers. why it is so important to prevent this.
When pumping oil, the drop in head and
You will find the correct answers in Check If gas is released at this point in the system, it efficiency is not quite so severe as with water
Yourself 4 on Page 66. will give rise to an effect known as cavitation. because the liquid is composed of mixtures
of different hydrocarbon compounds. The
The formation of bubbles is, in itself, quite bubbles which appear will consist of lighter
harmless. However, as the liquid containing hydrocarbons such as methane or ethane.
these gas bubbles, or cavities, passes through These can be more easily reabsorbed as the
A pump manufacturer will specify the NPSH the pump, the pressure will rise. Now we already pressure is increased. When pumping water
and maximum operating temperature required know that, if gas is released from a liquid for the bubbles are nearly always caused by the
for each pump to handle a given liquid the reasons I have described, an increase in water boiling at a reduced pressure. In this
effectively. The NPSH should be maintained pressure will drive this gas back into the liquid situation the bubbles collapse violently and
over the entire range of the pump again. each implosion is of a high intensity.
14
Petroleum Open Learning

Flow v Differential Pressure


Take another look at Figure 3. You will recall that
it illustrates the conversion of pressure energy
to kinetic energy, and the reverse, as a fluid
passes through a restriction. Remember also that,
because of turbulence, some pressure energy
is converted to heat energy. This conversion
is responsible for the pressure loss shown in
Figure 3.

Any pipeline will contain a whole series of


restrictions. These may be bends, changes in
diameter, obstructions and rough internal surfaces,
for example.

You will probably realise, therefore, that:

• at low flow rates the turbulence caused


by these restrictions may well be small,
therefore minimising the pressure loss

• at high flow rates the turbulence could


be very high, as will be the pressure
loss

We will now take a look at the relationship


between flow and differential pressure between
two points in, say, a pipeline.

Have a look at Figure 6.

15
Petroleum Open Learning

This shows the relationship between flow and (You will have noticed that, when the flow rate
differential pressure, both expressed as a doubled, the differential pressure increased by a
percentage of the maximum possible under those factor of 4).
particular circumstances.
• let us now increase the flow rate to 4 litres/
Test Yourself 5
We can see, for example, that 50% of the maximum min (40% of maximum). The differential
flow is equivalent to 25% of the maximum pressure rises to 16% of maximum, or 1.6
bar. Again, as the flow rate doubles, from In the example we have just used, if the
differential pressure.
2 to 4 litres/min, the differential pressure differential pressure fell from 70% to 40%
quadruples, from 0.4 to 1.6 bar. of maximum, what would be the change in
Now let us suppose that, in our pipeline, we can
flow rate, expressed in litres per minute?
generate up to 10 bar of pressure at the inlet
and deliver up to 10 litres of-liquid per minute. This relationship between flow and differential
pressure can be expressed as a mathematical The answer can be found in Check
Let us also suppose that, at the outlet, the liquid
equation : Yourself 5 on Page 66.
discharges into a pond at 0 bar.

We therefore have a differential pressure of 0- F = √DPxl0


10 bar and a flow rate of 0-10 litres/min. Let us
look at the conditions under different flows and where:
pressures.
F = flow rate as a % of maximum
• if we regulate the inlet flow to 1 litre/min DP = differential pressure as a % of
(10% of maximum) we could expect very maximum
little turbulence. From Figure 6 we can (√DP means the square root of DP)
estimate that the differential pressure will
be 1 % of maximum, or 0.1 bar, at this flow
rate. That ends our brief look at some of the key factors
which affect the design and operation of centrifugal
• if the flow is increased to 2 litres/min (20% pumps. Before we go on to the next Section,
of maximum), both turbulence and the however, try the following Test Yourself.
pressure drop will increase. At a 20% flow
rate, the differential pressure will rise to 4%
of maximum, or 0.4 bar.
16
Petroleum Open Learning

Summary of Section 1

In this section, we have looked at some of the scientific terms Net positive suction head (NPSH) was fully described,
and concepts which help us to understand the design and and we saw how important it was in relation to preventing
operation of centrifugal pumps. cavitation.

You will remember, for example, that both liquids and gases
are called fluids because they have the ability to flow. We We looked at differential pressure and flow. The
saw how fluids flow from high energy areas to low energy relationship was expressed as a graph, and also as a
areas. mathematical equation.

We examined the relationship between mass, force and


weight, and I tried to clear up some of the confusion which
exists in the common use of these words.
You are now ready to take a look at the construction and
operation of a centrifugal pump, and see how the terms and
Density and specific gravity were explained. concepts covered in Section 1 can be applied to the design
and performance of this type of pump.
I illustrated centrifugal force by asking you to visualise the
movement of a water drop on a spinning disk.

This led us to a description of kinetic energy, and how


kinetic energy and pressure energy can be interchanged.
We introduced the concept of conservation of energy.

17
Oil Pumping and Metering Petroleum Open Learning

Section 2 - Construction and Operation of Centrifugal Pumps


The centrifugal pump is the commonest form of pump in use
today. It is relatively cheap, easy to maintain and is to be found
almost everywhere when large flows are required.

We will first take a look at the basic configuration of a centrifugal


pump and then at the component parts, to see what they do and
how they work.

The type of pump illustrated in Figure 7 is one of the simplest. It


consists of:

• a casing, which contains and supports the rest


of the pump components. Access to the inside of
the pump is via a vertical split at the back of the
casing (not shown)

• a suction flange, which directs the liquid


entering the pump casing into the impeller

• an impeller, which imparts kinetic energy to the


liquid

• a pump shaft, connected through a coupling to a


motor which drives the shaft and the attached
impeller(s)

• a bearing housing, which supports the shaft

• a shaft seal, which prevents liquid escaping from


the casing along the shaft

• a discharge flange, which directs the liquid


away from the pump

18
Petroleum Open Learning

Figure 8 shows a cross section through a single


impeller pump, illustrating three more key items of
equipment:

• The wear rings, which act as seals between


the high pressure discharge side and the low
pressure suction side of the impeller

The wear rings are so called because they


wear in preference to the impeller. They are
‘sleeved’ on to the impeller, and may be
replaced when worn

• The balance holes, which allow the packing


to operate at suction pressure rather than
discharge pressure. This reduces the
differential pressure across the packer and
impeller, and therefore reduces the “thrust”
forces

• The packing, which prevents liquid escaping


from the casing

We will now examine some of these components in


more detail.

19
Petroleum Open Learning

Impellers The water pump on your car will probably be fitted


with a semi-open impeller similar to the one
We have already seen that a spinning disk can shown in the centre of Figure 9. The semi-open
impart kinetic energy to a drop of water on its impeller is a little more efficient and a little more
surface. A centrifugal pump, which is a dynamic expensive than the open impeller.
pump, does a similar job on the liquid it is pumping.
The pump then converts this kinetic energy into The car pump has to be reasonably efficient to
pressure energy before the liquid leaves the provide engine cooling by circulating water around
the engine and through the radiator. But, every
outlet.
bit of energy used in the water pump means that
The elements of the pump which impart kinetic there is less available to propel the car itself. The
energy to the liquid are called impellers. We will semi-open impeller is therefore a compromise
now look at the three basic kinds of impeller and between efficiency and cost.
see how they differ from each other. In the oil industry, the closed impeller is the one
All impellers are fitted with curved vanes which most often used. This is shown at the bottom of
spread out radially from the centre. The impellers Figure 9. It is more efficient than other types of
are attached to the pump shaft and rotate with it. impeller, but is also considerably more expensive.
Figure 9 shows the three most common types This is because special techniques are needed to
of impeller. weld the vanes to the inside of the shroud which
covers the impeller.
Your washing machine at home probably has a
pump with an open impeller similar to the one You should notice, in particular, the curve on each
shown at the top of the diagram. Open impellers impeller vane, and compare this shape with the
are cheap to make but they are inefficient. shape of the droplet trajectory in Figure 2. They
are very similar.
The one on your washing machine will be there
Vanes are designed in this way to impart the
to empty the machine. Washing machine pumps,
maximum amount of kinetic energy to the liquid
however, have to cope with debris - buttons, fluff,
being pumped, and to ensure that this liquid
coins and the like. An open impeller is ideal. It will leaves the impeller rim at a particular angle.
handle most foreign objects and, if it is broken, it
is cheap to replace. This angle will be matched by the shape of the
volute, or angle of the diffusers, depending
on the pump design. (I will talk about volute and
diffuser casings shortly).
20
Petroleum Open Learning

Impeller Speed The upper diagram shows a volute casing. In


this type of pump, the liquid leaves the tip of the
The type of impeller selected will depend on the impeller, and is thrown into a channel with an
planned speed of rotation, and the type and size increasing area of cross-section. Here the liquid
of pump. slows down and kinetic energy is converted into
pressure energy. areas
As a general rule: increase
constantly
The volute design ensures that it is aligned with
• low pressure, high capacity pumps will have the trajectory of the liquid leaving the impeller. This
large diameter impellers with a low rotating ensures efficient energy transfer and conversion.
speed
The liquid is then guided towards the pump
• highpressure,highcapacitypumpswillhave discharge flange.
small diameter impellers with a high rotating
speed The volute type of pump is the most common type
in use.

Pump Casings
We already know that the velocity of the liquid
increases as it passes across the impeller. We The lower diagram shows a diffuser casing. In
also know that, as the velocity decreases, the this type of pump, as the liquid leaves the tip of the
pressure will increase. Figure 10 shows the two impeller it moves through a set of angled vanes
main types of casing which allow this to happen known as diffusers. Again, these are lined up
within the pump. with the direction of the pumped liquid as it leaves
the impeller. The diffusers then guide the liquid
into the outer section of casing where its velocity
decreases and pressure increases before flowing
to the discharge flange.

21
Petroleum Open Learning

Bearings
Figure 11 is an illustration of a simple bearing
arrangement. The shaft is supported by two radial ball
bearing races, which allow it to rotate with minimum
friction.

Thrust force is a force which is directed along the axis


of the pump shaft. It arises because of the difference in
pressure between the discharge and suction sides of
the pump acting on the impeller. In Figure 11, the thrust
force will be from right to left, (from high pressure to low
pressure).

In this case, to counteract the thrust force, a ball bearing


race (the thrust bearing) is mounted between two
vertical plates. It allows the shaft to turn with a minimum
of friction as it takes up this thrust force.

The slinger rings (also called flinger rings) are two


slender rings, often of brass, which slide up and down
the shaft as it rotates. The slinger rings dip into the
lubricating oil and, as they turn, transfer oil onto the shaft.
The oil then runs along the shaft and contacts the faces
of the bearings. Centrifugal force throws the oil outwards
along the bearing faces to lubricate and cool them.

The oil in this type of bearing is either topped up through


an oil fill plug, as shown, or is automatically replenished
via an oil bottle arrangement.

22
Petroleum Open Learning

Seals In some cases a lantern ring is fitted between However, any leak would be dangerous when
sections of the asbestos packing so that any pumping oil or other hazardous liquid. In such
Figure 12 is an illustration of a typical packed liquid which has leaked along the shaft can be cases, a mechanical seal would probably be
seal. In this type of seal the packing consists of removed. used. A typical seal is illustrated in Figure 13 on
rings of asbestos rope which are impregnated with the next page.
graphite. The rings are placed around the shaft The problem with this type of shaft seal is that
and compressed into a packing gland by means small leaks almost always occur, whatever liquid is
of a gland follower, the pressure on which can being pumped. These leaks are usually necessary
be adjusted by four bolts. in order to keep the packing lubricated and to
prevent the shaft from overheating.

23
Petroleum Open Learning

The shaft enters the pump casing from the right hand
side of the diagram and passes through a stationary
seal. The stationary seal is fixed to the pump casing
and does not rotate.

Attached to the shaft is a rotary (or rotating) seal.


Leakage along the length of the shaft is prevented
by ‘O’ rings which seal the gap between shaft and
rotating seal. The ‘O’ rings turn together with the
shaft and rotating seal.

The sealing faces of the rotating and stationary


seals are usually of machined carbon or high grade
stainless steel which are polished to a mirror finish.
The two faces are held very closely together by a
spring and by the pressure of the liquid in the pump.

A small amount of the liquid being pumped is often


taken from the discharge of the pump, filtered,
and then returned through the mechanical seal via
the seal flush inlet. This liquid helps to keep the
mechanical seal clean, cool and lubricated.

24
Petroleum Open Learning

Pump Configurations In Figure 14a, a single pump is delivering 500 litres per minute with
a total head of 3.5 bar:
Figure 14 shows examples of how centrifugal discharge head - suction head = total head.
pumps may be configured to increase flow, or to
increase pressure. Figure 14b shows that, to increase the flow, two pumps arranged in
parallel are needed - that is, the pumps have a common suction
and a common discharge.

In this case, we can :

• run either pump on its own to produce a flow rate of 500


litres per minute and a total head of 3.5 bar, or,

• run both pumps together to produce a flow rate of 1000


litres per minute and a total head of 3.5 bar.

In Figure 14c, we can increase the pressure by running two pumps


in series. This means that the first pump discharges into the
suction of the second pump. In this case:

• both pumps must be run together

• the combination of both pumps will produce a flow rate of


500 litres per minute and a total head of 7 bar.

In most instances where high pressures are required, it is easier to


mount a number of impellers on a single shaft. These pumps are
called multi-stage pumps. They give us high flow rates, and a
gradual pressure rise over as many stages as required. Some main
oil pipeline pumps may have more than eight impeller stages.

25
Petroleum Open Learning

Centrifugal Pump Performance


Curves
Every centrifugal pump is designed and manufactured
for a specific purpose. This purpose is summarised in
a pump performance curve.

Figure 15 shows a typical performance curve which


gives us the following details about a specific pump:

• on the left hand side of the curve, there are


three vertical scales:

1. efficiency - from 0-100%. This compares


the power the pump is using to the work
it is achieving

2. power - from 0-24 kilowatts in this case.


This indicates the amount of power the
motor is consuming

3. total head - this indicates the pressure


which the pump can achieve

• on the top right hand side of the chart we


can see the required NPSH (net positive
suction head) in metres of liquid. (You will
recall that NPSH was described in Section 1
on Page 13)

26
Petroleum Open Learning

• the horizontal axis of the chart gives flow rate


in cubic metres per hour.

• in the body of the chart we find curves which Test Yourself 6 Test Yourself 7
show the relationship between:

1. NPSH and flowrate When pumping 20 cubic metres per hour the When the flow rate increases to 40 cubic
pump will: metres per hour this pump will:
2. efficiency and flowrate
• require a minimum of metres • require a minimum of metres
head of the liquid NPSH head of the liquid NPSH
3. power and flowrate
• develop metres total head of • develop metres total head of
4. total head and flowrate liquid liquid

• consume kilowatts of power • consume kilowatts of power

• operate at efficiency • operate at efficiency

Take a few minutes to study Figure 15 and then try


Test Yourself 6 and 7.

The answers to these can be found in Check Yourself 6 and 7 on Pages 66 and 67.

27
Petroleum Open Learning

Now, to summarise what we have covered so far in


Section 2, try Test Yourself 8:

Test Yourself 8
Indicate with a tick, to which part or parts of the pump the following items belong.

Item Casing Impeller Bearing Seal

Shaft sleeve 'O' ring


Shroud
Lantern ring
Wear rings
Flush inlet
Vane
Slinger ring
Balance holes
Gland follower
Volute
Ball bearing race
Diffuser

Check your answers in Check Yourself 8 on Page
67.

28
Petroleum Open Learning

A Centrifugal Pump Arrangement


Before we look at a typical oil pumping system, let us think about those items of equipment which you are
most likely to come across.

Figure 16 is an illustration of a typical


centrifugal pumping arrangement.

29
Petroleum Open Learning

The motor which drives the pump is called the The discharge of the pump is fitted with: There would be great turbulence inside a pump
main driver. In this case, the main driver is an with the impeller turning through liquid trapped
electric motor, but for bigger pumps it may be a • a discharge pressure gauge, which within the pump. The temperature would rise,
gas turbine or a diesel engine. indicates the pressure produced by the increasing the chances of cavitation.
pump
The motor has a set of local switches for starting In some instances, with large and powerful pumps,
and stopping. In some cases a local ammeter • a discharge check valve, which only damage can then occur in a matter of seconds. In
is fitted to check whether the motor is running allows flow in one direction, away from the smaller machines it may take much longer - but
properly. In many instances, the pump motor may pump.Thisvalve,therefore,preventsliquid damage will eventually occur.
also be started from a remote location, such as a flowing back through the pump, back-
To prevent this situation from happening, a
control room, either manually or via an automatic spinningitandcausingdamagetotheseals
minimum flow must be established and maintained
start system. and bearings of both pump and motor
through the pump at all times while running. This
minimum flow level is specified by the pump
The motor shaft is linked to the pump shaft via a • a discharge block valve, which can
manufacturer.
coupling, designed to transmit power from the be used to isolate the pump from the
motor to the pump, and to take care of any small downstream process, if required All centrifugal pumps which are at risk can be fitted
shaft misalignments which may occur. with a minimum flow system. This ensures that,
Minimum Flow System while the pump is running, there is sufficient liquid
The flow of liquid into the pump is through a flow to ensure that no damage occurs.
suction block valve, which can be used to All centrifugal pumps require one other item of
isolate the pump from the upstream process if equipment for their protection. In some instances, the minimum flow system
required. Occasionally, a strainer or filter (not shown) consists of a simple orifice plate, sized for the
may be fitted to the suction line, downstream of If we look back at the performance curve in Figure correct flow. The plate is inserted into a line
the suction block valve, to prevent debris from 15 we can see that, when the pump is running at through which is re-cycled a fixed flow from pump
entering the pump. zero flow, it is still using about 4 kilowatts of power. discharge to pump suction at all times.
We also know, from the performance curve, that
The pump casing is fitted with: In other instances a flow measuring device is
the pump efficiency will have fallen to zero.
fitted into the suction of the pump. This device
• a casing vent valve, used to bleed off any controls a flow control valve, inserted into a line
So, what has happened to the power we are using ?
gas or air in the pump before starting which re-cycles a fixed amount of flow. If the flow
falls below the pre-set minimum level, the flow
• a casing drain valve, used to drain liquid The answer, of course, is that it is converted into
control valve will open to restore flow rate to the
from the pump after shutdown heat energy.
minimum.
30
Petroleum Open Learning

A simple and very common device is illustrated


in Figure 17. It is called a minimum flow non-
return valve and serves the purpose of a check
valve and a minimum flow valve.

Ih Figure 17a, there is no flow through the main Figure 17b shows that there is some flow through In Figure 17c, all flow is through the main part
part of the valve, but the two smaller valves are the main part of the valve, but the two smaller of the valve and the two smaller valves are fully
fully open to let liquid flow to the minimum flow valves are still partially open, allowing some liquid closed. In this situation the pump is pumping at
system. flow to the minimum flow system. least a minimum flow through the main valve.

31
Petroleum Open Learning

Summary of Section 2
Ih this section, we have looked at the component parts and method of operation of a centrifugal pump.

These included: We examined Finally we looked at why centrifugal pumps are


fitted with a minimum flow system, ensuring that
• pump casing (volute or diffuser) • how different types of pump casing played they do not become damaged due to overheating
a part in converting kinetic energy into and cavitation.
• impeller and impeller wear rings pressure energy

• pump suction and discharge • how we can change flow and/or pressure
characteristics by changing pump In the next section, we will take a look at a typical
• shaft bearing systems with radial and configurations (parallel v. series) oil metering and sampling system.
thrust bearings
In particular, we looked at
• shaft seals
• the construction and interpretation of
a set of pump performance curves for a
typical centrifugal pump and how they
incorporate the concepts and ideas which
we had previously encountered

• a typical centrifugal pump arrangement


with its inlet and outlet lines and associated
equipment

32
Oil Pumping and Metering Petroleum Open Learning

Section 3 - Oil Metering and Sampling


We have considered the basic design and operation Multi-component liquids such as crude oil are The process of metering and sampling is therefore
of a centrifugal pump. slightly compressible. Increases or decreases in given a very high priority. Meters themselves are
the overall pipeline pressure will produce small checked regularly, using a permanently installed
Now it is time to take a look at crude oil metering, changes in the volume of oil contained within the meter prover. The meter prover itself is checked
metering systems, and sampling techniques. pipeline. The terms packing and unpacking are regularly to ensure that it, too, is accurate.
used to describe these small changes in volume.
There are four main reasons for metering and If they are ignored, apparent losses or gains in the To emphasise this point, try Test Yourself 9 :
sampling a flow of crude oil: pipeline inventory can accumulate.

1. to measure the amount of hydrocarbons


removed from the reservoir. This allows field The sampling and metering system is placed
production plans to be updated and revised. as late in the oil handling sequence as possible.
There are a number of reasons for this:
2. to determine the amount of each component
in a mixed oil stream. This is particularly • it should be downstream of any booster
Test Yourself 9
important where the production from separate pump which is fitted. (On many
On an offshore installation there is a 1.0%
oil fields are mixed (perhaps as part of pipeline installations, the crude oil passes through a
error in the volume of crude oil being
sharing agreements) prior to the point of sale. booster pump to raise the pressure prior to
metered. The installation produces 60 000
entering the metering and sampling section.
barrels of oil per day.
3. to measure the product for tax purposes. This This ensures that no gas or vapour will break
is called fiscal metering. out of the liquid whilst it is being metered and
If the price of crude oil is , say, US$25
sampled)
per barrel, what is the market value of
4. to ensure that no loss of product has occurred.
this discrepancy in the course of a year
In an offshore oilfield, the amount of metered • no further processing of the fluid occurs before
(assuming continuous production)?
offshore product, plus any losses or gains due export, and the fluid sampled and metered is
to packing or unpacking of the pipeline (see representative of the fluid being exported
You will find the answer in Check Yourself
below), is compared regularly with the amount
9 on Page 68.
of onshore metered product. • metering takes place downstream of water
removal. At a water content higher than about
1%, serious discrepancies occur in meter
accuracy which conflict with the objectives of
metering and sampling
33
Petroleum Open Learning

There are many methods used to measure fluid The most common differential pressure device is In order to measure the pressure drop, there should
flow. It is worth noting, however, that most flow one which uses a restriction, usually an orifice be pressure tappings on either side of the orifice
rates are arrived at indirectly by measuring some plate, in the pipeline. The pressure drop across this plate, as shown in Figure 18. These are usually
other property of the flowing fluid, and then relating restriction is measured. This pressure differential located:
the value of this property to flow rate by some form can then be related to flowrate by the use of, for
of calibration. This is true for the two most common example, calibration tables or graphs. A large • one pipe diameter upstream of the orifice plate
devices used for metering produced oil: amount of calibration data has been published on and a half diameter downstream
this.
the differential pressure meter or
• in the flanges which hold the orifice in the
The orifice plate is popular because it has no
the turbine meter orifice plate in the pipeline
moving parts and is very accurate if calibrated and
maintained correctly.
For example, as you will see later on:

in the differential pressure meter, it is a


pressure difference which is measured
directly

in the turbine meter, we measure the


frequency of electrical pulses

In this Unit, we will take a brief look at the differential


pressure meter, and how it operates. We will then
consider the turbine meter.

Differential Pressure Meters


Differential pressure metering is one of the oldest
methods of measuring flowrates. It is simple,
accurate, reliable and relatively inexpensive. It will
record volume flowrates (say, cubic meters per
day), but mass flowrates (say, tonnes per day) can
be calculated if the density of the oil is known.
34
Petroleum Open Learning

The first method provides more accuracy, but the Figure 19, demonstrates how the pressure • type of fluid
second method is most widely used. changes as fluid passes through an orifice plate.
The differential pressure is measured between • pipe diameter
in general, accurate metering can only be achieved points P1 and P2. Point P2 is positioned in line • orifice diameter
when the orifice plate is designed, fabricated and with the vena contracta - the point at which
installed with great care. fluid velocity is at its highest, and pressure at its • flow rate
lowest.
The most common type is the square-edge orifice The differential pressure thus created will depend • inlet pressure
plate, shown in Figure 18. mainly upon: The differential pressure thus recorded may then
be converted into a flowrate figure.
We must ensure that the flow entering the device
is steady and free of eddies which would affect the
accuracy of the meter. The orifice plate should,
therefore, be placed at a point where temperature
and pressure are constant. In addition, bends,
valves and other fittings upstream of the orifice
plate tend to disturb the flow pattern of the fluid
approaching the plate. To avoid this, it is common
practice to specify:

• a minimum length of straight pipe both


upstream and downstream of the orifice
plate
or
• a flow straightening vane to be fitted up-
stream of the plate

A flow straightening vane is a length of pipe with


a set of fins running along the inside. As the fluid
flows along this stretch of pipe, the fins straighten
the flow and prevent swirling. Flow straightening
vanes are also used upstream of turbine meters.

35
Petroleum Open Learning

Turbine Meters In most units, a magnetic pick-up system senses One of the major advantages of a turbine meter
the rotation of the turbine rotor. As each blade is in its use for producing additional flow data.
Turbine meters are the most popular method of passes the pick-up coil, an electric pulse is The electrical pulses generated can be fed into a
measuring produced oil. They are accurate, reliable generated. Each pulse is counted and, as each computer system, which can then perform other,
and are easily proved and adjusted. pulse represents a known volume of liquid, the total more complex, flow calculations. This additional
flow of oil can be calculated. In some cases, two information may be added to the final read-out.
Turbine meters consist of a straight flow tube within pick-up coils are installed, so that the two separate
which a turbine or fan is free to rotate. You can see this pulse counts may be compared with each other as It should always be remembered that the accuracy
in Figure 20. The flowing stream causes the turbine an additional check. of a turbine meter depends almost entirely on the
to rotate at a speed proportional to the flowrate. If the precision of the rotor and how consistently its
flow increases, the turbine will spin faster. If the flow speed of rotation can be related to flow. If the rotor
decreases the turbine will rotate more slowly. becomes damaged, worn or dirty, then its capacity
to measure flow accurately will suffer dramatically.

36
Petroleum Open Learning

Metering Systems
The component parts of a typical turbine
metering run are shown in Figure 21. These
consist of:
• A manually operated inlet block and bleed
valve, which allows the metering run to
be positively isolated from the rest of the
process upstream. The bleed facility allows
the space between the two valve seals to
be de-pressurised, proving that no liquid is
passing across the valve.

• A filter, to remove any particles which may


damage the measuring element. The filter
is fitted with a differential pressure switch
(PDS), which gives an alarm if the pressure
drop across the filter gets too high (due to
filter blockage).

• Flow straightening vanes, to remove


turbulence and any tendency for the fluid
to swirl.

• A measuring element, in this case a


turbine meter fitted with a pulse transmitter.
The electrical pulses produced may be
transmitted to the flow computer. (In the
case of an orifice plate metering system,
the differential pressure across the plate
produces an electrical signal, which may
also be sent to the flow computer.)
37
Petroleum Open Learning

A flow control valve, which controls the flow of Therefore, in many meter runs, but not shown in The basic principle on which a pipe prover works
liquid through the metering run. When there are Figure 21, you will find: is as follows:
two or more metering runs, a central metering
controller will apportion flow between the • a thermometer, which measures the A slightly oversize, elastic sphere is installed in a
different flow control valves to ensure that each temperature of the stream being metered special length of pipe. It is free to move within the
meter run is operating within its limits. pipe as it is pushed by oil flowing through. As it
• a pressure transmitter moves it forms a travelling seal against the inside
A motor operated outlet block and bleed of the pipe.
valve (MOV), which allows the metering run • an on-line densitometer
to be positively isolated from the rest of the The prover is connected in series with the meter
process downstream. This isolation is required to be proved. So, the volume swept out by the
when the meter run is out of service, or when it
Meter Proving sphere in a given time is identical to the volume
is being proved by the meter proving system. You saw, from Test Yourself 9, that small passing through the meter.
inaccuracies in measurement of oil can result in
A second, motor operated block and bleed considerable revenue losses. In order to minimise Two detectors are installed in the special pipe near
valve (MOV), which is opened when the meter any errors the meters are proved at regular intervals. each end. These emit a signal when the sphere
run is being proved. When this occurs the flow The term proving is used in the oil industry to refer passes them, which is transmitted to the pulse
is diverted through the second MOV to the to the calibration of oil meters. counter of the meter. When the sphere reaches the
meter proving system. first detector it starts the counter. When the sphere
The procedure involves comparing the indicated reaches the second detector it stops the counter.
In practice, the pressure, temperature and density (recorded) volume of oil passing through the meter
of the oil may change while the flowrate is being with the actual (true) volume as measured by a The pulses, and therefore the volume, recorded by
measured. To compensate for these changes, very accurate device known as a prover. From this the meter should be the same as the true volume
readings of the temperature, pressure and density comparison a correction factor can be obtained displaced by the sphere as it travels between the
are taken. This information is then fed, together with which is then used to convert the observed flow detectors. If it is not, the recorded volume and the
data from the flow measurement device, into the readings to true values. true volume are compared, to arrive at the meter
flow computer. Corrected values for volume flow factor.
rate, mass flow rate, etc., can then be computed This correction factor is known as the meter
and recorded. factor. The meter factor then is
accurately calibrated volume of prover
There are various types of meter prover, but the volume registered by meter
most common one is the pipe prover.
38
Petroleum Open Learning

Pipe provers usually consist of a U-shaped


or W-shaped length of pipe. Figure 22 is an
illustration of a bi-directional U-shaped meter
prover loop.

39
Petroleum Open Learning

A bi-directional U-shaped meter prover loop, The second pass is now completed as above, but A sampling system must therefore be installed
operates as follows : with the oil flow reversed. to. determine the precise nature of the liquid being
pumped.
The flow enters the meter prover through the meter The flow computer will then average the two
under test. Sampling systems have two main functions:
metered volumes from the first and second passes
and compare this average with the known volume. • sampling for metering
In the position shown, the oil flow is holding the
calibration sphere against the buffer. If the 4-way If the volume recorded by the meter under test is • sampling for analysis
diverter valve is now turned through 90 degrees, the same as the known volume then the meter has
been proved. Sampling for metering involves the use of an
the flow through the prover loop is reversed. This
online density measuring system. This system
reversed flow picks up the sphere and carries it
If there is a discrepancy between the measured continuously samples the fluid and passes the
round the prover loop for the first pass. Two sphere
volume and the known volume, the flow computer density results to the flow computer. The computer
detectors are mounted in the prover loop, and the
will calculate a correction factor and then apply then combines values for density, pipeline pressure
internal pipe volume between these detectors is
this to the meter under test. Another meter proving and temperature to calculate the mass flow.
already known.
run will then take place.
Sampling for analysis is carried out by a second
As the sphere passes sphere detector 'A', a
When the flowmeter reading (including any system. At regular intervals, a pump extracts a
signal to the flow computer records the flowmeter
correction factor) falls within 0.5% of the known small amount of the fluid being metered, and these
reading at that point.
volume, without adjustment, for at least five small samples are stored in a sample jar or similar
consecutive proving runs, it is classed as being vessel. Periodically, this combined fluid sample is
When the sphere passes sphere detector 'B', a
accurate. taken away to be analysed in detail.
new flowmeter reading is recorded.
An on-line basic sediment and water (BS&W)
The difference between these two meter readings, Sampling Systems system is also installed on most oil handling
representing the metered volume of the prover facilities. The BS&W analyzer ensures that the water
loop, is now computed and stored. It is not only important that the crude oil is
and solids content of the crude does not exceed
metered accurately. It is equally important to
pre-set limits (typically less than 1%") without a
The calibration sphere, at the end of the first pass, gather information on the nature of the oil being
warning being transmitted to the operator.
is now held against the other buffer. pumped. The chemical and physical nature of
the oil may change with time, as may the level of The automatic sampling systems described above
The flow computer now turns the 4-way diverter contaminants, such as water or solids, still present are usually backed up by samples taken manually
valve through another 90 degrees to start the after the separation process. by the operator, as a check on the automatic
second pass. systems.
40
Petroleum Open Learning

Before leaving this section, work through the following Test


Yourself. It will help you to recall the topics we have covered:

Summary of Section 3
Test Yourself 10
During this section we have looked at:
The following pieces of equipment are used in the metering and
sampling system I have described. Can you say briefly what their • the reasons why we need to meter the crude oil before
purpose is, and where in the system they are located ? it leaves the production facility

1. densitometer • the different types of meter which may be used


2. flow straightening vanes
• a typical meter run and what equipment it contains
3. 4-way diverter valve
4. vena contracta • a bi-directional meter proving loop and how it works

5. prover loop • a typical sampling system and the reasons for sampling
6. BS&W analyser the crude oil.
7. booster pumps
8. pick-up coil
9. orifice plate
10. sphere detectors
11. turbine meter We will now take a look at the final stage in an oil production
facility - the Pig Launcher.
12. block and bleed valves

You will find the answers in Check Yourself 10 on Page 69.

41
Oil Pumping and Metering Petroleum Open Learning

Section 4 - Pig Launching Facilities


In previous sections, we have looked at the By the way, there are two main explanations given • the brush pig is used for cleaning and de-
equipment used to pump the oil. I have described for the name "pig", both of which are equally waxing pipelines. (Scrapers may also be
typical metering systems and how we prove that unlikely ! included in the design). Brush pigs in liquid
they are accurate. Sampling and analysis of the service often incorporate a series of pipes
crude oil were touched upon. • The first is that the original pigs were made from which provide liquid channels through the pig
stuffed pigskins, sent through water pipelines centre. Some of the liquid behind the pig will
In Figure 1 we saw that the last stage in a crude oil to clear them out pass through the pipes and, because of the
production facility is normally the pig launcher -the angle at which these pipes are set, the pig
final item of equipment on the installation before • The second is that early pigs were made of rotates, thus improving the brushing effect. In
the oil enters the main oil pipeline. wood, with metal bands around them to help addition, the jetting action this causes ahead of
withstand constant rubbing against the wall of the pig stops a build up of debris at that point
The oil which flows through the pipeline may have the pipeline. As they travelled along the pipeline
a small amount of residual water in it. There may they "squealed like pigs" as the metal bands • the sphere is used mainly to de-water gas
be traces of sand, or wax may be deposited from scraped along the pipe pipelines but it is occasionally used for very
the oil as it cools down. All of these materials may light cleaning work on oil pipelines
settle out and affect the efficiency of the pipeline. I will leave you to choose which one you believe.
• the foam pig is most often used for the initial
Devices called pigs may then be pumped through Types of Pig de-watering and cleaning of pipelines. Any
the pipeline, from the pig launcher, to remove the welding rods, or other sharp objects which
water or sediments which have settled out from Pigs come in a variety of shapes and sizes may have been left in the pipeline, embed
the oil. depending on the service which they are intended themselves into the foam as the pig passes
to perform. by
Pigs should form a reasonably tight fit inside the
pipeline, in order that Figure 23 on page 43 illustrates a few of the • the foam brush pig is used in lightweight
designs available. Their main uses are as follows: cleaning service, usually on gas pipelines
• they perform their cleaning duties effectively
• the squeegee pig is often used for separating
• they are efficiently transported through the different liquids or gases when pipelines are
pipeline by the fluid flow being filled or emptied, or when the same
pipeline is being used for different products.
It may also be used for lightweight cleaning
duties and for de-watering gas pipelines
42
Petroleum Open Learning

Figure 23: Pig Designs

43
Petroleum Open Learning

A very specialised pig is shown in Figure 24. This


is the Kaliper pig or Linaiog pig.

As this type of pig travels along the pipeline, two


wheels, positioned near the centre of the pig, press
against the walls of the pipe and record how far the
pig has travelled.

At the same time a series of fingers, mounted at


the back of the pig, slide along the walls of the pipe
and measure its diameter.

The information thus collected is recorded on a


chart which is built into the pig. The chart can be
analysed on arrival, to reveal variations in internal
diameter (caused, perhaps, by dents or corrosion
pitting) and precisely where these variations
occur.

Pigs are becoming more sophisticated and, these


days, are capable of measuring and recording a
wide range of data related to the condition of the
pipeline and contents.

44
Petroleum Open Learning

Pig Launchers
We will how take a look at Figure 25, which shows the basic layout of a pig
launcher, and think about how it operates.

45
Petroleum Open Learning

Under normal conditions, the crude oil supply Pig Launching Problems In addition, it should be remembered that the
bypasses the pig launcher and flows through operation of pig launchers and pig receivers is a
valve X directly into the pipeline. It all looks pretty straightforward, so what can go major cause of explosions in the oil and gas
wrong? industry. You will understand, therefore, why the
• To load a pig into the pig launcher: launching and recovery of pigs is an operation which
Well, some pigs are very reluctant to leave the pig must be treated with a great deal of respect.
• valve B and valve C should be closed
launcher and it may take three or four attempts
• the pig launcher must be de-pressurised and at loading, to get them far enough into the pig Basic Rules for Pig Launching
drained of liquid launcher for them to leave.
• when these steps are completed, the pig Always bear in mind the following basic rules:
launcher door - door A - is opened and the pig Again, pigs can break up as they traverse the
placed inside the launcher pipeline. This may result in the non-arrival of a pig, • stick closely to your own laid down proce dures
and then damage to pigs which are sent down and do not take any short-cuts
• door A is then closed after it.
• the pig launcher is refilled with liquid and re- • during pig launching and receiving operations,
pressurised using the pressurising valve Pigs can stick in the pipeline. Some pig / pipeline do not assume that any event has occurred
combinations found onshore are so prone to or not occurred until you have checked and
• the pig launch indicator is re-set to record sticking that the pig is fitted with a radio transmitter double-checked thoroughly
when the pig passes that point to assist in locating the sticking point. When a pig
is stuck, the operator must decide whether to • always make sure that you are launching the
• valve B and valve C are then opened, and
launch another pig in an attempt to shift the first correct size of pig :
valve X slowly closed
one. If this doesn't work, you have two stuck pigs.
• the flow of oil is diverted through the pig Is it wise to try a third? too narrow, and it may not travel
launcher and this flow forces the pig into the
On occasion, foam pigs will leapfrog' each other too wide, and it may stick, blocking the
pipeline
inside the pipeline. Launched in the order 1,2,3, pipeline
• as the pig passes the pig launch indicator it they arrive in the order 1,3,2.
activates a "flag" which tells the operator that too long, and it may jam on a bend, again
the pig is in the pipeline Pigs may leave the launcher and enter the pipeline blocking the pipeline
without triggering the 'pig launched' signal; or arrive
• the operator can now open valve X, and close
at the other end of the pipeline without triggering too short, and it may hang on a bend allowing
valves B and C
the 'pig received' signal. the flow to bypass it
46
Petroleum Open Learning

• always ensure that the pig is properly positioned Safety Systems


in the launcher so that it will leave cleanly when
the flow is diverted You will note from Figure 25 that a number of
interlocks have been labelled. I do not intend
• always remember to re-set the 'pig launched* to go into any detail on these - this topic will
device before you launch the pig, otherwise be covered extensively by other Units in the
you cannot tell whether it has entered the Petroleum Processing Technology Series.
pipeline or not
As a simple illustration, however:
As I have already emphasised, opening and
closing pig launchers is potentially dangerous and, interlock A on the pig launcher door
because of this, most of the launching facilities interlock B on valve B (inlet to the pig launcher)
are fitted with safety systems. These prevent the interlock C on valve C (outlet from the pig
operator from opening the wrong valve or, worst of launcher) interlock D on the pig launcher low
all, opening the fauncher door whilst the launcher pressure switch
is open to the pipeline.
work together to ensure that the pig launcher
door cannot be opened unless

valve B is closed
valve C is closed
the pig launcher pressure is low

47
Petroleum Open Learning

Summary of Section 4 Test Yourself 11


In this section we have looked at:
1. Why do pig launcher systems present an explosion
• the reasons why we need to pig a pipeline hazard ?

• the different types of pig which may be used 2. What steps should always be taken before a pig
launcher door is opened ?
• a typical pig launcher and how to launch a pig

• and, very briefly, the need for safety systems

The answers are given in Check Yourself 11,


which you will find on Page 70.

We will now look at a typical Oil Pumping and Metering System and see how it
compares with what we have learned so far. Before that, however, try the following
Test Yourself :

48
Oil Pumping and Metering Petroleum Open Learning

Section 5 - A Typical Oil Pumping and Metering System


In this section we will take a look at a typical oil Booster Pumps Downstream of ESDV 1, the line branches into
pumping and metering system and see how it three, which provide suction to each of the booster
relates to what we have covered previously in this If we look at Figure 26, on the next page, we can pumps. It is normal practice to specify that the
Unit. see how the booster pump system works. piping configuration to the pumps is designed to
The crude oil flows from the 2nd stage separator distribute the oil flow evenly.
You may like to refer back to Figure 1, which
through an emergency shutdown valve, ESDV
shows the general layout.
1. Valve ESDV 1 is common to the suction of all The first valve on the suction of booster pump A is
three pumps we are using here. HV1. HV 1 is a hand operated valve and it is also
A few assumptions have been made in the design interlocked as an inhibit, to prevent the starting of
illustrated: ESDV 1 will be closed by remote signals if an booster pump A when it is closed.
emergency occurs. Typical emergencies would
• the production operation is offshore and the main be: Downstream of HV 1 and just upstream of the inlet
pipeline takes the oil to an onshore facility, where it to the pump is a T filter. This is usually a coarse
is treated further • a very low oil level in the 2nd stage separator screen, designed to prevent larger items of debris
(part of a process shutdown because only the oil (gloves, helmets, spanners, etc.) from entering and
• there is no crude oil buffer storage facility. process would be closed down if this occurred) damaging the pump.
Therefore, the separated crude oil is pumped
directly frohi the second stage separator, through • a fire in the wellheads area (part of an The filter is fitted with a differential pressure
booster pumps and pipeline pumps into ah export emergency shutdown, which would shut down switch (PDS) which incorporates a high
pipeline system all processes). differential alarm. This arrangement will give an
ESDV 1 also has an interlock (IL) which, if the alarm in the event of a high differential pressure
• the crude oil metering facility is located between caused by filter blockage. It should be noted that a
valve is in the closed position will prevent any of
the booster pumps and the pipeline pumps. As low or zero reading here may be caused either by
the booster pumps from starting.
previously explained, this location ensures that a clean filter or a ruptured filter!
there is a stable flow to the metering system and
It should be noted that, after the pumps are running,
that the pressure is sufficiently high to prevent any The discharge of booster pump A is fitted with
the closure Of ESDV 1 will not shut them down via
gas bubbles forming
the emergency shutdown system. It only acts as a pressure switch low (PSL) and a pressure
an inhibit to prevent the pumps starting in certain switch low-low (PSLL). PSL will give an alarm
circumstances. If ESDV 1 closes while the pumps and PSLL will cause the pump to shut down in the
are running, then the low-low pressure switch on event of low pressures.
the discharge of the booster pump (PSLL) would
shut down the pumps
49
Petroleum Open Learning

50
Petroleum Open Learning

After a shutdown, PSLL creates a potential problem. The discharge from pump A now passes through a
If the pressure at that stage is below the setting of hand-operated valve, before joining the flow from
PSLL, the pump cannot be re-started. A shutdown the other pumps. The combined flow then passes
signal is still being sent from the Pressure Switch through level control valve LCV 2. This valve
Low-Low. Something must be done to allow the controls the oil level in the 2nd stage separator. The
pump to restart. separator level controller will open this valve if the
level rises, and close it if the level falls. We can see
The problem is overcome by, automatically, that, in the event of a failure of supply to the 2nd
bypassing PSLL for 30 seconds when the pump stage separator, the valve would close completely
is started. This allows sufficient time to build and the booster pumps would go on to minimum
up enough pressure to re-set the switch. If the flow.
increasing pressure does not re-set PSLL before
the 30 seconds have elapsed, then the pump will After passing across LCV 2 the oil flows to the
shut down again. This system is called a time- sampling and metering systems.
pressure race, i.e., the pump is racing against
time to generate sufficient pressure to re-set the You should note that the booster pump system
switch. is designed so that its discharge pressure is high
enough to meet the required suction pressure at
The discharge of the pump is also fitted with a the main oil pipeline pumps, which we will look at
pressure switch high (PSH) and a pressure later.
switch high-high (PSHH). PSH will give ah alarm
and PSHH will cause the pump to shut down in
the event of high pressures, perhaps because of
problems downstream.

The discharge of the booster pump is fitted with a


minimum flow non-return valve (SV1), which we
have already described in Section 2, Page 31 and
Figure 17. To prevent the continuously re-cycled oil
from becoming progressively hotter, it is routed all
the way back to the 2nd stage separator via HV 2.

51
Petroleum Open Learning

Sampling System
Figure 27 shows the layout of a typical sampling system.

52
Petroleum Open Learning

You will see that a side stream is removed from the A BS&W analyzer checks for the basic sediment
inlet header to the metering system by one of two and water contained in the crude oil flow. Most
sample pumps, A and B. pipeline operations have a maximum specification
for BS&W which, typically, may be "not more than
This side stream is drawn through two continuous 1%". This means that no more than 1 % of the
sampling devices (A and B) where a small sample total volume pumped into the pipeline should be
is removed and stored. sediment and water.

As an illustration of the sampling routine: If an increased BS&W level occurs for any length of
time, the pipeline pigging programme is readjusted
• sample device A may take a composite sample to increase the rate of pigging. This is required to
of five litres per day prevent the sediments and water from blocking
and corroding the pipeline.
• sample device B may take a composite sample
of 35 litres per week

• one litre spot samples may be taken manu ally


by the operator, as a back-up, at twelve hour
intervals

After leaving the sample pumps, the sample


stream flows to two densitometers (A and B) and
a basic sediment and water (BS&W) analyzer
before returning to the inlet header.

A densitometer is designed to measure the


density of the sample stream fluid. It does this by
comparing this fluid with a reference, whose density
is known. The result is then passed automatically
to the flow computer.

53
Petroleum Open Learning

Metering System
In the metering system shown in Figure 28,
i have included just one meter run and a prover
loop. The meter run, which we could designate run
'A', is from upstream of the inlet block valve (HV 1)
to downstream of the outlet block valve (MOV 1).
In a complete system there would be three or more
parallel runs. I have indicated this in the drawing as
additional runs 'B' and 'C'. A single prover loop is
used and there are connections between each run
and the prover, enabling it to be placed in series
with any of the meters.

54
Petroleum Open Learning

When meter run 'A' is in service, the normal flow • the 4-way prover loop diverter valve (MOV 3) • operates MOV 3 again, to reverse flow through
pattern would be through: prover and start second (and any further) proving
• the inlet block valve (HV1) • the prover loop flow control valve (FCV 2) runs

• the filter (F) and from there to the pipeline oil pumps. • performs necessary calculations to obtain
meterfactor
• the flow straightening vanes Note the flow computer in the drawing. You will
remember from Section 3 that one of its jobs is A few other points to note are:
• the turbine meter to compare the volume indicated by the meter
with the true volume of the loop to obtain a meter 1. interlocks are fitted to MOV 1 and MOV 2 to
• the flow control valve (FCV1) factor. In addition, it ensures that there is equal ensure that these valves are at the right setting
flow between each of the meters being used. It (open or closed) before the meter proving starts
• the outlet block valve (MOV 1) does this by altering the settings of the appropriate
flow control valves. If meter run TV were in normal 2. pressure relief valve PSV1 is located
and from there to the pipeline oil pumps. service this would be FCV 1. If meter W is being downstream of the filter and upstream of the flow
proved however, the flow would be controlled via straightening vanes. If HV 1, MOV 1 and MOV 2
When the meter in run 'A' is being proved, the FCV 2. The flow reading from each meter is fed to are all closed for any reason, the pressure inside
flow would be through: the computer via a flow transmitter (Ft). the meter run may rise due to any temperature
increase. PSV 1 is fitted to relieve this pressure
• the inlet block valve (HV1)
3. to ensure the accuracy of the prover loop, the
So, when the meter in run 'A' is being proved, the
• the filter (F) sphere is always oversized by 1-2%. This ensures
flow computer:
a tight fit between the surface of the sphere and
• the flow straightening vanes the walls of the prover loop. The sphere is replaced
• closes MOV 1
on a regular basis, and it is normally the first item
• the turbine meter to be changed if the accuracy of the prover loop is
• opens MOV 2
suspect
• the prover loop block valve (MOV 2)
• transfers control of flow from FCV 1 to FCV 2
4. An independent contract company is often
• the 4-way prover loop diverter valve (MOV 3) used to prove the prover loop, say, on an annual
• allows flow to stabilise
basis
• the prover loop
• operates MOV 3 to start first proving run
55
Petroleum Open Learning

Oil Pipeline Pumps


If you look at Figure 29, you can probably see how the oil pipeline pumping
system works.

