1995 15 Fluenta Multiphase Flow Meter Tested and Marinised Olsvik Fluenta
1995 15 Fluenta Multiphase Flow Meter Tested and Marinised Olsvik Fluenta
1995 15 Fluenta Multiphase Flow Meter Tested and Marinised Olsvik Fluenta
'
North Sea
Measurement Workshop
1995
Paper 19:
Authors:
Kenneth Olsvik and Tor Widerae
Fluenta, Norway
Organiser:
Norwegian Society of Chartered Engineen
Norwegian Society for Oil and Gas Measurement
Co-organiser:
National Engineering Laboratory, UK
CONTENTS
SUMMARY
1. INTRODUCTION
4. REFERENCES
SUMMARY
This paper describes the technology and principle of operation ofFluenta's multiphase flow
meters, for both topside and subsea applications. It also deals with the basis of the design of
the subsea multiphase flow meter, and Fluenta's collaboration with Amerada Hess and
Kvremer FSSL (KFL). The paper further describes the qualification programme and test
results obtained at the National Engineering Laboratory (NEL) for this world first installation
of a subsea multiphase flow meter, the S:MFM I 000, and the benefits and savings this meter
will provide to the South Scott field where it was installed in May this year.
1. INTRODUCTION
The ultimate goal for multiphase flow metering has always been to use the technology in a
subsea environment. The technology do represents substantial operational and cost savings for
topside applications, but being able to implement this technology for subsea applications,
means that a higher proportion of proven hydrocarbon reserves found in economically
marginal fields can be developed and turned into profitable developments. This is due to the
. huge savings obtained by eliminating test flow lines and test separators.
In the next couple of years the oil companies have planned to develop a substantial part of
their new fields using this technology.
Amerada Hess Limited (AHL) is committed to the development of new technology to support
and enhance its business activities of finding, developing and producing offshore
hydrocarbons. To this end .Ma launched a Business Driven New Technology Initiative. This
seeks to identify key technology requirements. to initiate R&D projects to fulfil the~ and to
ensure that the results are commercially exploited.
Amongst the selected areas are health & safety. the environment, drilling costs, mini-field
developments, reservoir characterisation/performance and geological basin modelling.
One of the main requirements in the mini-fields area is the development of technology to
enable longer and more cost effective satellite tie-backs for small accumulations. This primarily
focuses on subsea technology, including separation, and multiphase transportation of produced
fluids.
Within this technology thrust, the possibly most significant gap in the armoury of available
technology is a reliable non-intrusive subsea multiphase flowmeter.
The applications are endless for a device shown to work satisfactorily in a subsea environment.
No more test separators, no more test flow lines. A subsea manifold could contain all the
valving required along with a multiphase flow meter, to remotely test each well in tum and still
produce via a single flow line back to a remote platform. The ultimate solution would be a
gathering platform with subsea satellites scattered over a large area collecting from small fields
and providing central processing.
There is no doubt that when the technology is proven the whole face of subsea oil production
from marginal fields will change.
In a unique joint venture Amerada Hess Limited and Amerada Hess Norge joined forces
(Norway supplying the funds and U.K. supplying project co-ordination and monitoring) in
backing one of the three phase race contenders.
Stage 1 was completed in 1992. This involved the purchase and testing of the basic fraction
meter version of the device. The meter was installed and tested in the new multiphase
laboratory at the National Engineering Laboratory in East Kilbride, near Glasgow.
The aims of Stage I were to prove that the technology works and determine by
. experimentation the limitations of use of the instrument. This was successfully done.
In Stage 2 of the project Amerada Hess purchased the three phase flowmeter
meter version of the Fluenta device - the MPFM 1900.
The MPFM 1900 was extensively tested at N.E .L. and has now been installed on the AHOOI
Floating Production Facility. The MPFM 1900 is installed upstream of the test separator
between two chokes which will allow direct comparison between the 2 technologies. The
signals were integrated into the AHOO 1 Metering Database system, a Eurothenn Maxi-Vis
system which provides comparison with the conventional metering system currently installed
on the test separator.