56
Petroleum Open Learning

It is rather similar to the booster pump layout, so we will


concentrate only on the important differences:

• When the minimum flow system is operating, crude oil is


re-cycled from the discharge to the suction of the pump
and is not routed back to a separator, as was the case in the
booster pump layout. (The separators are upstream of the
meters. Therefore, if the oil was re-cycled to the separators,
it would pass through the meters twice, which, of course,
would introduce errors into the flow measurements)

• However, because the pipeline pumps are transferring


a large amount of energy to the oil, this direct re-cycling
would result in a rapid and substantial temperature rise.
To prevent this from occurring, a re-cycle cooler is fitted
to cool the crude before it is returned to the suction of the
pipeline pumps. Offshore, the re-cycle cooler would often
use seawater as a cooling medium (as shown in Figure 29)
because it is cheap and plentiful

Fluid Coupling
Pipeline pumps have a variable speed drive. The speed at
which they operate is determined by the pipeline pressure
controller (which we will look at later). If the line pressure is too
low, then the controller increases the pump speed; if it is too
high, the pump speed is decreased.

This speed variation may be achieved by a fluid coupling


between an AC electric motor and the pump. Fluid couplings
are also known as hydraulic couplings.
Acommon design is shown in Figure 30
57
Petroleum Open Learning

Figure 30 is a three dimensional cut-away drawing The basis of operation is as follows: Pump Speed Control
of the coupling assembly. You will see that the
coupling comprises: • the inlet shaft turns the runner, and drives the The amount of oil transferred between the runner
oil circulating pump. Note that the runner turns and the impeller and, therefore, the main oil
• an inlet shaft, connected to the drive motor at 100% of the drive motor speed at all times pump.

• an outlet shaft, connected to the main pump • the cups on the runner pick up oil from the
outer perimeter of the scoop chamber and Figure 31, on page 59, shows a series of cross
The inlet shaft drives an oil circulating pump. The throw it into the receiving cups of the impeller. sectional diagrams through a fluid coupling,
oil path is from the reservoir, via a cooler and small The runner is therefore acting as a pump which help us to explain this mechanism of speed
holding tank, into the circulating pump section. control.
From the pump discharge, the oil flows to the • the oil striking the impeller cups turns the
scoop chamber impeller, which is acting as a turbine

You should take particular note of the components • the impeller then turns the main pipeline
labelled the runner and the impeller. They are pump
both of similar design and look like a ring of cups
attached to a wheel. The runner is a t the end of the
inlet shaft, and the impeller at the beginning of the
outlet shaft. Each turns independently of the other
within the casing. The only connection between
them is made by the circulating oil when the unit is
in operation-hence the term fluid coupling.

58
Petroleum Open Learning

59
Petroleum Open Learning

The position of the scoop tube will determine how Pipeline Pumping Pressure
much power is transmitted across the coupling.
Returning to Figure 29 again, the oil pipeline pump
• In Figure 31 a the scoop tube is at maxi- speed is controlled by the speed controller (SC)
mum extension, at a radius slightly greater which takes its signal from the pipeline pressure
than the outer boundary of the circulating controller.
oil. Therefore, all oil entering the scoop
chamber is 'scooped' away by the open If the pipeline pressure is too low, these controllers
tip of the scoop tube and returned to the will speed up the oil pipeline pump by shortening
reservoir. The scoop chamber is virtually the extension of the scoop tube.
empty, and no oil remains for the runner to
throw at the impeller. Power transmission If the pressure is too high, the controllers will slow
is therefore nil, and the main pipeline pump down the main pump by increasing the radius of
is stationary. the scoop tube.

• At an intermediate extension of the scoop Pressure Transmitter


tube, (Figure 31b), a ring of oil can accu-
mulate in the scoop chamber between the Finally, just upstream of the main outlet valve
tip of the scoop tube and the outer bound- ESDV 2 is a pressure transmitter (PT) which sends
ary. This limited volume of oil is now avail- a telemetry signal to the local control room, to the
able for the runner to throw at the impeller. shore (in offshore locations), and to other oilfields
An intermediate level of power can now be sharing the same pipeline facility.
transferred across the coupling to drive the
main pump. This safety feature is required to prevent over-
pressuring the pipeline.
• In Figure 31 c, the scoop tube is at mini-
mum ra dius, the oil retained within the
scoop chamber is at a maximum, and full
power transfer is taking place.

60
Petroleum Open Learning

Pig Launching Normal flow through the system would be: ESDV 2 and ESDV 3 are two emergency shutdown valves which
are interlocked with the ESD system to ensure that the pipeline
The pig launching facility is illustrated in Figure • through ESDV 2 pumps cannot be operated when these valves are closed.
32.
It is similar to the one I have described previously. • through MOV 1 On an offshore installation, ESDV 3 may be situated on the sea
bed. It is designed to ensure that ho oil can flow back to the
• through ESDV 3, and then installation in the event of platform malfunction. It is only operated
in extreme emergencies such as a fire or large oil leak.
• to the pipeline

61
Petroleum Open Learning

Now, with reference to Figure 32 again, we can 7. confirm that the pressure is off the pig launcher 13. open MOV 3 and then MOV 2. We open MOV
list the steps involved in launching a pig: by checking a pressure guage (PG). Then close 3 first because we do not want a sudden flow of
HV 3 oil through the pig launcher to try to force the pig
1. check that the pig is undamaged, the correct through MOV 3 as it is opening
size, and that the shape is undistorted. Spherical 8. open HV 4 to allow nitrogen (N2) to flow through
pigs may be passed through a sizing ring to make the pig launcher to remove hydro carbon gasses. 14. when MOV 2 is fully open, close MOV 1 to
sure that they are the right size Close HV 4 and HV 2 divert the flow through the pig launcher. We keep
closing MOV 1 until the flow launches the pig.
2. ensure that the pig signalling device (SX) has 9. open the pig launcher door When the pig enters the pipeline, it will hit the pig
been re-set, ready to tell us when the pig has been signalling device (SX). This will then tell us that the
launched 10. load the pig, ensuring that it is past the inlet pig has passed this point
from MOV 2
3. check that MOV 2, MOV 3 and HV 1 are closed 15. re-open MOV 1, close MOV 2 and MOV 3, to
so that we may de-pressurise th pig launcher 11. close the pig launcher door and purge air from return the system to normal
the launcher (with nitrogen) before re-pressurising.
4. begin the de-pressurisation process by the reason for purging is to prevent an explosion You should note that a pig should never be
opening HV 3, allowing pressure in the pig launcher when we bring the pressure up to normal operating launched without first ensuring that the pig receiver
to blow the oil it contains to the drain system level at the other end of the pipeline is ready to receive
it.
5. as the pressure falls, the high pressure switch (in our example, the purging operation is carried
(PSH) will show that the pressure is not high. out by re-opening HV 2 and then HV 4 this allows During all pigging operations, you should
Then the low pressure switch (PSL) will show a small amount of nitro gen to displace air to the follow the operational and safety procedures
that the pressure is low vent system via HV 2) laid down specifically for your equipment and
installation
6. when this situation is reached, we can open HV 2 12. when all air has been displaced, close HV 4
to the vent system and allow the pig launcher and and HV 2 and allow pressure to build up to the
vent system pressures to equalise. As this occurs, pipeline operating pressure by opening HV 1 as
the remainder of the oil will drain to the d r a i n this occurs, PSL will tell us that the pressure is not
system through HV 3 low and PSH will finally tell us that the pressure is
high. When these two switches have given their
indications, we will close HV 1

62
Petroleum Open Learning

Now, finally, try this Test Yourself, which covers some of the topics we
have discussed in Section 5.
Summary of Section 5

Test Yourself 12
In this section we have:

• looked at the main design features of a typical oil handling and


metering system
1. What do you understand by a time-pressure race ?
• examined the layout of a booster pump unit and, in particular, how it
may be controlled 2. In pig launching operations, what does the flag do ?

• discussed the key elements of a sampling system and noted that 3. What do we mean by the meter factor ?
density and BS&W are measured automatically
4. In the case of the booster pumps, why does the minimum
• worked through the operation of a metering system and, in flow system re-cycle oil back to the second stage separator,
particular, a meter proving loop and not directly to the booster pump inlet ?

• looked at a typical arrangement of the main pipeline pumps, and 5. The minimum flow system for the main pipeline pumps
compared this arrangement with that for booster pumps re-cycles oil directly to the pump suction. Why does this
arrangement differ from that for the booster pumps ?
• discussed the main design features of a fluid drive system, and how
it may be used to control pumping rate 6. What types of analysis does our sampling system perform
continuously on the oil flow ?
• described the procedure for launching a pig to the pipeline

You will find the answers to these questions in Check Yourself


12 on Page 70.

63
Petroleum Open Learning

Unit Summary

In the course of this Unit on Oil Pumping and Metering, we have:

• looked at some of the theories behind the operation of centrifugal pumps, including
the behaviour of fluids, centrifugal force and energy

• detailed the component parts of a centrifugal pump, and the role each plays in its
operation

• examined the main design features of a metering and sampling system, and how it
is controlled and operated

• familiarised ourselves with the layout and operation of a pig launching facility

• discussed the main design and operational aspects of a typical oil pumping and
metering system

Now go back to the Training Targets on Page 4 of this unit and satisfy yourself that you
are able to meet those targets.

64
Petroleum Open Learning

Check Yourself 1 Check Yourself 2 Check Yourself 3


Specific gravity of gasoline: Velocity of the car: 3 metres head of water exerts a pressure
of 0.3 bar.
= mass of five litres of crude oil S.G. of crude = 0.85
mass of five litres of water (reference) =180kph = 50 metres / sec
Head Pressure 3 metres of crude oil =
= 4.25 kg = 0.85 kinetic energy of the car: 0.3 x 0.85 = 0.25 bar.
5 kg
= 1/2 1000 kg x (50 m / sec x 50 m / sec)
= 1 250 000 joules 4.5 metres head of water exerts a
Specific gravity of the brine: pressure 0.45 bar.

= mass of five litres of brine Velocity of the truck: S.G. of brine = 1.1
mass of ive litres of water (reference)
= 30kph = 8.3 metres / sec Head Pressure 4.5 metres of brine =
= 5.5 kg = 1.1 0.45 x 1.1 = 0.5 bar.
5 kg kinetic energy of the truck:

= 1/2 x 20 000 kg x (8.3 m / sec x 8.3 m/sec)

= 688 900 joules

Therefore, the car has the greater kinetic energy

65
Petroleum Open Learning

Check Yourself 4 Check Yourself 5 Check Yourself 6


70% diffrential pressure = When pumping 20 cubic metres per hour
b. the static suction line pressure
84% flow = 8.4 litres/min this pump will:
c. the NPSH
40% differential pressure = • require a minimum of 3.4 metres
64% flow = 6.4 litres/min head of liquid NPSH
a. the pressure at which gas or
vapour is released
Therefore, the flow rate would fall by • develop 57 metres total head of
8.4 - 6.4 = 2 litres/min liquid
The NPSH represents the minium design
pressure to prevent gas or vapour release
• consume 8 kilowatts of power
and should therefore be above this gas /
vapour release pressure by a safe margin.
• operate at 72% efficiency approxi
mately
The static suction head pressure would
normally be maintained at about 10%
above the NPSH.

66
Petroleum Open Learning

Check Yourself 7 Check Yourself 8


Your answer should look like the following:
When pumping 40 cubic metres per hour this
pump will: Item Casing Impeller Bearing Seal

• require a minimum of 3.9 metres head of liquid shaft sleeve 'O' ring •
NPSH shroud •

• develop 47 metres total head of liquid lantern ring •


wear rings • •
• consume 11J kilowatts of power
flush inlet •
• operate at 85% efficiency
vane •
slinger ring •
approximately
balance holes •
gland follower •
volute •
ball bearing race •
diffuser •

67
Petroleum Open Learning

Check Yourself 9
1.0% of 60 000 bbl/day = 600 x 365 bbl/year = 219 000 bbl/year
(assuming, of course, that the installation produces at that rate without interuption).

at $25 per barrel, this error is valued at about $5.5 million per year.

This example emphasises very effectively the importance of accuracy in the metering process. You should note that the error
is equally undesirable, whether it involves an over-measurement or under-measurement of crude oil volume.

68
Petroleum Open Learning

Check Yourself 10
1. densitometer an instrument, installed between booster 7. booster pumps located upstream of the metering and
pumps and metering system, to measure sampling system (to discourage gas/
the density of the pipeline fluid. vapour breakout).

2. flow straightening vanes installed upstream of a flow meter to 8. pick-up coil part of a turbine meter, used to sense
smooth flow and prevent swirling and transmit speed of rotation.

3. 4-way diverter valve part of a meter proving loop, allowing flow to 9. orifice plate an essential part of the most common
be reversed for a second pass of the sphere. type of differential pressure meter.

4. vena contracta this is cheating a little bit - the vena 10. sphere detectors part of a meter prover loop, and
contracta is the point in the flow pattern signals the begining and the end of a
through an orifice plate where flow rate is prover run, allowing the meter reading
highest and pressure lowest to be recorded at thoses points.

5. prover loop a pipe loop of known volume in the meter 11. turbine meter the most common type of oil flow
proving system which allows accurate meter, located downstream of booster
calibration of the meter. pumps, filter and flow straightening
vanes.
6. BS&W analyser an instrument, installed between booster
pumps and metering system, to measure 12. block & bleed valves located at various places in a metering
basic sediment and water (BS&W) in the run, allowing the run to be positively
pipeline fluid isolated from the rest of the process.
The "bleed" facility allows the space
between the two valve seals to be
depressurised.

69
Petroleum Open Learning

Check Yourself 11
1. They are the only part of the pipeline 1. when restarting the booster pumps after 4. Re-cycling directly back to the pump section
system which is regularly opened to a shutdown due to low pressure, it will be would cause the oil to become progressively hot-
the atmosphere. necessary to by-pass the pressure switch- ter. Re-cycling to the separator will give the oil an
low (PSLL- see figure 26) for a short while. opportunity to cool down.
2. the pig launcher must be : This gives the pump sufficient time to build
up enough pressure to re-set PSLL. 5. The separators are upstream of the flow me-
a) isolated from the pipeline ters. If we re-cycled oil to the separators, it would
2. The flag is part of the pig launch indicator pass through the meters twice and give us a false
b) drained of liquids mechanism, and signals that the pig has flow reading.
passed that particular point in the system.
c) depressurised 6. a. density or specific gravity
3. The meter factor is a correction factor which
allows us to convert observed flow readings b. basic sediment and water (BS&W)
to true values.
samples are also taken for more detailed
It is calculated during the meter proving laboratory analysis.
procedure, by comparing the true volume of
liquid passing through the meter in a given
time, with the volume registered by the meter
in the same time :

meter factor = true volume of liquid passing through meter in a given time
volume registered by meter in same time

70
POL
Petroleum Open Learning

Oil Treatment
(Dehydration)
Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY
POL
Petroleum Open Learning

Oil Treatment
(Dehydration)
Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY
Petroleum Open Learning
Oil Treatment (Dehydration)
(Part of the Petroleum Processing Technology Series)

Contents Page
Visual Cues
Training Targets 4
training targets for you to
achieve by the end of the unit
Introduction 5

Section 1 – Emulsions – Their Nature and Occurrence 7 test yourself questions to see
What is an emulsion? how much you understand
The Creation of an Emulsion
Emulsion Stabiltiy
Emulsions and the Problem of Salt
check yourself answers to let
Section 2 – Principles of Emulsion Treating 14 you see if you have been think-
ing along the right lines
The Application of Heat
The Application of Electricity
The Application of Chemicals
Demulsifier Selection activities for you to apply your
Demulsifier Bottle Test
Equipment new knowledge
Test Procedure
Main Test
Injection of Chemicals
Water Washing summaries for you to recap
Settling on the major steps in your
progress


Petroleum Open Learning


Petroleum Open Learning

Contents (cont’d) Pages Visual Cues

Section 3 – Dehydration Systems and Equipment 32 training targets for you to


Settling Tanks achieve by the end of the unit
Wash Tanks
Free Water Knockout
Heater Treaters
Electrostatic Treaters test yourself questions to see
Desalting how much you understand

Section 4 – A Typical Dehydration System 45


Separation of Free Water Knockout check yourself answers to let
Crude and Emulsion Heating
you see if you have been think-
Electrostatic Dehydrators
Dilution Water System ing along the right lines

Check Yourself – Answers 58


activities for you to apply your
new knowledge

summaries for you to recap


on the major steps in your
progress


Petroleum Open Learning

Training Targets
When you have completed this unit on Oil Treatment (Dehydration), you will be able to :

• Explain what constitutes an emulsion.


• Describe how an emulsion is formed.
• Explain how residual water in oil can cause problems with salt content.
• Explain the basic principles of emulsion treating.
• List the basic properties of a demulsifier.
• Explain how a bottle test, used in demulsifier selection, is carried out.
• Explain how demulsifying chemicals are injected into a dehydration process.
• Describe the construction and operation of wash tanks and free water knockouts.
• Describe the construction and operation of heater treaters and electrostatic treaters.
• Explain the layout and flow through a typical dehydration plant.

Tick the box when you have met each target


Oil Treatment (Dehydration) Petroleum Open Learning
Introduction

Water is produced together with oil from most oil (Another unit in the Petroleum Processing The total production from an oil field flows from the
fields. This water, which may make up a very large Technology Series covers produced water treatment wells to the separation system. The function of this
percentage of the total production from a field, can in detail.) system is to separate the production into its individual
cause considerable problems. These problems I said that water is separated from the oil at the phases of oil, gas and water. The process is carried
include. first opportunity. But how is this done ? If you out in large vessels - the separators. A typical 3 phase
have completed previous units in this series you separator is shown in the diagram below.
• Corrosion - The produced water is very salty. If will be aware of the primary separation facilities in a
this water is allowed to remain in the oil it could production processing plant. Let’s look briefly at the
cause corrosion damage to pipes, vessels and system to refresh your memory.
other equipment

• Scaling - Salts are initially dissolved in the


water present in a reservoir. As conditions
change when this water is produced these salts
may be precipitated as solids and deposited as
scale. This in turn can reduce pipe diameters,
plug vessels and equipment and lead to lost
production

• Transportation - The oil will be transported


from the field by pipeline or tanker. Either way,
water in the oil will cause problems. Water in
the pipeline leaves less room for oil and results
in loss of pipeline efficiency. Water being sent
to a refinery can cause serious upsets in the
distillation process. Tankers will not accept a
cargo which contains more than a very small
percentage of water

In order to minimise the problems I have just


described, the water is separated from the oil at the
earliest opportunity. This separated water is then
treated before being disposed of.
Typical 3 phase separator


Petroleum Open Learning

The vessel is called a 3 phase separator because it This all seems fairly straightforward. However, there
separates the total flow stream into the three separate is a potential problem at this point. In order for this
streams of oil, water and gas. A 2 phase vessel would separation to occur the water must exist as free
only separate the stream into the liquid and gas water. In other words, the water must be present as a
streams. body of water. Or, if the water is present as droplets,
these must be large enough to fall through the oil and
I don’t intend to go through the construction and accumulate as a water layer. Unfortunately, some
operation of a separator at this point (another unit on water may be present in the Oil as very small droplets.
oil and gas separation is available from Petroleum These droplets are dispersed throughout the oil and
Open Learning). For the time being just look at the form an emulsion which can be very stable. Further
bottom left hand side of the vessel. treatment is then required on the oil to break down the
emulsion and separate the oil and water from each
This part of the separator is the liquid accumulation other.
section. The oil, water and gas stream has entered
the vessel at the inlet and been deflected at the This is what this unit is all about - the treatment
inlet deflector. The gas has passed towards the gas of oil to remove the final amounts of water after
outlet via straightening vanes and mist extractor, and primary separation. The treatment is often called Oil
the liquids have fallen into the liquid / accumulation Dehydration.
section.
I have divided the unit into four sections as follows:
This is where the separation of oil and water takes
place. But how does it occur? The water and oil • Section 1 covers emulsions. In this section we will
separate due to a difference in their densities. look at the nature of emulsions, how and why they
Providing the oil and water stay in the vessel for a form and what affects emulsion stability
sufficient period of time the bulk of the water can be
separated from the oil. This water is the produced • In Section 2 we will look at the basic principles of
water which has now to be disposed of. emulsion treatment

• In Section 3 we will examine the construction and


operation of equipment used in emulsion treatment
or oil dehydration

• Finally in Section 4 I will take you through a typical


dehydration system


Oil Treatment (Dehydration) Petroleum Open Learning

Section 1 – Emulsions – Their Nature and Occurrence

What is an Emulsion ? The dispersed water droplets are known as The Creation of an Emulsion
the internal or discontinuous phase. The
‘Oil and water do not mix’. This is an old saying oil surrounding the droplets is the external or The two liquids, oil and water, if they are in a pure
which is often quoted. In fact this is the basis of oil continuous phase. state could not form an emulsion. You could agitate
and water separation. If we were to shake up some the two liquids for ever, creating droplets of water in
oil and some water in a bottle and then let it stand When I defined an emulsion I said that a third the oil, but as soon as the agitation is stopped the
the following would happen; the water would sink substance is present in the mixture. This is a two would separate from each other. The reason for
to the bottom of the bottle and the oil would float on substance which separates the internal phase from this is that the liquids are not compatible. When they
top. When two liquids are not capable of being mixed the continuous phase and vice versa. It is known as are placed together in the same container they try to
we say that they are immiscible. an emulsifying agent. find a condition which will give the least contact area
between themselves.
However, oil and water can be made to mix under So, for an emulsion to form, there must be three
certain circumstances. This occurs when one of the components present. ie. The shape which has the least surface area for a
liquids is dispersed as fine droplets throughout the given volume is a sphere. So, a droplet of water
other and is stabilised. • water - which is the internal phase within a body of oil will assume a spherical shape.
This will ensure the minimum contact area between
Having said that, we could define an emulsion as • oil- which is the continuous phase itself and the surrounding oil. In addition, the droplet
follows: will try to make itself as small as possible. This
• an emulsifying agent also will reduce the contact area. But what has
An emulsion is a mixture of two liquids which the smallest surface area, a lot of small droplets
are usually immiscible. One of these liquids is In addition to the three components being present, or a single droplet with the same volume as the
dispersed throughout the other as small droplets they must be agitated for the emulsion to form. combined volume of the small droplets? Try the
and is stabilised by a third substance called an The individual components in themselves would following Test Yourself question which will show you.
emulsifying agent. never form an emulsion unless there was sufficient
agitation to disperse the water through the oil.
In oilfield emulsions the two immiscible liquids are However, no amount of agitation will form an
oil and water. Either one could be dispersed in the emulsion without the liquids being immiscible and an
other. The most common, however, is the situation emulsifying agent being present. This being so, let’s
where the water is dispersed in the oil. This is known look at the formation of an emulsion.
as a water in oil emulsion. Occasionally an oil in
water, or reverse emulsion will form but these are
much rarer. In this unit we will concentrate on the
more common one.


Petroleum Open Learning

From the answer to the question, you can see that

Test Yourself 1.1


the surface area of the larger droplets is smaller than
the sum of the surface areas of the combined smaller
droplets. We have already said that the water will try
to find a condition giving the least contact area. The
A sphere with a diameter of 15 mm has a tendency is for all the droplets of water in an oil water
volume of 1767 mm3. Five smaller spheres mixture to join together to form one body of water.
each having a diameter of 8.77 mm have
the same volume in total. Determine what What I have just said might indicate that we are
has the smallest surface area, the single unlikely to have a problem with emulsions. However,
droplet or the five smaller droplets. this is where the emulsifying agent, or emulsifier,
comes into the picture. This substance is essential to
the creation of an emulsion.
The formula for the surface area of a
sphere is πd2 A well known example of an emulsion which you
would find in the kitchen, is mayonnaise. The
basic ingredients for making mayonnaise are
vegetable oil and vinegar. If these two liquids are
whisked together they tend to mix. But, as soon as
the whisking is stopped, the oil and vinegar would
immediately separate. If eggs are slowly added
during the Whisking however, the mixture soon
takes on the familiar thick creamy appearance of
You will find the correct answer in Check mayonnaise – an emulsion.
Yourself 1.1 on Page 58.
In this case the emulsion is formed from two
immiscible liquids – vegetable oil and vinegar, then
subjecting them to violent agitation – whisking
– in the presence of an emulsifying agent – eggs.


Petroleum Open Learning

Emulsifying agents are always present in crude So, in the oilfield we have all the conditions • O
 il viscosity – In a thick viscous oil, water
oils produced from the reservoir. They include the necessary for the creation of an emulsion. The two droplets cannot move around very easily.
following substances. immiscible liquids, the presence of an emulsifying This means that there will be less chance of
agent but what about agitation? The very process the droplets meeting each other. Even if the
• asphaltines – a term given to a variety of producing reservoir fluids ensures that there is droplets which form during emulsification are
of compounds containing sulphur, agitation. Imagine the fluids flowing up the well relatively large, they will not be able to sink
nitrogen, oxygen, etc tubings, through chokes, via flowlines and headers through the oil and separate out. Therefore
into processing equipment. That certainly agitates the oil will be able to hold the water droplets
• resins the fluids. in suspension more easily.

• organic acids • Emulsifying agent – The type of emulsifier


Emulsion Stability will dramatically affect the stability of the
• metallic salts emulsion. However, an emulsifier which
The stability of an emulsion is a measure of its creates a stable emulsion in one situation
• silts resistance to being broken down into the separate could form a very loose emulsion under
components of oil and water. We can refer to an different circumstances. There are so many
• clays and many others emulsion as being tight (difficult to break) or loose variables in the conditions under which an
(more easy to break). Whether the emulsion is tight emulsion is produced, that it is impossible
These agents are known as surface active agents, or loose depends on a number of factors and we to state which agent creates the most stable
which means that they tend to alter the nature of can look at some of these now. emulsion.
the interface between the water droplets and the oil.
The emulsifier, which is present in the oil, migrates • Amount of water present - As the quantity of • 
Age of the emulsion - When the water and
to the interface and concentrates there. Emulsifiers water present in the mixture increases, more oil are first mixed together the emulsifying
are fairly neutral as far as an affinity for oil or water and more agitation is required to completely agent is evenly distributed throughout
is concerned. They neither like nor dislike the two emulsify it. If complete emulsification occurs the oil. It takes time for the ernulslfier to
liquids. They tend to form a barrier between the however, there will be a greater number of migrate to the interface between the oil and
water droplets and the surrounding oil. They form a water droplets present in a given volume. the water droplets. So initially the emulsion
type of skin round each droplet which prevents them Therefore, there will be a greater number of is relatively unstable. As time goes on and
from joining together. It is useful to imagine each collisions between the droplets, which gives the migration proceeds, the film is formed
water droplet being wrapped in a substance rather them a better chance of uniting, and then around the water droplets. With increasing
like clingfilm. Since the emulsifiers are moving separating from the oil. By and large, water time the film becomes thicker and tougher
around in the oil, they tend to carry the surrounded in oil emulsions with a high water content making it more difficult for the droplets
water droplets with them and keep the droplets tend to form less stable emulsions. to combine. This results in a more stable
floating in the oil. emulsion.


Petroleum Open Learning

• Size of water droplets - In general the size of Now try the following Test Yourself question before
the dispersed water droplets is a measure of we go on to discuss the problems of salt in crude oil.
its stability. If the agitation is such that very
small droplets are produced, the emulsion
will tend to be tight.

Figure 1.1 shows the difference between a tight and


loose emulsion with respect to their droplet size.
Test Yourself 1.2

Are the following statements True or False ?


TRUE FALSE

a. Mayonnaise is an example of an unstable emulsion.

b. If a mixture of oil and water is violently agitated a tight emulsion will form.

c. If, after agitation of oil and water in the presence of an emulsifying agent
small droplets of water are produced, the resulting emulsion will tend to be
a tight emulsion.

d. Emulsifiers are surface active agents which migrate to the interface


between oil and water and form a barrier between the droplets and the
surrounding oil.

You will find the correct answers in Check Yourself 1.2 on page 58

Figure 1.1

10
Petroleum Open Learning

Emulsions and the Problem of Salt The amount of salt in oil is usually quoted at a
refinery in units of pounds per thousand barrels
The water of the internal phase of an emulsion is ( PTB ). A limit of salt in crude of 50 PTB may be
invariably salt water, The salt content or salinity of established by the refinery, and any salt content
the water is expressed in ppm NaCI. This means above this would be unacceptable.
parts per million sodium chloride which is common
salt. This salinity can vary from field to field but could If we know the salinity of the residual water and the
be as much as 200 000 ppm. percentage BS&W we can determine the salt content
of the crude in PTB. The graph illustrated in Figure
Refineries cannot accept oil which has a high salt 1.2 can be used to determine salt in oil content, if
content as the salt breaks down during the refining the residual water percentage is just 0.1%.
process and causes considerable problems. Severe
corrosion, scaling and fouling of equipment and
pipework are just some of the undesirable effects of
salt in refinery operations.

The saltwater is removed as far as possible before


the crude oil gets as far as a refinery. Of course
this unit is all about the breaking down of emulsions
and the removal of the water. However, no matter
how efficient the dehydration process is, there will
usually be a very small amount of residual water in
the oil. This is expressed as the amount of base Figure 1.2 Salt in oil when 0.1% water remains
sediment and water ( BS&W ) as a percentage of
the total liquids. This residual water will vary with the
efficiency of the dehydration equipment but could
range from 0.1 to 0.3% BS&W.

Since the residual water is the salt carrier, the actual


amount of salt being transported in the crude oil will
depend on the salinity of the water and the amount
remaining after dehydration.

11
Petroleum Open Learning

Using Figure 1.2 try the following Test Yourself


question on salt contents.

Test Yourself 1.3


 fter dehydration, 0.1 % water remains in a
A
certain crude oil, and the salinity of this water
is 140 000 ppm NaCl. Would this crude be
acceptable to a refinery whose upper limit for
salt in crude is 50 PTB ?

Would the refinery accept the crude if the water


salinity is 100 000 ppm NaCI ?

You will find the correct answers in


Check Yourself 1.3 on Page 58.

You can see from the answer to the Test Yourself that
even with an efficient dehydration system it may be
necessary to reduce the salinity of the residual water
in crude in order to be able to export it for sale. In
Section 3 we will look at ways of doing this.

12
Petroleum Open Learning

Summary of Section 1

In this section we have been looking at emulsions in general, what they are and
how they are formed. I defined an emulsion as a mixture of two normally immiscible
liquids in which one of the liquids is dispersed throughout the other as small droplets.
It is stabilised by an emulsifying agent. I pointed out that in a water in oil emulsion
the water is the internal phase and the oil is the continuous phase. You saw that in
addition to the two immiscible liquids and an emulsifying agent being present, the
mixture must also be agitated for an emulsion to form.

We then went on to look at the way in which an emulsion is formed and I gave
an example of mayonnaise as a well known emulsion. In this case eggs form the
emulsifying agent. You saw what types of substances form emulsifying agents in the
oilfield and how the agitation occurs when reservoir fluids are produced.

We considered the difference between a tight and a loose emulsion and the various
factors which affect its stability.

Finally we looked at the problems of salt in crude oil. You saw that even if the residual
amount of water in oil is reduced to very low percentages, if the salinity of that water is
very high then there could be problems with the total amount of salt in the oil.

13
Oil Treatment (Dehydration) Petroleum Open Learning

Section 2 – Principles of Emulsion Treating

In theory, if an emulsion was allowed to remain in Don’t worry about this equation if your maths are a It would appear therefore that our emulsion treating
a vessel for an unlimited period of time it would bit rusty, this is the last you will see of it. However problem can be overcome if we can achieve the
eventually separate into oil and water. The droplets what it means is, that to increase the speed of set- following :
of water would fall through the oil and form a layer of tling we must do one of two things. We must either
water at the bottom of the vessel. In fact the settling increase the value of the factors on the top line of 1. Decrease the viscosity of the oil
process is the basis of all emulsion treating systems. the equation, or decrease the value of the factor on
The time required for this to happen we could call the the bottom line. How can we do this? 2. Increase the difference in density between
settling time. Unfortunately in petroleum producing the water and the oil
operations we just do not have this time, so, in order Let’s examine each of these factors in turn.
to separate the two liquids in an emulsion and allow 3. Cause the water droplets to join together
them to settle, we must assist the process. •  irstly g the gravitational constant. This, as
F and form larger droplets
its name states, is a constant and we can
Let’s start by having a look at a physical law regarding do nothing at all about this There is in fact a fourth factor we could add to this
the speed at which a suspended particle would fall list – time. If it were possible we could try to increase
through a continuous medium. It can be described •  econdly r which is the radius of the particle,
S the settling time available.
by an equation known as Stokes equation. This is in our case the water particle. We could try
written as : to increase the radius of the particles by Before we move on from here try the activity on the
causing the droplets to join together thus next page.
increasing their size and hence their radius
V=
2 g r 2 ( d 2- d 1)
• The expression ( d 2 - d 1 ) represents the
9N
difference in density between the water and
Where : the oil. We could try to increase this
V =
 velocity
•  inally N is the viscosity of the oil. To in-
F
g = gravitational constant
crease the speed of settling, this must be
r = radius of particle
reduced. We could certainly try to do that
d1 = density of continuous medium
d2 = density of particle
N = viscosity of continuous medium

14
Petroleum Open Learning

The Application of Heat In view of all this, most emulsion treating plants

Activity In fact heating the emulsion can assist not only in


use heat. You should note however, that heating
causes some vaporisation of the lighter components
item one in our list but in items two and three also. of the oil. If this is not contained, a reduction in
This is simply an activity to get you thinking gravity with a corresponding reduction in volume
about ways to decrease the viscosity of a You have already seen that increasing the will occur. This of course means loss in revenue.
liquid. temperature of the oil reduces its viscosity. This Also, as the temperature is increased, the likelihood
allows the water droplets to sink more rapidly of maintenance problems occurring in the plant
Imagine working in the kitchen and having to through the oil. and equipment will increase. This being so,
mix some treacle with dry ingredients in order other methods are used to assist the application
to prepare a particular dish. If you took the As the temperature is increased the difference in of a reasonable amount of heat in the treatment
treacle straight out of the tin it might prove density between the water and the oil also increases. process.
difficult to stir - it is thick or very viscous. What This occurs up to a temperature of about 8O°C.
could you do to make the treacle thinner? After that the effect of heat on density difference We can now go on to look at some of these other
One thing you might think about doing is to diminishes. methods.
warm up the treacle in a pan on the stove.
As the treacle gets hotter it will start to flow Finally, heating the emulsion promotes the
more easily. It gets thinner or its viscosity is combining together or coalescing of the droplets. It
reduced. does this in two ways. Firstly, having a hot emulsion
means that the water droplets move around much
more freely and collide with each other far more
frequently. If these collisions are forceful enough,
the film surrounding the droplets can be ruptured and
they will coalesce. Secondly, as the water droplets
are heated they will expand. This will stretch the
surrounding films and make them weaker enabling
them to be ruptured more readily.
From what you have just been thinking about,
it would appear that item one on our list can be
achieved by heating the emulsion, so let’s consider
this now.

15
Petroleum Open Learning

The Application of Electricity


The application of electricity in emulsion treating is an attempt to promote
coalescence of the water droplets. Let’s have a look at how this works. Before
we do this, try the following activity.

Activity

• All you need to perform this activity is a plastic ball point pen a piece
of woollen cloth and a source of running water e.g. the kitchen tap.

• R
 un the water from the tap as a small, thin continuous stream. Make
sure that the stream is not breaking up into droplets.

• H
 old the blunt end of the pen against the stream of running water.
Observe what happens to the stream.

• N
 ow rub the end of the pen against the piece of woollen cloth a few
times. (The blunt end of the pen not the metal ball point end).

• H
 old the end of the pen close to the stream of running water again
and observe what happens.

16
Petroleum Open Learning

What you should have noticed during the activity you Normally the dipoles are arranged randomly within
have just performed is the following; the first time the molecules as shown in Figure 2.2.
that you held the pen against the stream nothing
happened. The second time however, after rubbing
the pen against the wool, the stream of water bent
towards the pen.

Obviously some force acts between the pen and


the water after the pen has been rubbed. It is in
fact an electrostatic force. It occurs because in
rubbing the pen against the wool you charge the
pen electrically.

But how does that attract the water? The answer to


this lies in the way that the water itself behaves.

The water droplets in the emulsion are made up of


molecules which themselves are neutrally charged Figure 2.3 : Water Modules Attracted to the
electrically. However within the molecules is an Positively Charged Pen
arrangement of charges which is known as an
electric dipole. This has a positive and a negative
Figure 2.2 : Electric Dipoles in Water Molecules The phenomenon we have just been looking at can
end and is shown very simply in Figure 2.1.
Randomly Arranged be used in the problem of treating emulsions.

If the emulsion is passed through an electric field


between two electrodes, the water droplets are
When the charged pen is placed near to the water polarised. They are then stretched due to the polar
stream the molecules line up with their negative attractions which weakens the surrounding film.
ends being attracted towards the positively charged They are also attracted towards one or other of the
end of the pen. We can say that they become electrodes and tend to speed towards it. Because of
polarised. This has the effect of pulling the water the weakened film and the greater collision force as
towards the pen. Figure 2.3 shows this. they hurtle through the oil, the droplets unite more
readily to form the larger droplets necessary for
Figure 2.1 : Electric Dipole faster settling.

17
Petroleum Open Learning

Although the application of heat and the application of electricity are both commonly used in
dehydration they are rarely used by themselves. In order to assist in the process or to speed
it up, chemicals are invariably injected into the emulsion. We can look at this now but before
we do, try the following Test Yourself question.

Test Yourself 2.1

Are the following statements true or false? If they are false give the reasons why.
TRUE FALSE REASON
a) The speed at which a suspended particle would fall through a continuous
medium is described by Stokes equation.

b) D
 ecreasing the difference in density between water and oil in an emulsion
would assist in allowing the water to settle during treating.

c) Increasing the temperature of oil reduces its viscosity.

d) Electric dipoles in water molecules are normally arranged with their negative
ends all pointing in the same direction.

e) If an emulsion is passed through an electric field between two electrodes the


water droplets are polarised. This causes them to be stretched due to polar
attractions which weakens the surrounding film.

You will find the correct answers in Check Yourself 2.1 on Page 59.

18
Petroleum Open Learning

The Application of Chemicals Demulsifiers are in fact very similar to the emulsifying
agents which cause the emulsion to form in the first
The addition of a chemical to the emulsion helps to place. They are surface - active chemicals which,
cause coalescence of the water droplets. It does this when added to the emulsion, diffuse rapidly to the
by breaking the film surrounding the water droplets. interface. Once there, they attempt to neutralise the
Before it can do this it has to get to the interface effect of the emulsifying agent.
between oil and water droplets. It then has to gather
sufficient droplets together prior to coalescence. Having arrived at the interface, the demulsifier
This gathering together is called flocculation. In gathers together droplets of water by the action
addition to the above it must be able to remove any of flocculation. The demulsifier, which is now
solid particles from the interface and carry them concentrated on a droplet, has a strong attraction
away with the separated water. Chemicals which for other water droplets in the same condition. The
are able to do this are called demulsifiers. droplets tend to join up rather like a bunch of grapes.
If they collide with sufficient force the skin may be
We could say therefore, that good demulsifiers have ruptured and coalescence takes place. Sometimes
four basic properties: however they just nestle together and further action
is needed for coalescence.
•  hey are strongly attracted to the water / oil
T
interface The next action of the demulsifier is to attack the
films surrounding the droplets if they are still intact.
• They cause flocculation of the water droplets It does this by causing eruptions at the interface
which consequently ruptures the film. With no film
• They help to rupture the film surrounding the to prevent coalescence, the water tries to find a
droplets, promoting coalescence condition giving the least contact area with the
oil. The water droplets, which are close together
•  hey cause solid particles to be attracted to
T because of the flocculation, unite and form larger
the water so that they can be removed with and larger droplets.
the water
The action of flocculation and coalescence is
illustrated in Figure 2.4.
Figure 2.4

19
Petroleum Open Learning

You saw in Section One that emulsifiers can contain


solid particles such as sand, silt and clays etc. They
may come from the reservoir or be residues from
the drilling mud and so on. These particles help
to increase the stability of the emulsion and must
be removed if successful demulsification is to be
achieved. The demulsifier does this by wetting the
particles. At this point I will say a little more about
wetting and wettability.

Adhesion is the property by which particles of a given


substance stick together. Liquids will stick to some
solid substances more than others. For example, if
you were to dip a glass rod in a beaker of water and
then remove it, the rod would be wet. If however
you do the same thing in a beaker of mercury, when
you remove the rod no mercury would be clinging
to the rod. This shows that some water is more Figure 2.5
adhesive to the glass than to water itself. Mercury
however sticks to itself rather than the glass. We
could say that the glass is water wettable but not The demulsifer makes the particles water wet. it You can deduce from all this that the demulsifier has
mercury wettable. If you coated the glass rod with a has one end which is strongly attracted to the solild several jobs to do. It would be almost impossible to
greasy substance however, the glass rod would not particle and forms a coating on the particle. The other find a single chemical which could accomplish all
be wet by water. Figure 2.5 shows this in a simple end is strongly attracted to the water and will carry these actions. Therefore, Demulifiers are cocktails
diagram. the particle in the water. This means that when the of chemicals which are blended to give the best
water droplets coalesce and sink, the soild particles possible results for the type of emulsion being
will be carried out of the oil and can be disposed of treated.
with the water.

20
Petroleum Open Learning

Demulsifler Selection When all the details regarding the type of crude When carrying out the test several points must
and the nature of the produced water is known, the be adhered to regarding the sample of emulsion.
Just as there are many different types of crude oil search for the most effective demulsifier can be These are :
there can be many different types of emulsions narrowed down. A number of demulsifiers from a
formed. To obtain optimum dehydration the most supplier’s range would be chosen and subjected to • the sample must be free of any demulsifier
effective demulsifier for each type of emulsion must field tests. The most common type of test carried chemical
be selected. What is effective in one field may not out is known as the bottle test.
work in another. In fact, the addition of the wrong • the sample must be truly representative of
type of demulsifier could aggravate the situation Demulsifier Bottle Test the total production
and cause the emulsion to become more stable.
The bottle test is used to help to determine which • the sample must be tested as soon as
The selection of the demulsifier depends on a chemical can most effectively treat an emulsion from possible. Ageing of the emulsion sample
number of factors, including: a given field. The results of the test can also indicate could affect its reaction to the treatment
the optimum amount of demulsifier to be added, i.e,
• The type of crude oil produced the ratio of chemical to emulsion. Adding too much Equipment
can be as bad as adding too little.
• The nature and composition of the produced
water The following list of equipment needed for the test
The basic procedure for carrying out the test is fairly straightforward, however I have given a
• The type of dehydration process involves taking a representative sample of the brief description of the items which you may not be
emulsion to be treated from a point in the process completely familiar with.
• The point of chemical injection plant. The sample is placed in a calibrated bottle
and a specified amount of demulsifiers added. The • 12 Calibrated bottles. These are similar
• The temperature
sample is agitated, allowed to stand whilst settling to medicine bottles with graduations
• Whether other chemicals will be used which takes place and separation of water is observed marked in millilitres (ml)
may react with the demulsifier and measured. After a time, a sample from the oil
layer above the water is taken. This is processed • A 100 ml pipette. A glass instrument with
The above list is not exhaustive but it serves to in a centrifuge so that any emulsion, water and which accurate amounts of emulsion can
show just how difficult the choice of demulsifier can solids remaining in the oil can be determined. Let’s be taken
be. Service companies who specialise in supplying expand this procedure and go through the basics of
oilfield chemicals produce a range of demulsifiers a bottle test.
for the different situations encountered. Even so a
considerable amount of work must be done in the
field to ensure that the correct choice is made.

21
Petroleum Open Learning

• Micro pipettes calibrated in 0.001 ml Test Procedure • Record the total water content. This gives
divisions. Used for dispensing the a figure which can be used to compare the
demulsifying chemical in very small but demulsifying chemicals under test
accurately known quantities A representative sample of the emulsion to be
treated is taken in a suitable container capable of
• Graduated glass syringes of 50 and 100 ml holding at least 2 Iitres. Main Test
• A water bath with thermostatic control Before conducting the main test, the total amount of With the total amount of water and emulsion in the
water and emulsion in the oil must be determined. sample known, the main test can be carried out as
• A centrifuge with calibrated centrifuge tubes. This is done in the following manner: follows :
This is a machine which spins a number of
tubes at a very high speed. The centrifugal • Fill the centrifuge tubes with a solvent such • Fill the calibrated bottles with 100ml of
force causes the samples in the tubes to as xylene up to the 50% level then top up sample
separate into oil and water to 100% with the emulsion
• Label the bottles with details of type of
• An agitator. This is capable of agitating • Agitate the tubes to mix the contents demulsifier and amounts used
the samples of emulsion in the calibrated thoroughly
bottles. (Sometimes in the field the bottles • Heat the bottles in the water bath to the
are shaken by hand.) • Centrifuge the tubes for 10 minutes same temperature as that of the demulsifier
injection point in the field
In addition to the equipment I have just listed, • Note the emulsion and free water content
demulsifier chemical and solvents are required. The • Add the demulsifying chemicals in exact
demulsifier is usually used in testing in a diluted form • Add a few droplets of a slugging amounts using the micro pipette
called a solution. A typical 5% solution would be compound (this is a chemical which does
prepared by mixing 1 ml of concentrated demulsifier not over treat the emulsion and cause the • Screw the tops on the bottles and ensure
with 19 ml of demulsifying solvent. formation of a stable emulsion even if there is no leakage
excessive amounts of it are used)
• Agitate the bottles either by hand or using an
• Agitate to mix and heat in the water bath agitator, for a period of time which relates to
for 10 minutes at 60°C the intensity of agitation in the field

• Centrifuge again for 10 minutes. This


should totally break the emulsion. If not,
repeat adding more slugging compound

22
Petroleum Open Learning

• After agitation, immediately place the bottles The amount of chemical added in a treating system
in the water bath where the temperature is usually quoted in parts per million ( ppm ). This
has been adjusted to that of the settling means the volume of chemical used per million
temperature in the field volumes of emulsion throughput.

• Record the start of the settling time and allow In fact the treatment dosage is determined before
the samples to settle doing the main test to determine the most suitable
demulsifier.
• Record the amount of separated water and
emulsion at regular intervals If we use a demulsifier which is known to be
reasonably effective, then the test which I have just
• From the results, select the best performing described is carried out using different amounts of
samples the same demulsifier instead of different chemicals.
This time 6 bottles would be used. Knowing a typical
• From the best samples, remove the oil from dosing ratio, the bottles would have chemicals
just above the interface using a syringe. added at 0.5, 0.75, 1, 1.5, 2, and 4 times this
Leave an equal amount of oil above the oil amount. From the results of the test the optimum
water interface in each sample dosage is determined, and this figure is used for
further testing.
• With this oil conduct a centrifuge test, which I
described earlier, to determine the amount of The bottle test will then indicate which demulsifying
residual water in the oil if any chemical is going to be best for our particular
dehydration problem and what the optimum dosage
From this test the best performing chemical can rate will be.
be determined to treat the particular emulsion
problem. I now want to look at the actual injection of the
demulsifier into the process stream. However, before
Of course we not only want to know which type of doing so, try the following Test Yourself question.
demulsifier works best for a particular emulsion but
also, what is the optimum dosage.

23
Petroleum Open Learning

Test Yourself 2.2

The steps taken in carrying out a bottle test are listed below in the wrong order.
List the steps in the correct order starting with :

a. Fill the calibrated bottles with 100 ml of sample. 1

b. Add the demulsifying chemicals in exact amounts using the micropipette.

c. Agitate the bottles for a period of time which relates to


the intensity of agitation in the field.

d. Label the bottles with details of type of demulsifier and amounts used.

e. Heat the bottles in the water bath to the same temperature as that of the
demulsifier injection point in the field.

f. Record the amount of separated water and emulsion at regular intervals.

g. From the results select the best performing samples.

h. Record the start of the settling time and allow the samples to settle.

i. Screw the tops on the bottles and ensure there is no leakage.

j. After agitating immediately place the bottles in the water bath where the temperature
has been adjusted to that of the settling temperature in the field.

You will find the correct answers in Cheak Yourself 2.2 on page 59.
24
Petroleum Open Learning

Injection of Chemicals

Where is the best place to introduce the demulsifier


into the emulsion ? The answer to this question
is that there is no single best chemical injection
location. Each process system must be carefully
evaluated to determine the most effective point
of injection. We can look at a typical system and
identify some possibilities.

Activity

The following drawing Figure 2.6, is a simplified


layout of a typical production process. Study
the drawing for a few minutes and mark on it
the points where you think we could inject the
demulsifying chemical.

Figure 2.6

25
Petroleum Open Learning

The drawing shows the flow of fluids from the reservoir through
the separators. In such a system there are several locations
which seem to be suitable as chemical injection points. I would
suggest the following :

• Down hole in each individual well

• At the surface into each flowline

• Into the main header

• Into the separators

Let’s consider each of these locations.

In general the chemical should be introduced as far upstream as


possible. Doing this ensures that there is a minimum of time for
the emulsifying agent to create and stabilise an emulsion. It also
means that the demulsifier has maximum time to do its work. The
turbulence as the fluids flow up the wellbore through the surface
valves and pipework ensures the dispersal of the chemical.

Having said that, it would appear that injection downhole is the


most effective location. Many wells are equipped with facilities for
chemical injection. The most common method would be to have
a chemical injection valve installed in a side pocket mandrel in
the tubing string.

Figure 2.7 shows part of a simplified well completion drawing


with a chemical injection valve in a side pocket mandrel.

Figure 2.7

26
Petroleum Open Learning

The mandrel is a piece of tubing with a bulbous side


to it. A small tube, the pocket, is incorporated into this.
The chemical injection valve sits in this pocket. You
can see from the figure that a small line is fed to the
side pocket mandrel down the tubing / casing annulus
from the surface. Chemical being pumped down this
line is contained between the two seals which straddle
the inlet port in the pocket.

The valve itself looks like the one shown in Figure 2.8.

The chemical enters the valve through the inlet


port and is pressurised against the valve and seat.
The valve is being held on its seat by a spring. At a
predetermined pressure the valve will open allowing
the chemical to flow round the valves internal pathways
and out through the outlet port and into the well tubing.
Pumping the chemical under pressure through such
an injector ensures that it sprays into the produced
fluids and is thoroughly mixed downhole.

Although downhole injection is certainly carried out


in many locations it does present certain problems.
Each well has to be completed in such a way that
chemical injection valves can be installed in the tubing
string. This adds to the cost of the well completion and
introduces extra possibilities for mechanical failure
in the well. Maintaining the many injectors is time
consuming and costly.

Figure 2.8

27
Petroleum Open Learning

A more practical solution then, may be to inject the


chemical at the surface into each well’s flowline
near the wellhead. There would be less time for the
demulsifier to work and more time for stabilisation
than downhole but the costs would be less.

Injecting the chemical into the main header


ensures that it is introduced continuously into
the total production. It could be a relatively low
cost installation having a single injection point as
opposed to multipoint injection downhole or in the
flowlines. However there will be less agitation and
less time for the chemical to work.

If chemical is injected at the surface into the flowlines


or the header, it will be injected via an injection
quill. This is designed to ensure that mixing is as
complete as possible between the chemical and the
emulsion. Figure 2.9 shows a simple injection quill
arrangement.

A non-return valve fitted in the injection line will


prevent back flow and protect the line from well
fluids.

Injecting the demulsifier into the separators is rarely


The actual point is often a compromise which de- Before we finish this section on the principles of
considered. By the time the fluids reach this point
pends on the type of operation. Sometimes a few emulsion treating we should consider two other
there is very little time left for the chemical to work
wells would be treated downhole or at the flowline points, i.e. settling and water washing. Let’s look at
effectively. The fluid flow through the vessel is much
with additional injection at a single point in the main water washing first.
less turbulent thus the chemical is less effectively
header. The character of each well’s production
dispersed. The emulsion has also had more time to
must be determined so that the wells which contrib-
stabilise.
ute most to the emulsion problem can be selected
for downhole treatment.

28
Petroleum Open Learning

Water Washing Settling


This process is a mechanical means of reducing the water content of a destabilised emulsion. In such a I have already said that settling is common to all
system the emulsion is introduced to, or made to pass through, a large body of water. As it does so, each types of treatment of emulsions. At the beginning of
water droplet in the emulsion may be absorbed by contact with this large volume. This is referred to as this section I said that if an emulsion could be left for
water washing. For most effective absorption, the wash water should be exactly the same water as the a sufficient length of time the water droplets would
droplets. In fact the wash water is often free water which has already been removed from the emulsion. In sink and form a water layer at the bottom of any
a water washing facility the emulsion flows under a baffle in the treating vessel thus ensuring that it passes vessel. So, in addition to heat, electricity, addition of
through the wash water. Figure 2.10 illustrates this. chemicals and water washing, there must be some
time allowed for the water to form a layer from where
it can be drained separately from the oil.

We have covered quite a lot in this section but before


I summarise for you, try the following Test Yourself
question.

29
Petroleum Open Learning

Test Yourself 2.3

Read through the following statements and fill in the missing word / words from the list given below :

a. Heating oil tends to reduce its

b. The speed at which a suspended particle would fall through a continuous medium can be described by equation.

c. An electric dipole has a and a end.

d. If an emulsion is passed through an electric field between two . the water droplets become

e. When water droplets gather together we could say that occurs.

f. A demulsifier helps to remove solid particles from the emulsion by the particles.

g. A chemical injection valve could be situated in a side mandrel in the tubing string.

h. An injection is designed to ensure that mixing is as complete as possible between the chemical and emulsion.

LIST OF WORDS

POLARISED, QUILL, NEGATIVE, ELECTRODES, HEADER, VISCOSITY, STOKES, WETTING,

POSITIVE, ELECTROSTATIC, BOTTLE TEST, FLOCCULATION, POCKET.

You will find the correct answers in Check Yourself 2.3 on Page 59.

30
Petroleum Open Learning

Summary of Section 2

In this section I have taken you through the We further saw that the application of electricity can You saw that there are many different types of
principles of emulsion treating in a logical help in promoting coalescence of the water droplets. It demulsifiers available and careful selection of the
manner. We started by having a look at Stokes does this by polarising the droplets. This has the effect best one must be made for a particular emulsion
equation which is written as : of weakening the surrounding film and causing them treating application. I described for you the bottle
to be attracted towards the electrodes. Because of the test which is used to help determine which
2g r2 ( d2 - d1 )
V= weakened film and the collisions which occur as the demulsifier can most effectively treat a given
9N droplets move rapidly through the oil, the droplets unite emulsion. From there we moved on to look at the
and form the larger droplets necessary for faster settling. injection of the chemical. We saw that there are
We saw from the equation that to increase the several options for injection points. These could
speed of settling of droplets of water through a We then moved on to look at the application of be downhole, in the well flowlines,in the header
continuous medium we must do three things : chemicals which help to promote coalescence. These or in the vessels. We looked at the advantages
demulsifying chemicals are surface active chemicals and disadvantages of each of these.
1. decrease the viscosity of the oil which have four basic properties. They :
To end the section we had a brief look at water
2. increase the density difference between • are strongly attracted to an oil / water interface washing and settling. In water washing the
the water and the oil emulsion is made to pass through a large body
• cause flocculation of the water droplets of water which absorbs the water droplets from
3. cause the water droplets to coalesce and the oil. Time for settling as I have mentioned
form larger droplets. • help to rupture the film surrounding the water on several occasions is necessary for any oil
droplets dehydration process, but speeding up the settling
We saw that heating the oil helps to decrease time is what most of this unit has been about.
its viscosity and also increase the density • cause solid particles to be attracted to the water
difference between the oil and water. so that they can be removed along with the water

31
Oil Treatment (Dehydration) Petroleum Open Learning

Section 3 – Dehydration Systems and Equipment

Having looked at the basic principles of treating At least three tanks would be used. In operation, one Note that there is provision for injecting chemical on
emulsions let’s now go on to the practical application of the tanks would be in the process of being filled, and the offshore platform and also into the pipeline before
of these principles in the field. The dehydration one would be settling. The third, having had the settled the tankfarm.
equipment will rarely use just one of the principles water drained off, would be having its clean oil pumped
covered in the last section, but will use a combination to a tanker or a refinery. Figure 3.1 shows a simplified
of them. We can look at several types of treating version of such a system.
vessels. I intend to start with some rather basic
equipment and follow on with some which is slightly
more complicated.

Settling Tanks
All treating systems involve settling.
In some situations a simple settling
tank used in conjunction with chemical
injection could be all that is necessary.
In this case the tank must be big enough
to allow sufficient retention time for the
water droplets to sink to the bottom.

A typical example of such a system is a


tank farm at a shore terminal. Here the
total production from an offshore field is
transported via a subsea pipeline to very
large tanks at the terminal. Although the Figure 3.1
tanks are principally used for storage,
because of the amount of time that the oil
remains in the vessels, settling can take place.

32
Petroleum Open Learning

Wash Tanks Inside the main body of the tank a water layer is
maintained. This is the wash water. The spreader is
A wash tank is more likely to be found on older land designed so that the emulsion exits as several small
installations than offshore and can only be used with streams which rise independently through the wash
relatively low throughputs. It is basically a settling water. As these small streams rise, a certain amount
tank with a few refinements. Although you may not of de-emulsifying takes place as the water droplets
come across one of these vessels, it is worth having in the emulsion contact the large body of water. The
a look at its construction and operation as it uses clean oil will continue to rise whilst the water droplets
some basic principles of emulsion treating. A typical remain in the wash water.
wash tank is shown in Figure 3.2. Look at this now
and identify the various components. Any emulsion which has not broken down during its
passage through the water will form a layer on top of
The emulsion to be treated enters the unit through the water. Clean oil will form a further layer on top of
the inlet line and passes to a larger diameter pipe, the emulsion. Further breakdown of the emulsion will
the conductor. This vertical pipe may be mounted take place within the layer on top of the water. This
either inside the vessel as shown in the drawing, or layer will remain in the vessel for a relatively long
outside. Gas may be liberated from the emulsion at time so a certain amount of settling will take place.
this point. The conductor acts as a vertical separator Water will sink into the water layer with oil rising to
within the wash tank. Any gas is taken from the top join the oil layer.
of the conductor and passes through an equalising
line into the top of the wash tank. This equalises The wash water level in the tank is maintained by a
pressure between the conductor and the main level control valve in the water outlet line.
body of the tank. The gas-free emulsion then flows
down the conductor and is spread out through the In the system we have just looked at, the breaking
water layer at the bottom of the tank. A spreader down of the emulsion is achieved in two parts :
arrangement at the bottom of the conductor helps to
• water washing
do this.
• settling
Figure 3.2
A variation of these principles can be found in the
type of vessel known as the Free Water Knockout
which we will look at now.

33
Petroleum Open Learning

Free Water Knockouts


Strictly speaking, free water is produced water which will settle out of the oil within five minutes if the fluids are at rest. As such it is
not part of the emulsion and can be removed by gravity separation in a simple separator. It is important that this free water is removed
before the emulsion is further treated in a system such as the heater treater which we will look at shortly. A free water knockout drum,
as illustrated in Figure 3.3, will do this.

Figure 3.3 Free Water Knockout Drum

34
Petroleum Open Learning

This vessel is basically a horizontal 3 phase separa-


tor. The incoming fluids impinge on an inlet diverter
where initial separation of gas from liquids takes
Test Yourself 3.1
place. The liquids fall to the bottom of the vessel
and the diverter ensures that they pass through a
water layer which is maintained in the vessel. Thus
the liquids are water washed. The separated free The following terms apply to a wash tank, a free water knockout drum, both of these or neither.
water plus any water which has been washed out of Mark with a ✓ which.
the emulsion settles into the water layer. The level
of the water layer is maintained by an interface level
controller, operating a level control valve in the water
outlet line. The oil and emulsion flows over a weir into Terms Wash Tank Free Water Both Neither
the oil accumulation section from where it is taken Knockout Drum
under level control to the emulsion treating facility.
Inlet diverter
The treating facilities we have just been looking at
are fairly simple systems. We can now go on to look Gas Equaliser
at something a little more complicated. Before we do,
try the following Test Yourself question. Spreader

Water Layer

Weir

Conductor Pipe

Injection Quill

Water Level Control Valve

You will find the correct answer in Check Yourself 3.1 on Page 60.

35
Petroleum Open Learning

Heater Treaters This heat exchanger is of the shell and tube type. This vessel can include a number of elements such
In our case the medium to be heated, the emulsion, as :
We can now move on to a treatment facility which flows through the shell as shown. The heating
uses the application of heat to assist in the process. medium flows through the tubes. This could be hot oil • separation section
The heat may be applied prior to treating the which has been heated using waste heat from power
emulsion in a simple wash tank. In this case the generation turbine exhausts. • heating elements
heating could be carried out in a heat exchanger
similar to the one shown in Figure 3.4. Although the application of heat via an external • oil surge section
heating source which I have just described is
perfectly feasible, it is more common to incorporate • mist extractor
this into a vessel called a heater treater.
• coalescing section

• spreader

• oil collector

There are a number of different styles of heater


treaters so I will describe just one which is typical.
It is a horizontal vessel which looks rather like
a separator. Its internal features however are
completely different. Look at Figure 3.5 on the next
page, which shows the internals of a horizontal
heater treater vessel. Identify the internal features
of the treater then we will follow the flow through the
vessel.

Figure 3.4: Shell / Tube Heat Exchanger

36
Petroleum Open Learning

Figure 3.5 Heater Treater Vessel Figure 3.6 Mist Extractor

Flow, which is a mixture of oil, some gas, emulsion and free water enters the separation section of the treater,
where initial separation takes place. Any gas in the fluids is flashed off at this point and flows to the gas outlet
line. Before leaving the vessel the gas passes through a mist extractor. This is a device which ensures that
any small droplets of liquid which may have been retained in the gas stream are removed. A common type
of mist extractor is in the form of a knitted wire mesh. The droplets of liquid impinge on the mesh, form larger
droplets then fall into the liquid below. Figure 3.6 shows a simplified version of a mist extractor.

37
Petroleum Open Learning

The liquids fall down into the bottom of the separation As the heated fluids rise the water droplets coalesce
section and are deflected round the heating element and when they become large enough they fall
whilst doing so. Free water is separated here and through the rising continuous phase. The water
the water accumulates as a layer at the bottom of droplets accumulate at the bottom of the section and
the vessel. The water layer forms a water wash form another water layer. The height of this layer is
section which helps to remove unstabilised water maintained by a further interface level controller. This
from the emulsion. An interface between water and operates a level control valve in the water outlet line
oil / emulsion is maintained by an interface level from the coalescing section.
controller. This operates a level control valve in the
free water outlet line. The liquid which reaches the top of the vessel is
treated oil, which should be free from any water
The oil and emulsion then rise past the heating or emulsion. This is taken from the treater via a
element where the temperature is increased to the collector pipe and flows to the next part of the
optimum treating temperature. The heating element production process system.
may be simply a tube coil through which heating fluid
is being circulated. On some land locations a directly You saw in Section 2 that passing the emulsion
fired heating system may be used. through an electric field can help in the coalescence
of water droplets. We can now see how this is done
The heated oil and emulsion then flows over a in practice.
weir into the oil surge chamber. From here it flows
through a spreader arrangement into the coalescing
section of the vessel. The coalescing section is kept
completely full of liquid. Unlike the separators which
you are probably familiar with, the oil outlet is at the
top of the vessel rather than at the bottom.

The spreader ensures that the flow is distributed


evenly throughout the length of the section. If this
were not used the liquid flow could channel towards
the outlet and reduce the efficiency of the treater.

38
Petroleum Open Learning

Electrostatic Treaters
Electrostatic treaters are very similar in construction
to the heater treater we have just been looking at.
The main difference is that they incorporate high
voltage AC and / or DC electrostatic field in the
coalescing section. Figure 3.7 illustrates a typical
electrostatic treater. Study this for a while and note
the differences between this and the heater treater.

Figure 3.7 Electrostatic Treater Vessel

39
Petroleum Open Learning

In this vessel you will notice that the major difference Damage to the electrical system could occur if the
between it and the heater treater is the pair of level in the vessel were to go low enough to uncover
electrodes in the electric coalescing section. the electrodes. To prevent this happening a low level
shut down switch is incorporated into the emergency
The initial flow through the vessel is the same as shutdown system for the vessel.
that described earlier. However when the heated
emulsion rises through the coalescing section it
has to pass through the electric field created by the Desalting
electrodes. As it does so, the water droplets are
given an electric charge. The polarised droplets are As I pointed out in Section 1, crude oil which is
attracted to one or other of the electrodes and race contaminated with salt is unacceptable to a refinery.
towards it. As they move rapidly through the emulsion In production systems where salt in oil is a problem,
they collide with each other. The polarisation also something must be done to desalt the oil. Often the
weakens the film around the droplets so that as they dehydration process of chemical injection coupled
collide they readily coalesce. When the droplets are with heater treaters and / or electrostatic treaters will
large enough they sink to the bottom of the vessel be sufficient to accomplish the desalting. In some
forming a water layer. The oil / water interface level cases however, it may be necessary to inject fresh
is controlled by a level controller, operating a level water into the emulsion. This will dissolve the salt so
control valve in the water outlet line. that it can be removed together with the water in a
treating vessel.
The electrical system in an electrostatic treater
consists of a transformer and the two electrodes The desalting system which I have used to illustrate
which are suspended one above the other in the such a process, utilises a pre-heater, a fresh water
coalescing section. In some types of electrostatic storage tank, a fresh water injection pump and an
treater the distance between the electrodes can electrostatic desalter/dehydrator. It is the type of
be adjusted. This allows the voltage to be varied to system commonly found at a terminal where a fired
meet the requirements of the specific emulsion being pre-heater is used.
treated.
Look at Figure 3.8 on the next page, which shows
this system as a simple block diagram.

40
Petroleum Open Learning

Figure 3.8 : Desalting System

41
Petroleum Open Learning

The salt - contaminated oil passes firstly through


the heater where the temperature is raised to the
optimum treating temperature. The heater itself is
called an indirect heater. This is because the heat
from the burning fuel is not transferred directly to the
oil. It is transferred indirectly through a water bath in
the body of the vessel to the oil being heated as it
passes through tubes in the heater body.

Figure 3.9 shows such a heater.

The heater itself consists of the following items.

• heater shell

• firebox with burner

• flow tubes

The heat is generated by burning fuel gas or oil in


a burner. The hot flue gases flow through fire tubes
and are exhausted through the stack. This flow of hot
gases heats up a body of water contained in the shell
of the heater. The water in turn heats up the oil which
is flowing through the flow tube bundle.

Figure 3.9 : Pre-heater

42
Petroleum Open Learning

After being heated the oil flows towards the


dehydrator. In this case it is an electrostatic
dehydrator which operates in the manner described
earlier in this section. Before it gets to the dehydrator, Test Yourself 3.2
fresh water is mixed with the oil. The fresh water is
pumped from a storage tank to a spray injector in the
flowline. The salt in the oil is thus diluted by this fresh The following pieces of equipment could be found in a wash tank, a heater treater, an electrostatic
water which mixes with the very salty emulsion water. treater or all of them. Fill in the boxes with a ✓ to show which piece of equipment goes where.

The dilution water plus the emulsion water is finally


removed in the dehydrator and led off for disposal. Equipment Wash Tank Heater Treater Electroststic Treater
The oil leaving this vessel should be clean in terms of
salt content and water. Mist Extractor

Spreader
This completes Section 3, but before I summarise
what we have looked at in this section try the
Equalising Line
following Test Yourself question.
Weir

Electrodes

Conductor Pipe

Heating Element

Transformer

You will find the correct answers in Check Yourself 3.2 on Page 61.

43
Petroleum Open Learning

Summary of Section 3

We began this section by considering some basic equipment used for treating emulsions. Firstly, I pointed out that a simple settling tank could be used,
and illustrated this by showing you a tank farm system.

We then looked at facilities used to wash an emulsion. The simple wash tank was explained in detail and you saw that the breaking of the emulsion
is achieved in two parts i.e. water washing and settling. I also showed you a variation of the wash tank which is used to remove any free water prior to
emulsion treating. The vessel doing this is called a free water knockout drum.

Heater treaters’ vessels came next and we You saw how the treater vessel uses these elements Finally we saw that fresh water maybe injected into
looked at a typical treater vessel containing the to break down the emulsion so that the water can be an emulsion to reduce the amount of residual salt in
following elements : removed leaving clean oil. a produced oil stream.

• separation section We similarly went through the operation of an In the next section we are going to combine some
electrostatic emulsion treater vessel. You saw that in of these treatment systems and look at an overall
• heating elements operation it is very similar to the heater treater vessel. dehydration process.
The essential difference is the inclusion of a pair
• oil surge section of electrodes. These, when connected to a power
supply, create an electric field. The water droplets
• mist extractor passing through this field are polarised which causes
them to speed towards the electrodes, colliding as
• coalescing section they do so. The polarisation also weakens the film
surrounding the droplets so that when they collide
• spreader they coalesce more readily and sink to the water
layer.
• oil collector

44
Oil Treatment (Dehydration) Petroleum Open Learning

Section 4 – A Typical Dehydration System

In this final section we are going to look at a complete On the next page, I have included a simple block
dehydration and desalting system. diagram to show the system in its entirety. Study this
for a while and familiarise yourself with the equipment
The system I will use as an illustration includes two used and the flow paths through the system.
separators and a free water knockout (FWKO) drum
for initial separation. From the FWKO drum the crude We will now follow the flow through the system in
and emulsion is pumped via a pre-heater through more detail. Let’s consider the process, section by
a water bath heater and two stages of dehydration section.
to storage. As a means of reducing the residual
salt content, water is injected prior to dehydration.
This system is typical and does not represent any
particular system.

The function of this system is to :

• Separate free water from the incoming well


stream

• Treat the remaining emulsion and reduce the


residual water content to an acceptably low
level

• Reduce the residual salt content to within the


limits set by the purchaser

45
Petroleum Open Learning

Figure 4.1 : Typical Dehydration System

46
Petroleum Open Learning

Separation and Free Water Knockout


Look at Figure 4.2 which shows this part of the system. It is a straightforward separation process. The reservoir fluids which are
a mixture of oil, gas, free water and emulsion, flow to the first vessel in the system, the first stage separator. This is a 3 phase
separator which in our system is operating at a pressure of 10 barg.

The first chemical injection point is into the line entering the first stage separator. Demulsifier is injected here to give it as much
time as possible to take effect before the emulsion reaches the electrostatic dehydrators.

Figure 4.2 : Typical Separation and Free Water Knockout System

47
Petroleum Open Learning

In the first stage separator, the free water is The inlet flow into the FWKO drum consists of the Once again the interface level in the FWKO drum
separated and forms a water layer at the bottom liquids from the second stage separator plus the water is controlled by an interface controller (LC 04)
with oil floating on top. The level of the interface which has been removed in the first stage separator. operating LCV 04. The water which is removed in
between the two is maintained by an interface level It may seem strange removing water from the liquid this vessel consists of the water removed in the
controller (LC 01) which controls the interface level stream then recombining them at a later stage. first stage vessel plus any further free water which
control valve (LCV 01) in the water outlet line. This You will remember the reason for doing this if you has been washed out of the emulsion. This water is
water flows to the FWKO drum where it is used think back to our discussion on free water knockout routed to a produced water clean up facility prior to
as wash water. The oil plus emulsion flows to the facilities in Section 3. To make sure that you can recall disposal.
second stage separator under the control of the the process try the following Test Yourself question.
level controller (LC 02) operating LCV 02 in the oil
outlet line. The separated gas is taken from the top
of the vessel through pressure control valve (PCV
01) operated by pressure controller (PC 01) which
maintains the correct pressure in the vessel. Test Yourself 4.1
The second stage separator is a 2 phase vessel. It
is operating at a pressure of 3.5 barg maintained by Without referring to the notes make a sketch of a simple free water knockout drum
a pressure controller (PC 02) and a pressure control
valve (PCV 02).

Being a 2 phase vessel this separator has no oil


water interface control. All the liquids leave the vessel
via the oil outlet and flow to the FWKO drum. The
liquid level in the separator is maintained by the level
controller (LC 03) operating LCV 03 in the liquid
outlet line.

You will find the correct answer in Check Yourself 4.1 on page 62

48
Petroleum Open Learning

The pressure in the vessel is maintained just above Crude and Emulsion Heating Picking up the flow from the FWKO drum you will
atmospheric by PCV 03 operated by PC 03. see that there is further provision for the injection of
You will remember that heating an emulsion helps to demulsifier into the flowline upstream of the crude
The oil level in the vessel is maintained by a control enhance the dehydration process. In this section of pump. This ensures good mixing as the crude and
valve which is located downstream of the feed pump the plant the liquids are heated in two stages, first in emulsion flow through the pump. Downstream of the
to the heater section. This is operated by LC 05. We a shell and tube heater then in a water bath heater. pump is the control valve LCV 05. This valve controls
will look at this shortly. the oil level in the FWKO drum via LC 05.
Figure 4.3 shows this small section.
We can now move on to the next part of the plant
which includes the heaters. Before we do so
however, read through the last few paragraphs and
make sure that you understand the flow through the
separation section.

Figure 4,3 Typical Crude and Emulsion Heating System

49
Petroleum Open Learning

The pump transports the crude onwards through the


pre-heater. This heater is of the shell and tube type
which we looked at in Section 3. The heating medium
Test Yourself 4.2
is the treated crude which comes from the second
stage of the dehydration plant. Using this crude as
a heating medium means that energy is recovered I have listed the items of equipment and injection points in the initial flow path of our typical
which would otherwise have been wasted. dehydration system. These items are in the wrong order. Place them in the correct order, start-
ing with inlet.
From the pre-heater the crude flows through the
water bath heater. Gas from the plant is used as fuel a) Inlet 1
to fire the heater.
b) 1st stage separator
The temperature inside the heater is controlled by a
temperature controller (TC 01) which regulates the c) Injection of water from 1st stage separator
fuel supply through TCV 01.
d) Second demulsifier injection point
A safety shutdown system also protects the heater
if there should be a flame failure at the burner. e) FWKO drum
This stops supply of fuel gas, ensuring that there is
no dangerous build up of gas escaping from unlit f) Crude pump
burners.
g) 2nd stage separator

We can now look at the dehydrators themselves, h) Water bath heater


but before we do, try the following Test Yourself
question. i) LCV for FWKO drum oil level

j) Pre heater

k) Dehydrators

You will find the correct answers in Check Yourself 4.2 on Page 62.

50
Petroleum Open Learning

Electrostatic
Dehydrators
The crude enters this part of the process
from the water bath heater. Identify this
point on the drawing of this part of the
system in Figure 4.4.

The first thing you will see as you


trace the flow is an injection point for
water. This water is the reject water
from the second stage dehydrator.
It is injected at this point to help reduce
the salinity of the incoming water in the
emulsion and to water wash the emulsion.
Although the water used here is itself salty,
it is less saline than the incoming water.
The water is injected through nozzles
which ensure that it enters the main flow
as fine droplets. These droplets must
then combine with the water in the
emulsion which requires some form
of agitation. A mixing valve takes care
of this. The valve is a differential pressure
control valve (DPCV 01). Its controller
(DPC 01) maintains a pressure droplet
across the valve and this, together with
the plug and seat profile of the valve itself, provides
the necessary surface and energy for the agitation to
take place.

Figure 4.4 : Electrostatic Dehydrators

51
Petroleum Open Learning

After mixing, the flow of liquids enters the first stage


electrostatic dehydrator. This dehydrator works in
almost the same way as the one we looked at in
Section 3. The essential difference is that no heating
element is included in the dehydrator itself.
The heating of the emulsion is done prior
to the treater as we have just seen.

The flow at this point becomes


more complex so we will divide
it up and follow the flows of
oil / emulsion and water
separately. Let’s start with the
oil / emulsion.

These liquids enter the


dehydrator and follow a
similar flow path to the one
described in Section 3.
The instrumentation on the
vessel can be quite complex
but we can look at some of
the more important instruments.
To minimise the complexity of
Figure 4.4 we can look at this in
isolation in Figure 4.5.

Figure 4.5 : Electrostatic Dehydrator

52
Petroleum Open Learning

Look first to the right hand side of the vessel. There We can now go back to Figure 4.4 again and The dilution water is heated in the heater and then
you will see LG 01. This is a sight glass which gives continue to trace the flow. The oil / emulsion from the joins the oil entering the second stage dehydrator.
a visual indication of the interface level between first stage dehydrator passes to the second stage The reject water from this vessel is recycled to the
water and emulsion. An interface level controller vessel. Before entering this vessel more water is first stage dehydrator where it helps to dilute the
(LC 06) controls the water level through its control injected into the stream. This water is dilution water incoming water. It is pumped by the recycle pump
valve (LCV 06) which is situated downstream of the which is often supplied from specially drilled water through level control valve (LCV 07) which together
dilution water heater. Alarms are incorporated into wells. The heated dilution water is injected through with level controller (LC 07) maintains the water level
the interface level control instrumentation. These nozzles again, and a second mixing valve (DPCV 02) in the second stage dehydrator.
are designated level alarm high and level alarm low controlled by DPC 02 ensures correct agitation. The
(LAH &LAL) and will warn the operator if the level is second stage electrostatic dehydrator works in the
reaching potentially serious points. Separate level same manner as the first. It is also protected by the We have only one small section to look at now, the
switches (LSHH &LSLL) are tied into the shut down same type of instrumentation. dilution water system. Before we go on to this, try the
system of the plant. If the interface level should reach following Test Yourself question.
the set points of these instruments a shut down will The hot treated crude from this dehydrator, prior to
automatically be activated. being routed to storage, flows through the preheater
where it acts as the heating medium to raise the
Because it would be dangerous if the oil level temperature of the crude before it enters the main
dropped and uncovered the electrodes in this section water bath heater.
of the dehydrator, further level instrumentation
protects against this. A level transmitter in this section Tracing the water flows through Figure 4.4 we begin
activates an electrical power shut down if the oil with the reject water from the first stage dehydrator.
level drops below a pre determined minimum. This is This water flows firstly through the dilution water
shown on the drawing as (LT 01). heater where it acts as the heating medium for the
water from the wells. After passing through the level
control valve (LCV 06), the water is routed to a
produced water clean up facility prior to disposal.

53
Petroleum Open Learning

Test Yourself 4.3

Complete the following sentences with an appropriate word or phrase.

a) Upstream of the 1st stage dehydrator there is an injection point for reject water which comes from the

b) The injection water and water in the emulsion require agitation. This is taken care of by a mixing

valve which is a valve.

c) Dilution water is passed through a before joining the oil entering the 2nd stage

dehydrator.

d) It would be dangerous if the oil level dropped and uncovered the

in the dehydrator.

e) The dilution water heater uses from the

as its heating medium.

You will find the correct answers in Check Yourself 4.3 on Page 63.

54
Petroleum Open Learning

Dilution Water System In this system the dilution water is obtained from The water is produced from the wells to a storage
specially drilled water wells. In some areas where tank. It is pumped using submersible pumps which
Look at the final drawing in Section 4 Figure 4.6 fresh water sources are scarce, slightly salty brackish are driven by an electric motor. The level in the tank
which shows the dilution water system. water could be used. is maintained by the on / off operation of the water
well pumps using level switches (LSH 02 & LSL 02).

Figure 4.6 : Dilution Water System

55
Petroleum Open Learning

From the tank the water is pumped by the dilution


water pump to the second stage dehydrator.
Upstream of the pump, there is provision for injecting Test Yourself 4.4
chemicals. Scale inhibitor is added to the water
to prevent scale building up in the pipework and
vessels. An oxygen scavenger is also injected to Without referring to Figure 4.1 sketch a block diagram, illustrating the
reduce the dissolved oxygen content of the water and typical dehydration system which we have just studied.
reduce its corrosiveness.

The ratio of dilution water to crude / emulsion


throughput is carefully controlled. A typical figure
could be 1 : 20. The actual amount of dilution water
added is controlled by a flow control valve (FCV 01).
This is regulated by a flow controller (FC 01) taking
its signal from a flow transmitter.

Before being injected into the feed to the second


stage dehydrator, the dilution water is heated. This is
done in a shell and tube type heat exchanger which
uses the produced water as its heating medium.

This completes this unit on dehydration of crude


oil. Before I summarise Section 4, try the last Test
Yourself question.

You will find the correct answer in Check Yourself 4.4 on Page 64.

56
Petroleum Open Learning

Summary of Section 4

In this section I have taken you through a typical dehydration and desalting plant. It is an entirely hypothetical plant which is not intended to
represent any existing installation. It has simply been used to illustrate the principles which we discussed in the preceding sections.

In this plant we saw that the initial separation of water was accomplished using a three stage separation process. The first stage removed free
water which was used as wash water in the third stage or free water knockout. All the water removed in the FWKO drum was taken to a produced
water clean up facility prior to disposal.

The crude oil and remaining emulsion was then heated in a pre-heater and a water bath heater before entering the first stage of a two stage
dehydration and desalting process. These vessels were electrostatic units. Prior to the first stage the reject water from the second stage was
added to the feed. This helped to dilute the salt content of the produced water. The reject water from the first stage was combined with the water
from the FWKO drum and sent to disposal via the produced water clean up facility.

Before entering the second stage dehydrator, dilution water was added to the feed. This water can be obtained from water wells and heated by the
reject stream from the first stage in a heat exchanger located upstream of the injection point.

In the second stage dehydrator, the crude stream was finally treated to achieve the correct specification. The treated crude was then used as a
heating medium in the pre-heater prior to being sent to storage facilities from where it would be transported to the purchaser.

57
Petroleum Open Learning
Check Yourself – Answers

Check Yourself 1.1 Check Yourself 1.2 Check Yourself 1.3

Using the formula a. False – Mayonnaise once formed is very If the salinity of water is 140 000 ppm and 0.1% water
stable and is difficult to break down. remains in the oil, from the graph (Figure 1.2) the
surface area = d2 equivalent salt would be 57 PTB. Therefore it would
the S.A. of a single droplet is b. False – An emulsifying agent is also required. not be acceptable. If the water salinity was reduced
to 100 000 ppm the equivalent salt would be 39 PTB.
152 = 707 mm2 c. True. This would fall in the acceptable range.

the S.A. of each small droplet is d. True.


8.77 = 241.6 mm
2 2

total of 5 small droplets is

5 x 241.6 = 1208 mm2

single droplet has smaller surface area.

58
Petroleum Open Learning

Check Yourself 2.1 Check Yourself 2.3

a) True a. Heating oil tends to reduce its VISCOSITY.

b) False, we need to increase the density b. The speed at which a suspended particle would fall through a continuous medium can be
difference described by STOKES equation.

c) True c. An electric dipole has a POSITIVE and a NEGATIVE end.

d) False, they are arranged randomly d. If an emulsion is passed through an electric field between two ELECTRODES the water
droplets become POLARISED.
e) True
e. When water droplets gather together we could say that FLOCCULATION occurs.

f. A demulsifier helps to remove solid particles from the emulsion by WETTING the particles.

Check Yourself 2.2 g. A chemical injection valve could be situated in a side POCKET mandrel in the tubing string.

h. An injection QUILL is designed to ensure that mixing is as complete as possible between


The correct order is : the chemical and emulsion.

a, d, e, b, i, c, j, h, f, g

59
Petroleum Open Learning

Check Yourself 3.1


Your answer should look like the following:

Terms Wash Tank Free Water Both Niether


Knockout Drum

Inlet Diverter 4

Gas Equaliser 4

Spreader 4

Water Layer 4

Weir 4

Conductor Pipe 4

Injection Quill 4

Water Level Control Valve 4

60
Petroleum Open Learning

Check Yourself 3.2

Your answer should look like the following:

Equipment Wash Tank Heater Treater Electrostatic Treater

Mist Extractor 4 4

Spreader 4 4 4

Equalising Line 4

Weir 4 4

Electrodes 4

Conductor Pipe 4

Heating Element 4 4

Transformer 4

61
Petroleum Open Learning

Check Yourself 4.1 Check Yourself 4.2


Your sketch should look similar to Figure 3.3, a, b, g, c, e, d, f, i, j, h, k
which is reproduced below.

62
Petroleum Open Learning

Check Yourself 4.3


a) Upstream of the 1st stage dehydrator there is an injection point for reject water which comes from
the 2nd STAGE DEHYDRATOR.

b) The injection water and water in the emulsion require agitation, This is taken care of by a mixing
valve which is a DIFFERENTIAL PRESSURE CONTROL valve.

c) Dilution water is passed through a HEATER before joining the oil entering the 2nd stage
dehydrator.

d) It would be dangerous if the oil level dropped and uncovered the ELECTRODES
in the dehydrator.

e) The dilution water heater uses WATER from the 1st STAGE DEHYDRATOR as its heating
medium.

63
Petroleum Open Learning

Check Yourself 4.4


Your answer should
look like the following.

64
POL
Petroleum Open Learning

Petroleum Gas
Compression
Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY

1
Petroleum Gas Compression - Unit 1 - An Overview Petroleum Open Learning
(Part of the Petroleum Processing Technology Series)

Contents Page
Visual Cues
• Training Targets 1.2

• Introduction 1.3 training targets for you to achieve
by the end of the unit

• Section 1 - Compressor Applications 1.4

• Section 2 - Basic Principles of Compression 1.15 test yourself questions to see how
much you understand
Pressure - Volume Relationship
Temperature - Volume Relationship
The Combined Gas Law
Energy
check yourself answers to let you
see if you have been thinking along
• Section 3 - Types of Compressor 1.26 the right lines

The Compressor Family Tree


Positive Displacement Compressors
Continuous Flow Compressors
Compressor Selection activities for you to apply your new
knowledge

• Check Yourself - Answers 1.32


summaries for you to recap on the
major steps in your progress

1.
Petroleum Open Learning

Training Targets

When you have completed Unit 1 of the Petroleum Gas Compression series you will be able to :

• List the main uses of compressors in the petroleum producing industry.

• Perform simple calculations involving pressure, temperature and volume relationships.

• Explain how energy conservation principles are applied to compressor technology.

• Define compression ratio and compressor capacity.

• Describe the compressor family tree.

• List the factors which influence compressor selection.

1.
Petroleum Gas Compression - Unit 1 - An Overview Petroleum Open Learning

Introduction
In the petroleum industry, the two main hydrocarbon components dealt with are crude oil and natural gas.
These substances are produced together from the underground reservoir in varying proportions.

At the surface, the oil and gas are separated from each other. These two streams are then further processed
independently, where necessary.

The oil may then be pumped to a refinery, or a terminal This unit is the first in a series which covers the • Basic principles of compression - Here we
for onward transportation. subject of gas compression. In this one, we will have will look at the underlying theory of gas
a look at the basic principles of compression. The compression. In this section, you will also be
The gas which is separated from the oil may be second unit in the series will cover reciprocating introduced to some of the terms and
transported by pipeline for sale. It can also be used on compressors, the third will concentrate on centrifugal expressions commonly used in compression
site for a number of other applications. For instance: machines, whilst the fourth will be dealing with other technology
types of compressor.
• it may be injected back into the reservoir to • Types of compressor - This section will
help maintain the pressure there concentrate on the different types of
I have called this unit an overview. It is necessary compressor in common use and their
• it could be used in wells to assist them to flow, to introduce the subject in some detail before suitability or otherwise for specific applications
using a technique known as gas lift concentrating on specific types of machine. Therefore,
in the unit, we will cover material which is common to
• it may be used as fuel on the plant or platform all compressor technology.

• some of its constituents may be removed as a The unit is divided into 3 sections, and we will be
liquid in a gas liquids recovery plant looking at them in the following order:

But, however the gas may be used, it invariably will • Compressor applications -In this section
be at too Iow a pressure when it leaves the separation we will consider the reasons for, and uses of,
system. In order to transport the gas, or allow it to compressors in a petroleum producing
do useful work, it will usually need compressing to a operation
higher pressure. This requires the use of some kind of
gas compression plant.

1.
Petroleum Gas Compression - Unit 1 - An Overview Petroleum Open Learning

Section 1 - Compressor Applications


In this section we will consider typical compressor
applications on an offshore production platform. Of
course, what we cover would be applicable to an
onshore plant, but the emphasis here will be offshore

Let’s look first at a simplified flow diagram for an oil


and gas processing system.
Figure 1 shows such a diagram

1.
Petroleum Open Learning

You will see that, after gathering the well fluids But we are particularly interested in what happens to •
• Reservoir pressure
Reservoir pressure
together, the first part of the process is a separation the gas. Let us look a little more closely at this. We
system. Here the oil, water and gas which are will trace the gas flow through the various facilities, • • Ratios and
of oilgas
andvolumes produced
gas volumes (the (the
produced
produced from the reservoir are separated from each and will build up a simple picture of the gas process field Gas Oil Ratio - G.O.R.)
other. (This process is covered in detail in the Oil operation.
and Gas Separation Unit, which is also part of our • • Pressure requirements of platform gas
Petroleum Processing Technology Series of open A separation system may consist of a number of facilities
learning programmes.) After separation, the crude oil separators working in series. They operate at
is treated if necessary, metered and pumped away for successively lower pressures. The actual number of The system Ii am using as an example in this unit has
further processing. vessels and their operating pressures will depend on has
two stages
two stages
of separation
of separation
operating
operating
at 17atbar
17 and
bar 1.4
variables such as : and
bar respectively.
1.4 bar respectively.
The produced water is cleaned, and then disposed of.

1.
Petroleum Open Learning

In order to recombine the gas from each stage for


further use, the pressure of the gas leaving the 2nd
stage of separation must be increased to the pressure
of the 1st stage. A compressor is therefore required to
do this.

Figure 3 develops the gas flow diagram to include


this compressor.

1.
Petroleum Open Learning

After compression from 2nd to 1st stage separator pressure, all the gas is now at a pressure of 17 bar. However,
in our example, the gas will require drying {dehydrating} and will have some of its constituents liquefied in a gas
liquids recovery plant. This requires the gas to be at an even higher pressure. A further stage of compression is
therefore required at this point. In our hypothetical process plant the pressure is raised from 17 bar to 68 bar.

Figure 4 shows this further stage of compression and shows, in outline, the gas liquids
recovery system.

1.
Petroleum Open Learning

The residual gas, after dehydration and gas liquids


recovery, will be used for three things:

• gas lift

• gas export

• gas re-injection into the reservoir

In order to export the gas from the offshore location to


a terminal onshore a much higher pressure is now
required at the platform. Similarly, extra pressure is
needed to inject the gas into the well for gas lift. At this
point in our example, therefore, the pressure is raised
to 170 bar by further compression.

Figure 5 shows this.

1.
Petroleum Open Learning

Even at this high pressure, we would not be able to


re-inject the gas into the reservoir. The actual pressure
required to do this would depend on a number of
things such as :

• reservoir depth

• reservoir pressure

• type of reservoir rock

In our example, we need a pressure


of, say, 408 bar. So, yet another
stage of compression is required.

Figure 6 completes our simple


diagram of gas flow on a platform.

1.
Petroleum Open Learning

We have just been considering the gas flow on an We have just been looking at the compression of In all cases, the pressure required may be in the
oil production platform. Of course, the gas which is natural gas from the reservoir. However, there are a region of, say, 10 bar.
produced from the reservoir may not be associated number of other applications of compression which you
with oil. We may be looking at a gasfield. could come across in petroleum producing operations. A typical air compression package might consist of
Let’s consider some of these: the following items:
In the early days of production from a typical gas
field, the pressure of the gas at the surface will be Compressed Air Systems • air prefilter
sufficient to transport it to shore. As the life of the
field progresses, however, the natural pressure of Compressed air is required on a platform to supply • air compressor
the reservoir declines. A point is reached where this the following :
pressure is no longer sufficient to transport the gas to • wet air receiver
shore. • instrument air
• air drier
When this happens, it is necessary to install gas • works air
compression plant on the platform. • instrument air receiver
• inert gas generation
A schematic layout of such a system is shown in
The instrument air system requires a plentiful supply Figure 7.
of clean, dry air at an adequate pressure.

A works air system is required to drive tools and


equipment around the platform. It may not be quite
so demanding in terms of air dryness, but a plentiful
supply must be maintained.

Inert gas is required on the platform for purging


equipment and providing a blanket of non-flammable
gas in certain vessels. The gas used is nitrogen, which
is manufactured from atmospheric air on the platform.
The inert gas generator requires a supply of air at a
suitable pressure.

1.10
Petroleum Open Learning

1.11
Petroleum Open Learning

Refrigeration Systems
The gas liquids recovery system, which I referred
to earlier, requires gas to be chilled to a fairly low
temperature, i.e. -30°C or less. One way of doing this
uses a refrigeration system.

A refrigeration system which employs a compressor


is the vapour compression cycle process. In this
process, the refrigerant in the form of a vapour is
first compressed, then condensed to a liquid. This is
followed by expansion over a valve, and evaporation,
to achieve the necessary cooling effect. The process
then starts again.

It is not my intention to look at the process in any detail


in this unit. However, Figure 8 illustrates it in outline.

1.12
Petroleum Open Learning

Portable Air Compressors


It may be necessary, particularly during the
construction of a platform, to have access to
Test Yourself 1.1
compressed air in locations where there is no supply.
In this case portable compressors may be employed.
The following statements are either true or false. Indicate with a tick in the box provided which statements
belong in which category. If the statement is false, correct it.
HVAC Blowers
On an offshore platform, the accommodation will TRUE FALSE
require a heating, ventilation and air conditioning
(HVAC) system. This will necessitate the use of low
pressure/high volume compressors, which are often 1. Oil and gas are separated from each other in separators which
referred to as blowers. always operate at high pressures.

2. Some of the components of a gas stream leaving a separator can


be liquified.

3. Excess gas on a platform is disposed of by flaring.

4. The pressure required to inject gas into a reservoir depends on the


amount of gas liquids which have been recovered.

5. Compressed air is only required on a platform for inflating tyres on a


helicopter.

You will find the answers in Check Yourself 1.1 on page 1.32

1.13
Petroleum Open Learning

Summary of Section 1

In Section 1, I have tried to show you some applications of gas compression in petroleum processing.

In the section, you saw that gas which is separated from oil is usually at too Iow a pressure to be transported or
to do useful work. Compression facilities are required to raise the pressure of the gas for:

• gas liquids recovery

• gas lift

• gas export

• gas injection

You also saw that compressors are required for instrument and works air, and for refrigeration purposes.

I’m sure that you could think of a few more applications of compressors, but the ones I have just described are
the major ones in petroleum production operations.

You now have an overall impression of the way in which compressors are used. In Section 2 we will move on
to look at some of the basic scientific principles which affect the way in which compressors work.