The Maxi-Vis Metering Database on the AHOO I was reprogrammed to provide the data
reporting required to prove the device under real conditions rather than the controlled
conditions at N.E.L. This reporting includes duplicated well test reports, one report based on
the original test separator metering and a second report using the three phase meter data.
By installing the meter between the two chokes on the inlet pipework to the test separator we
were able to simulate either test separator conditions at the three phase meter or increase the
pressure closer to subsea pressures.
The final stage ofthe project is completed. This was split into two parts. The first was a study
of the requirements for a subsea meter. Do we want a meter that can be fitted by divers or
ROY? Do we want a complete skid designed including self-calibration checking facilities?. Do
we want a single meter on a subsea manifold or one meter per well? What will it cost? After a
great deal of thought a report was produced that provided the final specifications for a subsea
version of the Fluenta meter, which included for water-continuous measurement up to 100%
water in oil.
Part 2 of the project was to build a meter for installation and testing on the South Scott
Project.
KFL was selected to work with us on this phase of the project, and they also helped with the
funding. The Fluenta electronics were integrated with KFL's fourth generation subsea
electronics .
Both the MPFM l 900VI and MPFM 900VI are flow meters designed for accurate flow rate
metering of the oil, gas and water phases of a multiphase flow, without separation of the well
stream. The non-intrusive, real-time instruments require no by-pass line and no mixing device.
Both instruments divide the measurements into two; measurements of fractions, and
measurements of flow rates.
The fraction part is detennined in the same way for both meters: three independent equations
are needed and these are obtained by measuring the dielectric constant (permittivity) of the
mixture ( oiVgas/water) with the capacitance sensor, measuring the same mixture in a gamma
radiation path, in which the measured.absorption of gamma particles is a function of the
density of the mixture measured by the gamma densitometer, and using the third and last
equation that the sum of the three fractions will always be equal one.
For water-continuous mixtures an inductive sensor is used instead of the capacitance sensor.
The principle is basically the same, except that now conductivity rather than permittivity is
measured. Two toroids induce a constant current through the multiphase mixture while the
differential voltage is measured across two electrode plates placed between the toroids. As the
current is constant, the voltage drop is a measure of the electrical conductivity of the mixture,
which in tum is affected by the composition of the mixture (oiVgas/water). The conductivity
measured by the inductive sensor is thus a measure of the mixture (oiVgas/water) similar to the
pennittivity measured by the capacitance sensor used in oil-continuous mixtures
( oiVgas/water). The expression of this module is the capital letter 'T'.
An illustration of the measurement principle is shown below in figure 1.
/
= Math.
models oil
= c::>
tltltltl
dP
'water
1
The flow velocity can also be determined by the venturimeter. This gives redundancy in the
multiphase flow meter. The meter is named MPFM 1900V. 1£: as for the MPFM 900 series the
meter is to be used in water-continuous mixtures the inductive module will be included, and
the meter is designated the :MPFM 1900VI.
A more detailed description of the operating principles can be found in reference JI/ and /21.
The Fluenta "MPFM 1900VI is illustrated in the figure below.
lve.nturi
!Capacitance sensor
!Inductive sensor
This has been the philosophy behind Fluenta's development and so far we have made l S field
installations at many locations in Norway, France, Egypt, UK, and USA. We feel that
multiphase meters still have great potential for achieving better accuracy and reliability. Thus
we are constantly working on improvements together with companies such as Amerada
Hess/Conoco/Saga Petroleum and KFL.
Below we have listed some of the most important tests/qualifications carried out.
Very early in the development phase of the MPFM 1900 system, it was found extremely
difficult, not to say impossible, to use a mixer to generate a homogeneous mixture over more
than a very narrow band of flow conditions. The idea of including some kind of flow
conditioner as part of the multiphase meter was dropped, and it was decided to develop a
meter capable of a very wide range of operating conditions, requiring no flow conditioning.