1.14
Petroleum Gas Compression - Unit 1 - An Overview Petroleum Open Learning

Section 2 - Basic Principles of Compression


You saw in Section 1 of this Unit how important the Pressure - Volume Relationship • the pressure in this relationship has to be the
gas compression facilities are in a petroleum absolute pressure or the pressure above
producing operation. In this section we are going to The basic law which relates pressure and volume in absolute zero pressure. Absolute pressure
look at some of the basic principles of compression a gas is known as Boyle’s law. This law states means the pressure read on a gauge, plus the
- some of the rules which govern how gases behave that: pressure of the atmosphere :
when they are subjected to changes in pressure,
temperature and volume. At a constant temperature, the volume of a
given mass of gas is inversely proportional to absolute pressure = gauge pressure +
We will also look briefly at the application of energy its absolute pressure atmospheric pressure
conservation principles to gas compression.
This may be written as an equation, as follows:
Finally, I will introduce some of the terms and For example, when using bar as the unit of
expressions which are commonly used in P1V1 = P2V2 pressure:
compressor technology.
In this equation:
Let’s start by having a look at the relationship bara = barg + atmospheric pressure
between the pressure and volume of a gas. P1 is the initial pressure in absolute units

You should be aware that the following relationships V1 is the initial volume For our purposes we will take the atmospheric
apply to what is often referred to as an ideal gas. pressure to be 1 bar, although the exact figure
P2 is the final pressure in absolute units is 1.01325 bar.
In the real world, however, gases are not ideal and
their behaviour departs from the ideal situation. V2 is the final volume The inverse relationship in Boyle’s Law means that
if the volume of the gas is reduced, the pressure
The magnitude of this deviation depends upon the increases. Similarly, if the volume is increased the
nature of the gas, and the actual pressures and Two things should be noted here: pressure is reduced.
temperatures involved. More advanced calculations
can account for these deviations - for example, by • the process has to be at a constant This relationship can be shown in a simple
the use of compressibility factors. temperature. Any process which takes place illustration.
at a constant temperature is known as an
However. for our purposes, in this Compressor isothermal process Figure 9 overleaf shows this
Programme we are assuming ideal conditions.

1.15
Petroleum Open Learning

We can use the Boyle’s Law equation to determine


the change in volume of a gas with changes in
pressure and vice versa. I will show you a worked
example first. Then try Test Yourself 1.2 out for
yourself :

1.16
Petroleum Open Learning

EXAMPLE 1
BOYLE’S LAW CALCULATION

If 100 litres of gas is compressed from 10 barg to 15 barg, what will be it’s volume Test Yourself 1.2
after this compression at constant temperature.

First remember to convert the pressures to absolute units,


Using Boyle’s Law

P1 = 10 + 1 = 11 bara
P2 = 15 + 1 = 16 bara a) 300 litres of gas at 2 barg is
V1 = 100 litres compressed to 5 barg at constant
V2 = The volume after compression temperature. What will be its new
volume?
Boyle’s Law states

P1 V1 = P2 V2 b) The volume of a mass of gas at


20 barg is reduced from 1800 litres
We must rearrange this equation to 600 litres at constant temperature.
What will be its final pressure?
ie

V2 = P1V1
P2
You will find the answers in
V2 = 11 x 100 = 68.75 litres Check Yourself 1.2 on page 1.33
16

Now let’s look at the relationship between


temperature and volume.

1.17
Petroleum Open Learning

Temperature - Volume Relationship


Note that this process is at constant pressure,
The basic law in this relationship is Charles’s Law. Such a process is called an isobaric process.
This states that : Test Yourself 1.3 The Charles’s Law relationship means that, if
In a constant pressure process, the volume of a the temperature of a fixed mass of gas is
given mass of gas is directly proportional to its increased at a constant pressure, its volume
absolute temperature. State the following temperatures in absolute
units. will also increase.
Once again, we are dealing in absolute values - this Again a simple drawing, Figure 10, on the
time absolute temperature. next page, shows this.
1. 40ºC
• The absolute temperature is the temperature above
absolute zero.
• Absolute zero is the lowest point on the absolute 2. 100ºC
temperature scale, which is measured in units
called Kelvin (K) in the S1 system. In Imperial
system the units are known as Rankine (oR) 3. 0ºC
• On the Celsius temperature scale, absolute zero
(OoK) is at -273.15°C. (A convention, which is by
no means universally applied, is to omit the term
degree when using absolute temperature units).
• On the Fahrenheit scale, absolute zero (OoR) is at
-459.67”F

In most practical situations sufficient accuracy is You will find the answers in Check
achieved by using 273 as the conversion factor Yourself 1.3 on page 1.33
between Celsius and Kelvin, and 460 between
Fahrenheit and Rankine. So, to quote an absolute
temperature in Kelvin, when we are working with
Celsius units, we add 273 to the temperature in
Celsius.

i.e. 100°C = ( 100 + 273 )


= 373 Kelvin ( absolute)

Check that you understand this by doing the following


Test Yourself
1.18
Petroleum Open Learning

As before, the relationship can be written as an


equation for calculation purposes:

V1 V2
T1 = T2

This equation can be used to calculate the change


in volume with changes in temperature and vice
versa. Again, a worked example is given first. Test
Yourself 1.4 then gives you further practice at using
Charles’s law.

1.19
Petroleum Open Learning

EXAMPLE 2

CHARLES’S LAW CALCULATION


Test Yourself 1.4
If 100 litres of gas is heated from 40°C to 80°C at constant pressure, what
will be its final volume?

Again, remember to use absolute units.



V1 = 100 litres a) The temperature of 50 litres of gas is
T1 = 40 +273 = 313 K raised from 16°C to 38°C at constant
T2 = 80 +273 = 353 K pressure. What will be the volume of the
V2 = Final volume. gas at 38°C?

Charles’s law states b) 800 litres of gas at a temperature of


55°C is reduced ISOBARICALLY to a
V1 V2 volume of 700 litres. What will be the
T1 = T2 final temperature?

rearrange the equation

V2 = V1T2
T1

V2 = 100 x 353 = 113 litres


313

You will find the answers in


Check Yourself 1.4 on page 1.34

1.20
Petroleum Open Learning

The Combined Gas Law It reminds me of a bicycle pump. The piston within Energy
the cylinder is reducing the volume of gas. You will
The two laws of Boyle and Charles can be combined remember from Boyle’s Law that this reduction in Although the word energy is very commonly used, it
to allow us to relate all three variables of pressure, volume will increase the pressure. is difficult to define precisely. We could say that a
volume and temperature. person has energy if he or she has the capacity to
If, in Figure 9, we : do things or influence events. In science, a system
This combined gas law may be written as: has energy if the objects in the system can do things
• remove the compressed gas from the cylinder and possibly affect other objects.
when the piston is at the bottom of its stroke
P1V1 P2V2 Consider for a moment the following example.
T1 = T2 • refill with low pressure gas as the piston
moves upwards again Supposing you were standing next to a cricket ball
which is resting on the ground. The ball isn’t doing
You will remember that the relationships we have we have a simple reciprocating compressor. much or affecting anything else. If you picked it up
just been looking at apply to what is called a perfect then threw it at the nearest window, however, it
gas. In reality, however, gases are not perfect, but This is, in fact, what a bicycle pump is. We will look would certainly be doing something. It would also
real. They do not behave exactly as you would at the construction and operation of reciprocating be affecting the window. If you picked it up then
expect from these (perfect) gas laws. For practical compressors in Unit 2 of this series on compressors. dropped it on your toe, it would affect your toe.
purposes, however, we can use these gas laws to
perform basic calculations. The point of this, is that you would have given the
ball energy by lifting it, and then throwing it.
We have just seen how pressure and temperature Now let’s look at another scientific concept, namely, These are two forms of energy.
alter as we reduce the volume of a fixed mass of energy.
gas. But what does this mean as far as Lifting the ball gives it potential energy. From its
compression is concerned? position in your hand, it was then able to fall and
affect your toe.
The fact that the pressure will increase as the
volume of a gas is reduced may suggest to us how Throwing the ball gave it energy of motion. This is
we could make a simple compressor. Look back to called kinetic energy.
Figure 9 on page 1.16. What does the series of
simple drawings remind you of ?
There are many other forms of energy and I have
listed some in the following table.

1.21
Petroleum Open Learning

FORM OF ENERGY EXAMPLES


Potential a weight lifted above the ground
Test Yourself 1.5
Kinetic any moving object
What forms of energy are indicated by the
following:
Chemical gas, oil, coal, etc. They can be burnt to provide
heat energy
1. A separator operating at 40 bar and 80°C.
Magnetic available when two magnets repel or attract
each other
2. A pallet suspended from a crane hook.
Electric available from an electric socket
3. A bottle of propane gas.
Heat a tank of hot water has more energy than one
4. An object being pumped along a pipeline.
full of cold water
5. A boiler full of steam.
Pressure a vessel at high pressure has more energy than
one at a low pressure

You will find the answers in


Check Yourself 1.5 on page 1.35

TABLE 1. SOME FORMS OF ENERGY

The list above is not exhaustive and there are other forms of energy.

1.22
Petroleum Open Learning

This is a fundamental principle of science - the Compression Ratio


principle of conservation of energy.
A compressor must have the capability to take in
This principle can be applied in the design of gas at a certain pressure and deliver it at a higher
equipment which is used to change one form of energy
into another.
pressure. The relationship between a compressor’s
intake and delivery pressure is known as its
Test Yourself 1.6
compression ratio.
Imagine having a machine which can rapidly speed
up a mass of gas. Energy would be added to the gas For instance, supposing a compressor takes in gas
in the form of kinetic energy. This energy would have at a pressure of 20 bara, and delivers it at 60 bara. a) A compressor takes in gas at 20 bara
come from the conversion of some of the energy being The ratio between delivery and intake pressure is and delivers it at 70 bara. What is its
used to drive the machine. Now, if the gas is rapidly 60 / 20 = 3. compression ratio?
slowed down, its kinetic energy is reduced. But the
total energy in the gas must remain the same, so the In this case, the compression ratio of the machine b) The same machine running under the
kinetic energy must be converted into some other is 3. Note how, once again, the units of pressure same conditions takes in gas at
type of energy. In a compressor the conversion is to are absolute units, i.e., bara. 40 bara. What will be the delivery
pressure energy and some heat energy. pressure?
If an identical compressor takes in gas at 30 bara
This is the principle of operation of a centrifugal and delivers it at 90 bara its compression ratio is
compressor, which is the subject of Unit 3 of this also 3, i.e. 90/30.
compressor series.
This ratio can be used to make comparisons
between compressors of different types.

Make sure you understand this concept by


Let’s move on now to look at some terms and trying the following Test Yourself.
expressions which you need to become familiar with You will find the answers in
during your investigation of gas compression. Check Yourself 1.6 on page 1.35

1.23
Petroleum Open Learning

Compressor Capacity A Compressor can be described as having an FAD


of 30m3 per hour at 7 bara. This means that it takes
When considering the performance of different in 30m3 of air per hour at atmospheric conditions
compressors, it may be useful to compare the and delivers it at 7 bara.
amount of gas compressed in a certain time for
each. Mass Flow Rate
The volume of gas passing through the machine in a A compressor requires some form of driving motor
given period of time is called the compressor or engine. It is important to know the power
capacity. requirements of such a machine for a given
compression unit. This power depends on the mass
The gas, however, is being compressed as it flows of gas compressed per unit time rather than the
through, and its volume is being reduced. We must capacity of the compressor. So another term you
be careful, therefore, Where we measure the volume may come across is that of mass flow rate. It is
flowing through, or capacity. quoted in units of pounds per second or kilograms
per second.
The compressor capacity is defined as the
volume of gas compressed and delivered per
unit time, expressed in terms of inlet conditions
of temperature and pressure.

The most common type of compressor is an air


compressor which takes in air from the atmosphere.
Because of this, gas compressor capacity is often
quoted in terms of air at conditions of atmospheric
pressure and 15˚C.

You will often see compressor capacity referred to


as the Free Air Delivered (F.A.D.)

1.24
Petroleum Open Learning

Summary of Section 2

In this section we have had a brief look at some basic principles of compression.

First we considered the basic gas laws which relate pressure, volume and temperature.
You saw that the relationship between pressure and volume at constant temperature
can be expressed by the equation:

P1V1 = P2V2 We next considered the principle of energy • compressor capacity - the volume of
conservation. Here you saw that there are many gas compressed per unit time, expressed
forms of energy including kinetic ( the energy of in terms of inlet conditions of temperature
This is known as Boyle’s Law. motion) and pressure energy. You saw that energy and pressure
cannot be destroyed, but only converted into another
type. I pointed out that gas can be compressed by • mass flowrate - the actual mass of gas
Charles’s Law relates volume and temperature increasing its kinetic energy, then converting this compressed per unit time
when the pressure is constant: energy into pressure energy.

Finally we noted some compression terms.


V1 V2 These were:
T1 = T2
• compression ratio - the ratio of the
discharge to the inlet pressure of a
You also saw that these equations can be compressor
merged into the Combined Gas Law.

In the next section we will go on to look at some


different types of compressor in common use

1.25
Petroleum Gas Compression - Unit 1 - An Overview Petroleum Open Learning

Section 3 - Types of Compressors


Up to now in this Overview Unit you have seen that
there is a considerable requirement for compression
plant in a petroleum producing operation. You have
also looked at the basic principles of gas
compression. But what kind of machine is used to
achieve the required increase in gas pressure? In
this section we are going to look at the various types
of compressor which are available, and for what
application they are most suited.

The Compressor Family Tree


Compressors can be classified into a number of
categories according to the way they work. This
compressor family tree is shown in Figure 11.

1.26
Petroleum Open Learning

You can see from Figure 11 that all compressors fall Positive displacement compressors will develop In the more complex types of compressor, the
into one of two main groups: sufficient pressure to overcome any resistance to chamber is equipped with valves on the inlet and
flow and the operational limits are essentially outlet to control the flow of the gas being compressed.
• Positive Displacement/Intermittent Flow determined by the driver power and the strength of
Compressors (these are commonly known as the compressor parts. The operation of these valves is linked to :
Positive Displacement Compressors and we
will use this term throughout the remainder of From the family tree we can see that positive • the motion of the piston or diaphragm
this Unit) displacement compressors fall into two types. They
are: • the rise and fall of the pressure in the chamber
• Continuous Flow Compressors
• Reciprocating Compressors Rotary Compressors
We will start our look into the compressor family tree
by talking about positive displacement compressors. • Rotary Compressors Rotary compressors have a variety of uses in the oil
and gas industry.
Positive Displacement Reciprocating Compressors
Compressors In this type, the displacement of the fluid is produced
Reciprocating compressors play a very important by the rotation of one or more elements within a
A positive displacement compressor works on the role in the oil and gas industry and for this reason stationary housing.
principle of pushing a gas from a vessel by partially, Unit 2 of the compressor series is dedicated to
or completely displacing its internal volume. them specifically. The most common types of rotary compressor found
in the oil and gas industry are the:
This is usually achieved by mechanical means or, Reciprocating compressors come in all shapes and
less frequently, by a second fluid. sizes and fall into two types: • screw compressor

Because the vessel is alternately emptied and • piston type • lobe compressor
refilled the flow is intermittent. The intermittent
flow into and out of the compressor causes the • diaphragm type • sliding vane compressor
pressure to pulsate on both the inlet (suction) and
outlet (discharge) sides. The action of the fluid-transferring parts is the same • liquid ring compressor
in each. A piston or diaphragm is made to pass, or
flex, back and forth in a chamber. These compressors will be explained further in Unit 4
of this series.

1.27
Petroleum Open Learning

Continuous Flow Compressors There may be as few as one impeller, or as many as Fluidic Compressors
twenty or more impellers, on a shaft. The shaft may
From the compressor family tree we can also see be rotated at speeds which exceed 30 000 rpm. Fluidic compressors, including the Ejector and
that the second group of compressors are the Diffusion Pump types, will be covered in Unit 4.
Continuous Flow Compressors. When the gas leaves each impeller it is allowed to
slow down. As this happens, kinetic energy is
In these compressors the movement imparted to the replaced by pressure energy. You will remember
gas is continuous and constant. Continuous flow this from Section 2 of this unit. Before moving on, have a go at the following Test
compressors fall into two types, which are: Yourself question.
Dynamic compressors are classified according to
• Dynamic Compressors the manner in which the gas flows through the
compressor. Within this category are:
• Fluidic Compressors
• Centrifugal Compressors - where, in each
stage, the gas flows radially outwards
Dynamic Compressors
• Axial Flow Compressors - here, the gas flows
Dynamic compressors have a system of elements along the line of the shaft
(called impellers) which are arranged on a shaft.
The impellers rotate with the shaft and impart • Mixed Flow Compressors - a combination of
energy to the gas by increasing its velocity. centrifugal and axial types
The amount of energy which is imparted to the gas
by a dynamic compressor is mainly determined by : Centrifugal Flow compressors (commonly referred to
as centrifugal compressors) are dealt with
• the design of the impellers comprehensively in Unit 3 of the Compressor Series
• the number of impellers used Axial and Mixed Flow compressors will be covered
in Unit 4.
• the speed at which the impellers rotate

• the density of the gas which is being


compressed

1.28
Petroleum Open Learning

Compressor Selection
As you have just seen, we can choose from a

Test Yourself 1.7


variety of compressors in order to perform the task
of raising the pressure of a gas. The choice of
compressor for a particular application will often be
Indicate whether the following compressors are positive displacement machines or based on the two factors which we looked at in
continuous flow machines. Section 2 :

• compression ratio
Positive displacement Continuous flow
• capacity

1. Double acting reciprocating However, many other factors may influence this
compressor choice. Some of these I have listed below:

• nature of the gas - hot or corrosive gases


2. Mixed flow compressor may restrict the choice because of the
requirement for special sealing or lubricating
systems, or special materials used in
3. Axial compressor construction

• reliability - for continuous running applications


4. Screw type compressor
• costs - not only the initial capital costs, but
service and maintenance costs may have to
5. Sliding vane compressor be considered

• power availability - the power available to


6. Centrifugal compressor drive the compressor could influence the
choice of machine

You will find the answers in Check Yourself 1.7 on page 1.36

1.29
Petroleum Open Learning

Of course, compression ratio and capacity are of


critical importance when choosing a machine.
Figure 12 shows typical pressure and capacity ranges
over which various types of compressor usually
operate. You should note that the Figure shows very
approximate ranges and some compressors may be
capable of operating outside the ranges indicated.

1.30
Petroleum Open Learning

Now use Figure 12 to answer the following Test


Yourself question.

Summary of Section 3

In section 3 we looked at the compressor family tree and at the different groups and types of compressor.
Test Yourself 1.8 We then discussed the different compressors which may be found on an oil production facility. The different
characteristics which placed them within certain groups and types were examined. We then looked at each
type of compressor found within each group.
From Figure 12 decide what type of compressor Finally, we considered, briefly, compressor selection.
would be suitable for the following applications.
You have now completed this Overview Unit in the Petroleum Gas Compression
1. Delivering 85 m3 / hour at 350 bar Series. Unit 2 of the Series will examine Reciprocating Compressors in detail.
2. Delivering 15,000 m3 / hour at 70 bar

You will find the answers in


Check Yourself 1.8 on page 1.36

1.31
Petroleum Open Learning

Check Yourself - Answers

Check Yourself 1.1


1. FALSE Separators are operated at various pressures depending upon
- reservoir pressure
- ratios of oil and gas produced
- pressure requirements on the platform
- number of stages of separation.

2. TRUE

3. TRUE

4. FALSE The pressure required depends on - reservoir depth - reservoir pressure - type of reservoir rock.

5. FALSE Compressed air is required on a platform for



- instrument air
- works air
- inert gas generation

1.32
Petroleum Open Learning

Check Yourself 1.2 Check Yourself 1.3

1. 313 Kelvin
Using Boyle’s Law
2. 373 Kelvin
P1V1 = P2V2
3. 273 Kelvin
and remembering to work in absolute units.

a) P1 = 2+1 = 3 bara b) P1 = 20 + 1 = 21 bara


V1 = 300 litres V1 = 1800 litres
P2 = 5+1 = 6 bara V2 = 600 litres
V2 = ? P2 = ?

rearrange the equation rearrange the equation

V2 = P1V1 P2 = P1V1
P2 V2

V2 = 3 x 300 P2 = 21 x 1800
6 600

= 150 litres = 63 bara

= 63 - 1

= 62 barg

1.33
Petroleum Open Learning

Check Yourself 1.4

a) Using Charles’s Law b) Using Charles’s Law

V1 V2 V1 V2
T1 = T2 T1 = T2

and remembering to use absolute temperature and remembering to use absolute temperature values.
values.

V1 = 800 litres
V1 = 50 litres T1 = 55 + 273 = 328
T1 = 16 + 273 = 289 V2 = 700 litres
T2 = 38 + 273 = 311 T2 = ?
V2 = ?
rearrange the equation
rearrange the equation
T2 = V2T1
V2 = V1T2 V1
T1
T2 = 700 x 328
V2 = 50 x 311 800
289
= 287K
= 53.8 litres
= 287 - 273

= 14oC

1.34
Petroleum Open Learning

Check Yourself 1.5 Check Yourself 1.6

1. Pressure and heat energy. a) Compression Ratio = delivery pressure


intake pressure
2. Potential energy.
= 70 bara = 3.5
3. Pressure and chemical energy. 20 bara

4. Kinetic energy

5. Pressure and heat energy. b) Delivery Pressure = intake pressure x compression ratio

= 40 x 3.5

= 140 bara

1.35
Petroleum Open Learning

Check Yourself 1.7 Check Yourself 1.8


1. Positive displacement. 1. Reciprocating compressor.

2. Continuous flow. 2. Centrifugal compressor.

3. Continuous flow.

4. Positive displacement.

5. Positive displacement.

6. Continuous flow.

1.36
POL
Petroleum Open Learning

Petroleum Gas
Compression
Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY

2
Petroleum Open Learning
Petroleum Gas Compression - Unit 2 - Reciprocating Compressors
(Part of the Petroleum Processing Technology Series)

Contents Page

Training Targets 2.2 Visual Cues


Introduction 2.3 training targets for you to achieve
by the end of the unit

Section 1 – Basic Theory 2.4


Operating Principles
Capacity and Compression Ratio
Compressor Performance test yourself questions to see how
much you understand
Section 2 – Design and Construction 2.17
Cylindars
Pistons and Piston rings
Compressor Valves
Stuffing Box
check yourself answers to let you
Crankshaft, Connecting Rod, Crosshead and Piston Rod
see if you have been thinking along
the right lines
Section 3 – Auxiliary Systems 2.29
Cooling System
Lubrication System
Suction and Discharge Piping System
Drive Coupling activities for you to apply your new
knowledge

Section 4 – Operation of Reciprocating Compressors 2.37


A Typical Gas Compression System
Alarm and Shutdown Systems summaries for you to recap on the
The Main Operational Checks on a Reciprocating Compressor major steps in your progress

Check Yourself – Answers 2.44


2.
Petroleum Gas Compression - Unit 2 - Reciprocating Compressors Petroleum Open Learning

Introduction

Training Targets

When you have completed Unit 2 of the Petroleum Gas Compression series you will be able to :

• Explain the basic operating principles of a reciprocating compressor.

• Describe the construction of a reciprocating compressor.

• Explain the function and operation of the principal components of a reciprocating


compressor.

• Describe the layout and operation of the auxiliary systems associated with a reciprocating
compressor.

• Explain a basic reciprocating compressor alarm and shutdown system.

• List the common operating checks carried out on a reciprocating compressor.

2.
Petroleum Gas Compression - Unit 2 - Reciprocating Compressors Petroleum Open Learning

Introduction
As I previously explained, the compressor requirements of an oil or gas production system are dependent upon
many variables.

Each system will have its own characteristics and show In this Unit, we will be looking at the construction and
detailed design differences. operation of typical reciprocating compressors.

You saw in Unit 1, that compressors can be classified The Unit is divided into four sections:
as either continuous flow or positive
displacement machines. The reciprocating Section 1 covers the basic operating theory of
compressor is the most common of the positive reciprocating compressors.
displacement type. This is the one we are going to
concentrate on in this unit. In Section 2, we will look at the design and
construction of a typical machine.

Section 3 will describe a range of auxiliary equipment


Reciprocating compressors are designed to operate
over a wide range of capacities and pressures. Small
portable machines may be adequate for the delivery and
of small volumes, at pressures of, say, 1.5 bar. Large
industrial units may be required to deliver
several thousand cubic metres per hour, at pressures in Section 4, we will be looking at basic compressor
approaching 1000 bar. operations.

2.
Petroleum Gas Compression - Unit 2 - Reciprocating Compressors Petroleum Open Learning

Section 1 - Basic Theory


Operating Principles When the air pressure in the cylinder is greater than
the pressure in the bicycle tyre, the air flows into
All positive displacement compressors operate by : the tyre. A small non-return valve on the bicycle tyre
prevents the air from flowing out of the tyre back into
• creating a low pressure space into which gas the cylinder. The whole cycle may now be repeated.
may flow
All reciprocating compressors work in a similar way to
• closing the entrance to this space the bicycle pump.

• displacing the enclosed gas with a mechanical


device (thereby increasing the pressure)

• opening the exit to the space, allowing the Figure 1 on the next page shows the main
compressed gas to leave components of a reciprocating compressor. Take a look
at the Figure and try to become familiar with the names
The simplest form of reciprocating compressor in of the various parts.
common use is the Bicycle Pump. In this type of
compressor, a small washer, the piston, is pushed
back and forth inside a tube which is called the
cylinder. As the piston moves backwards it creates a
low pressure space inside the cylinder. The washer is
then distorted and allows outside air to
flow past it into the cylinder.

When the piston reaches its furthest point of backward


travel, the washer again flexes, and seals
the gap between the piston and cylinder.

Now, when the piston moves forward, the volume in


the cylinder is reduced and the air is compressed.

2.
Petroleum Open Learning

2.
Petroleum Open Learning

You can see from Figure 1 that the flow of gas through
the compressor is controlled by valves. These act as
non-return valves to permit flow in one direction only.

These valves are often called ‘check valves’. They are


positioned in the inlet (suction) and outlet
(discharge) of the compressor. Gas enters the
cylinder through the suction valve and leaves through
the discharge valve,

• Suction valves open when the cylinder


pressure is lower than the pressure of the gas
to be compressed

• Discharge valves open when cylinder pressure


is higher than the pressure of the system into
which the gas is to be discharged

Reciprocating compressors are classified as :

Single Acting

or

Double Acting

Let’s look at the way in which each of these work. First


the single acting compressor. Figure 2 shows the
flow of gas through this type of machine.

2.
Petroleum Open Learning

The section of the cylinder nearest the crank is called


the crank end. The section of the cylinder furthest
away from the crank is called the head end. Only
the space at the head end of the cylinder is used for
compression.

In the single acting compressor, the back stroke is the


suction, or intake stroke.
The forward stroke is the compression or discharge
stroke.

In a double acting compressor there is a suction


stroke and a discharge stroke each time the piston
moves either backwards or forwards.

This is illustrated in Figure 3.

2.
Petroleum Open Learning

The back stroke is the compression stroke at the crank


end of the cylinder and the suction stroke at the head
end of the cylinder.

As can be seen, these roles are reversed during the Test Yourself 2.1
forward stroke.

Now, why not try the following Test Yourself:


Is a bicycle pump a single or double acting
compressor?

You will find the answer to Test Yourself 2.1


on Page 2.44

2.
Petroleum Open Learning

Figure 4 is a more detailed drawing of a double acting reciprocating compressor.


Study the drawing for a while and identify its components.

2.
Petroleum Open Learning

Capacity and Compression Ratio


Compressors are used to increase the pressure of

Activity
gases and transfer these compressed gases to a
higher pressure system.

The volume of gas moved by the compressor in a Take a bicycle pump and, without connecting it to a When pumping before connection to the tyre, you
given period of time is called its capacity. But, as the bicycle tyre, pump it ten times. will notice no temperature increase. This is because
gas is being compressed during transfer, its volume is you are displacing air into the atmosphere without
reducing, and we need to be careful at what point we increasing its pressure.
measure this capacity. Now put your hand on it and test the temperature.
What do you notice?
Capacity is measured as the volume of gas entering After connecting to the tyre, however, you should
the compressor in a given time period. have noticed a sharp increase in temperature as the
After you have done this, connect the pump to a pressure in the tyre increases.
The amount by which a compressor increases the bicycle tyre and pump it another ten times.
pressure of the gas is called the compression ratio.
it is defined as :
What do you notice about the temperature this time?

Discharge Pressure
Suction Pressure Repeat this a few times while the pump is still
connected to the tyre. After every ten strokes,
check the temperature of the pump by feeling it with
For example, if the suction pressure is 10 bara and the your hands.
discharge pressure is 40 bara, the compression ratio is
40/10 or 4. It is usually expressed as 4 : 1 or 4 to 1.

2.10
Petroleum Open Learning

The activity you have just completed demonstrated


a basic fact regarding compressors. When a gas is
compressed, its temperature increases. The higher
the compression ratio, the larger the
temperature increase. Test Yourself 2.2
This temperature increase has a detrimental effect
on both the efficiency of the compressor and its
mechanical reliability. Because of these considerations, Diesel engines are classed as ‘compression-
the temperature rise is restricted to ignition’ engines - in other words, the heat
certain limits (typically about 200°C, although higher generated by compression of the fuel/air mixture
temperatures may be experienced). also ignites this mixture.

One of the basic means of limiting the temperature rise In a particular diesel engine, the compression ratio
is to limit the compression ratio to about 6 to 1. is 20:1. Air is taken in from the atmosphere at a
pressure of 1 bara.
If the required final discharge pressure cannot be met
by this compression ratio, then compression is carried What will be the pressure of the air/fuel mixture in
out in a number of stages. Machines capable of doing the engine cylinder when maximum compression is
this are called multi-stage compressors. They have reached?
coolers to reduce the temperature of the gas between
each stage.

I suggest you have a go at the following Test Yourself,


to underline the points covered above.

You will find the answer to Test Yourself 2.2 on


Page 2.44

2.11
Petroleum Open Learning

Compressor Performance

The performance of a reciprocating compressor can


be represented by a pressure/volume (PV) diagram.
One of these is shown in Figure 5, which illustrates
the relationship between the cylinder pressure of a
compressor and the cylinder volume enclosed by the
piston, for a single-acting compressor.

2.12
Petroleum Open Learning

Horizontal distance in the PV diagram represents a When the cylinder pressure drops slightly below the
change in volume produced by the movement of the suction pressure (at point B), the suction valve opens.
piston in the cylinder. Vertical distance on the diagram Curve AB on the PV diagram represents the pressure
represents a change of pressure in the cylinder caused fall and volume increase as the piston begins to move
by the movement of the piston. back on the suction stroke.

As the piston moves back and forth in the cylinder, The opening of the suction valve is represented on the
the volume and pressure in the cylinder changes. We PV diagram by point B.
will now use the PV diagram to follow these changes.
The diagram shows a complete compression cycle, As the piston moves further back in the cylinder, gas
consisting of one backward stroke and one forward flows in through the suction valve. This is represented
stroke. on the PV diagram by the line from B to C. The end
of the suction stroke is represented by point C. At this
Point A represents the end of the compression stroke point, the piston reverses its direction and begins the
and we shall use this as our starting point. compression stroke.
(The piston is designed so that it cannot touch the end
of the cylinder. The small space which is left between As soon as the piston begins to move in the opposite
the piston and the end of the cylinder is called the direction the gas begins to be compressed. Cylinder
clearance space. At the end of every stroke there is a pressure rises above suction line pressure and the
small amount of gas left in the clearance space.) suction valve closes.

As the piston begins to move back in the cylinder, on As the piston continues to move forward in the cylinder
the suction stroke, the gas remaining in the the gas pressure increases and, at point D, the gas is
cylinder expands. compressed to a level slightly higher than the pressure
of the gas in the discharge system. At this point the
As the gas expands, the pressure in the cylinder discharge valve opens.
decreases.
For the rest of the stroke, D to A, gas is forced out
through the discharge valve and into the high pressure
discharge system.

2.13
Petroleum Open Learning

Test Yourself 2.3

The following statements are in the wrong order.


Place them in their correct sequence, starting with No.1:

1. Piston begins suction stroke. .... 1 .... 8. Gas remaining in cylinder expands and discharge valve closes. ............

2. Gas flows from cylinder into discharge line. ............ 9. Suction valve closes. ............

3. Piston reverses direction at end of suction stroke. ............ 10. Gas in cylinder is compressed to above suction line pressure. ............

4. Cylinder pressure rises above discharge line pressure. ............ 11. Discharge valve opens. ............

5. Cylinder pressure falls below suction line pressure. ............ 12. Suction valve opens. ............

6. Gas flows into cylinder from suction line. ............

7. Piston reverses direction at end of discharge stroke. ............

The answers to Test Yourself 2.3 will be found on Page 2.44

2.14
Petroleum Open Learning

Summary of Section 1
In this first Section of Unit 2, we have looked at the basic theory of operation for positive displacement
compressors.

We started by looking at how a reciprocating Using diagrams, you looked at the flow of gas through
compressor operates, and compared this operation to single acting and double acting
that of a bicycle pump. compressors.

The main components of a reciprocating compressor


were described, and you identified these on a simple
line diagram.
Next, we went on to consider compressor capacity
and compression ratio. We saw how the flow of gas
through the compressor could be represented by a
You saw that reciprocating compressors can be pressure/volume diagram.
classified as either :

• single acting,

or We will now go on to Section 2, which examines the


design and construction of reciprocating compressors.
• double acting But first, by way of a little revision, try Test Yourself
2.4.

2.15
Petroleum Open Learning

Test Yourself 2.4

Indicate whether the following statements apply to a single-acting, or double-acting reciprocating compressor, or both.

Single- Double- Single- Double-


Acting Acting Acting Acting

1. Only the space at the head end of the cylinder is 6. The suction valve opens every second stroke.
used for compression.

2. There is a suction and a discharge stroke each 7. There is a suction and a discharge valve at each
time the piston travels the length of the cylinder. end of the cylinder.

3. A suction valve is open during each suction stroke. 8. The forward stroke is the compression,
or discharge stroke

4. During each stroke (forward and backward) a


discharge valve is open.

5. The back stroke is the suction, or intake stroke.

You will find the answers to Test Yourself 2.4 on Page 2.45

2.16
Petroleum Gas Compression - Unit 2 - Reciprocating Compressors Petroleum Open Learning

Section 2 - Design and Construction


In this Section, we are going to have a look at the The piston is fitted with piston rings, which we will
principal components of a reciprocating compressor. look at later. Because these will cause wear, the
We will see how they are constructed and exactly what cylinders are commonly lined with a smooth bored
they do. liner, which can be replaced when it becomes worn.

Figure 6 is a drawing of a cylinder, liner, piston and


I have listed below the components which we will piston rings.
consider in the section :

• Cylinders

• Pistons and piston rings

• Compressor valves

• Stuffing box and packing

• Crankshaft, connecting rod, crosshead and


piston rod

Take another look at Figure 1 on Page 2.5. See how


many of the components listed above you can identify-
they are not all labelled ! When you have done that, we
will look at each item on the list in turn.

Cylinders
You have seen in previous illustrations that the
cylinder in a reciprocating compressor is considerably
more substantial than a bicycle pump. However, it is
still basically a tube in which a piston slides back and
forth.

2.17
Petroleum Open Learning

Wear occurs where the piston rings rub against the Cylinder Cooling
liner. To avoid this wear forming a ‘shoulder’ or ‘step’,
counter bores are machined into the liner. (Look at On compressors with lower compression ratios, the
Figure 6 again). A counterbore is a small increase in cylinders may not require cooling. In most cases,
cylinder bore diameter, made just above the point at however, the temperature rise across the machine
which the end piston rings stop and reverse direction. requires that the cylinders are cooled.

Liners are usually pressed or expanded into place in On smaller machines, the cooling is done by blowing
order to avoid slippage which could result in knocking air across fins which are attached to the outside of the
and excessive wear. cylinder. Most reciprocating compressors, however,
use a liquid cooling system.
Cylinder Lubrication
The cylinders are surrounded by cooling jackets,
In low pressure/low temperature applications, the through which a coolant solution is circulated. This
cylinders may not require lubrication. In this case, the solution is usually a mixture of water and glycol, which
pistons may be fitted with self-lubricating piston also acts as an anti-freeze agent.
rings, made of nylon or teflon.

However, in most compressor applications, cylinder


lubrication is required to prevent excessive overheating
or wear. In such situations, a boundary layer You can see the cooling jacket round the cylinder in
lubrication system is usually installed. This injects Figure 7, on the next page.
small droplets of oil into the cylinder, to be distributed
by the movement of the piston rings. This type of
lubrication prevents the formation of an oil mist in the
gas leaving the compressor.

2.18
Petroleum Open Learning

The coolant fluid is circulated to prevent localised hot spots and to take
away unwanted heat generated by compression. This removal of unwanted
heat improves compressor efficiency.

2.19
Petroleum Open Learning

Pistons and Piston Rings To prevent or minimise gas leakage between the piston
and the liner, piston rings are provided to make a
Pistons are most commonly made from a solid seal. They fit into grooves cut in the side wall of the
casting. The piston rod, often made of stainless steel, piston. The piston rings also serve to carry some heat
is tapered where it passes through the piston. It is then from the piston to the cylinder wall.
secured against the shoulder by a lock nut. This is
illustrated in Figure 8. The clearance between piston and cylinder wall must
be:

• small enough to prevent the back-flow of gas


across the piston

• small enough to permit adequate support of


the piston rings

• large enough to prevent the rings from sticking


to the cylinder and causing excessive friction

All piston rings are designed to wear more rapidly than


the cylinder liner, which should be true and free from
scores.

2.20
Petroleum Open Learning

Figure 9 shows two types of piston ring. Compressor Valves All valves have certain features in common:

There are several types of valve used in reciprocating • a valve seat which provides a pressure tight
compressors. There is no significant difference gas seal.
between suction and discharge valves, and they both
operate in a similar manner. • a valve plate or other device to seal across the
valve seat.

• a spring or other mechanism to hold the valve


plate on the seat in the closed position.

• a cover to contain the springs and prevent the


plate from moving too far.

A typical valve is shown in Figure 10.

During operation the rings must move out against the


cylinder to effect a ring-to-wall seal, and the gaps in
the rings allow them to do this. The sealing effect is
aided by the piston and rings expanding out towards
the cylinder wall as the compressor reaches operating
temperature.

2.21
Petroleum Open Learning

The valve plates are in the form of rings connected by Remember that a suction valve is properly installed Stuffing Box
webs and are held lightly against the seat by a set of when you can depress the plate in towards the centre
small leaf or coil springs. of the cylinder, and a discharge valve is installed In order to prevent leakage of compressed gas from
properly when you can depress the plate away from the cylinder past the piston rod, some form of seal
To open the valve, the gas must overcome the the centre of the cylinder. is required. The most common type of seal is the
pressure of the gas behind the plate and the light stuffing box.
tension of the springs. It must be emphasised that any loose material such
as screws or nuts falling into a cylinder can cause The stuffing box consists of a series of seal elements
Any tendency of the valve to slam or flutter can often very severe damage. Hence compressor valves are each containing a pair of seal rings. Figure 11, on the
be controlled by changing the tension of the valve installed with through-bolts, lockscrews or jackbolts to next page, shows the arrangement of a seal element
springs. hold the valve assembly together. with a type of seal ring known as the TR type.

Compressor valves are among the most important


parts of a reciprocating compressor and the following
points should always be born in mind:

1. A worn or damaged valve allows gas to leak


back.

2. When a valve leaks, the gas returning through


the valve is hotter. Valve leakage can often be
detected by an increase in temperature at the
valve.

3. The sudden, chilling effect of cold liquid on


a hot valve can break the valve plate. Hence
the requirement for liquid-free gas in the
compressor.

4. Dirt or frozen deposits can foul or damage a


valve and prevent it from seating properly.

2.22
Petroleum Open Learning

The TR type seal element consists of two rings which are distinguished from each The T ring haste function of preventing gas leakage. The R ring protects the T ring
other as follows and helps to dissipate heat.

• the internal ring (T) is fitted first and has tangential cuts The two rings are assembled with staggered cuts and a dowel (not shown) provides
for their correct positioning.
• the external ring (R) is fitted last and has radial cuts

2.23
Petroleum Open Learning

Another type of stuffing box seal element is the TT


type as shown in Figure 12. In this, both rings have
tangential cuts.

The ends of the seal ring segments, in both the TR


and TT types, are not in contact. This allows them to
compensate for the progressive wear of the rings by
gradually moving closer together.

A spiral spring, assembled on the groove drawn round


the edge of each ring, keeps the segments together.

2.24
Petroleum Open Learning

Figure 13 is a drawing of a typical Stuffing Box. The series of sealing elements are held in position by
the stuffing box end cover. This is secured by long stud
bolts. Piping for the entry and exit of lubricating oil, and
the venting of gas are built into the end cover.

Crankshaft, Connecting Rod,


Crosshead and Piston Rod
The drive motor (either an electric motor or an internal
combustion engine) imparts a rotary motion to the
drive shaft. This is converted to reciprocating motion by
the crankshaft, connecting rod and crosshead.

Crankshaft
The crankshaft is made of forgeable carbon steel,
machined throughout. It is provided with a single
crank and is suitably counterweighted to limit the
dynamic load on the foundation.

The crankshaft ends are equipped with bearings of the


bush type. They are fitted on the crankcase sidewalls.

2.25
Petroleum Open Learning

The crankshaft, connecting rod, crosshead and


piston rod are shown in Figure 14.

2.26
Petroleum Open Learning

Connecting Rod Crosshead


This is made of high strength pressed steel. The crosshead connects the piston rod to the
connecting rod of the crankshaft. It is equipped with
Both ends of the connecting rod are equipped with shoes which permit it to slide back and forth within
heavy duty sleeve bearings. Figure 15 shows the the crosshead guides. (See Figure 15)
connecting rod at the crosshead.
The connecting rod is moved by the crankshaft. As
the crankshaft rotates, the connecting rod
reciprocates.

Piston Rod
Piston rods are usually made of stainless steel.
They are accurately ground and have no taper
within their length of travel.

The piston rod screws into the crosshead and is


secured in place by a locking device.

A slinger ring prevents oil from the crankcase being


carried out by the piston rod and reaching the
cylinder. It is installed on the piston rod, as you can
see in Figure 14.

Now that you are familiar with the components of a


typical reciprocating compressor, have a go at the
following Test Yourself before moving to the next
Section.

2.27
Petroleum Open Learning

Test Yourself 2.5 Summary of Section 2

1. What is the purpose of a counterbore in a In this Section, we have looked at the component parts of a reciprocating compressor.
liner?
We saw how the space between the cylinder liner and the piston is sealed by the piston rings.

2. Why are liners pressed or expanded into You will have noted how the piston is lubricated and how the cylinders are cooled.
place?
The construction of compressor valves was described, and how they operate to maintain the flow
of gas through the compressor.
3. How does a boundary layer lubrication
system work ? We have looked at the different types of seal used in the stuffing box, and how the stuffing box
prevents gas from escaping from the compressor along the piston rod.

4. How does a cooling system improve the You saw how the crankshaft, crosshead and connecting rod convert the rotary motion of the driver to the
efficiency of the compressor? reciprocating motion required by the compressor.

5. What are the two functions of piston rings?

6. Which components convert rotary motion to


reciprocating motion?

In the next Section, we will take a look at the auxiliary systems which are used with reciprocating
You will find the answers to Test Yourself compressors.
2.5 on Page 2.45

2.28
Petroleum Gas Compression - Unit 2 - Reciprocating Compressors Petroleum Open Learning

Section 3 - Auxiliary Systems


In this Section, we will be looking at a number Cooling System The coolant solution (usually a mixture of water and
of auxiliary systems which are associated with an antifreeze, such as glycol) is circulated through the
reciprocating compressors. These are: You saw earlier that, as gas is compressed, its cylinder coolant jackets. This prevents the formation of
temperature increases. The compressor’s cooling local hot spots, and provides for an even distribution of
• Cooling System system removes some of the heat generated by heat. The heat is carried away from the cylinder by the
compression (heat of compression) and also protects coolant, in a closed loop thermosyphon system.
• Lubrication System the piston and cylinder from becoming overheated.
• The cylinder jackets are connected by pipes to
• Suction and Discharge Piping System Figure 16 is a simple line drawing of a compressor an expansion tank, which allows expansion
cooling system, cooling the cylinder and stuffing box. of the coolant solution as it heats up during
• Drive Coupling compressor operation. (This tank is provided
with a vent and a level gauge.)

• A thermosyphon effect is obtained when


the coolant is warmed by the heat from the
compressor: the cold (and therefore heavier)
coolant flows from the bottom of the tank to
enter the bottom of the cylinder jacket,
while hotter (and therefore lighter) coolant
is displaced from the cylinder jacket back to
the expansion tank. The return line to the tank
is near the top, but below the liquid level.

• The warm coolant loses heat from the sides of


the tank to the atmosphere and, when cold,
falls to the bottom of the tank.

• As long as the circuit is kept full of coolant, the


coolant will keep flowing around the system.

This limited circulation system gives adequate cooling


for a process compressor handling high pressure, low
temperature gas.

2.29
Petroleum Open Learning

Lubrication System

2.30
Petroleum Open Learning

Figure 17 is a line drawing of a typical lubrication Oil from the crankcase sump first passes through the
system which supplies lubricating oil to the following coarse strainer. This strainer is removable so that it
parts of the compressor : can be cleaned.

• Crank mechanism The oil is then drawn into the pump suction. The pump
increases the pressure of the oil and
• Piston rod packing discharges it to the oil cooler. From the cooler outlet
the oil flows, via fine filters, to two separate lubrication
• Crosshead systems:

The lubricating oil forms a surface film which reduces • crosshead and stuffing box
friction and, therefore, wear between the moving
compressor parts. • crankshaft frame

The lubricant also has a cooling function. Some of


the heat generated by friction is carried away by the
lubricating oil.

The lubrication system supplies filtered oil at the


required pressure and temperature to the compressor
frame.

The most common form of lubrication is a forced feed


type. Here, the oil is pumped under pressure to the
required parts. The pressure is supplied by means
of an electric motor driven pump. A standby pump
is usually provided in order to achieve uninterrupted
operation. This can be seen in Figure 17.

The lubricating oil is collected and stored in the


crankcase sump. The sump is equipped with a heater,
level sight glass, coarse strainer and a drain.

2.31
Petroleum Open Learning

Suction and Discharge Piping System


Figure 18 shows a typical piping system for a single stage reciprocating compressor.

2.32
Petroleum Open Learning

A liquid knockout drum is installed in the suction Such pulsations may also cause starvation on the To save space, volume bottles can be replaced by
piping to remove any entrained liquid from the process suction side of the compressor. The effective capacity pulsation dampeners.
gas. A drain is provided to allow any accumulated of the cylinder may be reduced by as much as 25% by
liquid to be removed. operating without a suction volume bottle. The most common pulsation dampener is the baffle
type.
The knockout drum is one of the most important The capacity of a suction volume bottle is normally not
items of equipment in the piping system. Liquids less than seven times the total cylinder capacity for all Figure 19 shows a typical baffle type pulsation
are incompressible fluids and, if they enter the cylinders served. The bottles are usually located close dampener.
compressor, even in very small amounts, they to their cylinders.
could cause the cylinder to rupture.

In addition, cold liquid mist entering a hot compressor


can seriously damage the suction valves.

A strainer is fitted in the suction piping downstream of


the knockout drum. This strainer is normally installed
for start-up purposes to prevent hard pieces of scale,
welding beads, etc., left over from construction and
maintenance, from entering the compressor and
causing damage.

The suction piping transfers the process gas to the


inlet of the compressor via a suction volume bottle.

The purpose of the suction volume bottle is to act


as a reservoir which damps down pulsations in the
inlet gas. Such pulsations are due to intermittent flow
through the compressor and, if they happen to match
the natural frequency of vibration in the pipework, can
cause serious damage.

2.33
Petroleum Open Learning

The gas leaves the compressor via a discharge A compressor vent line is fitted on the compressor A compressor bypass line is sometimes used to
volume bottle. The purpose of the discharge side of the discharge block valve. The compressor vent transfer discharge gas back into the compressor
volume bottle is to prevent excessive momentary line is used: suction piping and reduce the efficiency/capacity of
discharge pressures or pulsations. Large discharge fixed speed machines.
pulsations can result in severe overloads to the • to depressurise the compressor after shutdown
compressor and pipework and may also reduce
effective cylinder capacities. • to purge the compressor of flammable gas Drive Coupling
before maintenance
Again, the discharge volume bottle may be replaced by
a pulsation dampener. • to purge the compressor of air before start-up Reciprocating compressors are normally driven by an
internal combustion engine or an electric motor, which
All volume bottles are equipped with drains and is connected to the crankshaft by means of a drive
pressure taps for checking pressure losses. They For start-up and maintenance purposes, the most shaft and a direct coupling.
must be located so that they can be easily removed for common purge gas is nitrogen. The purge gas is:
inspection or possible repair. A direct coupling will only accommodate small
• injected into the suction line inaccuracies in the alignment of the drive and
The compressor discharge piping transfers the crankshaft - both the motor and the compressor must
compressed gas to the process equipment. • allowed to flow through the compressor be accurately positioned to achieve an acceptable
alignment. This is usually ensured by using a common
A non-return valve is fitted in the discharge pipe close • vented from the system via the vent line base for the driver and the compressor. This common
to the compressor. The function of the non-return valve base is called a bed·plate.
is to prevent high pressure gas from the downstream During start-up, the vent line is also used to purge
process equipment flowing backwards through the the nitrogen from the compressor casing with the gas The bed-plate is accurately machined to ensure that
compressor when it is not operating. (The compressor which is to be compressed. On a typical oil production it is level, and the two machines are positioned by the
discharge valves should prevent the backflow of gas platform, the vent line is routed to the platform flare use of dowels.
through the compressor. The non-return valve is fitted system.
as an added safeguard.) A small clearance is maintained between the two
The compressor casing is protected against excessive halves of the coupling to avoid imposing any end thrust
A block valve is also fitted to the discharge of the pressure by a pressure relief valve fitted in a branch on the motor bearings.
compressor. This valve is used for compressor pipe which is connected to the compressor discharge
isolation. The discharge block valve should always line. To prevent accidental isolation of this relief
be opened before the compressor is started up. valve, it is always fitted on the compressor side of the
discharge block valve.

2.34
Petroleum Open Learning

The rotational direction of the crankshaft is important


and the motor rotation should be checked to make
sure that it matches that of the machine, before the two
are coupled together.
Test Yourself 2.6

1. Why do we mix glycol with the water in the cooling system?


You have now completed Section 3 of this Unit on
reciprocating compressors, which dealt with the 2. What makes the water circulate through the cooling system?
auxiliary systems. The following Test Yourself will help
to reinforce your understanding of the topics covered. 3. Where is the lubricating oil collected and stored?

4. What is the most common form of lubrication for a reciprocating compressor


stuffing box?

5. Why is there always a liquid knockout drum installed in the suction piping ?

6. Are liquids compressible?

7. On a typical oil production platform where would you expect the compressor
vent line to lead to ?

8. How do we ensure that the driver and compressor are


accurately aligned?

You will find the answers to Test Yourself 2.6 on Page 2.46

2.35
Petroleum Open Learning

Summary of Section 3

In this Section on auxiliaries, we have looked at :

• the compressor cooling system

• the lubrication system

• the suction and discharge piping system

• the driver coupling

You saw how the cooling system supplied cooling The suction and discharge piping system was Finally we looked at the driver coupling and saw how
liquid to the cylinders and stuffing box. Use of the examined. We saw how the knockout drum prevented the use of a common bed-plate for the driver and the
thermosyphon effect to achieve circulation of this liquid from entering the compressor. Volume bottles compressor reduced the problems of alignment.
cooling liquid was also explained. (or pulsation dampeners) were used to reduce
pulsations caused by the intermittent flow of gas
into and out of the compressor. We noted the use of
We then looked at the lubrication system and saw the vent line and saw how the pressure relief valve
how the lubricating oil was stored, filtered and then was always positioned on the compressor side of the
pumped to the crank mechanism, piston rod packing discharge block valve.
and crosshead. In the next Section, we will look at the operation of a
typical gas compression system using reciprocating
compressors, together with alarm and shutdown
systems and some of the main operational checks.

2.36
Petroleum Gas Compression - Unit 2 - Reciprocating Compressors Petroleum Open Learning

Section 4 - Operation of Reciprocating Compressors


In this, the final Section of the reciprocating I have divided the Section into the following topics : A Typical Gas Compression System
compressor unit, we will be looking at the
operation of the compressor. • a typical gas compression system Figure 20 is a line drawing of a
separation and gas compression system which uses
• alarm and shutdown systems reciprocating compressors.

• the main operational checks on a


reciprocating compressor

2.37
Petroleum Open Learning

It is intended to illustrate a typical reciprocating • After treatment the gas is passed through A minor process disturbance maybe any process
compressor installation, but is not meant to represent the 2nd stage suction knockout drum before variable (temperature, pressure, level, flow, etc.) which
any specific plant. being fed into the 2nd stage of the two stage is too high or too low. This will not be a dangerous
compressor. situation, but it has the potential to become dangerous
Follow this illustration using the description below. if not attended to. For example:
From the diagram you can see that: • Here, the gas is compressed to meet the
requirements of a gas-lift system or of sales • Suppose there is a low liquid level in
• Low-pressure gas from the 2nd stage gas. the coolant tank of a compressor. There is no
of an oil/gas separation system is immediate danger of overheating. If the
passed through a booster compressor • Note that, in this two stage system, both operator reacts quickly to top up the tank
suction knockout drum.This drum removes compressors are driven by the same motor. with coolant, the immediate problem is solved.
any entrained liquids from the gas before it is (Clearly, however, the operator must find out
fed into the booster compressor. (The suction Alarms and Shutdown Systems why the coolant level fell in the first place.)
knockout drum is sometimes referred to as the
suction scrubber.) We should now look at how we control the When this type of disturbance occurs, the control
compression process and how we protect the system will generate an alarm. The setting of the alarm
• The booster compressor increases the equipment from damage. status usually gives the operator sufficient time to react
pressure of the gas from the 2nd stage and correct the problem before the situation becomes
separator to that of the 1st stage. Generally, all process controls are designed to inform dangerous.
the operator automatically if anything goes wrong.
• After passing through the booster compressor Process control systems normally work on two levels:
the gas is cooled before it joins with gas from
the 1st stage separator. • minor process disturbances
Major Process Disturbances or
• The combined gas stream is then passed
through another suction knockout drum to
• major process disturbances, or emergency Emergency Incidents
incidents
remove any entrained liquids before being fed
into the suction of the 1st stage of a two stage A major process disturbance may be any process
compressor. variable which is so high or low that the system has
reached a potentially dangerous condition.
• This 1st stage increases the gas pressure to a
level which allows it to be treated, say, in a gas
liquids recovery plant.

2.38
Petroleum Open Learning

One example may be where the low coolant level In both cases, there is an immediate danger to the We will now take a closer look at the gas compression
problem described above has not been dealt with : process, and the control system would generate a system shown in Figure 20. We can see that there are
shutdown and alarm which would automatically a number of suction knockout drums.
• The low-level alarm from the coolant tank has shut down the compressor to prevent damage to the
already warned the operator of a minor problem. equipment. If the liquid level were too high in any of these suction
A temperature measuring device in the remaining knockout drums, an alarm would be sent to the main
coolant will have warned him that the coolant In both minor and major process problems, the alarm control room.
temperature was rising. If he failed to react to normally consists of a flashing light and a ‘beeper’
these two minor alarms then, before the coolant which draws the operator’s attention to the problem. It
started to boil, the control system would generate is called an audio/visual alarm system. If the operator failed to stop the liquid level rising any
a shutdown and alarm. The shutdown and higher then, before the liquid was carried over into
alarm sequence will automatically shut down The flashing light normally lights up behind a glass the compressor, where it would cause damage, the
the compressor safely, to prevent damage to the plate which has the number and name of the particular shutdown and alarm would be activated by a high-
equipment. alarm written on it. The beeper is normally common to high level switch which responds to a high-high liquid
all the alarm systems. level in the knockout drum.
An emergency incident may be any situation which
would cause immediate danger to the system being The alarm light will continue to flash and the The control system would then :
controlled. It may be directly related to the system, or beeper to ‘beep’ until the operator ‘acknowledges’
have nothing at all to do with it : the problem by pressing a button. • shut down the compressor

• An example of an emergency incident which When the problem has been ‘acknowledged’ in this • give an alarm to the operator
is directly related to the system being controlled way, the beeper stops sounding and the light stops
would be where fire or smoke had been detected flashing but stays alight. This reminds the operator • indicate that the compressor had been shut
in the immediate area. that the problem still exists. The light will not be down because of a high-high level in a
extinguished until the problem is resolved and the particular suction knockout drum.
• An example of an incident not directly related alarm has been re-set.
would be where there was a failure of a utility
system, such as instrument air.

2.39
Petroleum Open Learning

In addition to these high level alarms and high-high High temperature alarm and high-high temperature High discharge temperature alarm and high-high
level shutdown and alarms, the following alarms shutdown and alarm on the lubricating oil system discharge temperature shutdown and alarm
and shutdowns are fitted to most reciprocating
compressors : If the lubricating oil gets too hot, it will become less If the gas discharge temperature is too high then
viscous and will be unable to lubricate the bearings damage may occur to the compressor, either because
Low lube-oil pressure alarm and low-low lube oil and pistons effectively. the lubricating oil becomes too thin, or the temperature
pressure shutdown and alarm rating of the downstream pipework is exceeded.

If the lubricating oil pressure is too low, then the The driver which is driving the compressor will also
compressor will not be lubricated properly and be fitted with its own alarm, and shutdown and
excessive wear, or even a piston seizure, may result. The following alarms and shutdown and alarms are alarm, system. This system is normally tied into the
fitted on the piping into and out of the compressors: compressor system and is classed as a local alarm
High vibration alarm and high-high vibration or local shutdown and alarm, because it operates
shutdown and alarm Low suction pressure alarm and low-low suction in conjunction with the compressor, without being
pressure shutdown and alarm installed on it.
If the compressor vibrates too much, this indicates
excessive wear, poor alignment or incorrect operation. If the suction pressure is too low, then the compressor In addition to all the shutdown and alarms which may
Excessive vibration will damage bearings, valves, cannot achieve the discharge pressure required. be fitted to the compressor, its adjacent pipework and
pistons and cylinder walls. its driver, there maybe other emergency situations
High discharge pressure alarm and high-high which will shutdown the compressor.
High temperature alarm and high-high temperature discharge pressure shutdown and alarm
shutdown and alarm on the coolant system A prime example of such a condition would be a fire
If the discharge pressure is too high, then the pressure in the compressor area. Under these conditions, it
If the coolant system gets too hot, it will be unable to rating of the equipment maybe exceeded, or the would be unwise to keep the compressor running and
cool the compressor effectively, and damage to the compression ratio, and therefore the gas discharge therefore it would be shut down by a fire and gas
pistons and cylinders will result. temperature, will rise. alarm and shutdown system.

2.40
Petroleum Open Learning

Another type of alarm maybe fitted to the compressor, The Main Operational Checks on a • the liquid is unable to flow through
called an inhibit alarm. Inhibit alarms are fitted to the discharge valve fast enough to reduce the
prevent the compressor from being started under Reciprocating Compressor pressure
certain conditions but, once the compressor is running,
the inhibit alarms will not stop the compressor. Having looked at how a reciprocating compressor • the pressure continues to rise until the engine
system is controlled and shut down, we now need to stalls (or the cylinder head blows off!)
Two examples of inhibit alarm conditions are: consider how this system should be operated.
Check that the suction and discharge pipelines are
A limit switch on the discharge valve from the The ‘golden rules’ for operating a reciprocating lined up correctly
compressor compressor are as follows :
We must make sure that the compressor has an
Before the compressor can be started, this switch uninterrupted supply of gas to the suction and that,
may need to be in the ‘ON’ position, showing that the after compression, the gas is able to flow away from
valve is fully open. If an operator were foolish enough Before Starting the Compressor the compressor to its intended destination.
to close the discharge valve after the compressor
had been started, the compressor would shut down Check that the compressor is purged of all air Check that dependent systems are operational
because of a high-high pressure condition, not
because the switch had been moved to the ‘OFF’ If the compressor is not completely purged of air then Before starting the compressor, we need to be sure
position. it may act as a ‘compression-ignition’ engine (for that it is not going to shut down because of a lack of
example, a diesel). This means that, when the first gas, because the main driver has run out of fuel, or for
A low temperature switch fitted to the lubricating compression stroke occurs, the heat of compression other reasons not directly related to the compressor
oil tank may ignite the air/gas mixture in the cylinder and an itself.
explosion will occur.
If the lube-oil is too cold at start-up, then it would be Check that the discharge valve is fully open
too viscous to circulate around the compressor and Check that the suction line is free from liquids
protect the bearings. The compressor is therefore This ensures that pressure built up in the compressor
inhibited from starting until the lube-oil reaches a Liquids are incompressible. If there is liquid in the is allowed to flow away without interruption.
minimum temperature. Once the compressor is cylinder when the piston starts a compression stroke:
running, however, the lube-oil will be heated by the
compressor and its temperature should not fall. • the pressure rises rapidly

2.41
Petroleum Open Learning

Check that the discharge relief valve Is operational Check that the cooling system is operating When the Compressor is Running
correctly
This must be checked very carefully. If any piping in Check that the pressures, levels, flows and
the system is wrongly aligned, or if any of the high- We should check that: temperatures are within operational limits
high pressure shutdown systems are not working,
then it is this valve which provides us with adequate • there is sufficient coolant in the tank These checks must be made frequently, say, at least
protection against a pipeline rupture or damage to the once every two hours. They form the bulk of a typical
compressor. • any coolant added to the system is of the operator’s working day. The successful operation of
correct type and concentration any process will depend on repeated checks of this
nature, to ensure that nothing is amiss with the system,
Check that the lubricating oil system is operating • pumps, where fitted, are running or ready to or with the equipment.
correctly run when the compressor is started
Get to know the characteristics of each
We should check that: compressor set

• there is sufficient lube-oil in the tank Each compressor set has its own particular operating
Check that no current alarm or shutdown characteristics. These characteristics consist, not only
• any lube-oil added to the system is of the conditions exist (including inhibit alarms) of data which can be measured (by reading gauges,
correct type and grade level indicators, and so on) but of less ‘scientific’
Even if the compressor controller allowed us to start up information such as the noise made by the equipment.
• pumps, where fitted, are running or ready to the compressor with a high liquid level in the suction
run when the compressor is started knockout drum, it would be unwise to do this. If the The operator should know when the machine ‘sounds
level increased as we started, the compressor would right’. Each compressor makes a different noise and,
be shut down by the high-high level condition. with practice and familiarity, a change in this noise can
be the first warning that something is going wrong.

If you are Involved in compressor operations


you should become completely familiar with
the equipment under your control. The specific
operating procedures should be followed and safe
working practices adopted at all times.

2.42
Petroleum Open Learning

Test Yourself 2.7 Summary of Section 4

In the final Section of this Unit on reciprocating compressors, we have concentrated on the operation and
control of the system.
1. In Figure 20, where does the Booster
Compressor take gas from? The Section was split into three parts :

2. In Figure 20, what is the produced gas finally


used for? • In the first part we looked at a Now that you have completed Section 4, you have
typical gas compression system come to the end of Unit 2 of the compression
3. Would a high-high level in a compressor using reciprocating compressors. I described programme. I must emphasise once again
suction drum be classed as a minor process each component for you, and how the system that this unit is not meant to take the place of
disturbance? overall was operated and controlled. specific manufacturers guidelines or operating
Instructions. It is intended to give you a good basic
4. What normally happens when the operator • We then went on to look at the various grounding in the design, construction and operation of
‘acknowledges’ an alarm? alarms, and shutdown and alarms which reciprocating compressors.
would be incorporated into such a
5. What does ‘high vibration’ indicate in a system. We saw why each particular alarm
compressor? and shutdown was fitted and what it was
there to protect.
6. Why are inhibit alarms fitted, and what makes
them different from other alarms? • Finally, we reviewed the main operational
checks which we would expect to make on
a reciprocating compressor system. We saw Now go back to the Training Targets and satisfy
why the checks were made and what action yourself that you have met these targets.
the operator was expected to take.

You will find the answers to Test Yourself 2.7 on Page


2.46

2.43
Petroleum Open Learning

Check Yourself - Answers

Check Yourself 2.3


Check Yourself 2.1
The steps should be in the following order:
A bicycle pump is a single acting compressor.

1. Piston begins suction stroke. 9. Suction valve closes.

8. Gas remaining in cylinder expands and 4. Cylinder pressure rises above discharge
discharge valve closes. line pressure.

5. Cylinder pressure falls below suction 11. Discharge valve opens.


Check Yourself 2.2 line pressure.
2. Gas flows from cylinder into discharge line.
20 bara 12. Suction valve opens.

6. Gas flows into cylinder from suction line. 7. Piston reverses direction at end of
discharge stroke.
3. Piston reverses direction at end of
suction stroke.

10. Gas in cylinder is compressed to above


suction line pressure.

2.44
Petroleum Open Learning

Check Yourself 2.4 Check Yourself 2.5


Single- Double-
Acting Acting
1. To prevent the formation of shoulders in the
liner
1. Only the space at the head end of the cylinder is used for
compression. a 2. To avoid slippage of the liner, resulting in
knocking and excessive wear
2. There is a suction and a discharge stroke each time the
piston travels the length of the cylinder.
a 3. A boundary layer lubrication system injects
small droplets of oil into the cylinder. The oil
3. A suction valve is open during each suction stroke. is distributed as a thin layer by the movement
a a of the piston rings
4. During each stroke (forward and backward) a discharge
valve is open. a 4. The cylinder cooling system improves
compressor efficiency by removing unwanted
5. The back stroke is the suction, or intake stroke. heat of compression
a a
6. The suction valve opens every second stroke. 5. To prevent or minimise gas leakage between
a the piston and the liner
7. There is a suction and a discharge valve at each end
of the cylinder. a To carry some of the heat from the piston to
the cylinder wall
8. The forward stroke is the compression, or discharge stroke.
a a 6. The crankshaft, crosshead and connecting
rod assembly

2.45
Petroleum Open Learning

Check Yourself 2.6 Check Yourself 2.7


1. So that the water will not freeze in cold weather 1. From the second stage of the oil/gas separation system

2. The thermosyphon effect. Warm (lighter) water rises, 2. As lift gas and/or as sales gas
cold (heavier) water sinks
3. No. It is a major problem. If the level gets any higher, the liquid may
3. In the crankcase sump enter the compressor and cause damage. A shutdown and alarm
will be generated
4. The most common form of lubrication is a drip feed type
4. The beeper stops sounding and the light stops flashing but stays
5. To remove any entrained liquid from the process gas and prevent the alight to remind the operator that the problem still exists
possibility of serious damage due to liquids entering the compressor
5. It is a sign of excessive wear, poor alignment or incorrect operation
6. They are generally considered to be incompressible
6. Inhibit alarms are fitted to prevent the compressor from being started
7. The flare system under certain conditions. Once the compressor is running, the
inhibit alarms will not stop the compressor
8. By mounting them on a common bed-plate

2.46
POL
Petroleum Open Learning

Petroleum Gas
Compression
Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY

3
Petroleum Gas Compression - Unit 3 - Centrifugal Compressors Petroleum Open Learning
(Part of the Petroleum Processing Technology Series)

Contents Page
Visual Cues
• Training Targets 3.2

• Introduction 3.3 training targets for you to achieve


by the end of the unit
• Section 1 – Basic Theory 3.4
Operating Principles of a Centrifugal Compressor

• Section 2 – Design and Construction 3.11 test yourself questions to see how
Compressor Casing much you understand
Rotating Assembly
Bearings
Diffusers and Diaphragms
Compressor Seals
check yourself answers to let you
see if you have been thinking along
• Section 3 – Auxiliary Systems 3.32 the right lines
Seal Oil System
Lubrication System
Cooling System
Drivers and Couplings
activities for you to apply your new
• Section 4 – The Operation of Centrifugal Compressors 3.42 knowledge
Performance Characteristics
Operating Problems, Alarms and Shutdowns
The Main Operational Checks on a centrifugal Compressor
summaries for you to recap on the
• Check Yourself – Answers 3.55 major steps in your progress

3.
Petroleum Open Learning

Training Targets

When you have completed Unit 3 of the Petroleum Gas Compression series you will be able to:

• Explain the basic operating principles of a centrifugal compressor.

• Describe the construction a centrifugal compressor.

• Explain the function and operation of the principal components of a centrifugal compressor.

• Describe the layout and operation of the auxiliary systems associated with a centrifugal compressor.

• Explain a basic centrifugal compressor alarm and shutdown system.

• List the common operating checks carried out on a centrifugal compressor.

3.
Petroleum Gas Compression - Unit 3 - Centrifugal Compressors Petroleum Open Learning

Introduction
In Unit 1 of this Compressor programme, you saw that a centrifugal compressor is a member of the dynamic branch of the continuous flow
family of compressors.

Dynamic compressors increase the pressure of gas in a different way to that of the positive displacement machines. A reciprocating compressor
reduces the volume of a trapped mass of gas to increase its pressure. A dynamic machine, however, uses the principle of energy conservation
to achieve pressure increase.

There are two main types of dynamic compressor, and these are :

• centrifugal (radial flow) compressors


The Unit is divided into four sections.
• axial flow compressors
Section 1 covers the basic operating theory of a centrifugal compressor.
We are only going to look at centrifugal compressors in this unit. However, in Unit
4 of the compressor programme, we will look at axial flow machines in a little In Section 2, we will look at the design and construction of a typical machine.
more detail.
In Section 3, we will concentrate on the auxiliary equipment
Centrifugal compressors are large capacity, continuous flow machines with a very
smooth output. When run at their optimum speed and loading, they are vibration and
free and have few moving parts.
Section 4 looks at centrifugal compressor performance and operations.
They are capable of delivering very large volumes of gas. In days gone by,
they tended to be used for lower pressure applications than reciprocating
compressors. However, modern machines are capable of delivering gas at
pressures in excess of 700 bar.

3.
Petroleum Gas Compression - Unit 3 - Centrifugal Compressors Petroleum Open Learning

Section 1 - Basic Theory


Operating Principles of a The compressor consists of an impeller mounted on a
shaft which can be rotated at high speeds.
Centrifugal Compressor
This assembly is enclosed in a casing which
In the Introduction I said that a dynamic machine uses incorporates a diffuser. Also built into the casing are
the principle of energy conservation to increase the the suction and discharge ports, i.e. the inlet and outlet
pressure of a gas. This means that one type of energy of the compressor.
is converted to another. In this case it is the energy
a gas has due to its velocity which is converted into We will be looking at the components of a compressor
pressure energy. in much more detail in Section 2. For the moment we
will concentrate on this rather simple machine.
The compression of gas in a centrifugal compressor is
a two part process. Look again at Figure 1 and follow the flow of gas
through the compressor from the inlet or suction port to
• In part one, gas enters the machine and it is the outlet or discharge port.
speeded up or accelerated.This increases
the energy of the gas by giving it kinetic Gas enters the compressor through the suction port
energy or energy of motion. A certain amount and is directed to the inlet or eye of the impeller.
of pressure energy is also added at this
time. As the impeller rotates, the gas is forced to rotate with
it, causing the following effects :
• In part two, the gas is rapidly decelerated.
This converts kinetic energy into more Centrifugal force causes the gas to flow from the eye
pressure energy. to the outside or rim of the impeller. As the rim of
the impeller is travelling faster than the eye, the gas
speeds up as it moves outwards.
Look at Figure 1 which shows a simple centrifugal
compressor with its main components.

3.
Petroleum Open Learning

Figure 2 shows the effect in graphical form. Again, this effect is shown graphically in Figure 3.

The impeller is fitted with blades which act rather like These two forces, the centrifugal force and the lift
airplane wings. These create a lift force which helps force, accelerate the gas and raise the pressure. This
to force the gas from the low pressure at the eye to is the first part of the process.
the higher pressure at the rim.

3.
Petroleum Open Learning

When the gas leaves the impeller at the rim, it enters As the gas flows through the diffuser, the increase in flow area causes the gas to slow down.
the diffuser. This part of the compressor is a flow This reduction in velocity means that the gas loses its kinetic energy.
channel. It is designed in such a way that the flow
area is gradually increased along its length. Figure 5 shows this reduction in velocity.

The diffuser shown in Figure 1 is called a volute


diffuser. A volute is an increasing spiral shape, as
illustrated in Figure 4. You will recognise the
shape if you look at shellfish such as whelks or
winkles.

Because the total energy of the gas cannot be reduced, the fall in kinetic energy must be compensated
for by a rise in some other type of energy. In this case the pressure energy of the gas is increased.

3.
Petroleum Open Learning

The next Figure, 6, shows, in graphical form how the


pressure of the gas increases as it flows through the
diffuser.

3.
Petroleum Open Learning

So, you have now seen how the two parts of the process produce an overall pressure increase from the inlet to
the outlet of the compressor. Figure 7 shows the overall process from inlet to outlet, again in graphical form.

3.
Petroleum Open Learning

The compressor we have just been looking at is The compression ratio which can be achieved Now, before moving on to the next section, have a
called a single stage compressor. It has one largely depends on the centrifugal force which is go at the following Test Yourself question.
impeller and one diffuser. It will have a capability of developed by the impeller. This in turn depends on
increasing the pressure of gas by a fixed amount. three things:
You will remember from Unit 1 of the compressor
series, that this increase is known as the
compression ratio.
• speed of rotation
Test Yourself 3.2
• impeller shape
To remind yourself of compression ratio, have a go
at the following simple Test Yourself question. • impeller diameter Of the following 10 items, only 5 belong to a
centrifugal compressor. Indicate with a tick in
There is a practical limit to some of these. For the box provided which items are centrifugal
instance, the diameter of the impeller and its speed compressor components.
Test Yourself 3.1
of rotation will determine how fast the rim of the
impeller is travelling. This cannot be allowed to
exceed the speed of sound in the gas it is Impeller
compressing. If it does so, serious damage to the Diffuser
a) If a compressor takes in gas at a machine may result as the sound barrier is broken.
pressure of 25 psia and delivers it at a Piston Rod
discharge pressure of 90 psia, what is Cylinder
In order to achieve greater pressure increases, Shaft
its compression ratio? multiple impellers and diffusers can be fitted inside Crank
one casing. Each impeller and diffuser unit is called Casing
b) If a compressor has the same a stage and such a machine is called a multi-stage
compression ratio as the one in part (a) Blades
compressor. Cross Head
and it takes in gas at 30 psia, what is
its discharge pressure? Clearance Pocket
In a multi-stage compressor, the gas passes
through an impeller, then a diffuser, is then directed
to the eye of the next impeller, and so on.

You will find the answers in Each stage is, in effect, a compressor in its own
Check Yourself 3.1 on page 3.55 right. In the case of a five stage compressor there
are actually five compressors within one casing.
You will find the answers in
Check Yourself 3.2 on page 3.55

3.
Petroleum Open Learning

Summary of Section 1

In this Section on the basic theory of centrifugal compressors, you saw that the principle of energy conservation
is used to raise the pressure of gas.

You saw that, within the compressor, the gas is first accelerated to give it kinetic energy and some pressure
energy. It is then slowed down and the kinetic energy is converted into more pressure energy.

We looked at a simple compressor made up of the following parts:

• impeller

• shaft

• diffuser

• casing

• inlet port

• outlet port

It was pointed out that a single impeller has restrictions which can limit the amount of pressure increase it can
produce. In order to achieve greater pressure increases, multi-stage compressors should be used.

In the next Section we will look in more detail at the components of a typical multi-stage centrifugal compressor.

3.10
Petroleum Gas Compression - Unit 3 - Centrifugal Compressors Petroleum Open Learning

Section 2 - Design and Construction


In this Section we are going to look at the component
parts of a centrifugal compressor. We will see how
they are constructed and what their function is in the
operation of the machine.
I have listed below the components which we will
consider.

These are:

• casing

• rotating assembly - consisting of

shaft

impeller

balance piston

• bearings

• diffusers and diaphragms

• seals

Before we proceed, take a look at Figure 8 overleaf,


and identify these components.

3.11
Petroleum Open Learning

3.12
Petroleum Open Learning

Of course, a compressor has to be powered by


something. The motor which powers the compressor
is called the main driver or simply, the driver. In
this Section we will not concern ourselves with the
construction or operation of the driver.
However, centrifugal compressors are orientated, or
lined up, relative to the main driver, so it is worth
identifying the orientation at this point. The end of the
compressor nearest to the driver is called the drive
end or inboard end. The end of the compressor
furthest from the driver is called the non-drive end or
outboard end.

Let’s move on to the components now.

Compressor Casing
The compressor casing is used to house the
component parts of the compressor. It may be
either horizontally split or vertically split, depending
upon the design and application.

Horizontally Split Casings


Figure 9 shows a compressor with a horizontally
split casing. The casing is made in two halves
which are then bolted together along a horizontal
join.

The internal assembly of the compressor may be Horizontally split compressors are mainly used in
removed only after the two halves of the compressor lower pressure service.
casing have been unbolted and lifted apart.

3.13
Petroleum Open Learning

Vertically Split Casings


A vertically split casing consists of a barrel shape
with end caps bolted onto each end of the barrel.

The internal assembly of the compressor may be


withdrawn from the non-drive end of the casing after
the end plate has been removed.

Vertically split casing compressors are often called


barrel compressors. They are mainly used in high
to medium pressure service.

Figure 10 shows a vertically split compressor


casing.

3.14
Petroleum Open Learning

Rotating Assembly
You will remember that this assembly consists of a
shaft, impeller(s) and a balancing piston. Let’s take
a look at each of these components in turn.

The Shaft
The compressor shaft is the heart of the
centrifugal compressor. It carries the impellers and
balance piston and they all rotate together within the
compressor casing.

Figure 11 indicates the basic shape of the


compressor shaft and the relative positions, on the
shaft, of the various components.

To minimise vibration, the rotating assembly is finely balanced and not allowed to run near any critical
speeds.

Every rotating assembly has a number of speeds, where it will reach a peak of vibration. These speeds are
called the critical speeds.

Critical speeds can be calculated as the compressor is designed and built. The normal running speed of a
compressor is set to avoid being near a critical speed. If any rotating equipment is run at a critical speed,
damage can be caused to bearings within seconds as the vibration causes metal to metal contact.

It is normal practice to go through critical speeds as quickly as possible when accelerating the compressor
to its normal running speed.

3.15
Petroleum Open Learning

The Impeller
The impellers provide the driving force for the gas
as it flows through the compressor. They are fixed
to the compressor shaft and rotate within the
compressor casing.

The impellers impart kinetic energy to the gas by


increasing its velocity within the compressor casing.

Impellers are available in a variety of designs such


as open, semi-open or closed, and the different
types are shown in Figure 12.

3.16
Petroleum Open Learning

The open type impeller consists of a number of You will remember from Section 1, that the impeller
blades attached to a shaft. It rotates within the cannot be rotated at too great a speed - a speed
casing of the compressor and the gas is constrained which would result in the outer circumference of the
between the open impeller blades and the casing impeller exceeding the speed of sound in the gas it
wall. is compressing. This is due to the possibility of
damage being caused as the sound barrier is
The semi-open impeller, in addition to the blades, broken and turbulence is created within the casing.
has a disc or hub to which the blades are attached.
The hub stiffens the assembly, and helps confine Remember also that, to overcome this restriction, a
the gas to the blade area. multi-stage compressor may be used to achieve a
given compression ratio.
The closed type of impeller is the most common in
large compressor applications, and this is the one To illustrate this further, if each stage has a
we will concentrate on. compression ratio of 1.35 then a five stage
compressor would have an overall compression
It has blades of course, and also a hub. At the front ratio of 6.75.
of the impeller, however, the blades are attached to
a cover or shroud. The Balance Piston
The shroud also confines the gas to the blade area The last component of the rotating assembly is the
and provides stiffening. balance piston.
Closed impellers are made of forged steel. The In a compressor the pressure at the inlet is
blades may be welded to both the disc and the obviously less than the pressure at the outlet. This
shroud. Alternatively, the blades may be machined difference in pressure across a compressor acts on
from a solid disc and then welded to the shroud. the impellers and shaft to create a thrust force.
The force tends to push the rotating assembly
Figure 13 shows a closed impeller. towards the inlet, or suction end, of the compressor.
A force pushing along the line of the shaft is called
an axial force.

3.17
Petroleum Open Learning

In Figure 14, we can see how the thrust force which If nothing was done about this, severe damage could
is generated will be imposed upon a single impeller. be done to the machine as the rotating assembly tried
to move axially within the casing.

One way to minimise rotating assembly movement is


to install thrust bearings in the compressor. We will
be looking at how these bearings work shortly.

On a large compressor, the amount of thrust force


developed is extremely high. If a thrust bearing was
the only device installed to prevent thrust forces from
damaging the compressor, then the bearing would
probably be bigger than the compressor itself.

To reduce the effects of the thrust forces and hence


reduce the thrust bearing requirement, a balancing
piston is fitted to the shaft. It is designed to reduce the
thrust forces to an acceptable level.

Figure 15 shows a typical balancing (or balance)


piston assembly which is fitted to the high pressure
end of the rotating assembly.

3.18
Petroleum Open Learning

The balance piston is a solid drum which is


attached to, and rotates with the shaft. The high
pressure gas at the discharge of the compressor,
acts on the inboard side of the drum. Low pressure
gas from the suction side of the compressor is fed to
the outboard side of the drum. The pressure
differential across the drum is maintained by having
seals between the drum and the casing.

Figure 16 shows how the low pressure gas from the


suction side is fed to the outboard side of the drum
via a balancing line.

The pressure differential across the drum produces an axial force which opposes the thrust
exerted by the unbalanced forces acting on the impellers.

3.19
Petroleum Open Learning

Bearings
The rotating assembly of a compressor needs to be
supported within the casing and allowed to rotate
freely. This means that some form of bearing is
required. The bearings used for this application are
usually referred to as the main bearings or journal
bearings.

Also, as you have already seen, a thrust bearing is


required, but let’s look at the journal bearings first.

Look back to the drawing of the shaft. It is Figure 11


on Page 3.15. You will see that there are just two
locations for journal bearings. They are at the extreme
outboard and inboard ends of the shaft.
The bearings can be of several different types, but the
one I will describe is known as a tilting pad journal
bearing.

It is a development of a simple sleeve type journal


bearing which is shown in Figure 17.

3.20
Petroleum Open Learning

The sleeve type journal bearing consists of a simple


housing forming a sleeve around the shaft. As the
shaft rotates, it causes the film of oil to form a
wedge which holds the shaft and housing apart.

The tilting pad type bearing is a development of


this. It has a number of pads which are located on
fixed pivots attached to the stationary housing.

Figure 18 shows the arrangement of the pads in the


bearing.

As the shaft rotates, the film of oil again forms a


wedge between shaft and pad. In addition, the
tilting pads give the bearing self aligning properties.
These compensate for any slight misalignment of
shaft and bearing. They also tend to distribute
evenly the loads which are created when the
rotating assembly is spinning.

And now, let us take a look at thrust bearings.

3.21
Petroleum Open Learning

You have already seen how the balancing drum helps to reduce the thrust forces to acceptable levels. The bearing has a collar which is fitted to and rotates
Any residual thrust, however, must be taken up by a thrust bearing. with the shaft. Located in recesses machined into a
fixed or non-rotating thrust surface, are a number of
A typical thrust bearing is shown in Figure 19 and is called a pivoting pad thrust bearing. metal shoes or pads.

The collar rotates against the pads which are free to


pivot. An oil wedge forms between the collar and the
pads, as in the tilting pad journal bearing.

Figure 19 shows a thrust bearing which is capable of


taking up thrust in one direction only. Most thrust
bearings, however, can take up thrust axially in both
directions. This requires two fixed or stationary shaft
thrust surfaces containing two sets of pads.

Diffusers and Diaphragms


From Figure 8 we can see that the diaphragms and
diffusers are non-rotating parts of the compressor.

From Section 1, you will remember that gas leaving


the impeller at the rim enters a flow channel called the
diffuser. In the simple, single stage compressor which
I used as an example, the diffuser was in the shape
of a volute. In a multistage compressor, having a
volute after each stage would be unwieldy. Therefore,
in this case, the flow channels are formed by having
diaphragms, which form part of the casing, and
separate the stages. The adjacent walls of individual
diaphragms form a diffuser passage.

3.22
Petroleum Open Learning

Figure 20 demonstrates the layout of the


diaphragms and diffusers in a horizontally split.
multi-stage centrifugal compressor.

3.23
Petroleum Open Learning

The diaphragms also guide the gas through return


passages to the eye of the next impeller.

Although not shown in Figure 20, the diaphragms


also carry labyrinth seals which prevent the back
flow of gas along the shaft from the high to the low
pressure sides of the impellers. We will now have a
closer look at seals.

Compressor Seals
Compressor seals can be divided into:

• internal seals, which are designed to prevent


the movement of gas within the compressor
casing

• external seals, which are designed to prevent


the escape of gas from the compressor casing
to the atmosphere.

Internal Seals
Let us consider internal seals first.

Labyrinth seals are the most common form of


internal seal. They consist of a series of teeth,
across which the gas would have to flow, in order to
escape from a high pressure area into a low
pressure area. Figure 21 shows a labyrinth seal
and how it works.

3.24
Petroleum Open Learning

In a labyrinth seal the teeth are most commonly The type of seal selected depends upon the job it The most common devices used for external sealing
machined into sleeves which are attached to the has to do. However, because of the imperfect are liquid film seals.
casing or diaphragm, and are a very close fit with sealing nature of a labyrinth seal, it is never used as
the revolving shaft. the external shaft seal when compressing Liquid film seals are commonly used as external
flammable or dangerous gases. seals on high pressure, heavy duty compressors as
They operate by maintaining a minimum gap they provide complete sealing capabilities. There
between the shaft and the seal, and creating Typical uses of labyrinth seals in a centrifugal are, however, other types of seal in use. You will
turbulence in the gas as it passes across each seal compressor are : come across carbon ring seals, mechanical
surface. contact seals and increasingly, dry gas seals. In
• as interstage seals to prevent the flow of high this Unit, we will concern ourselves only with the
This turbulence converts pressure energy into heat pressure gas from the tip of the impeller to its liquid film seal and the dry gas seal.
energy, and pressure is therefore lost across each eye
stage of the labyrinth. Figure 22 on the next page shows a liquid film seal
• to maintain the differential pressure across the and the way in which it works.
The teeth of the labyrinth are machined into soft balance piston
alloys of aluminium or lead. In order to minimise the
gap between shaft and seal they are often allowed • to control the escape of compressor gas into
to rub against the rotating element when being the sealing system of an oil film seal
bedded in.
• to control the loss of lubricating or seal oil
The number of teeth on the labyrinth will vary from along a shaft
as few as two to well over 20. The number used will
depend upon:
External Seals
• the type of gas being compressed
Now let us look at external seals.
• the level of sealing required
When flammable or dangerous gases are being
• the differential pressure across the seal compressed it is important that the gases do not
escape from the compressor. To prevent this from
occurring, the gap between the compressor shaft
and the compressor casing, at each end, is sealed.

3.25
Petroleum Open Learning

Liquid Film Seals


The liquid film seal uses oil as the sealing medium.
Oil is pumped between the shaft and two tight fitting
sleeves which are anchored to the compressor
casing. The two sleeves form an inboard sealing
element and an outboard sealing element.

The oil flowing across the outboard sealing element


does not come into direct contact with the
compressed gas and is returned directly to a seal
oil reservoir.

The oil flowing across the inboard sealing element


comes into contact with the compressed gas. Some
of the gas may dissolve in the oil, and the oil is
therefore routed to a de-gassing system before
being returned to the seal oil reservoir.

The liquid film seal has no touching parts in its


assembly. It is therefore not prone to wear, and is
very reliable.

3.26
Petroleum Open Learning

Dry Gas Seals


The dry gas seal is a recent development which is
becoming increasingly popular as an external
compressor seal.

Figure 23 illustrates the construction and operation


of this type of seal.

The seal consists of :

• a rotating tungsten ring, attached to the


compressor shaft

• a static carbon ring, attached to the casing

The static carbon ring is pushed towards the rotating


tungsten ring by a set of coiled springs. When the
compressor is at rest, the two faces touch to form a
gas tight seal. Each contact face is machined to a
high degree of flatness so that this sealing effect
can be achieved.

3.27
Petroleum Open Learning

A series of spiral grooves are cut into the face of the Pressure of gas in the sealing dams forces the carbon Now have a go at the following Test Yourself question.
tungsten ring, as shown in Figure 24. When the ring away from the tungsten ring, against the coiled
shaft begins to rotate, gas is pulled into these springs. At this point, the two surfaces are no longer
grooves, and forced towards the centre of the ring. A in contact and, therefore, no frictional heat is being
tiny area of high pressure gas is created at the generated. This means that no cooling is required
inside end of each groove, called a sealing dam. under normal running conditions.

If we were handling a non-flammable gas, some of the


gas being compressed would be used to create the
seal. The small amount of leakage through the seal
could then be vented.

Figure 23, however, shows the seal arrangement used


for flammable or toxic gases.

A labyrinth seal is mounted between the gas being


compressed and the dry gas seal. Nitrogen is injected
into this space at a pressure slightly higher than the
gas being compressed. Some nitrogen may leak back
into the compressor, but no toxic or flammable gas can
escape across the dry gas seal to atmosphere. When
used in this way, the nitrogen is called a buffer gas.

Dry gas seals give a reliable level of sealing without


the ancillary equipment associated with liquid film
seals- tanks, pumps, filters, de-gassers, and so on.
Apart from cost savings, the weight of this equipment
could be as high as seven or eight tons. This will be a
very important factor on an offshore production facility,
and the use of dry gas seals in this environment is
expected to increase.

3.28
Petroleum Open Learning

Test Yourself 3.3


Read through the following statements and fill in the missing words from the list given below:

1. In a . . . . . . . . . . . . . . . . . . . .. split casing, the casing is made in two halves which are


bolted together.

2. At a rotating assembly’s............................... speed it will reach a peak of vibration.

3. In a closed impeller the blades are attached to both the................... and the.........................

4. The thrust force acting on the rotating assembly is taken up by two items. They are the
..................... and the ...................................

5. The ........................guide the gas through the return passages to the eye of the next
impeller.

6. The most common internal seal is the............................seal.

LIST OF WORDS

HUB, DIFFUSERS, CRITICAL, SHROUD, THRUST BEARING,


LABYRINTH, BALANCING PISTON, HORIZONTALLY.

You will find the answers in Check Yourself 3.3 on page 3.55

3.29
Petroleum Open Learning

Summary of Section 2
In the Section we have just worked through, we looked at the component parts of a centrifugal
compressor.

These items are : From there we moved on to look at the rotating Next we looked at diaphragms and diffusers.
assembly which consists of a shaft, one or more These are the non-rotating parts of the compressor
• casing impellers and a balancing piston. The impellers are which reduce the velocity of the gas leaving the
mounted on the shaft and rotate with it. You saw that impeller, thus raising its pressure. They also guide
• rotating assembly (shaft - impeller-balance there are a number of basic types of impeller open, the gas from the outlet of one impeller to the inlet
piston) semi-closed and closed. The balance piston is also eye of the next.
mounted on the shaft and its function is to reduce
• bearings the thrust forces which can act on a centrifugal Finally we considered seals. You saw that they
compressor. I pointed out that it does this by using a are either internal or external seals. The labyrinth
• diffusers and diaphragms pressure difference across it to counteract the main types are commonly used as internal seals.
thrust.
• seals Liquid film seals are most often used for external
Bearings were the next thing we considered. We sealing purposes, although dry gas seals are
looked at journal bearings and thrust bearings. Journal becoming increasingly popular. Both types were
You saw that the casing can be horizontally split or bearings are used to support the rotor and allow it considered in detail. You should be aware,
vertically split and I illustrated the difference between to rotate freely. Thrust bearings are used to prevent however, that other designs maybe used as
the two. movement of the shaft in an axial direction. In both external seals.
cases the tilting or pivoting pad type bearing was
illustrated.

3.30
Petroleum Open Learning

Before moving on to Section 3, where we will be


looking at auxiliary systems, take some time to try
the following Test Yourself.

Test Yourself 3.4


State whether the following components
are part of : 1. Inboard element ...................
a) the rotating assembly 2. Labyrinth teeth ...................
b) the sealing system 3. Impeller rim ...................
c) the casing and non-rotating assembly 4. Diffuser ...................

5. Thrust collar ...................

6. Diaphragm ...................

7. De-gasser ...................

8. Journal bearing ...................

9. Balancing piston ...................

10. Inlet port ...................

You will find the answers in Check Yourself 3.4 on


page 3.56

3.31
Petroleum Gas Compression - Unit 3 - Centrifugal Compressors Petroleum Open Learning

Section 3 - Auxiliary Systems


In this, the third section of the Unit, we will be looking Figure 25 also shows how the oil flows through the
at the auxiliary systems associated with centrifugal seals. Note how part of the oil in each seal flows
compressors. These are: outwards past the outboard sealing element. This oil
does not come into contact with gas and, therefore,
• seal oil system can flow directly back to the seal oil reservoir. The oil
which flows inwards past the inboard sealing element
• lubrication system becomes contaminated with gas. In order that this oil
can be used again it must be cleaned. It is therefore
• cooling system routed to a de-gassing system before returning to the
reservoir.
• driver and coupling
(We will talk about the reference line very shortly).
Seal Oil System
Let’s start with the seal oil system.

As we saw in the last Section, high pressure, heavy


duty compressors are often fitted with liquid film seals.
These are designed to prevent any gas from leaking to
the atmosphere from the shaft ends.

Look at Figure 25. This is a simplified view of a


compressor which shows the liquid film seals at each
end of the shaft. Each seal will be the same type as
the one illustrated in Figure 22 on Page 3.26.

3.32
Petroleum Open Learning

3.33
Petroleum Open Learning

The system which supplies the oil to the seals consists


of the following items.

• a seal oil reservoir

• one main and one auxiliary seal oil supply


pump

• filters

• coolers

• a header tank

• seal oil traps (de-gassing system)

Take a look at Figure 26 which shows the seal oil


system in simplified form.

3.34
Petroleum Open Learning

The oil is pumped from the reservoir via the filters


and coolers to the header tank.
Lubrication System consider the two as separate systems.

A typical lubrication system would consist of the


We can now move on to the lubrication system.
A pressure reference line takes gas from the seals following components :
and feeds it to the header tank. Variations in this
All rotating machinery must have a lubrication
gas pressure at the seals are, therefore, mirrored in • an oil reservoir
system of some form or another. Compressors are
the tank.
no exception. The lubricant, in this case oil,
• a main and auxiliary pump
performs the following functions:
However, the oil pressure at the seal must always
be greater than the gas pressure there, otherwise • coolers
• separates moving parts
gas will escape. This is achieved by always
positioning the header tank above the compressor, • filters
• removes heat generated by friction
and maintaining the seal oil at a fixed level in the
tank with a level controller. The seal oil pressure, The system layout is shown in Figure 27, overleaf.
• reduces metal wear
therefore, will always exceed the gas pressure at Take a look at this now and identify the
the seal by an amount equal to the static head of oil. components.
• protects metal surfaces from corrosion
From the header tank the oil flows to the seal.
The lubrication system normally provides a flow of
As you have just seen, some of the oil then flows
oil to the journal and thrust bearings of the
outwards between the shaft and the outboard
compressor. In addition, it provides lubrication for
sealing element to the reservoir. The rest flows
the main driver, gear box and other accessories.
across the inboard sealing element, to the reservoir
via the de-gasser
Most lubricating oils, usually abbreviated to lube
oils, are refined from crude oil. However, to give
Follow carefully the layout of the seal oil system in
them their special properties, chemical additives
Figure 26 and ensure that you are familiar with the
may be mixed with them.
components and the method of operation.
Each compressor will have a lube oil which is
As you look at Figure 26, remember the comment I
specified for that particular machine. Care must be
made on Page 28 about the weight of a liquid film
taken to ensure that no other lubricant is used, in
seal system - that it could easily reach several tons.
order to prevent the possibility of damage to the
compressor.
It will be clear why lighter systems, such as dry gas
seals, often prove more popular in an offshore
In some compressors, the lubrication system may be
environment.
combined with the seal oil system which we have
just been looking at. Here, however, we will
3.35
Petroleum Open Learning

The lube oil is pumped from the reservoir through


coolers which reduce the temperature of the oil
to its correct operating value. From there it passes
through filters which remove any dirt particles. The
cooled, clean oil then flows under pressure to each
bearing through pipework, grooves and channels.
After performing its lubricating job, the oil flows back
to the reservoir under gravity. The system pressure is
maintained by having a pressure controller in the line
between the filters and the compressor.

Note that a header tank is incorporated into the


system. Under normal running conditions the tank is
kept topped up with oil from the main feed line. A small
amount of oil will overflow back to the reservoir.

If the main lube oil pumps fail, the compressor will shut
down. When this happens, the bearings are lubricated
using a gravity feed from the header tank. This ensures
that there will be no damage caused to unlubricated
bearings during the time that the compressor is rolling
to a halt.

Before moving on to the next part of Section 3,


in which we will look at cooling systems, try Test
Yourself 3.5.

3.36
Petroleum Open Learning

Test Yourself 3.5


Indicate if the following statements are true or false. If false give the correct answer.

a) The external seals of a high pressure compressor are usually labyrinth seals.

b) In a seal oil system a pressure reference line takes gas from the seals and feeds it to a
header tank.

c) In a seal oil system the oil which comes in contact with the gas is routed to the reservoirs via
the filters.

d) Lubricating oil helps to remove heat generated by friction.

e) In a typical lube oil system the filters remove dirt particles from the oil.

You will find the answers in Check Yourself 3.5 on page 3.56

3.37
Petroleum Open Learning

Cooling System
Compression generates heat. It is often necessary
to cool the compressed gas for safe and efficient
operation of the plant. In many installations, there
may be a number of compressors working in series.
If the hot, compressed gas from the first compressor
were passed directly to the suction of the next
machine, overheating and damage could occur.

In such a situation, an aftercooler may be installed


downstream of each compressor.

Figure 28 shows a typical aftercooler for a


centrifugal compressor, and its position in the
system.

3.38
Petroleum Open Learning

Large compressors will consist of two or more sets


of impellers mounted on a common shaft. In order
to reduce the temperature increase within such a
machine, the gas may be discharged after the first
set of impellers, cooled, and directed to the suction
of the next set.

In this case, the gas passes through an intercooler


as it flows between the two sets of impellers.

This arrangement is shown in Figure 29.

3.39
Petroleum Open Learning

Drivers and Couplings In some instances, it may be necessary for the


compressor to run at a variable speed, even though
In the final part of this Section, I want to have a brief the driver may be a constant speed machine such
look at drivers and couplings. as an electric motor. In these instances, a variable
speed fluid coupling may be fitted between the main Test Yourself 3.6
Centrifugal compressors used in oil and gas field driver and the gearbox to permit this.
operations are normally driven by gas turbines
or electric motors. These are connected to the In this short piece on drivers and couplings I have not In the following list of components, some belong
compressor by means of drive shafts and couplings. tried to describe the components in detail. However, it to the compressor auxiliary systems, some do
is important that you know the terminology associated not. If not, to which system do they belong?
Some drivers, which rotate at high speeds, may be with them.
coupled directly to the compressor. However, it is 1. Impeller
often necessary to incorporate some kind of speed
increasing gearbox between driver and compressor. 2. Lube oil cooler

Couplings are used to connect two shafts together and 3. Coupling


transmit the rotary motion of one to the other. Great
care is taken to align the driver shaft to the compressor 4. Driver
shaft. However, due to variations in loading and
possible uneven heating of materials and equipment, 5. Balancing piston
small misalignment conditions can occur.
Before I summarise this Section, attempt 6. Diaphragm
Because of this a flexible coupling is required. Test Yourself 3.6.
You should not misunderstand the word ‘flexible’. 7. Intercooler
A compressor coupling may be over 10 inches in
diameter, and appear to be far from ‘flexible’ when 8. De-gasser
handled.
9. Reservoir

10. Thrust collar

You will find the answers in Check Yourself


3.6 on page 3.56

3.40
Petroleum Open Learning

Summary of Section 3

In this Section on auxiliaries, we have looked at seal oil systems, lubricating systems, cooling
systems, drivers and couplings.

Starting with seal oil systems, you saw that the oil is pumped from a reservoir, through coolers and
filters, to a header tank. The level in the tank maintained the correct oil pressure on the seal.
Some of the oil, which becomes contaminated with gas, is cleaned in a degassing system before
being used again. The rest of the oil goes straight back to the reservoir.

Turning our attention to the lubricating system, you saw that the system is very similar to the seal
oil system. Again, coolers and filters are used and the oil is continually circulated round the
system and back to a reservoir. The lube oil not only lubricates the compressor bearings, but also
the driver and gear box.

A cooling system may be required to reduce the temperature of the gas for safe and efficient
operation. You saw that this system may take the form of an intercooler, or an after cooler.

Finally, in the Section, we had a brief look at drivers and couplings. The compressors may be
driven by electric motors, or gas or steam turbines. They may be directly coupled or be driven via
a gear box. Couplings connect the shafts of the driver and compressor. These couplings are
called flexible couplings which are capable of taking up any small misalignments between the
shafts.

In the final Section of this Unit on centrifugal compressors, we will take a look at compressor
operations.

3.41
Petroleum Gas Compression - Unit 3 - Centrifugal Compressors Petroleum Open Learning

Section 4 - The Operation of Centrifugal Compressors


In this, the final Section of the centrifugal
compressor unit, we will be looking at the operation of
the compressor.

I have divided the Section into the following topics:

• compressor performance characteristics

• operating problems, alarms and shutdowns

• operational checks

Let’s start by considering the performance


characteristics.

Performance Characteristics
The performance of a centrifugal compressor can
be shown on a set of operating curves. These are
graphs prepared individually for each compressor.
They show the range of flows, heads, efficiencies
and speeds within which a particular compressor is
capable of operating. In other words, they indicate
the performance of the compressor under different
operating conditions.

Figure 30 is a simple graph which shows a single


operating curve for a specific compressor.

3.42
Petroleum Open Learning

You can see that: You will notice three other lines on Figure 30, marked
as:
• the vertical axis gives the actual compression • optimum efficiency
ratio (or head) as a percentage of the design
compression ratio • approximate surge limit

• the horizontal axis gives the inlet flow (capacity) • stonewall (choke) line
as a percentage of the design capacity
We will look at all of these in some detail later.
The solid line curve which passes through the design
point shows the relationship between head and inlet In fact, the performance of a compressor is usually
flow when the machine is running at 100% of its design expressed by a family of curves. The shape and
speed. position of the curves depend on a number of things.
These include the design, size, speed and the number
The design point for any compressor is the point at of impellers.
which the machine is :

• running at 100% of its design speed

• compressing 100% of the design capacity or


inlet flow
A typical set of curves is shown in Figure 31, overleaf.
• producing 100% of the design compression
ratio

At this point the compressor is operating at 100%


efficiency. As you can see from the operating curve,
any changes to the speed, flow or pressure will remove
it from this point.

3.43
Petroleum Open Learning

In Figure 31, the curves represent the relationship


between head and flow for a number of different
speeds.

The efficiency of the machine at various heads, flows


and speeds is indicated in the graph as a series of
ellipses representing lines of constant efficiency. You
will also see that optimum efficiency at varying
speeds is represented by a single line running through
the ellipses.

The machine is capable of operating at conditions


anywhere within the envelope of the curves. This
means that the compressor is able to operate at
conditions away from its design point, but at a reduced
efficiency.

The boundaries of this envelope to the left and right


are labelled approximate surge limit and stonewall
(choke) line. I will have more to say about these lines
shortly.

The upper boundary of the envelope is determined by


the maximum speed at which the impeller is capable of
rotating. This in turn depends on its size and
construction.

The lower boundary is not really a problem for the


compressor. If the surge and stonewall lines were
extended down, they would meet at the point of zero
flow and zero head. So the compressor could operate
at much lower speeds. This lower limit is usually
determined by the minimum speed at which the driver
can be operated.

3.44
Petroleum Open Learning

Let’s now look at the two boundaries of the compressor performance curves shown in Imagine that the compressor is running at constant
Figures 30 and 31. speed at Point 1 on the curve. This means that it will be
delivering a certain pressure (P) with a corresponding
These are the approximate surge limit and stonewall. First of all, surge limit. inlet flow (F). Everything is normal.

Compressor Surge Supposing, however, that there is a sudden increase in


downstream resistance which reduces the inlet flow to
The surge limit is the minimum flow for a given speed at which the compressor can maintain Point 2 on the curve.
stable operation. At inlet flow rates to the left of this line, the operation of the machine
becomes unstable. Don’t forget that, if the compressor is running at a
constant speed, the pressure and flow values must lie on
We can analyse a compressor characteristic curve to see how surge in a compressor occurs. the curve.

Take a look at Figure 32 which shows a curve of discharge pressure against inlet flow for a At this point the compressor will continue to operate in a
constant speed. You should note that I have deliberately exaggerated the shape of this stable manner. Even though there is a pressure increase
curve in order to make the following explanation of surge more easily understood. in the downstream pipework, the compressor discharge
pressure has risen to overcome this.

Now look at Figure 33

3.45
Petroleum Open Learning

Supposing, now, that the compressor is operating at In major cases of surge there are complete reversals of The effects of compressor surge are well known and
Point 3 on the curve which is right at the peak. flow which result in a massive shunting of the rotating predictable. Because of this, all high pressure
assembly. Each cycle may occur over twice per second compressors are fitted with anti-surge control
Now, if there is a reduction in flow due to an increase and on every cycle the whole rotating assembly (which loops which prevent the situation from occurring.
in downstream resistance, the operating point moves may weigh in excess of two tons) shunts inside
to Point 4. the compressor casing as the thrust forces change The anti-surge control loops operate by controlling
direction across the compressor. the flow of gas through the compressor to a pre-set
This point is in a region where the compressor actually minimum.
produces less pressure than at Point 3. The machine Each shunt can throw enormous stresses on thrust
now cannot produce the pressure necessary to bearings, journal bearings, seals and shaft
overcome the downstream resistance. couplings.

When this condition occurs, the flow momentarily The ultimate result of compressor surge is the rapid
reverses its direction so there is no forward flow. The backward and forward movement of the rotating
operating point now moves to Point 5. assembly, and the resulting vibration may:

With no flow through the machine, the discharge • destroy internal labyrinth seals
pressure is reduced. This then allows the gas flow rate
to build up again towards Point 6. • destroy journal and/or thrust bearings

But, at that flow rate, the pressure delivered by the • seize the rotating assembly in its bearings
machine is less than that required to overcome the
downstream resistance. The operating point then • stall the main driver
moves along the curve towards Point 3 again and,
once beyond there, the cycle is repeated. • shatter the shaft coupling and/or gear box

The cycling I have just described causes oscillations of • shatter the impellers and/or the compressor
the gas flow in the compressor and pipeline, which is casing
known as surge.

3.46
Petroleum Open Learning

Figure 34 is a drawing of a simple anti-surge controller which re-cycles gas from the You can see from the drawing that the anti-surge controller is
discharge of the compressor back to the suction line. connected to a control valve. The valve is in a pipeline which
is connected between the compressor suction and discharge
lines outside the machine.

The controller senses and compares the rate of gas flow into
the compressor, and the pressure rise across it. If the flow rate
at a given pressure rise is less than the setting of the controller,
the control valve is opened. This allows enough gas to flow
from the discharge to the suction of the compressor to keep the
suction flow rate above the surge value.

Note that the recycled gas is cooled in a heat exchanger. This


is to prevent a rapid increase in temperature as hot discharge
gas is fed to the suction and further heated as it is compressed
again.

In view of the damage which may result from surge, most


centrifugal compressors are fitted with an approaching surge
alarm and an anti-surge shutdown, in addition to an
anti-surge controller. These systems work as follows :

The anti-surge controller should normally prevent surge from


occurring. If it fails to operate, then the approaching surge
alarm will be activated.

The approaching surge alarm will alert the operator to the fact
that the anti-surge controller has not managed to rectify the
situation. If the situation is not corrected then the anti-surge
shutdown is activated.

The anti-surge shutdown device is set to activate just before the


point at which the compressor will enter surge. When this
safety device operates, the compressor will be stopped.

3.47
Petroleum Open Learning

Stonewall or Choke Operating Problems, Alarms


and Shutdowns
You will remember that the second boundary line
on the compressor performance graph was called
the stonewall (choke). Figures 30 and 31 both Test Yourself 3.7 Just like any other machinery or process plant,
compressors are protected from malfunction or
show this. damage by instrumentation systems. These
a) List 3 possible consequences of surge systems will generate alarms and compressor
Stonewall is the point at which the flow rate through shutdowns if dangerous situations should arise.
the compressor approaches the speed of sound in in a compressor.
Let’s have a look at some of these situations now.
the gas it is compressing. As this occurs, the shock
waves generated result in a choking effect, which
prevents the gas from building pressure. b) What measurements are compared by Excessive Compressor Speed
an anti-surge controller?
The symptoms that the compressor is approaching If the compressor is being driven by an electric
a stonewall condition are a rapidly falling discharge motor, rotational speed is normally fixed and normal
head coupled with very little extra flow. c) What determines the lower speed limit running speed cannot be exceeded.
of a centrifugal compressor?
Stonewall limits the maximum flow which the If the compressor is being driven by a steam or gas
machine can achieve at a given speed. turbine, normal running speed could be exceeded
which might result in severe damage to the system.
In these cases the compressor, and the main driver,
are normally protected by :

• a speed governor on the main driver which is


Before moving on, have a go at Test Yourself 3.7. set to control the compressor at a maximum
of 100% normal running speed

• an electronic speed sensor on the main


driver which will shut down the compressor if
its rotational speed exceeds 105% of normal
running speed

You will find the answers in


Check Yourself 3.7 on page 3.57

3.48
Petroleum Open Learning

• a mechanical trip which flies out from the shaft The results are varied but may include: The liquid which is separated from the gas,
of the main driver and trips the fuel supply, accumulates in the bottom of the drum. If this liquid
if the rotational speed of the compressor • The compressor may slow down or stop and level rises beyond a certain point, there is a danger
exceeds 110% of normal running speed the main driver may stall of it being carried over with the gas. To prevent this,
a low liquid level is maintained in the vessel by level
Ingress of Liquids • The strain on the gear box may cause it to fail control instrumentation. Any further rise in level,
or burst perhaps through a level control malfunction, will
Ingress of free liquids into the compressor can be trigger an alarm at a pre-set value. If the level still
extremely dangerous. The liquids could be in the • The compressor shaft may be bent by the continues to rise, the instruments will cause a
form of mist or droplets, or in the form of larger sudden and uneven forces which occur. This shutdown of the compressor plant.
slugs. bending of the compressor shaft may often
result in a compressor casing failure
If liquids enter a centrifugal compressor in mist or
droplet form, there may be erosion of the impellers, • The seal system may fail due to an over
diaphragms and casing. pressuring of the system

This erosion may create a loss of efficiency, or The most common result is that considerable
problems of vibration as the impellers become damage to the compressor’s shaft, shaft coupling, Figure 35, on the next page, shows a simple layout
unbalanced. impellers and casing will occur. of compressor and knock out drum.

In extreme cases, the erosion will result in both In order to prevent liquids entering the machine, all
impeller and casing failure. centrifugal compressors are fitted with suction
knock-out drums. These are pressure vessels
If the liquids are in the form of slugs when they enter located in the pipework leading to the suction of the
the compressor then, as liquids are both compressor. They are designed to separate any
incompressible and denser than gases, the effects free liquid from the suction gas stream.
are likely to be sudden and dramatic.

3.49
Petroleum Open Learning

3.50
Petroleum Open Learning

temperature alarms and shutdowns.


Vibration In addition to the alarms and shutdowns which I
have just described, other alarms may be fitted to
If the compressor starts to vibrate, the rotating the compressor. These are called inhibit alarms.
impellers could rub against the casing. This could Test Yourself 3.8 They are fitted to prevent the compressor being
started under conditions which may be detrimental
damage the machine, and in severe cases, cause
catastrophic failure. to the machine. Once the machine is running,
however, the inhibit alarm will not cause a
The vibration could be caused by surge, corrosion Describe with the aid of a simple sketch an shutdown.
products being deposited in the compressor, erosion anti-surge controller.
of impellers, or misalignment of the driver and I have included below, as examples, a few inhibit
compressor. alarm situations :

To protect the compressor, vibration monitors are • An inhibit which will not allow the compressor
installed at various points on the machine. At pre- lube-oil pump to start until the lube-oil
determined vibration levels, the monitors will initiate temperature is at a pre-set level.
alarms, followed by shutdowns.
You will find the answer in • An inhibit which will not allow the compressor
Check Yourself 3.8 on page 3.57 to start until the lube-oil temperature is at a
pre-set level.
Surge
• An inhibit which will not allow the compressor
We looked at the problem of surge a little earlier in to start until the lube-oil reservoir is full.
this Section. Check that you remember what protects
the machine in the case of surge by attempting Test • An inhibit which will not allow the compressor
Yourself 3.8. Lube Oil Pressure and Temperature to start unless the suction valve is open.

If the pressure on the lube oil system is too low, the


compressor will not be lubricated properly.
Excessive wear on bearings would then occur. To
prevent this, pressure sensors initiate an alarm,
followed by a shutdown, if the pressure falls below a
set value.

Similarly, if the lube oil gets too hot, it will become


too thin and lose its lubricating properties.
Temperature sensors will initiate high lube oil
3.51
Petroleum Open Learning

The Main Operational Checks on a Check that the valves in the suction and We should check that:
discharge pipelines and the anti-surge system
Centrifugal Compressor are in the correct position, i.e., open or closed. • there is sufficient lube-oil in the reservoir

We have looked at how the system is controlled, We have to make sure that the compressor has an • any lube-oil added to the system is of the
and what alarm and shutdown systems are uninterrupted supply of gas to the suction. We also correct type and grade
commonly installed. Now we can look at how the have to make sure that the gas is able to flow away
compressor should be operated. from the compressor to where it is intended to go. • main lube oil pumps are running and header
The anti-surge line should also be checked to tanks are full
The golden rules for operating a centrifugal ensure that the control valve is fully open when the
compressor are: compressor is being started. • lube-oil flows are within normal limits

Before Starting the Compressor Check that dependent systems are operational • auxiliary lube oil pumps are available for use

Check that the compressor is purged of all air We have to check that the compressor is not going Check that the seal oil system is operating
to shut down because of a lack of gas, or because correctly
If the compressor is not completely purged of air, the main driver has run out of fuel, or for any other
the gas/air mixture may burn or explode when the reason which is not directly related to the We should check that:
compressor is started. compressor itself.
• there is sufficient seal oil in the reservoir
Check that the suction line and compressor Check that the discharge valve is in the correct
casing is free from liquids position • any seal oil added to the system is of the
correct type and grade
Liquids are incompressible. If we try to compress Some centrifugal compressors are started up with
them, the pressure increase maybe so high that the the discharge valve closed and allowed to go • main seal oil pumps are running and header
compressor is damaged. through a re-cycle stage before they are put on line. tanks are full

The high rotational speeds of centrifugal Check that all relief valves are operational • seal oil flows are within normal limits
compressors also means that they develop high
torque when starting. Trying to spin a casing full of Check that the lube-oil system is operating • auxiliary seal oil pumps are available for use
liquid may stall the main driver during the start correctly
sequence.

3.52
Petroleum Open Learning

Check that the lube-oil and seal oil cooling • A high vibration shutdown by-pass. As the If you are involved in compressor operations
medium is available compressor starts up it may pass through two you should become completely familiar with the
or more critical speeds which will cause high equipment under your control. Your specific
We should check that: vibration. Unless the high vibration shutdown operating procedures should be followed, and
is by-passed, the compressor will be shut safe working practices adopted at all times.
• the lube-oil and seal oil coolers are operating down as it passes through these critical
correctly speeds.

• the temperatures are being properly When the Compressor is Running


maintained
• Check that the pressures, levels, flows and
Check that no current alarm or shutdown temperatures are within operational limits.
conditions exist These checks must be made at regular
intervals, at least once every two hours.
In addition to the alarm and shutdown instrumentation
(including inhibit alarms), centrifugal compressors are • Check that the lube-oil and seal oil levels
often fitted with various by-passes which allow them to are maintained. If it becomes necessary to
start up. The most common bypasses are: top up the systems with oil, the following
points should be noted:
• An anti-surge shutdown by-pass. Surge is
a function of low flow, and the anti-surge Always use the correct type of oil for each system.
shutdown operates when the flow through the
compressor is low. Before start-up there is, Always ‘top up’ the system with clean and dust free
of course, no flow. Therefore, the anti-surge oil of identical grade.
shutdown must be by-passed during start-up,
to prevent it shutting down the compressor I have described a few checks that an operator of
due to the temporary, low flow conditions compression plant should carry out as part of his
which exist at this time. routine duties. The list is by no means exhaustive
and is certainly not meant to be taken as an
operating procedure.

3.53
Petroleum Open Learning

Summary of Section 4
In the final Section of this Unit on centrifugal compressors, we have concentrated on the operation
of the system.

The section was split into three parts: • Finally we had a look at some of the
operational checks which should be carried
• In the first part we looked at compressor out before a compressor is started and when
performance characteristics. You saw that the machine is running.
the performance can be illustrated by a set
of operating curves. These curves show the Now that you have completed Section 4, you
range of heads, flows, efficiencies and have come to the end of Unit 3 of the
speeds within which a compressor can compression programme. I must emphasise
operate. once again that this unit is not meant to take
the place of specific manufacturers
We studied a set of typical curves and used guidelines or operating instructions. It is
them to explain surge and stonewall. You intended to give you a good basic grounding in
saw how a compressor is protected against the design, construction and operation of
surge by an anti-surge system. centrifugal compressors.

• We then went on to consider some Now, go back to the Training Targets and satisfy
operating problems, alarms and shutdowns. yourself that you have met those targets .
We looked at some typical problems and
saw how a compressor is protected against
the damage that they can cause.

3.54
Petroleum Open Learning

Check Yourself - Answers

Check Yourself 3.1 Check Yourself 3.2 Check Yourself 3.3


The items marked with a tick are centrifugal 1. In a HORIZONTALLY split casing,
a) Compression ratio
compressor items. The others belong to a the casing is made in two halves
discharge pressure
reciprocating compressor. which are bolted together.
= suction pressure

Impeller √ 2. At a rotating assembly’s
90 psia = 3.6
= 25 psia CRITICAL speed it will reach a
Diffuser √ peak of vibration.
Piston Rod 3. In a closed impeller the blades are
b) Discharge pressure attached to both the HUB and the
Cylinder SHROUD

= Suction pressure x Compression ratio
Shaft √ 4. The thrust force acting on the

= 30 psia x 3.6 rotating assembly is taken up by
Crank two items. They are the

= 108 psia BALANCING PISTON and the
Casing √ THRUST BEARING
Blades √ 5. The DIFFUSERS guide the gas
through the return passages to
Cross Head the eye of the next impeller.
Clearance Pocket 6. The most common internal seal is
the LABYRINTH seal.

3.55
Petroleum Open Learning

Check Yourself 3.4 Check Yourself 3.5 Check Yourself 3.6


a) FALSE 1. Impeller NO - Rotating assembly
The external seals are usually liquid
1. Inboard element (b) film seals or other, equally efficient types. 2. Lube oil cooler YES

2. Labyrinth teeth (b) b) TRUE 3. Coupling YES



3. Impeller rim (a) c) FALSE 4. Driver YES
The oil which comes into contact with
4. Diffuser (c) the gas is routed via the DE-GASSERS to the 5. Balancing piston NO - Rotating assembly
reservoirs.
5. Thrust collar (a) 6. Diaphragm NO - Non rotating assembly
d) TRUE
6. Diaphragm (c) 7. Intercooler YES
e) TRUE
7. De-gasser (b) 8. De-gasser YES

8. Journal bearing (c) 9. Reservoir YES

9. Balancing piston (a) 10. Thrust collar NO - Rotating assembly

10. Inlet port (c)

3.56
Petroleum Open Learning

Check Yourself 3.7 Check Yourself 3.8


Using a sketch similar to that in Figure 34, your description should have been as
a) Any three of the following : follows :

An anti-surge controller compares the inlet flow to the compressor


- destroy internal labyrinth seals with the pressure rise across the compressor. If the inlet flow rate is
too low for a particular pressure rise the controller opens a valve in a
- destroy bearings line between compressor suction and discharge.

- stall main driver This allows gas to flow from the discharge to the suction side of the
compressor to keep the suction flow rate above that at which surge
- shatter coupling and/or gear box will occur.

- shatter impeller

b) Inlet flow rate, and pressure difference between suction


and discharge.

c) The minimum speed at which the driver can be operated.

3.57
Petroleum Open Learning

3.58
POL
Petroleum Open Learning

Petroleum Gas
Compression
Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY

4
Petroleum Open Learning
Petroleum Gas Compression – Unit 4 - Other Types of Compressor
(Part of the Petroleum Processing Technology Series)

Contents Page

• Training Targets 4.2 Visual Cues

training targets for you to achieve


• Introduction 4.3 by the end of the unit

• Section 1 – Other Types of Positive Displacement Compressor 4.4


test yourself questions to see how
much you understand
Rotary Screw Compressors
Lobe Compressors
Sliding Vane Compressors
Liquid Ring Compressors
check yourself answers to let you
see if you have been thinking along
the right lines
• Section 2 – Other Types of Continuous Flow Compressor 4.11

Axial Flow Compressors
Mixed Flow Compressors activities for you to apply your new
Fluidic Compressors knowledge
Ejectors

• Check Yourself - Answers 4.18 summaries for you to recap on the


major steps in your progress

4.1
Petroleum Open Learning

Training Targets

When you have completed Unit 4 of the Petroleum Gas Compression series you will be able to :

• List the main types of compressor used in the petroleum industry other than centrifugal and
reciprocating compressors

• Describe the construction and operation of four types of rotary positive displacement compressors

• Describe the construction and operation of the axial and mixed flow of dynamic compressors

• Describe the construction and operation of the ejector type fluidic compressor

• Explain the principal uses of the compressors listed

4.2
Petroleum Gas Compression – Unit 4 - Other Types of Compressor Petroleum Open Learning

Introduction
In Units 2 and 3 of this compressor programme we From the family tree you can see that we must
concentrated on two types of compressor : consider a few more machines to complete our study
of compressors.
• reciprocating
Test Yourself 4.1 In this Unit, therefore, we will tie up the loose ends in
• centrifugal the programme on compression by taking a brief look
at these other types.
These are the common types of compressor found in Test a typical compressor family tree, naming the
petroleum production operations. types of compressor in each main branch. I do not intend to go into nearly so much detail as I did
in Units 2 and 3. In fact, we will limit ourselves to just
However, they are by no means the only cornoressors two sections.
in use. You will remember from Unit 1 that we
classified compressors into a number of different In Section 1 we will look at the basic construction
categories and represented this as a family tree and operating principles of another type of positive
displacement compressor, namely the rotary
compressors. (In the family tree, we also included the
diaphragm type of reciprocating machine. However,
Check your recall of Unit 1 now by trying the following You will find the answer in Check Yourself 4.1 on I do not intend to say anything further about this
Test Yourself question. page 4,18 compressor.)

Section 2 will concentrate on the other continuous


flow machines. These are the axial and mixed flow
dynamic and the fluidic types.

4.3
Petroleum Gas Compression – Unit 4 - Other Types of Compressor Petroleum Open Learning

Section 1 - Other Types of Positive Dislacement Compressor


You will remember from the family tree that the two Take a look at Figures 1(a) and 1(b) which show, in Figure 1(a) is a plan view and Figure 1(b) is a
branches of the positive displacement arm led to the simplified form, a rotary screw compressor simplified end view of the machine.
reciprocating machines and the rotary machines. In
this section we will be having a look at the rotary
types. These include:

• screw type

• lobe type

• vane type

• liquid ring type

Let’s now look at each of these in turn.

Rotary Screw Compressors


The double rotary screw type of compressor is the
most common form of rotary compressor used in the
oil and gas industry. In this machine, two intermeshing
screws are used to trap and compress a gas in the
space between the screws.

The double rotary screw compressor is a positive


displacement machine which delivers a constant
volume at variable pressures. It has a single stage
compression ratio of approximately 4 to 1.

4.4
Petroleum Open Learning

The compressor casing contains two precision The incoming gas is trapped between the compressor
machined, helical screw rotors. One of these rotors has casing and the tips of the lobes. As the lobes are fitted
splines, giving it a male profile. These splines mesh close together, no gas can escape backwards between
with grooves in the other, female, rotor. The rotors are the lobes themselves. The trapped mass of gas is
usually of small diameter. This small diameter allows then pushed forwards towards the delivery end of the
shaft speeds of up to 20 000 rpm. compressor, with no reduction in volume.

The screws rotate together in opposite directions.


The splines and grooves mesh to form a series of
pockets which travel one following the other, towards Think for a moment about this process and then do the
the discharge end of the machine, The volume in each following Test Yourself
pocket reduces, thereby increasing the pressure. as it
moves from suction to discharge,

In most designs the rotors require no lubrication within


the compressor chamber and can produce an oil free
compressed gas,
Test Yourself 4.2
Rotary screw compressors are often found in
compressed instrument air service,
What is the fundamental difference in method of
compression, between a lobe type compressor and
Lobe Compressors the other two positive displacement compressors
we have looked at up to now, ie, the reciprocating
type and the screw type rotary?
Lobe compressors are often used where large volumes
of relatively low pressure gas are required,

Figure 2 shows how a twin lobe compressor works


and how the gas flows through the compressor.
The casing of the compressor encloses two rotors,
each of which has a Figure of eight shape. At each
end of the figure of eight is a lobe. These specially
shaped rotors are geared together externally to You will find the answer in Check Yourself 4.2
ensure correct meshing. They rotate in opposite on page 4.19
directions within the casing.
4.5
Petroleum Open Learning

From the answer to this Test Yourself you can see that
the back pressure in the delivery pipework determines
the outlet pressure of this compressor.

The machine we have just been looking at is a two


lobe compressor. However, lobe compressors may be
found with two, three or four lobes per rotor.

Because of these low pressure applications often


referred to as blowers.

Lobe compressors can be found on petroleum


croductlon installations as Fuel Gas compressors, or
as Booster compressors and other lower pressure
applications.

Sliding Vane Compressors


Sliding vane compressors are designed to supply a
constant volume and pressure of gas.

Figure 3 shows the basic construction of a sliding


vane compressor. Familiarise yourself with the
components, then go through the following description
of how the machine works.
Within the casing of the compressor is mounted a rotor or drum. The drum is designed to rotate eccentrically
within the casing.

The drum carries a number of vanes. These are free to slide back and forth within slots machined into the rotor.
As the drum rotates. the vanes are thrown out by centrifugal force so that they contact the casing. Sometimes
spring loading assists in this.

4.6
Petroleum Open Learning

The sliding vane compressor operates by trapping Liquid Ring Compressors


the gas in a succession of pockets whose volume is
gradually reduced. The pockets are formed by the The liquid ring compressor is more commonly
inside of the casing, the outside of the drum and a called a vacuum pump, It is used extensively in
pair of sliding vanes. injection water treatment plants, It is capable of
creating a vacuum which assists in the
Gas enters the casing through the inlet nozzle and de-oxygenation of water.
is trapped in a pocket where the vanes are at their
maximum extension. As the drum rotates the The unit consists of a cast iron casing which
vanes are pushed back into the drum as the contains a multi-bladed impeller. The impeller is
distance between casing and drum decreases. This mounted on a shaft supported at each end on
then reduces the volume of the pocket. The bearings. The shaft is fitted eccentrically to the
pressure of the gas within the pocket increases until casing so that the blades of the impeller are nearer
the volume of the pocket is at its minimum. At this to the casing at one point of the rotation.
point the gas is discharged through the delivery
nozzle.

Sliding vane compressors are very prone to damage


and breakdown if any liquids or solid particles enter
the gas stream. The vane tips erode and gas can
then leak backwards, past these tips.

They are occasionally found in Heating, Ventilation Figure 4, on the next page, shows diagramatically
and Air Conditioning (HVAC) Systems, In this the construction of a liquid ring compressor.
service are used both as air blowers and as
refrigerant compressors.

4.7
Petroleum Open Learning

4.8
Petroleum Open Learning

The liquid ring compressor operates in a similar


manner to the sliding vane type of compressor. The
essential difference is the way in which the reduction
in volume of the pockets is achieved.
Summary of Section 1
You will remember that this reduction in pocket
volume is accomplished, in a sliding vane
compressor, by the vanes being pushed back into
the drum. The liquid ring compressor uses an In this section you have seen how a selection of rotary compressors work. These were:
advancing ring of water to create the same effect.
• screw type
Look at Figure 4 and see how this works.
• lobe type
Water is injected into the unit and centrifugal force
causes it to form a liquid ring which is pinned to the • vane type
inside of the casing.
• liquid ring type
Air is drawn a pocket formed between the
impeller blades and the inner surface of the water You saw the lobe type of compressor was the only one which did not trap gas and then reduce its volume. In
ring. This occurs where the blades are furthest from fact the trapped gas in such a machine is pushed towards the discharge at an almost constant volume.
the casing and the pockets have their greatest
volume. As the impeller rotates, the blades get
nearer to the casing wall. The ring of water We did not go into a great deal of detail regarding the construction of rotary compressors. However, Section 1
advances into the pocket, and pocket volume is should have given you an idea of how these machines work and what their main applications are.
reduced. This increases the pressure of the air in
the pocket The compressed air / water mixture is
then ejected at the discharge port.

Have a look at Figure 4 again, and try to visualise


what is going on inside the compressor, Before moving to Section 2, have a go at the following Test Yourself to check your
understanding of Section 1.

4.9
Petroleum Open Learning

Test Yourself 4.3


Which type of positive dlsplacernent compressor do the following statements refer to?

1. The male splines mesh with the female grooves.

2. Springs assist in throwing the vanes outward

3. These compressors are for low pressure applications and are sometimes called blowers

4. The piston moves back and forth within a cylinder

5. The drum is mounted eccentrically within the casing.

6. This compressor is more commonly called a vacuum pump.

7. The two rotors have a figure of eight shape.

8. The small diameter of the rotors allows shaft speeds of up to 20 000 rpm.

You will find the answers in Check Yourself 4.3 on page 4.19

4.10
Petroleum Gas Compression – Unit 4 - Other Types of Compressor Petroleum Open Learning

Section 2 - Other Types of Continous Flow Compressor


The continuous flow branch of the family tree In the case of an axial flow compressor a large number As the impellers rotate, the shaped rotor blades induce
was divided into dynamic machines and fluidic of blades, called rotor blades, are attached to an the gas to flow across them, parallel to the shaft.
machines. Unit 2 of the Compressor Series impeller. There may be as many as fifteen or more
concentrated on one of the dynamic machines, namely impellers attached to a shaft. The impellers and the The rotor blades accelerate the gas towards the
the centrifugal compressor. In this section we will have shaft are contained within a casing. stator blades. The shape of these stationary blades
a brief look at the other dynamic types of compressor, causes them to act as diffusers, and slow down the
the axial flow and mixed flow compressors. The rotor blades run between fixed blades, called gas. This converts kinetic energy into pressure
stators, mounted inside the compressor casing. energy. The stator blades then direct the gas
Finally we will consider just one type of fluidic machine, towards the next set of rotor blades. and so on.
that is the ejector. Figure 5 shows the component parts of an axial flow
compressor The most common use for axial flow compressors in
Once again we will not go into any great detail but the oil and gas is as inlet air compressors
will only cover the basic construction and operation of on gas turbines. In this service they are ideally suited
each unit. to provide the large amounts of low pressure air which
are required to enable the engine to function.

Axial Flow Compressors


Figure 6 : On the next page shows the Axial
Like centrifugal compressors, axial flow compressors Compressor, in a Gas Turbine layout
are dynamic machines. They impart kinetic energy to
the gas by increasing its velocity. This kinetic energy
is then converted into pressure energy by allowing the
gas to slow down.

In the case of an axial flow compressor, the flow of


gas stays parallel (axial) to the shaft. The simplest type
of axial flow compressor is the propeller. Air is drawn
into the propeller from the front and thrown backwards
through the propeller.

4.11
Petroleum Open Learning

4.12
Petroleum Open Learning

Mixed Flow Compressors


Mixed Flow compressors utilise both centrifugal and

Test Yourself 4.4


axial flows within the same compressor casing,

All centrifugal compressors have an


element of mixed flow. In these machines, the gas
flows axially along the shaft into the inlet of each
impeller before being accelerated by centrifugal List 5 errors which appear in the following sentences.
force.
Axial flow compressors are dynamic machines. They impart potential energy to the gas by increasing
Mixed flow compressors utilise a form of axial flow it’s velocity. This energy is to converted pressure energy as the gas is slowed down. The flow of gas
blades called inlet guide vanes installed upstream stays parallel to the shaft. A number of stator blades are attached to the impeller. They rotate between
of the first centrifugal impeller. These guide the gas fixed blades mounted inside the compressor casing. Rotor blades act as diffusers to slow down the gas
into the eye of the first impeller in the most efficient after it has been accelerated. Axial compressors are often used as inlet air compressors on gas turbine
direction. The inlet guide vane often has a variable installations.
pitch facility which allows the operator to control the
direction of this inlet flow.
Mixed flow compressors utilise axial flow and reciprocating flow within the same machine. Compressors
Before we move on to have a look at Fluidioc which are designated mixed flow utilise inlet guide vanes to guide the gas into the axial flow impeller,
compressors, try the following Test Yourself question
about dynamic compressors of the axial and mixed
flow types.

You will find the answers in Check Yourself 4.4 on page 4.19

4.13
Petroleum Open Learning

Fluidic Compressors
In Fluidic compressors, a carrier fluid is accelerated in
order to transfer its energy to another fluid.

Fluidic compressors have no pistons, valves, rotors or


any other moving parts.

They are relatively inefficient but are often favoured


for use with corrosive materials or in inaccessible
situations,

Ejectors (also called Eductors) and Diffusion


Pumps are the two most common types of fluidic
compressor. However, because diffusion pumps are
seldom found in petroleum producing operations, we
will not be considering them in this Unit.

Ejectors
Ejectors (or Eductors) are names used to describe
equipment which operates to the Bernoulli Principle.

Bernoulli was a scientist who discovered that there is a


relationship between pressure and velocity of a fluid as
it flows across a restriction.

Take a look at the two diagrams in Figure 7.

Figure 7(a) shows fluid flowing across a restriction.

Figure 7(b) illustrates this relationship in the form of a As the fluid enters the the pressure falls At the exit from the restriction, the reverse
graph. and the velocity increases. This occurs as occurs, as you can see.
pressure energy is converted kinetic energy.

4.14
Petroleum Open Learning

Figure 8 shows a restriction built a of equipment called an ejector A fluid, called the driving fluid, is pumped, at high
This restriction, similar to the one shown in Figure 7, is called a venturi tube. velocity, through the venturi tube of the ejector via
an inlet nozzle. The underside of the ejector is
connected to the gas inlet. As the driving fluid enters
the venturi tube, its velocity increases and its pressure
falls. This reduction in pressure pulls gas from the gas
inlet and entrains it in the driving fluid.
This can create a vacuum at the gas inlet and, in fact
this type of equipment can also be called a vacuum
pump.

As the mixture of driving fluid and gas leaves the


venturi tube, its velocity falls again and the pressure
increases to that of the discharge pipework.

Ejectors have many applications in petroleum


production. For example:

• To increase the vacuum which is applied to


the de-aeration tower of a water treatment
facility (in conjunction with a liquid ring
compressor)

• To pull gas into a water stream in order to


create a foam in water clean-up facilities

4.15
Petroleum Open Learning

Summary of Section 2

In this short section we have had a look at some of the other compressors which fall into the
continuous flow category. These were:

• axial flow compressors

• mixed flow compressors

• ejectors

You saw that the first two were dynamic machines. They work in a similar manner to the centrifugal
compressor which we considered in Unit 2 of this Compression Series

The axial flow compressor uses rotor blades which act as impeller blades, and stator blades which act as
diffusers. The flow of gas through such a machine is parallel to the shaft, hence the name axial flow.

The ejector, however, works in a different way and has no moving parts. It relies on a driving fluid creating a
low pressure zone. This type of equipment is often used in water injection and produced water applications.
Further units in the Petroleum Processing Technology Series will explore these applications more fully.

Now, before leaving the Unit, try the final Test Yourself question.

4.16
Petroleum Open Learning

Test Yourself 4.5

Decide to which type of compressor and which branch of the family tree (positive displacement or continuous
flow) the following compressor components belong. I have done the first one as an example.

1. piston rod reciprocating compressor positive displacement

2. splined rotors ............................................. .......................................

3. inlet guide vane ............................................. .......................................

4. crosshead ............................................. .......................................

5. venturi tube ............................................. .......................................

6. water ring ............................................. .......................................

7. figure of 8 rotor ............................................. .......................................

8. drum and vanes ............................................. .......................................

9. dry gas seal ............................................. .......................................

10. stator blades ............................................. .......................................

You will find the answers in Check Yourself 4.5 on page 4.20

4.17
Petroleum Open Learning

Check Yourself 4.1

Sketch a typical compressor tree, naming the types


of compressor in each main branch.

Your sketch will look something like the family tree


shown opposite.

(You will remember that you first saw this diagram in


Unit 1 on Page 1.26).

4.18
Petroleum Open Learning

Check Yourself 4.2 Check Yourself 4.3 Check Yourself 4.4

With the piston and screw types, the 1. screw type 1. Kinetic energy is imparted not potential
compressor traps a mass of gas and reduces energy
its volume, to achieve compression, The lobe 2. vane type
type compressor simply pushes a fixed volume 2. Rotor blades not stator blades are attached
of gas into a higher pressure discharge. 3. lobe type to the impeller

4. reciprocating type 3. Stator blades act as diffusers, not rotor


blades
5. vane type
4. Mixed flow compressors use axial flow and
6. liquid ring type centifrugal flow, not axial and reciprocating.

7. lobe type 5. Inlet guide vanes guide the gas into the
first centrifugal impeller not the axial flow
8. screw type impeller.

4.19
Petroleum Open Learning

Check Yourself 4.5

1. piston rod reciprocating compressor positive displacement

2. splined rotors screw type positive displacement

3. inlet guide vane mixed flow continuous flow

4. crosshead reciprocating positive displacement

5. venturi tube ejector continuous flow

6. water ring liquid ring positive displacement

7. figure of 8 rotor lobe type positive displacement

8. drum and vanes sliding vane positive displacement

9. dry gas seal centrifugal continuous flow

1O. stator blades axial flow continuous flow

4.20
POL
Petroleum Open Learning

Process Flow & P&ID’s


(Process Engineering Drawings)
Part of the
Petroleum Processing Technology Series

OPITO

2
THE OIL & GAS ACADEMY
Process Flow & P&IDs
Process Engineering Drawings
(Part of the Petroleum Processing Technology Series)

Contents Page

BOOK 2
* Section 1 - Symbols 3
1.1 Structure
1.2 Pipeline Symbols
1.3 Other Lines
1.4 Pipeline Numbering and Identification
1.5 Product Designations
1.6 System Numbering
1.7 Pipeline Numbering
1.8 Pipeline Specifications
1.9 Insulation
1.10 Pipe Fittings
1.11 Valves
1.12 Valves used for “On/Off” Service
1.13 Valves used for “Control” Service
1.14 Valves used for “One-Way” Service
1.15 Valves used for “Special” Duties
1.16 Valve Actuators
1.17 Tanks and Pressure Vessels
1.18 Tank and Vessel Fixtures and Fittings
1.19 Filters
1.20 Pumps and Compressors
1.21 Metering Devices
1.22 Heat Exchangers
1.23 Other items of Equipment
1.24 Equipment Identification
1.25 Instrument

* Section 2 - Practical Application of Symbols 40


2.1 Plate Type Heat Exchanger
2.2 Valve Interlocks

* Section 3 - Piping and Instrument Diagrams 44

Figures 7 to 17 to be used with Book 1 Sections 2, 3 and 4

1.
Section 1 - SYMBOLS
1.1 STRUCTURE
In this Section we will look at the different symbols which may be used on Piping and Instrument
Diagrams (P&IDs) (also called Process Engineering Drawings). Although BS1553 provides the
specifications for symbols, you will find many variations in the different P&IDs you will come across.
The symbols given in this appendix include the BS1553 specifications, and some of the most common
variations of symbols used in the oil and gas industry.

The Appendix has been broken down into different categories. They are :

• pipeline symbols

• pipeline numbering and identification

• pipe fittings

• valves

• valve actuators

• tanks, separators and common pressure vessels

• filters

• pumps and compressors

• metering devices

• heat exchangers

• other items of equipment

• equipment and instrument identification

In each section there is an example of the British Standard symbol, an example of any common variation
of the symbol, and a brief description of the main points of the item.

After studying this Appendix, and applying the knowledge learned in the POL Unit, you should be able to
navigate your way through any P&ID.

It should be appreciated that P&IDs are NOT scale drawings. However, the actual pattern of pipeline
connections, pipe fittings, valves, instruments etc WILL be accurate.

2.
1.2 PIPELINE SYMBOLS
In general, the symbols used to identify the different types of pipe will be as laid out below :

Usually includes all piping above 2” diameter. In the diagram


the flow is indicated as being from left to right.

Usually includes all piping less than 2” diameter and all utility
piping. The direction of flow is not normally indicated.

Most control valves are pneumatic (air), powered by


Instrument Air.

Most wellhead valves and pipeline Emergency Shutdown (ESD)


Valves are hydraulically powered. The hydraulic fluid may be
water or oil based.

Flow from the horizontal pipe joins the flow in the vertical pipe.

The two pipes are crossing each other on the drawing.

At the indicated point the pipeline material specification


changes.

At the indicated point the pipeline leaves one Module or Plant


Area (M8) and enters another Module or Plant Area (M10).

3.
1.3 OTHER LINES

A low voltage electric signal. Usually to and from controllers or


instrument switches.

A signal to and from computer based instruments.

1.4 PIPELINE NUMBERING AND IDENTIFICATION


All pipelines will have a unique identification number. As a general rule the identification number will
indicate :

• the pipe diameter Note : ID size given up to 12” dia. OD size given for over 12”dia.

• the product which is carried within the pipeline

• the system number in which the pipeline is installed

• the pipeline sequential number which identifies the particular pipeline within each system

• the pipeline specification (ie the pipeline pressure rating and the material from which the pipeline
is made)

• the type of insulation applied to the pipeline

4.
1.5 PRODUCT DESIGNATIONS
A few of the more common pipeline Product Designations are listed below, with a few alternative types of
identification: -

• PL = Process Liquid may also be HC = Hydrocarbons, PO = Produced Oil

• PG Process Gas

• FG = Fuel Gas ; -

• Al = Instrument Air, IA sometimes used

• AP = Plant Air, PA sometimes used

• CW = Cooling Water, may also be CM = Cooling Medium, SW = Seawater

Figure 17 provides a comprehensive list of product designation codes.

1.6 SYSTEM NUMBERING


The different systems within each process are normally identified by a number, eg The Wellheads and
Manifolds System may be System 01. The Crude Oil Separation System may be System 02. The
Produced Water Treatment System may be System 12. The Instrument Air System may be System 62. An
instrument air line which passes through the crude oil separation area and the produced water area, will still
be designated a System 62 line.

1.7 PIPELINE NUMBERING


A different number is usually given for each length of pipeline. The number will usually change when
something happens to change the nature of the pipeline, eg When the pipeline specification changes
across an emergency shutdown valve, when the diameter of the pipeline changes across a reducer etc.

1.8 PIPELINE SPECIFICATIONS


The method of identifying the pipeline specification can be extremely complex, due to the large number of
pipeline materials and pressure ratings. A common system is where letters indicate the pipeline material
and a specific number indicates the pressure rating. A few examples are :

• CS1 = Carbon Steel - Schedule 40

• SS2 = Stainless Steel - Schedule 80

• CU1 = Copper - Schedule 40

5.
1.9 INSULATION

Pipeline insulation, when provided, will mainly be for:

• H = Full Heat Conservation

• HE = Full Heat Conservation with electrical trace heating

• Z = Cold Conservation

• P = Personnel Protection

• F = Frost Protection

• FE = Frost Protection with electrical trace heating

• A = Acoustic Protection

• R = Fire Proofing

Combinations of insulation class may also be used. A pipeline designated as insulation class P/A would
indicate that the pipeline required personnel protection insulation and acoustic protection.

1.10 PIPE FITTINGS


Any item of equipment which can be attached or connected into a pipeline may be classed as a pipe
fitting. The majority of the examples given below will be found somewhere on most oil and gas production
and treatment plants. A few specialised items have also been included.

A flange is a fitting which is welded or screwed on to the end of the pipe. The
flange allows the pipe to be joined up to another pipe, a pipe fitting or item of
equipment. If an item of equipment is shown without a flange attachment it most
often indicates that the item is welded or screwed into the pipe.

The flanges are joining two pipes. A gasket placed between the two flanges
ensures that the joint is sealed. Most pipe fittings and items of equipment are fitted
into the pipeline between two flanges.

Insulating gaskets, bolt sleeves and washers are installed to insulate one flange
from the other. Fitted to sections of pipeline which are protected from corrosion by
impressed current cathodic protection systems, or where a difference in metallurgy
could start the corrosion process, eg Carbon steel pipe / bronze valve.

Used to terminate a pipe in situations where there are no plans to extend or fit
anything to the end of the pipe in the foreseeable future.

Used to terminate a pipe in situations where there may be a reason to extend or fit
something to the end of the pipe in the future.

6.
Used to terminate a pipe in situations where there may be a reason to extend or fit
something to the end of the pipe in the future, but in situations where it may not be
possible to depressurise the pipe or take it out of service.

Used to terminate a pipe in situations where there will be a regular requirement to


connect a hose to the pipe.

Flat solid plate. Fitted in locations where a positive isolation is required to prevent
flow through the pipe.

Flat plate with central hole. Fitted in locations where the installation of a pipe
spade blind may be required.

A combined pipe blind and ring spacer. Fitted where frequent positive isolation
may be required. The example shows that the blind is normally in the open
position.

The example shows that the blind is normally in the closed position.

A Filter. Fitted where fine screening is required but where frequent changes are
not expected.

7.
Coarse filter. Fitted to protect equipment from construction debris (eg gloves,
welding rods etc).

Basic symbol for filters / strainers that are usually fitted with
mesh baskets or cartridge type filters. Sometimes provided
with pressure differential indicator (PD1).

Alternative symbol.

Coarse Filter. Also called a witches hat. Fitted to protect equipment from
construction debris (eg gloves, welding rods etc).

Fitted where a pipeline size change is required.

8.
1.11 VALVES
Before looking at the symbols used to identify the different types of valve we will consider the different
services which the valve may be required to perform. The three main service requirements are :

• ON/OFF SERVICE : For on/off service the valve should ensure full flow when fully open and a leak
free shut-off when fully closed.

• CONTROL SERVICE : In control service the valve should be able to control the flow of fluid through
the valve in accordance with the requirements of the design. The valve should also be able to give a
leak free shut-off when it is fully closed.

• ONE WAY SERVICE : Valves are required which ensure that flow is maintained in only one direction.
They should allow free flow in the direction required but give a leak free shut-off in the reverse
direction.

We can see from the various requirements of each service that we will require different types of valve. I will
explain the basic design features of the different types of valve.

On most P&IDs each type of valve is given a different symbol. The type of valve selected will mainly depend
on the operating conditions, product and type of service. Other factors such as cost, weight and
maintenance requirements will also be considered.

In some cases the different types of valve may not be indicated on the P&IDs. When this occurs, generic
valve symbols as shown below are sometimes used.

Valves which are normally in the OPEN position will not usually be coloured in. The
letters “NO” (indicating normally open) may also be printed next to the valve.

Valves which are normally in the CLOSED position will usually be coloured in. The
letters “NC” (indicating normally closed) may also be printed next to the valve.

9.
1.12 VALVES USED FOR “ON/OFF” SERVICE
The following valves are the ones normally selected for ON/OFF service. They generally allow a free flow
when the valve is fully open and a leak free shut-off when the valve is fully closed.

Most common of all valves. Used at all pressures. Gate valves are not very good
in dirty service as debris may damage sealing surfaces or accumulate at the
bottom of the valve to prevent closure. The gate valve must never be used for
control service as the flow across the valve will cut away the sealing surfaces.

Note : The gate valve symbol may be used as a common symbol for all valve
types.

Alternative symbol.

Used at all pressures. Some designs have sealant injection points to improve
shut-off capability. Not very good in dirty service because debris may damage
seals. Specialised internal designs may allow the valve to be used for flow control
with relatively low pressure drops.

Alternative Symbol.

Mainly used in medium and low pressure service. Most designs have sealant
injection points to improve shut-off capability. Not very good in dirty service
because debris may damage seals.

Mainly used in specialised high pressure service. A derivative of the ball


valve which has a turning / sliding action which pushes the ball against the
sealing surface. Must be facing correct way in line for best results.
Performs fairly well in dirty service.

Used mainly in low pressure dirty services. Care should be taken not to
over tighten the valve and damage the flexible diaphragm.

10.
1.13 VALVES USED FOR “CONTROL” SERVICES
The following valves are able to control the flow of fluid through the valve and, in most cases, give a leak
free shut-off when fully closed.

Used at all pressures. The most common of all control service valves. Good in
dirty service. Different internal designs can cope with all service and pressure
requirements.

Used at all pressures. A derivative of the globe valve. Used for very fine flow
control (eg sample points). Useless in dirty service.

Alternative symbol.

Mainly used at high pressures. Derivative of the globe valve. Reduced turbulence
within the valve gives better flow than the globe valve.

Used for high pressure drop service. A derivative of the angle valve.

Mainly used in low pressure and low pressure drop services. Some designs are
directional in order to improve sealing. Not to be relied upon for tight shut-off.

Alternative symbol.

11.
1.14 VALVES USED FOR “ONE-WAY” SERVICE
Valves used for ONE-WAY service are called CHECK VALVES, NON-RETURN VALVES or ONE-WAY
VALVES. The following valves ensure that flow is maintained in only one direction and, in most cases, allow
free flow in the direction required but give a leak free shut-off in the reverse direction. The three main
versions are :

• SWING CHECK VALVE : A flat circular plate is hinged so that it lifts to allow flow past the plate in one
direction but falls down to seal against the valve seat when the flow is reversed. The Swing Check
Valve is the most common type of valve in ONE-WAY service.

• TILTING PLATE CHECK VALVE : A flat circular plate is hinged with a slight offset from the central
position. The offset position of the hinge results in the valve opening in one direction but sealing in the
opposite direction. Mainly used in high pressure, high flow gas service.

• BALL CHECK VALVE : A free moving ball is contained within a cage. The ball lifts away from the
seat to allow forward flow but falls back into the seat when the flow is reversed. Mainly used in low
flow liquid service.

• PISTON CHECK VALVE : A free moving piston slides up and down inside a cage. The piston lifts
away from the seat to allow forward flow but falls back into the seat when the flow is reversed. Mainly
used in low flow, high pressure liquid service.

Any of the above types may be spring loaded to assist the sealing function when the flow is reversed.
Swing check valves may also be fitted with :

• a device which allows the check valve to be screwed down to enhance the tight shut off capabilities,

• a hydraulic dampener which prevents slam shut closure.

Indicated flow is from left to right.

Alternative symbol. Indicated flow is from left to right.

Alternative symbol. Indicated flow is from left to right.

Main flow is indicated from left to right. Recycle flow is indicated vertically.
Used to provide centrifugal pumps with a discharge check valve which also
incorporates a minimum flow facility.

Check valve may be screwed down to enhance tight shut-off. Occasionally


found on high pressure pumps or compressors which share common
headers with other pumps or compressors. Indicated flow is from left to
right.

12.
1.15 VALVES USED FOR “SPECIAL” DUTIES
The following valves are designed to cope with certain special requirements. They are all specialised
derivatives of valves which have already been described.

Used to protect vessels and pipes from over pressure. Derivative of the
angle valve. Standard type is spring loaded to ensure that valve lifts at
pre-set pressure underneath the valve seat. Balanced models are available
which compensate for any difference in pressure downstream of the valve.

Alternative symbol.

Used to protect vessels and pipes from over pressure in high pressure high
volume services. A small spring loaded pressure relief valve (the Pilot)
activates to allow the main valve to open.

Used to protect tanks and low pressure vessels from over pressure and
vacuum conditions. Allows air to move into and out of the tank or vessel in
response to changing internal pressure. Derivative of the globe valve which
uses weighted valve seats.

Used to prevent excess flow to or from tanks and vessels. Usually


positioned where piping failures could occur which may have extremely
hazardous consequences (eg flexible hoses carrying hydrocarbons).

13.
Not strictly a valve but used for pressure relief service in a similar
manner to a pressure safety valve. Often positioned beneath PSV’s
to protect them from corrosive process fluids. The rupture disc is
also used as a last resort over pressure protection device in critical
services, such as the shell side of a shell / tube heat exchanger.

A single inlet is split into two outlets. The internal design may be :

ON/OFF - depending upon the valve position, the flow may be


through either outlet.
SHARED - depending upon the valve position the flow may be
through either or both outlets.

Two inlets may be diverted to two outlets. Specialised valve used on


meter prover loops.

May be derivatives of ball valves (as illustrated) gate valve or plug


valves. An internal section of the valve is sandwiched between two
independent seals. The internal section is connected to a bleed valve.
The bleed valve may be opened to drain the internal section, and
prove that there is no migration of fluid across the two sealing sur-
faces. Double block and bleed valves are most commonly found on
metering systems.

Used for maintaining levels in storage tanks such as, potable water,
diesel etc.

14.
1.16 VALVE ACTUATORS
Valve Actuators are the devices which move valves to the desired position. As the majority of valves are
hand actuated the most common valve actuator is the Process Operator! Other common actuators which
may be found are :

• DIAPHRAGM ACTUATORS : A flexible diaphragm is moved in and out by pneumatic or hydraulic


pressure. Almost all control valves are pneumatically powered diaphragm actuated valves.

• PISTON ACTUATORS : A piston is moved in and out of a cylinder by pneumatic or hydraulic


pressure. Almost all well safety valves and many sub-sea safety valves are hydraulically powered
piston actuated valves.

• MOTOR ACTUATORS (Also called MOTOR OPERATED VALVES): An electric motor, suitably
geared. Large valves in non-critical service are often fitted with a Motor Actuator.

• SOLENOID ACTUATORS : A solenoid is an electro magnetic device with limited movement. Almost
all instrument air dump valves will be solenoid actuated.

The type of actuator will normally be indicated on the Piping and instrument Diagrams, but the type of valve
being actuated may not be specified. The main valve actuator symbols are indicated below :

If no symbol is attached to the valve it will also be a hand operated


valve.

May be pneumatic, hydraulic or electric motor actuated by manual


operation only.

Similar to above, but can be automatic operation as well.

The illustration shows that the actuator is pneumatically


powered.

15.
The illustration shows that the actuator is hydraulically powered.

The illustration shows the electric power signal to the motor actuator.

The illustration shows the electric power signal to the solenoid actuator.

As well as identifying the type of actuator the Piping and Instrument Drawing will also indicate the main
characteristics of the valve in the FAILURE mode :

SOV with a local reset facility to allow the signal to be reinstated to the
valve.

The valve can be opened by hand against the power being exerted by the
closing spring if required. Occasionally a hand actuator is fitted with a
clutch
which can be used to both open and close the valve. Beware of leaving
these valves in the hand actuated position when normal operations
are resumed as they will not operate in response to the automatic signal.

16.
The letters FO” indicate that, in the event of a pneumatic power failure
(ie an instrument air failure) the valve will move to the OPEN position.

An alternative signal. The arrow indicates that the valve will move to the
OPEN position.

This PCV uses the flowing product to act on the diaphragm to control the
downstream pressure requirements.

Same as above but controlling upstream pressure requirements.

17.
The configuration indicated below is typical for a diaphragm operated control valve in critical service.

Under normal operating conditions :

the ESD system ensures that there is a supply of electrical power to the solenoid operated valve

the solenoid valve is energised and in the normal position

the instrument air (I/A) supply is routed to the diaphragm valve through the solenoid operated three
way valve

(The normally closed section of the three-way valve is shaded in to indicate that the normal flow is
through the two open sections).

the control valve is in the normal position.

If the ESD system is activated :

the ESD system removes the electrical power to the solenoid operated valve

the solenoid valve is de-energised and moves to the failure position

the three-way valve changes position to :

- close off the supply of air from the instrument air system, and

- vent the instrument air from the diaphragm actuator

(The curved arrow shows the route the air takes when the solenoid valve is in the failure position),

the control valve moves to the failure position

(In the example, the control valve will fail to the CLOSED position, as indicated by the downward
pointing arrow).

Study this system carefully. Ensure that you understand the relationship between the ESD System,
the Solenoid Operating Valve and the Control Valve.

18.
1.17 TANKS AND PRESSURE VESSELS
We will now look at a few of the tanks and pressure vessels used in the oil and gas industry. We will start by
looking at the various types of tanks and then move on to pressure vessels and a few specialised vessels.
The examples I have given cover most of the varieties of tanks and pressure vessels which may be
encountered.

Seldom used in the oil and gas industry, and then only in water
service. Open pits may be found on drilling rigs in mud service.

Used offshore for the bulk storage of liquids. Used onshore for the
bulk storage of low or non-volatile liquids. Cone roofed tanks in
flammable liquid service are often gas blanketed. Often constructed
with a cone bottom or a sump to allow complete emptying of the
contents.

Seldom (if ever) used offshore. Used onshore for the bulk storage of
volatile liquids. The roof floats on top of the product and reduces
product losses by evaporation. (The first 1,000,000 barrel tanks ever
constructed were floating roof tanks).

Used in low, medium and high pressure storage or process services.


As a general rule when the vessel is in process service the function
will be part of the name, eg separator, knock-out drum, surge
drum etc. When the vessel is in storage service it is generally called
an accumulator or a bullet.

19.
Used in low, medium and high pressure process services, eg knock-out
drum, surge drum etc. Very occasionally used in storage service.

Used for the storage of liquefied low vapour pressure gases (eg butane),
sometimes referred to as a Norton Sphere. Not found offshore.

A vertical pressure vessel used for all types of distillation, fractionation,


rectification and stripping services. The most popular types of trays are
valve trays but sieve trays and bubble cap trays may also be used.
Used extensively offshore in water deaeration and glycol dehydration
systems, and onshore in refineries and gas processing plants, for the
separation of a wide range of hydrocarbon products.

A vertical pressure vessel used for all types of distillation, fractionation,


rectification and stripping services. The most popular types of packing
are raschig rings, pall rings, ceramic or plastic balls and berl saddles.
Used extensively offshore in water deaeration and glycol dehydration
systems and onshore in refineries and gas processing plants for the
separation of a wide range of hydrocarbon products.

20.
A horizontal pressure vessel which utilises gravity and a (relatively) long residence time (3 minutes) to
separate gas and water from produced oil. The separated water is retained to the left of an internal weir.
The separated oil flows over the internal weir to the oil outlet. The two liquid outlets are fitted with vortex
breakers which prevents oil being drawn into the water outlet stream, and gas from being pulled into the
oil stream. The separated gas leaves the top of the vessel after passing across a de-mister pad. The de-
mister pad removes any entrained droplets of oil.

A vertical pressure vessel which utilises gravity, centrifugal force and a


reduced upward velocity to separate a small amount of liquid from a gas
stream. The inlet is deflected around the inner walls of the vessel to create
centrifugal force. The liquids strike the vessel wall and then drain down into
the bottom of the vessel, where they are removed. The liquid outlet is fitted
with a vortex breaker, which prevents gas from being pulled into the liquid
stream. The gas leaves the top of the vessel after passing across a
de-mister pad. The de-mister pad removes any entrained droplets of oil.

A vertical, or horizontal, pressure vessel which uses centrifugal force.


Cyclones may be used to separate :

- two immiscible liquids of different densities (eg oil/water separation)

- solids from a gas (eg dust extraction from a boiler flue)

- gas from a liquid (eg mud de-gasser)

- solids from a liquid

The oil/water cyclones used to separate oil from produced water are called
hydro-cyclones. They are becoming more popular in produced water
systems.

21.
L18 TANK AND VESSEL FIXTURES AND FITTINGS
All tanks and vessels are normally fitted with at least one manway / access
hatch to provide access for personnel. Some are straightforward flanged
connections, some are hinged, and others are provided with small cranes to
make them easier to remove. As a general rule they are 24” in diameter and
are fitted with an internal grab handle to make access easier.

Tanks are often fitted with propeller mixers / agitators to ensure that the tank
contents are kept mixed, or to keep any solids in suspension.

1.19 FILTERS
In the section on pipe fittings we saw four types of coarse filtering devices called strainers. These are also
filters. I have classed them as pipe fittings because they are usually fitted as part of a pipeline, rather
than as an individual item of equipment. Below we can see a number of filters which are fitted as items of
process equipment.

An extremely coarse filter which is used to prevent large items of debris


entering a pipeline. Strum boxes may be fitted to the inside of dirty service
tanks. They are most often found on the inlets to pumps in raw water
service which take water from the sea or from a river.

Often found in raw water filtration service, downstream of the supply


pump. Can be cleaned whilst still in service. A motor driven internal
scraper revolves inside the basket and diverts filtered water back over
the filter mesh to the backwash outlet. The cleaning cycle is usually
activated by a timer or high differential pressure. The flanged top
allows the filter basket and cleaning head to be maintained.

22.
Extremely efficient method of filtration. Found in all types of
service where fine filtration is required. Filter medium is graded
layers of coarse and fine sand (sand filter) which may be
overlaid with a layer of anthracite (dual media filter). The filter
is cleaned by backwashing with filtered water. Some models
have a scouring system which injects air or gas into the
backwash stream to increase the cleaning efficiency. A
manway allows access for the filter media to be changed.

Most often used where extremely fine filtration is required. The


fluid to be filtered flows across a set of cartridges. The
cartridges normally consist of a metal or plastic support cage
which is wrapped with layers of fine cloth or fibres. The filter
cannot normally be backwashed or cleaned whilst it is in service
and is therefore most often used as a polishing filter. The
flanged top connection allows the filter cartridges to be
replaced.

23.
1.20 PUMPS AND COMPRESSORS

The most common pump used in the oil and gas industry. The pump may
be used in almost any service. Capacity may range from a few cubic metres
per hour to around a thousand cubic metres per hour. The centrifugal pump
is used mainly in constant pressure/variable volume services. Extremely
high pressures can be achieved by the use of multi-impeller pumps.
Regardless of the size of pump, the symbol almost always remains the
same.

Alternative symbol. In this instance the pump is being driven by an electric


motor as indicated by the letter “M”.

Alternative symbol. In this instance the pump is being driven by a gas


turbine. (If the main driver is not indicated on the Piping and Instrument
Driagram then almost certainly an electric motor is being used.)

A variation of the centrifugal pump. The sump pump is designed to hang


vertically below the level of the liquid being pumped. Smaller models are
often driven by small pneumatic (air powered) motors.

A multi-stage (often over 34 stages) centrifugal pump usually fitted to high


volume wells which will not flow without mechanical assistance. Sizes may
vary from as low as 200 barrels per day to over 20,000 barrels per day.

Electrical Submersible Pump (ESP)

24.
Used mainly in services which require relatively low flows at high
differential pressures.

An alternative symbol. For chemical injection service it may be provided


with a variable flow feature or in multi-head arrangements whereby six or
more pumps are powered by a single motor.

Used in low flow and relatively low differential pressure service. Excellent
type of pump for dirty services.

A positive displacement pump with a rotary action, (ie uses meshed gears,
screws or lobes to generate the pressure and flow). Often used in services
where a relatively high pressure is required, and the liquid to be pumped is
clean, (eg lubricating oil, seal oil etc).

Either rotary or reciprocating action. Installed where there may be a regular


requirement to unload barrels or empty small sumps.

Used exclusively onshore to pump low volume wells which will not flow
without mechanical assistance.

The reciprocating pump which is driven by the Beam Pumping Unit.

25.
Can be classed as a pump or as a compressor depending on the fluid
being handled. The device is used for a variety of low pressure / high
volume services. The motive power may be a high pressure gas or a
high pressure liquid (usually air and water respectively).

The most common type of compressor found in the oil and gas industry.
The compressor may be used in almost any service. Capacity may range
from a few cubic metres per hour to many thousands of cubic metres per
hour. The centrifugal compressor is used mainly in constant pressure /
variable volume services. Extremely high pressures can be achieved by
the use of multi-impeller compressors. Regardless of the size of
compressor, the symbol almost always remains the same.

Used mainly in services which require relatively low flows at high


differential pressures.

Alternative Symbol.

A low pressure / large volume compressor used for such services as


ventilation or air conditioning.

Normally used in similar service to the centrifugal blower where a higher


pressure is required or where the suction may be under a partial vacuum.

A positive displacement compressor with a rotary action, (ie uses meshed


gears, screws or lobes to generate the pressure and flow). Often used in
instrument air or plant air where a relatively high differential pressure is
required for low flows.

26.
It should be noted that in a number of cases the symbol for a particular pump and a particular compressor
is identical, (eg reciprocating pump and reciprocating compressor.) When this occurs reference will have to
be made to the identification lettering of the equipment to establish the type of equipment. As a general rule,
pumps will be identified with the letter “P” and compressors will be identified with the letter “C” or “K”.

As already indicated, the type of equipment used to drive the pump or compressor may be indicated. The
method of identification may be as illustrated below.

Often used to drive emergency generators, firewater pumps etc.


The Piping and Instrument Diagram may also include such items of
equipment as fuel tanks, exhaust spark arrestors etc.

Often used to drive power generation facilities, large capacity


centrifugal pumps and compressors and occasionally emergency
generators. The Piping and Instrument Diagram may also include
fuel gas supply equipment, exhaust waste heat exchangers etc.

1.21 METERING DEVICES

Used in both liquid and gas service. The differential pressure across the restriction is
measured and used to calculate the amount of fluid flow. Light in weight,
relatively cheap to produce, easy to install and maintain. The orifice plate is the
most common type of metering device in use.

Used in both liquid and gas service. The differential pressure across the
restriction is measured and used to calculate the amount of fluid flow. More
accurate than the orifice plate but heavier and more expensive. Mostly used
where a high pressure drop across the measuring device cannot be tolerated
(eg compressor suction lines).

Used in both liquid and gas service. Uses the ram effect of the fluid hitting
the end of an open pipe to generate a differential pressure which is
measured and used to calculate the amount of fluid flow. Not very accurate
when compared to the orifice plate and the venturi but can cope with large
variations of flow. Often used in flare headers.

27.
Used in liquid service. Extremely accurate. Individual compartments fill and
empty as the liquid passes through the meter. The number of compartments
filled and emptied gives an accurate measure of the liquid passing through
the meter. Positive displacement meters are used on garage forecourts and
are used to calibrate meter prover loops.

Used in both liquid and gas service. The fluid flow spins a turbine (ie a
propeller). The number of times that the turbine rotates is an indication of
the amount of fluid passing the turbine. The rotation of the turbine is
measured and the fluid flow calculated from the measurement.

Used in both liquid and gas service for the accurate measurement of small
flow rates. The fluid flows upwards through a conical tube. A ball or small
conical weight is suspended by the flow. The flow is measured in relation to
the height at which the ball or weight is suspended.

1.22 HEAT EXCHANGERS


As the name implies, a HEAT EXCHANGER is an item of equipment which is specifically designed to
exchange heat between two substances. Heat exchangers are most often named in accordance with their
function. They will be called coolers, heaters, chillers, reboilers etc, depending upon their function.

With shell and tube heat exchangers, the following general rules will apply :

• the high pressure fluid will be routed through the tube side of the heat exchanger

• the fluid most likely to cause fouling will be routed through the tube side of the heat exchanger

• the most corrosive fluid will be routed through the tube side of the heat exchanger

In situations where both fluids fall into one or more of the above categories the designer will compromise to
give the best operating results. An example of this situation is where a high pressure gas is being cooled by
seawater. The gas is at high pressure and should be routed through the tube side of the heat exchanger. The
seawater is corrosive and likely to cause fouling and should also be routed through the tube side of the heat
exchanger. In this particular case the high pressure gas would probably be routed through the tube side of
the heat exchanger.

Most common type of heat exchanger. The tubes are


indicated by the kinked line. The shell is indicated by
the circle.

28.
Alternative symbol. This type of exchanger may also be
called a “U” Tube Heat Exchanger.

Two shell and tube heat exchangers are often


banked together. This system is often used where
a single off the shelf heat exchanger would be too
small and / or where room is at a premium.

With this particular kettle type heat exchanger, the


shell side fluid enters as a liquid and is partially
vaporised within the shell. The vapours leave the
shell side from the top. The remaining liquid flows
over an internal weir and leaves the shell under
level control. Reboilers are normally this type of
heat exchanger.

29.
With this kettle type heat exchanger, the shell side
fluid enters as a liquid and is fully vaporised within
the shell. The vapours leave the shell side from
the top. Chillers are normally this type of heat
exchanger.

Used in a variety of services where the ambient air


temperature provides sufficient cooling for the
process. The tubes are normally wrapped with
light alloy fins to assist in the heat transfer
process. The term forced draft is used because
the air is pushed (forced) across the tubes.

The term induced draft is used because the air is


pulled (induced) across the tubes.

30.
To control the heat exchange rate on an air cooler the flow of air across the cooler is increased or
decreased. The easiest way to do this is for the Operator to switch the fan on and off. This gives very
coarse control. Other control methods are illustrated below.

A temperature controller adjusts a set of louvres to


increase or reduce the amount of air flowing across
the tubes.

A temperature controller adjusts the speed of the motor


to increase or reduce the amount of air flowing across
the tubes.

A temperature controller adjusts the angle of the blades


on a fixed speed fan to increase or reduce the amount of
air flowing across the tubes.

Electric heaters may be provided where the amount of


heat required is extremely low or in areas where a naked
flame may be hazardous. Electric heaters are commonly
controlled by a device called a thyristor. The thyristor is
basically a very rapid switching mechanism which can
adjust the amount of time that the heater is switched ON
or switched OFF.

31.
Commonly found in low pressure services. This type of
heat exchanger is extremely efficient and is therefore
relatively small and light for the amount of heat which can
be exchanged. Because of these features the plate heat
exchanger is becoming very popular offshore. It is prone
to fouling but can easily be dismantled for cleaning.

Used for heavy duty service both onshore and offshore. May
be used to provide heat for a wide variety of purposes from
steam generation to reboiler service. The fired heater may burn
a variety of fuels ranging from gas to heavy fuel oil.

32.
1.23 OTHER ITEMS OF EQUIPMENT
There is a wide variety of equipment which may be found on an offshore installation or an onshore facility.
I have selected a few items to illustrate the variety of items which may be shown on almost any Piping and
Instrument Diagram.

Fitted to the ends of pipelines for the launching and


recovery of the pigs or spheres used to clean the
pipeline.

Fitted to the inlet of instrument air compressors, gas turbines, diesel en-
gines or any other location where a limited amount of filtration is required
prior to the air entering the process.

Fitted to tank vents, low pressure flares, gas vents or any other
location where a flammable atmosphere may be discharged to the air.
Flame arrestors work by removing heat from the flame front to prevent
the flame migrating into the pipe to which they are fitted. If the flammable
atmosphere is ignited, the flame arrester prevents the flame from back
flashing into pipes and vessels.

May be an identical symbol to that used for a flame arrestor. Fitted to the
exhausts of internal combustion engines such as diesel engines on emer-
gency generators, mobile compressors etc. They prevent the emission of
sparks from the exhausts and enable the engines to be used in potentially
hazardous areas.

Fitted wherever there is a need to reduce or muffle the noise of an


expanding gas. They may be extremely large (eg when fitted to the exhaust of
a gas turbine) or very small (eg when fitted to the exhaust of an air operated
valve).

Fitted upstream of meters (especially turbine meters) to reduce the swirling of


the fluid being measured and thereby increase the accuracy of the metering.

33.
1.24 EQUIPMENT IDENTIFICATION
Each item of equipment, and each instrument, will have a unique identification. The identification will
normally be a combination of letters and numbers. The most common uses of identification letters are
shown below. As with the symbols, the identification of the equipment may differ from project to project.
On one system item “E-101” may refer to Heat Exchanger 101 on another project “E-101” may refer to
Engine 101. I have illustrated the most popular usage of the letters.

It should be noted that:

• First is the most popular usage and Second is an alternative usage.

• A Special Item is almost anything which does not appear regularly on the Piping and Instrument
Diagram (eg an insulating gasket, a temporary strainer etc).

• In some instances two letters may be used (eg GT = Gas Turbine or DE = Diesel Engine).

34.
1.25 INSTRUMENT IDENTIFICATION
On the Piping and Instrument Diagrams the instruments are identified by their Location, by their Function
and by their Number. The basic identification symbol for an instrument is a circle which encloses the other
information.

The actual point at which the instrument is connected to the process will be indicated by a line between the
circle and the process. The actual location of the instrument itself will be indicated by the design of the circle.

eg pressure gauge, sight glass, thermometer etc.

eg compressor lube oil pressure gauge, pump suction valve status light etc.

eg compressor low lube oil pressure switch, pump suction pressure switch etc.

eg vessel level indicator controller, ESD valve position light etc.

eg compressor anti-surge by-pass switch, pump low flow shutdown by-pass


switch etc.

Contained within a Computerised System (ie a Central Computer Control


and Monitoring System [CCCMS] or a Supervisory Control and Data
Acquisition [SCADA] System.): eg vessel level, valve status, compressor speed
etc. The information can normally only be accessed via a computer VDU screen or
a print-out facility.

35.
The Function of the instrument is indicated by a series of letters contained within the top half of the circle.
Before we look at a few examples take the time to study the matrix laid out below. The matrix identifies the
most common usage of letters which are used as identifiers.

36.
From the matrix you can see that:

• a First Letter “L” identifies a Level,

• a Second Letter “G” identifies a Gauge.

Therefore “LG” identifies a Level Gauge.

We can also see that:

• “LALL” is a Level Alarm Low-Low, and

• “FIC” identifies a Flow Indicator Controller.

Try a few combinations yourself before moving on to the next page.

37.
In the section on piping I indicated that there was a System Unit Number for each pipe. In most instances
the System Unit Number will be carried on into the instrument identification numbering system. A few
examples are indicated below :

Locally mounted pressure gauge, number 01, installed in System 10.

Local control panel mounted temperature indicator, number 18, installed in


System 20.

Control room panel mounted flow indicator controller, number 10, installed in
System 20.

Computerised level indicator controller, number 02, installed in System 40.

38.
Section 2 - PRACTICAL APPLICATION OF SYMBOLS

2.1 PLATE TYPE HEAT EXCHANGER


We will now look at a small section of a P&ID to see a practical application of some of the symbols. Take the
time to study Figure 18 on page 42, before carrying on.

The first thing we can see is that the Piping and Instrument Diagram is of a Plate Type Heat Exchanger ~,
numbered “E-1001 A”. This tells us that it is in “System 10”. From the letter “A” we can assume that there is
at least one other heat exchanger in identical service, (ie somewhere else is heat exchanger “E-1001B” and
maybe “E-1001C”, “E1001D” etc.)

The piping commodity letters “PO” indicates that the Production Header, and all associated pipework, is in
Produced Oil service. E-1001 A is supplied with heating medium from a Heating Medium Supply Header
which indicates that the function of E-1001 A is to heat up the produced oil.

The temperature of the oil leaving E-1001 A is measured by a locally mounted temperature element
(TE-1014). The electronic signal from TE-1014 goes to a control room panel mounted temperature indicator
controller (TIC-1014). The temperature indicator controller is set to control the produced oil temperature at
65°C. The electronic signal leaving TIC-1014 goes to a locally mounted temperature relay (TY-1014). The
temperature relay converts the electronic signal to a pneumatic signal (as indicated by the letters “I/P”). The
pneumatic signal flows through a solenoid operated valve (SOV-1014) to a control valve (TV-1014) to adjust
the amount of heating medium entering the heat exchanger.

The electronic signal from TE-1014 is also used to generate a temperature alarm high (TAH-1014) set at
70°C and a temperature alarm low (TAL-1014) set at 60°C.

A locally mounted temperature switch high-high (TSHH-1011) offers another level of protection against
over temperature. TSHH-1011 is set at 75°C. If the produced oil temperature reaches 75°C TSHH-1011 will
activate an alarm (TAHH-1011) on the control room panel and also send a signal to the Shutdown System.

Another input to the Shutdown System is generated by a locally mounted pressure differential switch low
(PDSL-1027). A low differential pressure could indicate that a leak has occurred within the heat exchanger. If
a low differential pressure occurs PDSL-1027 will activate an alarm (PDAL-1027) on the control room panel
and also send a signal to the Shutdown System.

If the Shutdown System is activated by TSHH-1011, or by PDSL-1027, the power to SOV-1014 will be
removed. SOV-1014 will fail to the vent position (as indicated by the small curved arrow). The air from
TY-1014 would be isolated and the air would be vented from TV-1014. TV-1014 would fail to the closed
position as indicated by the letters “FC”

Other features include:

• Spectacle blinds are fitted on the produced oil and heating medium lines to allow E-1001 A to be
isolated for maintenance.

• Pressure relief valve (PSV-1073) set to relieve at 12 barg pressure. PSV-1073 protects the produced
oil side of E-1001 A against over-pressure.

39.
We have seen that TV-1014 is indicated as being a fail closed valve by the letters “FC”. The valve
downstream of PSV-1073 is indicated as being “LO”. This indicates that the valve must be locked open to
ensure that PSV-1073 can operate properly at all times. Valves may be indicated as being :

• FC = Fail Closed (when the motive power is removed the valve moves to the closed position)

• FO = Fail Open (when the motive power is removed the valve moves to the open position)

• FIS = Fail In Situ (when the motive power is removed the valve stays in the last position requested
by the controller)

• LO = Locked Open (a physical barrier prevents the valve from being closed)

• LC = Locked Closed (a physical barrier prevents the valve from being opened)

• CSC = CAR Sealed Closed

• CSO = CAR Sealed Open

CAR sealed is an abbreviation of Customs and Revenue sealed. It comes from situations where
Customs and Revenue Officers seal valves to ensure that they are not operated without authorisation.
(eg whisky distilleries, bonded stores etc). The name has been adopted by the oil and gas industry to
indicate those valves which should only be moved from the sealed position in an emergency situation
or with the proper authority.

40.
41.
2.2 VALVE INTERLOCKS
Another feature which may be encountered is where two or more valves are interlocked. The most
common example of interlocked valves is to be found where two PSV’s are used to protect an item of
equipment (eg a vessel or pipeline). To ensure that the equipment is protected at all times, the interlock
system is designed such that at least one PSV is operational at all times. In the illustration :

• Valves A & B and Valves C & D may all be open.

• Valves A & B may be closed only if valves C & D are open.

• Valves C & D may be closed only if valves A & B are open.

42.
Section 3 - PIPING AND INSTRUMENT DIAGRAMS

Figures 7 to 17 to be used with Book 1 Sections 2, 3 and 4.

Figure 7 - PL - 0101 - 01 DP PLATFORM - MAIN DECK LEVEL 1 - PLOT PLAN

Figure 8 - EL - 0206 - 01 UQ PLATFORM ELEVATION - LOOKING NORTH

Figure 9 - FD - 0002 - 01 PROCESS FLOW DIAGRAM - SEPARATION - YEAR 1

Figure 10 - FD - 0002 - 02 PROCESS FLOW DIAGRAM - GAS TREATMENT AND


COMPRESSION - YEAR 1

Figure 11 - PD - 0002 - 01 P&ID - TOPSIDE PRODUCTION WELLHEAD TYPICAL

Figure 12 - PD - 0016 - 01 P&ID - FIRST STAGE PRODUCTION SEPARATOR

Figure 13 - PD - 0021 - 01 P&ID - OIL BOOSTER PUMP - P-0101A

Figure 14 - PD - 0030 - 01 P&ID - LP COMPRESSOR SUCTION COOLER AND DRUM - TRAIN A

Figure 15 - PD - 0031 - 01 P&ID - LP COMPRESSOR - TRAIN A

Figure 16 - PD - 0001 - 02 GENERAL LEGEND FOR P&IDs

Figure 17 - PD - 0001 - 01 GENERAL LEGEND FOR P&IDs

43.
POL
Petroleum Open Learning

Produced Water
Treatment
Part of the
Petroleum Processing Technology Series

OPITO
THE OIL & GAS ACADEMY
Petroleum Open Learning

Designed, Produced and Published by OPITO Ltd., Petroleum Open Learning, Minerva House, Bruntland Road, Portlethen, Aberdeen AB12 4QL

Printed by Astute Print & Design, 44-46 Brechin Road, Forfar, Angus DD8 3JX www.astute.uk.com

© OPITO 1993 (rev.2002) ISBN 1 872041 85 X

All rights reserved. No part of this publication may be reproduced, stored in a retrieval or information storage system, transmitted in any form or by any means, mechanical,
photocopying, recording or otherwise without the prior permission in writing of the publishers.
Petroleum Open Learning

Produced Water Treatment Petroleum Open Learning

(Part of the Petroleum Processing Technology Series)

Contents Page Visual Cues

* Training Targets 4 training targets for you to


achieve by the end of the unit
* Introduction 5
test yourself questions
* Section 1 - The Problems Associated with Produced Water 6 to see how much you
understand
The Mechanics of Water Production
Corrosion Problems
Scale Problems
Transportation Problems check yourself answers to
Disposal Problems let you see if you have been
thinking along the right lines
* Section 2 - The Basics of Produced Water Treatment 15
Primary Separation activities for you to apply
Gravity Separation your new knowledge
Coalescence
Short Distance Gravity Separation
Gas Flotation
Centrifugal Force Separation
Chemical Treatment summaries for you to recap
on the major steps in your
progress

1
Petroleum Open Learning
Petroleum Open Learning

Produced Water Treatment Petroleum Open Learning

(Part of the Petroleum Processing Technology Series)

Contents (cont’d) Page Visual Cues

training targets for you to


* Section 3 - Produced Water Cleaning Equipment 23 achieve by the end of the unit

API Separators
Plate Interceptors (or Separators)
Oil / Water Filters Coalescers test yourself questions
Gas Flotation Units to see how much you
Hydrocyclones understand
Use of Chemical Additives

* Section 4 - A Typical Produced Water System 36 check yourself answers to


let you see if you have been
Tilting Plate Separators thinking along the right lines
The Flotation Unit
Chemical Dosing Package
Produced Water Caisson
activities for you to apply
your new knowledge
* Test Yourself - Answers 47

summaries for you to recap


on the major steps in your
progress

3
Petroleum Open Learning

Training Targets

When you have completed this unit on Produced Water Treatment you will be able to :

• List the sources of produced water

• Describe the mechanics of water production

• Explain what problems can arise from the production of water

• Explain the basic principles which govern the separation of oil from produced water

• Describe the construction and operation of 5 types of oily water clean up facility

• Explain the requirement for chemical injection in a produced water treatment system

• Describe the flow of water and separated oil through a typical produced water treatment facility

Tick the box when you have met each target.

4
Produced Water Treatment Petroleum Open Learning

Introduction
In the vast majority of oil fields, water production unit, we will be looking at the produced water
becomes a problem as the field gets older. Towards handling system of an oil production facility. Before
the end of their useful lives some oil wells may be we examine a typical system, however, I think we
producing 95% of their total liquid as water. This should look at where the water comes from and the
produced water may be extremely salty and likely to problems it poses in a little more detail. So, I have
be of little value to the operator. It is removed from split the unit into four sections as follows:
the oil stream during primary separation and by
other facilities, and has then to be disposed of.
However, we are talking of a great deal of water in • In Section 1 we will look at the sources of
some cases. How do we dispose of it, and where do produced water and the problems which may
we put it? be encountered if we fail to treat it.

• Section 2 will cover the basic principles


involved in the treatment and clean-up of
Offshore, the obvious place would be into the sea. produced water.
Dumping this produced water directly from
separators into the sea, would however, soon have • In Section 3 we will examine the construction
the operator in trouble with the authorities. Even and operation of produced water clean-up
after initial separation the water still contains oil in equipment.
small amounts. Serious environmental pollution
would build up if oil contaminated water were to be • Finally, in Section 4, I will take you through a
dumped directly to the sea. typical produced water handling facility which
may be found on an offshore production
platform.
Onshore, disposal wells may have to be drilled, into
which the produced water can be injected for
disposal. This also may have its problems. Oil in
the water, or fine solids, could plug the injection
wells in a very short time. Although produced water treatment applies to both
onshore and offshore locations, I will be
concentrating on the offshore situation in this unit.
So the water which is produced with, and separated However, most of what I have to say would apply to
from, the oil in an oilfield must be cleaned prior to both.
disposal. This is what this unit is all about. In the
5
Produced Water Treatment Petroleum Open Learning

Section 1 - The Problems Associated with Produced Water


Water is present in some form, in most oil reservoirs The Mechanics of Water Production
before any production takes place.
Look at Figure 1 which shows a cross section through a typical water drive reservoir.
There are however, many different types of
reservoir. In one very common one, the oil
accumulates above large volumes of water, which is
usually salty. This water is what remains of ancient
seas from an earlier period of Earths history.

This body of water is called an aquifer, and the


reservoir is known as a water drive reservoir.

In addition, a considerable amount of water may also


be found as small droplets distributed throughout the
oil (and gas) in a reservoir. For reservoir
engineering purposes this water is called connate
water or interstitial water. We will just call it
formation water.

During production, further injection water may


be pumped into the reservoir to assist in pressure
maintenance.

Any of these types of water may eventually find their


way into the oil wells and be produced to the surface
along with the oil. It is all then called produced
water.

Before we look at the problems which can be caused


by this produced water, let us first consider how the
water gets into the producing wells,

6
Petroleum Open Learning

You can see that the oil lies above the aquifer and the The high pressure water in the aquifer,
PERMEABILITY
well is taking oil which is not contaminated with water. therefore, will tend to displace the oil
Permeability is a measure of the ability of a fluid to
The point at which the oil and water touch each other towards the low pressure areas
flow through the rock from one pore to another. In
is called the oil water contact. surrounding each well bore.
order for it to be able to do this, the pores must be
interconnected.
The oil is able to flow through the reservoir rock
towards the well because the rock is porous and Permeability is measured in d’arcys- named after
permeable. These are probably the two most a French engineer who studied the flow of liquids
important properties of reservoir rocks. through filters. He found that the flow increased in
proportion to the pressure increase. However he also Figure 2 on the next page, shows the
discovered that the flow was affected by the thickness, situation with just one producing well
POROSITY or viscosity, of the fluid. and one water injection well.
Porostiy is the property of the rock which enables it
to hold fluids within itself. The oil, gas and water are Generally there is a wide spread of permeability in
contained in tiny holes in the rock called pores. reservoir rocks.
Sandstone is a common reservoir rock. It is made up
So, the rock properties of porosity and permeability
of grains of sand which are cemented together at the
points where they touch. Between the sand grains are allow the oil to flow towards the producing wells. But
void spaces - the pores. what causes the oil to flow through the reservoir?

The ratio of the volume of the pores to total rock Let’s look at that now.
volume expressed as a percentage is the rocks
porosity. You are probably aware that fluids always flow from
areas of high pressure to areas of low pressure:
This means that, if you have a sandstone reservoir
with a porosity of 25%, for every 4m3 of reservoir rock, • The oil producing wells create areas of low
1m3 consists of holes and 3m3 solid sand grains. pressure in the surrounding reservoir rock as
the well is opened at the surface and oil flows
Another common reservoir rock is limestone. This is
into the well
a rather brittle rock which contains lots of tiny cracks
and fissures. These tiny cracks give the limestone its
porosity. • The aquifer is usually at a relatively high
pressure. In addition, the injection of
water into the aquifer is intended to maintain
the reservoir pressure
7
Petroleum Open Learning

If this situation remained constant there would be no


problem. However the situation does not remain
constant. Think about this and try to answer the
following Test Yourself question.

Test Yourself 1
As oil is removed from the reservoir what
will happen to the position of the oil water
contact ?

You will find the answer to Test Yourself 1 on


Page 47

8
Petroleum Open Learning

Over a period of time, as the oil water contact


approaches the well intakes, the water will start to
flow preferentially to the oil wells. This occurs
because the water is much less viscous than the oil
and therefore flows more easily through the rock,
bypassing the oil.

The water is said to finger through the oil.

Figure 3 shows water starting to follow these


preferential routes through the reservoir rock.

9
Petroleum Open Learning

Once water starts to break through to the The water cut from a particular well or field depends
producing wells, it tends to be produced in ever on a large number of factors. These include:
increasing amounts.

The ratio of water produced to total production is


Test Yourself 2 • The geology and porosity of the reservoir rock

called the water cut and is expressed as a • The size and, particularly, the vertical
percentage. To make sure you understand this, a) If a well produces 3975m /d of oil and
3 thickness of the reservoir
have a go at the following Test Yourself question. 795m3/d of water, what is the water
cut. • The degree of fracturing of the oil field

b) A well produces a total of 875m3/d • The position and depth of the producing wells
liquids and the water cut is 20%. What in relation to the oil water contact
is the oil production from this well.
• How long the reservoir has been producing oil
c) What is the water cut of a well if the
total production is 556m3/d., and the
oil production is 397m3/d Actual water cuts vary tremendously, of course, but
can be as much as 99%. Imagine a field which
produces a total of 15,900m3 of liquid per day with a
water cut of 60 %. This means that 9540m3 of water
are produced every day.

This can pose significant operating problems and


these are what we will look at now.

You will find the answers to Test Yourself 2


on Page 47

10
Petroleum Open Learning

Corrosion Problems One of the ways of reducing corrosion damage is to Transportation Problems
separate the water from the oil at the earliest
We said earlier that the aquifer water can be very opportunity and dispose of it. In fact this, together The produced oil may have to be transported from an
salty. Injection water, being in the main sea water, is with the separation of gas, is one of the first offshore location to a shore based refinery or tanker
salty as well. It follows then that the produced water processes in a production facility. This however terminal. There are two ways of doing this. If the field
will be salty also. In fact the saltiness, or salinity, gives rise to another problem - one of disposal. We is large and the economics justify it, the best way is by
of produced water is usually considerably more than will look at this shortly. pipeline to shore. However, some fields are too small
that of normal sea water. To put it in perspective: to justify the expense of a pipeline or are too far from
shore. In this case the oil is loaded into a tanker at
• fresh water from streams, lakes etc. usually Scale Problems the point of production via a tanker loading facility.
contains less than 0.2 % salt Either way, water in the oil to be transported can
Salts are initially dissolved in the water present in a cause problems:
• sea water has an average salt content reservoir. As conditions change when this water is
of 3.5% produced, the salts may be precipitated as solids • The obvious one we have looked at already,
and deposit as scale. that of corrosion. Salt water in pipelines or
• produced water can contain up to 15% salt tanker loading units can corrode facilities
This can reduce pipe diameters, plug vessels and rapidly. I don’t think I need to elaborate on that
Pure water in itself is not particularly corrosive. equipment which in turn can lead to lost production. at this time.
However, up to a point, the more saline it becomes Once again, removal and disposal of produced water
the more corrosive it is. can help prevent the problems of scaling. • If the oil is going down a pipeline, excess water
reduces the efficiency of the line, leaving less
If the produced water is allowed to pass through all space for oil.
the surface processing equipment to the oil
transportation system, it could cause considerable • Water being sent to a refinery with the oil can
corrosion damage to pipes, vessels and other cause serious upsets in the distillation process.
equipment. Refinery operators usually limit the amount of
salt and water which they will accept.
In fact, corrosion costs the petroleum industry
millions of pounds annually. It makes sense to try to • When loading oil to a tanker there are laid
reduce this expense. down limits of water in oil which it is permitted
to take. If more than, say, 0.5 % of the cargo
loaded is water, then the producing company
can face severe penalties.

11
Petroleum Open Learning

It would seem from the foregoing that one of the Sewers are not built for these amounts of water and For instance, discharging produced water into the
first things which we must do on a production facility would be overloaded, in addition to suffering sea in the UK sector of the North Sea is subject to
is get rid of the water. This is of course what pollution problems at the outfalls. compliance with the following conditions at present:
happens. On most installations water is separated
from the produced oil in the first process system. You may have thought of drilling wells and injecting • On average, discharged water must not
The separated water has no value and has to be the water back into a reservoir. This is in fact done. contain more than 30 ppm by weight of oil.
disposed of. But how? This brings us to our final But the water has usually to be treated before it can
problem, that of produced water disposal. be injected. It may have to be filtered and dosed • An oil content of up to 100 ppm is allowed in
with chemicals to make it suitable for injection. individual monitoring samples
Disposal Problems
If the production facilities are located offshore the • For a normal month of sampling, not
When trying to decide how to get rid of the water we problem of disposal may seem easier. Why not just more than three samples (4% a month) may
must consider first of all the location of the dump it into the sea? Unfortunately it is not quite exceed the limit of 100 ppm.
production facility. that simple.
• Regular monitoring of effluent discharged
Think for a moment and try to decide how you After initial separation, the produced water is still from each platform to the sea is a stipulated
would dispose of 1590m3 per day of produced water likely to contain a considerable amount of oil in the requirement
from a site on land. form of small droplets. The actual amount will vary
from installation to installation but could be of the • Samples should be taken at 0700 and 1700
You may have come up with one of the following: order of 150 ppm. hrs each day

• Dump the water into lakes or rivers (The unit ppm means parts per million. In other The figures of 30 ppm is the one currently in force and
words, in every million drops of liquid 150 of them is constantly under review. It is not inconceivable that
• Dump the water into sewers would be oil, the rest water.) it could be reduced even further at some future date.

Both of these solutions would be totally This may not seem very much, but if those
unacceptable. quantities were dumped into the sea, an oil slick So, bearing in mind that I said that we would
would soon form and pollution would occur. concentrate on an offshore location, it would seem
In the first case, pollution of the fresh water by the that our biggest problem is getting the oil-in-water
Salts in the produced water would cause damage to In most countries the removal of oil from produced concentration down to acceptable limits.
the environment and could destroy wildlife. Drops of water before dumping it into the sea is a legal
oil in the water would also cause considerable requirement. The quality of produced water This is what we will concentrate on for the rest of this
environmental pollution. disposed of in this way is subject to strict control. unit.

12
Petroleum Open Learning

Before going through the summary of this section,


try the following Test Yourself question,

Test Yourself 3
Are the following statements true or false?

True False
a) Permeability is a measure of the ability of a rock to
allow fluids to pass through it.

b) The oil water contact within a reservoir tends to move


down as the production from the field proceeds.

c) Seawater usually has a greater salt content than


produced water.

d) Produced water can cause loss of efficiency in


pipelines.

e) In the UK sector of the North Sea discharged water


must not contain more than 130ppm by weight of oil.

You will find the answers to Test Yourself 3 on Page 47

13
Petroleum Open Learning

Summary of Section 1
During this section I have tried to introduce you to We then moved onto the problems of water
the problems arising from the production of water production and saw that they could be
with oil. classified as :

• corrosion problems
We started by looking at the sources of produced
water and you saw that it can be from the aquifer, • scale problems
formation water or the injection water which is
used to maintain reservoir pressure. • transportation problems
We then looked at the mechanics of water
production and considered the rock properties of • disposal problems
porosity and permeability which allow fluids to flow
through a rock. We concentrated on disposal problems offshore
and I indicated that dumping water into the sea is
the easiest option but this is often governed by
You saw that the relatively high pressured water legislation. I gave as an example that the
underlying the oil pushes the oil towards the average oil in water content permitted to be
wellbores. However the water may eventually discharged into the UK sector of the north sea
start to finger through the oil and be produced in must not exceed 30 ppm.
ever increasing quantities. You discovered that
very large quantities of water may be produced In the next section we will go on to look at some
and I defined the ratios of oil and water production of the basic theory behind produced water
as the water cut. treatment. In particular we will concentrate on the
removal of oil.

14
Produced Water Treatment Petroleum Open Learning

Section 2 - The Basics of Produced Water Treatment


In this section we are going to look at some of the
methods which could be used to treat produced
water. The water can contain dissolved gases and
solids and also some suspended solid particles such
as sand.

However, we are going to concentrate on the


removal of oil from water, so that it can be dumped
to the sea. Let’s start with the separation of water
from the main oil stream. That is, after all, the first
part of the treatment programme.

Primary Separation
The total production from an oil field flows from the
wells to the primary separation system. The function
of this system is to separate the production into its
individual phases of oil, gas and water. The process
is carried out in large vessels - the separators. A
typical 3-phase separator is shown in Figure 4.

15
Petroleum Open Learning

The vessel is called a 3-phase separator because it What you have read in the answer to Test Yourself 4
separates the total flow stream into the three is exactly what happens in the separator. The water
individual streams of oil, water and gas. A 2-phase and oil separate due to a difference in their densities.
vessel would separate the stream into the liquid and
gas streams.
Test Yourself 4 Providing the oil and water stay in the vessel for a
sufficient period of time, the bulk of the water can be
I don’t intend to go through the construction and If you shake a mixture of oil and water in separated from the oil. This water is the produced
operation of a separator at this point. A programme a beaker and allow it to stand for a period water which has now to be disposed of.
on Oil and Gas Separation is also available in the of time, what will happen to the two
Petroleum Processing Technology Series. substances ? Although primary separation is quite efficient, oil may
remain in the water as small droplets. These have
Briefly, however, the oil, water and gas stream Can you explain your answer? also to be removed. We can now look at some ways
enters the vessel at the inlet and is deflected by the of doing that.
inlet deflector. The gas passes towards the gas
outlet via straightening vanes and mist extractor and
the liquids fall into the liquid accumulation section.

This is where the separation of oil and water takes


place. But how does it occur? Think about it for a
moment then answer the following Test Yourself
question.

You will find the answer to Test Yourself 4 on


Page 47

16
Petroleum Open Learning

Gravity Separation However, if the smaller droplets will not rise,


something else must be done to remove them. Why
The primary separation we have just looked at is an not try to combine them into larger drops, which will
example of gravity separation. Oil and water then rise? This is indeed done by using various
separate because of the difference in their density or types of coalescer.
specific gravity. Most crude oils are less dense than
water so they tend to float on top of water. Coalescence

Even where the amount of oil in the water is To coalesce simply means to join together or unite.
minimal, given sufficient time and under the right
conditions, the crude oil will float to the surface of the Entrained oil droplets in the water which are too
water where it can be removed. small to rise rapidly by gravity, can be coalesced in a
number of ways. One way is to pass the oily water
There is, however, a theoretical lower limit to the through a specially developed cartridge. This is
size of crude oil droplets which will rise freely made of a porous plastic medium such as
through the water. Oil droplets which have a polypropylene or polyurethane foam. When in use
diameter of less than, say, 5 microns will not rise the oily water flows to the centre of the cartridge and
through the water, but will stay in suspension out through the walls, where coalescence takes
indefinitely. A micron is one millionth of a metre. place.

In practical terms, the limiting droplet size in an oil/ The larger oil droplets then rise to the surface of the
water gravity separator is much higher, and in the water by gravity as before,
range of 50 to 150 microns. This is because of such
factors as turbulence, limited retention time, and Figure 5 shows the cartridge coalescer principle.
so on.

It would appear then that a certain amount of


produced water clean up can be done in a simple
tank where the water stays long enough for the oil
droplets to rise to the surface and be removed.

17
Petroleum Open Learning

Short Distance Gravity Separation Gas Flotation


Under constant conditions, a drop of oil Flotation is a process which has long been
suspended in water tends to rise at a fixed used for cleaning industrial waste water.
rate. This rate is called the terminal The operating principle depends on
velocity. The time required for separation increasing the buoyancy of entrained oil or
therefore depends on the distance the drop solid particles, enabling them to rise more
has to rise to reach the surface. freely through the water. This increase in
buoyancy is achieved by attaching gas
In order to speed up this process, various bubbles to the suspended particles.
types of equipment have been developed
which reduce the distance the particles The gas bubbles are generated by :
have to travel.
• dissolving gas in the water under
This type of equipment provides closely pressure and then releasing the
stacked parallel plates or tubes through pressure prior to entering the
which the water flows without cleaning unit
turbulence. As the oil droplets rise
within these confined spaces they have • mechanically introducing the gas into
only a short distance to travel before the water.
reaching a solid surface. where they
concentrate and coalesce. It is the second of these methods which is
more commonly used in the oil industry.
Figure 6 shows an end view of a simple
plate pack.

18
Petroleum Open Learning

Figure 7 shows oil droplets with gas bubbles In a flotation cell, the oil and gas mixture accumulates on the surface of the water
attached rising through water. as a layer of oily froth. This is skimmed from the top of the water to a channel which
directs the oil to a recovered oil system. The skimming may be over a simple
adjustable weir. Alternatively a system of paddles may be used to sweep the oily
froth continuously from the surface of the water.

Figure 8 shows a much simplified version of a flotation cell. I will describe this in
much more detail in Section 3.

19
Petroleum Open Learning

Centrifugal Force Separation This is the principle of a hydrocyclone, another Chemical Treatment
piece of equipment used in the separation of oil from
You are probably familiar with the principle of water. Sometimes emulsions form in the produced water
centrifugal force, but let’s just remind ourselves of it in which are very difficult to break down.
a rather simple way. Figure 9 illustrates this principle. An emulsion is a stable mixture of two or more
immiscible liquids, one dispersed in another, in the
Imagine a spinning disc, similar say to a record on a form of very small droplets. There are two distinct
record player. If you dropped a liquid onto the disc types of emulsion. They are:
near the centre it would be flung to the outer rim of the
disc. The force which causes this to happen is • Water-in-oil emulsions where a small amount
centrifugal force. of water is dispersed in a larger amount of oil

If the liquid was pumped into a container, in such a • Oil-in-water emulsions where a small amount
way that it was made to swirl within that container, the of oil is dispersed in a larger amount of water
centrifugal force would cause a vortex to be formed.
The first type is the more common, but oil in water
(A familiar example of a vortex is the cone-shaped emulsions can occur and may be a problem in the
whirlpool which forms above the plug hole when water treatment of produced water.
runs out of a bath).
In order to break down this type of stable emulsion,
If a mixture of two liquids, of different densities, was chemicals are injected. These chemicals, called
pumped into the container, centrifugal forces would demulsifiers, help the oil droplets to coalesce and
tend to separate the two liquids: separate from the water.

• the liquid having the lower density would Demulsifiers are usually used in conjunction with
migrate towards the middle of the vortex some other form of water clean up facility.

• the liquid with the higher density would migrate


towards the outside of the vortex
The oily water is introduced continuously to the unit. Before summarising Section 2, try Test Yourself 5,
If the two liquids are water and oil, it would be the oil A low pressure outlet is connected to the centre. which will help you bring together the basic
which would migrate towards the centre of the Through this outlet the oil is continuously removed. principles of water treatment we have covered in
container. Clean water is thrown towards the outside wall of the this section,
unit, where it leaves by a separate outlet.
20
Petroleum Open Learning

Test Yourself 5
In the first column of the table below I have listed the following terms: porous medium, plate pack, oil droplets rising,
demulsifier, finely dispersed bubbles, vortex. Each is associated with one or more methods of oil removal from
produced water. Put a tick in the appropriate column (s) to show which one(s).

Short Distance Centrifugal


Gravity Coalescence Gravity Gas Force
Separation Separation Flotation Separation

porous medium

plate pack

oil droplets rising

demulsifier

finely dispersed bubbles

vortex

You will find the answers to Test Yourself 5 on Page 48

21
Petroleum Open Learning

Summary of Section 2
In this section we have been looking at some of the basic principles which govern the separation
of oil from water in a water clean up facility.

First of all we considered the primary separation of the water from the main oil stream. You saw
that this was a simple gravity separation process. You also saw that gravity separation is the
basis of most produced water treatment facilities.

However, in order to speed up the process or make it more efficient you saw that other types of
treatment could be undertaken.

We considered:

• coalescence

• short distance gravity separation

• gas flotation

• centrifugal force separation

You also saw that chemicals may have to be injected into the produced water to assist in
separation, particularly if an oil-in-water emulsion has formed.

22
Produced Water Treatment Petroleum Open Learning

Section 3 - Produced Water Cleaning Equipment


In this section we are going to have a look at the However, the time available on an offshore production The produced water enters the unit on the left
construction and operation of equipment which can facility is very limited and residence times are hand side. As it enters the chamber it hits a small
be used to clean up produced water. A variety of extremely short. As API Separators require relatively stilling plate. The stilling plate distributes the
such equipment is available, but we will concentrate long residence times, there are very few of them found incoming liquids evenly over the width of the
on the most common: offshore. I have included the API Separator because separator and reduces turbulence and mixing.
they are very common in land installations and
• API Separators contain features which are found in other devices. Underneath the stilling plate is a sludge/debris
trap which will catch small solids as they sink. The
• Plate Interceptors (or separators) The API Separator is basically a very large open tank sludge outlet is normally designed so that it can
or pond which permits a long residence time for the oil easily be freed if it should become blocked.
• Filter Coalescers to separate from the water.

• Gas Flotation Units Figure 10 is an illustration of a typical API Separator.

• Hydrocyclones

The simplest of the above pieces of equipment is the


first on the list so let us start with this.

API Separators
As we have seen, the most common way of
separating oil and water is by the use of gravity
acting on the density difference between the two
liquids.

Time is also required for this process to work


effectively. Each separator is designed to retain the
liquid mixture within it until separation has been
accomplished. This time is known as the residence
time or retention time.

23
Petroleum Open Learning

The water flows: The adjustable oil skimmer is normally set at Coalescing devices provide a solid surface which
between 1/4th to 1/8th of an inch above the level of can be contacted by small oil droplets. An
• from the stilling plate the water. accumulation of these oil droplets creates a thick oil
film which becomes a source of large drops.
• above and below an intermediate baffle API Separators require careful adjustment of the Eventually enlarged drops of oil break loose from the
(water with oil in it will rise above the baffle skimmer to remove as much of the oil as possible, solid surface. These large drops separate from the
water with no oil in it will fall under the baffle) but without removing any water. Slight changes in water phase much faster than the original small
flow will raise or lower the height of water falling over droplets.
• under the final baffle the weir and, if the oil/ water interface is disturbed,
water could slop over with the oil. Coalescing surfaces come in two basic forms:
• over the outlet weir
As I said earlier, API Separators are not suitable for • Plate interceptors
• out of the separator via the water outlet. offshore applications. There, more efficient means
of oil and water separation are necessary. Facilities • Cylindrical cartridges (called “fixed media”
The level in the separator is controlled by adjusting are needed which are designed to reduce the cartridges)
the height of the outlet weir. residence time required for efficient oil/water
separation to take place.
The oil flows:
This reduction in residence time is important. For a
• from the stilling plate given flow rate of fluid the residence time can only
be extended by increasing the volume of the
• above the intermediate baffle separator. Offshore, where space is at a premium,
this is extremely difficult.
• to the surface where it forms a layer of oil on
top of the water One way of reducing the required residence time is Let’s look at a couple of plate interceptors first.
to include some form of coalescing device in the Two main types are in use, the parallel plate and
• into an adjustable oil skimmer water cleaning unit. the corrugated plate interceptor.

• out of the oil skimmer via the oil outlet

24
Petroleum Open Learning

Plate Interceptors (or Separators) Now follow the flow with me with reference to
Figure 11.
Figure 11 shows a typical parallel plate interceptor. Plate interceptors work on the
principle of short distance gravity separation, which we looked at in Section 2. • The produced water flows into an inlet
Have a look at the figure now and try to visualise the flow through the unit. chamber which is equipped with a sludge/
debris trap.

• Any gas in the produced water stream then


rises, and leaves the separator by a vent.

• The produced water then flows through the


parallel plate pack.

The pack consists of a number of parallel steel or


plastic plates connected together with small gaps
between them. It is set at an angle of about 45˚ to
the horizontal and the produced water flows
downwards, towards the right of the illustration.

Flow between the plates is far more streamlined than


in the inlet chamber. In addition, the distance
through which oil droplets have to rise before
reaching a surface is much smaller.

Oil droplets coalesce on the underside of the plates


and slide upwards and backwards against the flow of
the water. The oil then breaks free and rises to the
surface where it forms a layer on top of the water.
The oil is removed from the surface by a skimmer.

The water carries on towards the outlet chamber,


then doubles back over the outlet weir which
controls the height of the liquids in the separator.

25
Petroleum Open Learning

The water then flows away, under gravity, out of the


separator.

A parallel plate separator will often reduce the amount


of oil in the produced water from 5 000 ppm to :

• 100 ppm of oil (with a residence time of,


say, 10 minutes)

or

• 50 ppm (With a residence time of 30 minutes)

Their main disadvantage is that frequent cleaning is


required to remove solids which stick to the plates.

A variation of the parallel plate separator is the


corrugated plate interceptor which we can look at
now.

The plate pack is installed in the same manner as the


parallel plate interceptor pack, at an angle of 45°, and
the oil and water flows are identical.

Figure 12 illustrates the major difference between the


two types of separator - in this type the plates are
corrugated. In the tops of the corrugations are slots
which enhance the oil removal process.

Oil coalesces on the underside of the plates, and


accumulates beneath each corrugation. From there it
migrates backwards and upwards until, eventually, it
forms a layer of oil on the surface, as before.

26
Petroleum Open Learning

A corrugated plate separator will often reduce the Oil/Water Filter Coalescers In our example in Figure 13, a set of cartridges
amount of oil in the produced water from 5 000 ppm would be mounted on a deck-plate within a vertical
to 30 ppm oil with a residence time of 5 to 10 In an oil / water filter coalescer the oil / water pressure vessel. Note that, for simplicity, I have only
minutes. separation is achieved by coalescence of dispersed shown one cartridge in the vessel, and omitted the
oil droplets within specially designed coalescer deck-plate.
This type of separator has been adapted for use on cartridges.
offshore systems and has even been tried in high
pressure systems at the wellhead so as to separate
production water before dissolved gas is released.

The corrugated plates can be made of


polypropylene, polyvinyl chloride, stainless steel or
carbon steel.

As we have seen, both the parallel plate separator


and the corrugated plate separator have their
plates tilted at an angle of 45°. Because of this,
both types are often referred to as tilting plate
separators.

When you are satisfied that you understand the


construction and principle of operation of the plate
interceptors, we can move on to another type of
coalescer unit.

27
Petroleum Open Learning

The flow through the unit is as follows: Before moving on to the next part of this section,
have a go at the following Test Yourself question.
• oily water enters the base of the vessel

• is distributed to the cartridges

• flows radially outwards, through each


cartridge into the main part of the vessel
Test Yourself 6
As the fluid passes through the cartridges, the oil Fill in the missing word or words from the following sentences.
droplets are forced into close contact with each other.
They coalesce within the cartridge wall and rise from a) In an A.P.I. separator the produced water enters the unit and hits a small ................... .......................
the outer cartridge surface to the top of the vessel. which distributes the incoming liquids.
These droplets then form an oil layer which is
discharged through the oil outlet. (Have another look b) In a plate interceptor the inlet chamber is equipped with a ...............or ...................trap.
at Figure 5 on Page 13.)
c) The outlet ..................... controls the height of the liquids in the unit.
An oil/water interface level is maintained in the
vessel by a level controller (LC), which controls flow d) The pack consists of a number of ......................plates.
from the oil outlet.
e) Tilting plate separators may have .....................or .......................plates.
An oil/water filter coalescer will often reduce the
amount of oil in the produced water from 5 000 ppm f) The oily water flows radially outwards through the ..................... where the oil droplets .......................
to between 2 and 15 ppm.

Under severe operating conditions, the cartridges


may not last very long.

However cartridge life can be improved by : You will find the answer to Test Yourself 6 on Page 49

• pre-filtering the oily water


We can now take a look at a produced water clean-up facility which has a
• steady state operation somewhat different operating mechanism. This is the gas flotation unit.

• scale inhibition, where required 28


Petroleum Open Learning

Gas Flotation Units


Let us first of all take a look at the flow
You will remember from Section 2 that the principle of operation of a gas flotation unit (also called a depurator) of produced water through the unit.
is that the oil droplets, and any suspended solids, are assisted to the surface by small gas bubbles. The oil and
suspended solids are then skimmed off as a froth. The produced water:

Part of a typical flotation unit is illustrated in Figure 14. Study the illustration carefully. • enters the inlet chamber

• flows out of the inlet chamber,


under a baffle, and enters the first
flotation cell where most of the oil
is removed from the water

• flows over the top of an internal


baffle and enters the second
flotation cell where the rest of the
oil is removed from the water

• flows out of the second flotation


cell, under a baffle, and enters
the exit chamber as clean water

• flows out of the exit chamber into


the suction of the recirculation
pump

• leaves the recirculation pump and


goes to the water outlet or back
into the two flotation cells

Make sure you can follow the flow of the


water before proceeding.
29
Petroleum Open Learning

We will now look at how the water is aerated. The oil and gas then form a layer of foam on top of Take a look at Figure 15. This shows a side view of
the water. the flotation cells. You can see the layer of oily foam
You can see in Figure 14 that the recirculation pump on top of the water.
takes water from the exit chamber. The water can In the type of unit we have been looking at, a venturi
go in one of two directions: was used to create the gas bubbles. As you saw in To the side of the unit is a collecting box called a
Section 2, other methods of creating the bubbles launder. The oily foam spills over a weir into the
• water outlet may be used. launder where it separates into oil and gas.
But what happens to the oil ?
• back to the flotation unit The gas flows back into the flotation cells and the oil
collects at the bottom of the launder. The oil level is
The level in the exit chamber is controlled by a level controlled by a level controller (LC) which opens
controller (LC) which opens and closes the valve in and closes a valve on the oil outlet line.
the water outlet.

If the level rises the valve will open and allow water
to leave the system. If the level falls, the valve will
close and retain water in the system.

The bulk of the water (up to 70% of the design


throughput) is recirculated back to the two flotation
cells.

As the water enters each cell it passes through a


venturi. This is a device which uses the flow of
water to create a low pressure area. Gas, from the
area above the water, is sucked into the venturi and
mixes with the water. The gas and water mixture is
then discharged at the outlet at the bottom of each
cell.

The gas drawn into the venturi results in millions of


tiny bubbles being released at the bottom of the
flotation cell. These bubbles attach themselves to oil
droplets in the water, and carry them to the surface.
30
Petroleum Open Learning

In general:

• In some large units there maybe as many as


ten flotation cells
Test Yourself 7
• The residence time in a flotation unit may be
as low as three minutes Correct the following sentences, which all refer to a flotation unit.

• A flotation unit may reduce the amount of oil in


the produced water to around 10 ppm a) Oil spills over a weir into the exit chamber.

• Flotation units may also remove most of the b) Water flows under the middle (internal) baffle, in a two cell unit
solids suspended in the produced water
c) The recirculation pump takes its suction from the launder.

A problem associated with flotation units is that they d) Gas is introduced to the top of the cells as finely dispersed bubbles.
are difficult to control. The size of the bubbles
affects the efficiency of the unit and it may take e) Gas from the area above the water is sucked into a venturi and mixed with the oil.
many hours to set up each venturi to give the
optimum operating conditions. f) In the launder, oil and water separate and the oil level is controlled by a level controller.

If you are happy with the construction and operation


of flotation units, try the following Test Yourself
question.

You will find the answers to Test Yourself 7 on Page 49

31
Petroleum Open Learning

Hydrocyclones The hydrocyclone has a cone shaped liner within a


pressure shell. The cone shaped liner of the unit is
The most important development in oil / water treatment in recent years is the fabricated as a thin walled vessel.
hydrocyclone. This is a unit which uses centrifugal force to separate oil and water.
Oily water enters the unit through an inlet into the
Figure 16 is an illustration of a hydrocyclone. liner. The inlet is designed to spin the water as it
enters the swirl chamber. Inside the swirl chamber
the flow forms a vortex which passes along the
length of the liner.

32
Petroleum Open Learning

The water / oil mixture is accelerated to high velocity The major factors influencing the performance of a Use of Chemical Additives
and strong centrifugal forces develop. This hydrocyclone are:
promotes oil / water separation. The more dense Chemicals are increasingly used in produced water
water phase moves to the wall of the liner and • The specific gravity difference between the oil treatment facilities. I don’t intend to go into the
displaces the lighter oil phase towards the centre of and water. The greater the difference, the chemistry of the way that they work. However, I
the vortex. greater the potential for rapid separation. think that you should be aware of the main types of
chemical used and what their function is.
The water continues to flow to the water outlet along • The oil droplet size. Larger droplets move
the sides of the liner. The oil flows in the opposite more rapidly towards the central core. Emulsion breakers or demulsifiers. These
direction, via the low pressure central core, to be chemicals assist in separating oil/water emulsions.
removed at the oil outlet. The oil stream is called the • Temperature. This affects both density and They break down the mechanisms which cause the
reject oil stream. viscosity. Higher temperatures increase the emulsion to form.
potential for easy separation and therefore
Total residence time of the liquid in the hydrocyclone hydrocyclones are most often installed Flocculation and flotation agents. These
is about two seconds. Hence the equipment can upstream of any produced water coolers. chemicals act as seeds around which small solid
be compact. particles or oil droplets may collect. The increased
• Higher flow rates. These increase the intensity size assists in the separation process.
The capacity of a hydrocyclone is dependent on the of the centrifugal separation forces.
pressure drop between inlet pressure and reject oil Corrosion Inhibitors. These chemicals help
stream pressure. The reject oil stream will not be 100% oil. It will be a prevent corrosion of the vessels and pipework of the
mixture of oil and water. This mixture is then fed produced water system.
Hydrocyclones are usually mounted as groups in back into the main process, where the oil is
parallel to increase capacity. Each unit can be recovered. Biocides. These chemicals kill bacteria in the
opened up or closed in. This enables the operator to water. They are used to prevent the formation of
maintain optimum conditions during varying flow Hydrocyclones have a weight / efficiency ratio which slimes.
rates. makes them attractive for use offshore.
Oxygen scavengers. These chemicals are used to
In addition, units can be installed in series to remove residual oxygen in water. This also helps to
increase oil removal. The water leaving one prevent corrosion.
hydrocyclone enters the next, and so on. Using this
system oil concentrations as low as 5 ppm can often Scale inhibitors. These chemicals prevent
be achieved. dissolved solids coming out of solution and being
deposited on pipework, etc., as scale.

33
Petroleum Open Learning

In all instances, great care is needed to ensure that:

• the chemicals used are harmless to the


environment
Summary of Section 3
• one chemical will not counteract the effects of
another chemical used in the process In this section we have looked at different types of oil/water
separators which may be found on an oil production facility.
• the chemicals used will not affect downstream
processing We have looked at the construction and operation of :

In all cases chemicals should be introduced to the • API separators


system at the correct dosage rates. Too high a
dosage rate is often worse than no dosage at all. • parallel plate separators

• corrugated plate separators

• oil / water filter coalescers

• gas flotation units

• hydrocyclones

We have also taken a brief look at chemicals which may be used in a Produced Water System.

In the final section of this unit we will consider a typical produced water handling system.
Before you move on to that however have a go at the following Test Yourself question.

34
Petroleum Open Learning

Test Yourself 8
State whether the components listed on the right are part of :

a) An A.P.I. separator List of components

b) A tilting plate separator 1. swirl chamber

c) An oil / water filter coalescer 2. adjustable oil skimmer

d) A flotation unit 3. baffle

e) A hydrocyclone 4. weir

5. cartridge

6. level controller
Note: some of the components are found in
more than one type of unit. 7. sludge trap

8. cone shaped liner

9. venturi

10. recirculation pump


You will find the answers to Test Yourself 8 on Page 49

35
Produced Water Treatment Petroleum Open Learning

Section 4 - A Typical Produced Water System


In this final section, we are going to look at a complete
produced water system.

The system I will use as an illustration includes two


tilting plate separators, a flotation unit and a produced
water caisson. It is fairly typical of produced water
handling systems which you might find offshore.

In our example, the function of the system is to


remove:

• oil

• entrained gas

• fine solids

from the produced water streams. This is achieved by


short distance gravity separation and flotation.

Take a look at Figures 17. This is a simple block


diagram of the overall system. Study this for a moment
and get a general idea of the relationship between the
various subsystems.

As we work through the system in detail, however, we


will need a more complex drawing. This is Figure 18,
entitled “A Typical Produced Water System”. In order to
allow easy reference to this drawing, I have included it
as a separate sheet in your pack.

Take this out now and study it for a few minutes.

In order to help you, I have also included a symbol key,


which includes those symbols used in the drawing. 36
Petroleum Open Learning

Look at Figure 18. You should be able to identify Tilting Plate Separators The diagram indicates that the chemical enters the
four separate parts of the system. They are: produced water line via an injection quill.
The produced water enters the system from the
• the two tilting plate separators (TP-01 and main oil, gas and water separation facilities. Find Figure 19 is an illustration of an injection quill. It is
TP·02) in the upper left hand corner of the the entry point in Figure 18 and follow the flow. designed to ensure that the chemicals are efficiently
figure mixed with the water flow.
The first thing you will see is that a chemical is
• the flotation unit (DP-01) in the upper right injected at this point. The chemical being injected is
hand corner a demulsifier which assists in the separation of the
oil and water. You will remember we discussed the
• the chemical dosing system in the bottom use of demulsifiers in Section 2.
left hand corner

• the circulation and recycle pumps in the


bottom right hand corner

We will look at each of these sections in turn as I


guide you through the flow diagram. Because the
pumps are part of the flotation unit equipment we will
study these two sections together.

Not shown in Figure 18 is the produced water


caisson where the treated water enters the sea.
We will look at that later as a separate item.

37
Petroleum Open Learning

The chemical injection line is protected from any Any gas which is released from the produced water At this point. let me say a few words about
back flow by a non-return valve (Figure 18). is vented off to the vent header. Note that in our shutdown systems.
example, there is no valve on the vent line. There
Downstream of the injection quill, oily water from the are no pressure relief valves on TP-01 so, to prevent All offshore production facilities are protected
platform drains system joins with the produced an accidental over-pressure situation, the vent line is by a safety shutdown system. This comprises
water. fitted without an isolation valve. dedicated sensors, actuators, valves,
pipework, etc. They are installed to enable a
Just after this connection is a sample point. In TP-01 is also fitted with a nitrogen purge safe and effective shutdown of plant and
order to obtain a true sample, this connection connection. If TP-01 is shut down, the vessel can be equipment in a controlled manner. The whole
is placed after the point at which the two purged with nitrogen. The nitrogen ensures that an system is called an emergency shutdown
streams combine. air/gas flammable mixture cannot occur, by system (E.S.D. system).
sweeping out any flammable gases before the
Beyond the sample point another line enters the vessel is opened for maintenance. Levels of shutdown may be designated.
produced water line. It is protected from back-flow These depend on the degree of hazard 10
by a non-return valve. This line is the discharge line TP-01 is fitted with a level gauge (LG-01) so that personnel, plant and the environment. Less
from the float recycle pumps P-02A and P-02B. the operator can check the liquid level in this serious hazards may only require the
We will be looking at these two pumps later. separator. Connected to the level gauge is a level shutdown of individual items of plant or
switch high (LSH-01). If LSH-01 is activated it will : equipment. Severe hazards, however, may
The produced water then splits into two individual necessitate a total platform shutdown.
lines, one to each of the two tilting plate • sound an alarm in the Control Room via level
separators. Each line is fitted with a butterfly valve alarm high (LAH-01) If our platform had four designated levels of
which is indicated as being LO. This means that the shutdown, with level 1 being the most severe
valves are normally locked open to make sure that • cause a shutdown of the produced water hazard level, then the produced water
TP-01 and TP-02 are not accidentally isolated. system via the E.S.D. system shutdown would probably be a level 3. This is
indicated in Figure 18.
The tilting plate separators are identical, so we will
just look at TP-01. You will appreciate that E.S.D. systems are
very complex and I do not intend to talk about
In our example, TP-01 has three sets of corrugated them any further here. Other units in the
tilted plates. The produced water enters the side of Petroleum Processing Technology Series
TP-01 via three separate connections. This ensures will describe these systems in much more
that turbulence is reduced at these points by detail.
reducing the produced water flow rate.

38
Petroleum Open Learning

Now back to the tilting plate separators. You will The Flotation Unit
notice that a water jetting hose connection is
fitted. This allows a high pressure water hose to be The water which leaves the tilting plate separators
connected into the system for washing out any should have a fairly low oil content - say 60-80 ppm.
sludge which collects in the sludge trap of TP-01. We now have to polish the water to reduce the oil
Each trap is also fitted with a connection to allow the
liquidised sludge to drain away to disposal.
Test Yourself 9 content to less than 30 ppm (the standard for our
system). This is done in the flotation unit.

You will remember from Sections 2 and 3 that With reference to the tilting plate separators You should remember the principle of operation of
separation of oil from the produced water takes in our typical system, see if you can answer such a unit from Section 3. If you need to refresh
place in the tilting plate separator. These two liquid the following questions. your memory, do that now before continuing.
phases are discharged via separate lines.

The recovered oil flows from TP-01 and joins with a) What is the function of an injection Referring back to Figure 18, you will see that, from
the oil line from TP-02. The combined oil stream quill ? the tilting plate separators the water can be routed
then flows to a slop oil tank. The oil from the slop either:
oil tank will be pumped back into the primary b) Why is the sample point downstream
separation system. of the connection for the oily drains • to flotation unit (DP-01)
system?
The water from TP-01 passes through a butterfly • to a by-pass round DP-01
valve before joining with the water line from TP-02. c) What system is used to ensure that
The combined water line then takes the water the tilting plate separators are not The reason for the by-pass is to allow the flotation
towards the flotation unit DP-01 accidentally isolated? unit to be taken out of service for maintenance etc.

That completes our look at the tilting plate d) What is the nitrogen purge If you look at the system you can see that there are
separators. Before you move on to look at the connection used for? two tilting plate separators, but only one flotation
flotation unit however, have a go at the following unit. Normally both separators are in use.
Test Yourself question.
We can, however, keep one separator on line whilst
we clean and maintain the other one. Any loss of
efficiency would be taken care of by the flotation unit.
You will find the answers to
Test Yourself 9 on Page 50

39
Petroleum Open Learning

But what happens if we have to clean or maintain the If LSH-03 is activated it will start float recycle pump The oil from the float re-cycle pumps is fed into the
flotation unit? P-02A or P-02B. The pump running light (XL-03 or produced water line downstream of the chemical
XL-04) will light in the control room to alert the injection point and downstream of the sample point.
We have already seen that corrugated plate operator that the pump is running. If it was fed into the line upstream of :
separators, (The type we have selected), can reduce
the oil content of produced water to 30 ppm. So, if The oil is pumped back into the inlet line to the tilting • the chemical injection point - it would get a
both tilting plate separators are on line, it would be plate separators. From there it is recovered again second dose of chemicals
possible to by-pass the flotation unit, temporarily, so and discharged to the slop oil tank.
that it, too, could be maintained. • the sample point - it would affect the amount
If LSL-03 is activated it will stop P-02A or P-02B. of oil being measured as entering the system
Under normal conditions the water from the The pump running light (XL-03 or XL-04) will for treatment
separators flows through a butterfly valve towards extinguish in the control room to alert the operator
the flotation unit. Downstream of the butterfly valve that the pump has stopped. Neither of these conditions are desirable.
is a second chemical injection point where
demulsifier can be injected from the chemical The oil leaving the launder is filtered before it enters The exit chamber of the flotation unit is fitted with a
dosing package. the suction of the float recycle pumps. Each pump is level gauge (LG-04) to allow the operator to check
fitted with a discharge pressure relief valve the level of water in the exit chamber. Connected to
The chemically treated water enters the inlet (PSV-01 and PSV-02). The PSV is fitted on the the level gauge is :
chamber of the flotation unit and from there into each pump side of the discharge Isolation valve.
aeration cell in turn. Our unit has four separate • a Level Switch High (LSH-04)
aeration cells. Within each cell oil foam accumulates If the discharge pressure exceeds a pre-set value,
on top of the water. A motor (M) is indicated on the say 30 psi, then PSV-01 or PSV-02 will lift. This will • a Level Switch Low (LSL-04)
left hand side of the unit. This motor drives a set of relieve the discharge pressure by circulating oil back
paddles which skim the oil foam from the surface of to the suction of the pump. If LSH-04 is activated it will :
each cell into the launder.
• sound an alarm in the Control Room via Level
The launder is fitted with a level gauge (LG-03) to Alarm High (LAH-04)
allow the operator to check the level of oil in the
launder. Connected to the gauge is : • will cause a level 3 shutdown

• a Level Switch High (LSH-03) A level 3 shutdown generated by LSH-04 would shut
down the flow of produced water which is leaving the
• a Level Switch Low (LSL-03) primary separation system.

40
Petroleum Open Learning

If LSL-04 is activated it will stop the hydraulic


circulation pumps (P-01 A or P-01 B). The pump
running light (XL-01 or XL-02) will extinguish in the
Control Room to alert the operator that the pump has
stopped.

The exit chamber of the flotation unit is also fitted


Test Yourself 10
with a level transmitter (LT-05) which feeds a
signal to a level controller (LC-05). The level Which of the following components are not part of the flotation unit.
controller opens and closes a level control valve
(LV-05) to maintain a constant level in the exit
chamber. skimmer motor

float recycle pumps


The water which flows through LV-05 is the treated
water leaving the system. There is a sample point plate pack
just downstream of LV-05. This is the position where
the final water quality is checked. launder

pump running light

filter cartridge

hydraulic circulation pumps

level gauge

discharge pressure relief valve

exit chamber level switch low

chemical injection point

inner liner
At this point I think that you should go back over
what we have looked at up to now in this section,
then have a go at Test Yourself 10.
You will find the answers to Test Yourself 10 on Page 50

41
Petroleum Open Learning

Lets now continue with the flotation unit by looking If there is a problem in the fuel gas system the In normal operation:
at the gas aeration of the water. pressure of the gas blanket could fall and affect the
operation of the produced water handling system. • the electrical signal to the solenoid of XY-01
Water is fed into the gas aeration side of the flotation is live
unit by either hydraulic circulation pump P-01A or However, tied into the fuel gas line, downstream of
P-01B. These pumps take their water supply from PCV-01 is an automatic nitrogen back-up system. • the air flows through XY-01 and maintains
the exit chamber of DP-01, and the water is then The nitrogen back-up facility is activated pressure on the actuator of XV-01
filtered before being recirculated to DP-01. automatically if the fuel gas pressure fails below
a pre-set value. • the valve stays closed
The combined discharge of P-01A and P-01B splits
into four separate lines, one for each of the flotation • limit switch ZSL-01 is activated and the
cells. The operator can balance the flow of water to The system comprises: signal valve closed is indicated in the
each cell, by adjusting the opening or closing of the Control Room.
individual inlet butterfly valves. • flow orifice (FO-01) - a flow orifice is a small
plate with a precision drilled hole in the centre If the low fuel gas pressure condition is activated
A fuel gas line provides the gas required for -the size of the hole determines the amount then the electrical signal to the solenoid is made
aeration via PCV-01. PCV-01 is a forward of flow through the orifice dead.
pressure control valve. This means that it controls
the pressure downstream of where it is installed. • emergency shutdown valve (XV-01) - this When this occurs:
PCV-01 maintains a blanket of gas on the flotation has an FO indication underneath. This means
unit with a constant pressure of a few inches water that the valve will fall open if the instrument • XY-01 changes position
gauge. air pressure is lost.
• the air supply to XV-01 is cut off
The water flowing through a venturi causes the gas • solenoid valve (XV-01) - a solenoid valve is
to be sucked into the water. It is then released as a valve which is opened or closed by an • the air supply to XY-01 is vented to
tiny bubbles. (If you are having difficulty visualising electro-magnet atmosphere
this, go back to the description of flotation units in
Section 3 and refresh your memory.) • limit switches (ZSH-01 and ZSL-01) The result of these actions is that XV-01 will open.
When this occurs the movement of the valve will
activate ZSH-01 which will signal to the control room
Solenoid valve XY-01 is fitted into the instrument air that the valve has opened.
supply to XV-01. This valve is a three-way valve.

42
Petroleum Open Learning

Nitrogen will then be supplied to DP-01. The amount Chemical Dosing Package These pumps are fitted with discharge pressure
of nitrogen will be regulated by the orifice size of relief valves PSV-03 and PSV-04. The PSV’s are
FO-01. This package consists of : fitted on the pump side of the discharge isolation
valves.
There is another connection to DP-01 which we • demulsifier drum D-01
have not yet mentioned. This is a third chemical If the discharge pressure exceeds say, 30 psi, then
injection point positioned on the top of the unit. It • chemical dosing pumps P-03A and P-03B PSY-03 or PSY-04 will lift. This will relieve the
enables demulsifier to be injected into the main body discharge pressure by circulating chemical back to
of DP-01. Demulsifying chemicals are pumped into D-01 from the demulsifier drum.
drums with a small hand pump via the hose
connection. A level indicator (LI-06) allows the Chemical dosing pump P-03A supplies the
operator to stop filling the drum when the correct demulsifier to the tilting plate separators. Pump
level is reached and to monitor the level of P-03B supplies the demulsifier to the flotation unit.
We have completed our look at the flotation unit. demulsifier in the drum during normal operations.
By the time the water leaves the unit it should have The flow of demulsifier leaving the pumps is
an oil content down to specification. To achieve this, If the operator fails to re-fill the drum when a low monitored by a sight glass (SG) in each of the lines.
we have injected demulsifier into: level is reached, a level switch low (LSL-07) will
activate. That completes our look at the chemical dosing
• the tilting plate separators package.
This will ;
• the flotation unit
• activate a low level alarm (LAL-07) in the
control room to warn the operator

We will now take a look at the chemical injection • shut down the chemical dosing pumps
dosing package, which is the final part of Figure 18.
The chemical dosing pumps are reciprocating / The treated water has finally to be disposed of to the
positive displacement / variable stroke pumps. (In a sea. This brings us to the final part of this section,
reciprocating pump a piston moves backwards and the produced water caisson.
forwards inside a cylinder. In a variable stroke
reciprocating pump the amount of liquid pumped is
controlled by changing the length of the stroke of the
pump cylinder.)

43
Petroleum Open Learning

Produced Water Caisson


Basically, the produced water caisson is a long
residence time vertical separator. It is designed to
remove any traces of oil which may be left in the
produced water when it leaves the flotation unit.

Figure 20 is an illustration of the produced water


caisson.

44
Petroleum Open Learning

The caisson is a pipe of about 1.2m diameter which When the oil in the sump reaches a pre-set high
hangs from the lower deck level of the platform into level it will activate a level switch high (LSH-09).
the sea. The bottom of the caisson may be 30m or When activated LSH-09 will start the oil sump
so below sea level. The actual length of the caisson pump. The oil is pumped to the slop oil system,
will depend on water depth and the height of the from where it will be returned to the primary
lower deck above sea level, separation system.
Test Yourself 11
The bottom of the caisson is open to the sea and When the oil in the sump reaches a pre-set low
the level of water inside the caisson will go up and level it will activate a level switch low (LSL-09). With reference to the system you
down with the rise and fall of the tide. When activated, LSL-09 will stop the oil sump have been following in Section 4, answer the
pump. following questions,
The produced water line from the flotation unit
enters the top of the caisson at deck level. This line a) In the automatic nitrogen purge system
extends down into the caisson to a point where its on the flotation unit, what controls the
end is under the level of the water, even at low tide. flow of gas?
The produced water discharges into the caisson at You have now completed the section on a typical
this point. produced water system. Have a go at the final Test b) What is used to skim the oil from the
Yourself question before going through the section surface of the water into the launder?
The produced water may stay in the caisson for summary.
upwards of 30 minutes, This is a much longer c) Where does the separated oil from the
residence time than anywhere else in the system tilting plate separators go to?
and traces of oil will be able to float to the surface,
where they can accumulate. d) Why is there a by-pass around the
flotation unit?
Inside the caisson is a small sump with a weir set
at a level which is just above the highest high tide e) Where does the oil accumulate in the
level. The oil which has accumulated on top of the produced water caisson?
water can then spill over the weir into the sump.

You will find the answers to Test Yourself 11


on Page 50.

45
Petroleum Open Learning

Summary of Section 4
In this section we have looked at how a typical produced water system operates,

I have described a system which includes:

• tilting plate separators

• flotation unit

• chemical dosing

• produced water caisson

and which combine to treat produced water for dumping to the sea.

As you worked through the section you followed the main flow lines and Although the system I described is similar to many you would come
traced the path of the water and oil. I pointed out the points where chemical across offshore, it is a hypothetical one. You must remember that, if
is injected into the system and where sampling takes place. you are working on a produced water system you should become familiar
with the layout and operating procedures of that particular system.
You also discovered the function and operation of the instrumentation
associated with such a system and the safety features involved.

Now go back to the training targets for this unit and make sure that you
have met those targets,

46
Petroleum Open Learning

Check Yourself - Answers

Check Yourself 1 Check Yourself 3


a) True
The water underlying the oil pushes the oil towards the
producing wells. The aquifer expands to fill the space left by the
b) False
oil which has been removed. The oil water contact therefore will
rise up the reservoir towards the producing well intakes.
c) False (produced water can contain up to 5 times
the amount of salt present in sea water.)

d) True

Check Yourself 2 e) False (the current figure is 30 ppm)

a) total production = 4770m3/d (3975+ 795)

Therefore water cut = 795 x 100% = 16.67%


4770

b) oil production is 80% ( 100 - 20) 875m3/d


Check Yourself 4

Oil and water have different densities. Water is the most
Therefore oil production = 875 x 80 = 700m3/d
dense, and would sink, allowing the oil to float on top.
100
The two liquids would separate from each other and the oil
c) water production = 159m3/d (556 - 3971)
would tend to float on top of the water.

Therefore water cut = 159 x 100% = 28.6%
556

47
Petroleum Open Learning

Check Yourself 5
Your answer should look like the following:

Short Distance Centrifugal


Gravity Coalescence Gravity Gas Force
Separation Separation Flotation Separation

porous medium √

plate pack √

oil droplets rising √ √ √ √

demulsifier √ √ √ √ √

finely dispersed bubbles √

vortex √

48
Petroleum Open Learning

Check Yourself 6 Check Yourself 7 Check Yourself 8


1. e
a) stilling plate a) Oil spills over a weir into the launder.
2. a
b) sludge debris b) Water flows over the middle (internal) baffle in
a two cell unit.
3. a-d
c) weir
c) The recirculation pump takes its suction from
4. a-b-d
d) parallel the exit chamber.
5. c
e) flat or corrugated d) Gas is introduced to the bottom of the cellsas
finely dispersed bubbles.
6. c-d
f) cartridge, coalesce
e) Gas from the area above the water is sucked
7. a-b
into a venturi and mixed with the water.
8. e
f) In the launder, oil and gas separate and the
oil level is controlled by a level controller.
9. d

10. d

49
Petroleum Open Learning

Check Yourself 9 Check Yourself 10 Check Yourself 11


a) The injection quill is designed to ensure that plate pack a) The size of the hole in the flow
the chemicals are efficiently mixed with the orifice.
main water flow. filter cartridge
b) A motor driven set of paddles.
b) In order to ensure that a true sample is inner liner
obtained, i.e. after the two streams have c) To a slop oil tank and from there to
mixed together. the main separation system.

c) The inlet valves have a lock open facility. d) To allow the unit to be taken out of
service for maintenance whilst the
d) In order that nitrogen can be introduced to plate separators are still working.
sweep out any flammable gases before the
vessels are opened for maintenance. This e) On top of the sea water from where it
ensures that an explosive air gas mixture spills over a weir into a sump.
cannot form.

50

You might also like