In order to limit the range of different flow regimes through the meter, the installation of the
meter has been restricted to be vertical with upwards flow, at a distance of S - 20 diameters
downstream of a blinded T. This installation is shown in Figure 3. ·
At the project stage when the design of a multiphase meter must be fixed, there is
nonnally some uncertainty in the process data, and the flow conditions cannot always
be confirmed.
Estimates of flow regimes are very uncertain, particularly for a location where fully
developed regimes are not likely to occur.
The meter may have to operate on a number of wells with very different flow
conditions.
The MPFM 1900 VI has a measurement range of 0 - 100% watercut and 0 - 100% gas
fractions.
The multiphase meter has a turndown ratio of at least 1to10. For low gas fractions, the
· minimum multiphase velocity is LS m/sec. For high gas fractions the minimum velocity is 2.5
m/sec. Upper velocity is less critical. For design purposes 1.5 m/sec and. 2.5 m/sec should be
used, depending on gas content. As upper limit is less critical, velocities above 15 m/sec
through the meter are acceptable. The velocity range is indicated in Figure 5.
l:'
u0
15
-- ....... ....... -
~
12
- ........ .......
-
'tJ
:;
er..!!
9 -- ....... --
- - Upper limit
Lower limit
--
=
C'll
u
'E
E
6
........ ......
........
.......
Cl> 3 ........
c.
:= ,.. .......
en .......
0
-.. ........
0 5 10 15 20 25
Superficial gas velocity (m/s)
Different flow regimes and changes in the composition of oil, water and gas make it impossible
to give one fixed value for the measurement uncertainty. For this reason we have given
realistic values of the performance for different ranges of gas fractions. Values are given for
liquid flow rate, gas flow rate, watercut and gas fraction. At the outer limits of the operating
range, uncertainties might be slightly greater.
On the basis of the results of ongoing qualification tests, new performance specifications
for the meter have been de.fined. Due to requests from potential users the flow rate
uncertainties are specified as values relative to actual volumetric flow rates. Uncertainties
in fractions are expressed as absolute deviations.
Uncertainties in volumetric flow rates are given as relative uncertainty to actual volumetric
flow rates. (not to tqtal flow rate)
The project to develop a subsea multiphase flowmeter, SMFM 1000 started in 1994 as a
cooperative project between Fluenta and KFL, with Amerada Hess and KFL funding the
project. Ref /3/.
Sensor electronics
Power suppl
Sella! conm.
The meter system has been designed in accordance with the following specifications:
Environmental
Water Depth 670 metres maximum seawater. (All components are easily upgradeable for
2000 m water depth)
Ambient Temp. -5 to +30°C
MTBF
The system has been designed to have an overall MTBF as high as practically possible, with a
design aim of 15 years.
The standard version requires diver-intervention. Alternative designs for ROV maintenance
and retrieval of electronics unit or guidewire line installation and retrieval of the complete
assembly are available ·
Materials
Process fluid-wetted parts are made of super duplex: stainless steel. Seawater-wetted parts are
made of AISI 316 stainless steel, or carbon steel, protected by a paint system designed for
long-term subsea use.
Serial Link
Communications between the Subsea Production Control System and the meter take
place via a serial link through a cable that also supplies the power. The
communication protocol between the meter and the Subsea Production Control
System is Modbus.
Power Supply
3.2 Qualification of the SMFM 1000 at the National Engineering Laboratory (NEL)
WA'TER Oil
Nl'TROOEN
SUPR.Y
H'
lVRSllE METERS
1r
1\JRBllE METERS
. __ _TESTSECllON
' - - - - - 0 0 - -.....----1 .......
Some carry-over from the separator is inevitabl~ and both the oil and water feeds :from the
separator were fast-loop sampled to measure the composition of each process stream.
Conditioning circuits allow operation of the facility at temperatures between 10°C and S0°C,
to within 1°C, and the operating pressure may be up to 10 bar gauge. Temperature and
pressure were also measured at several points in the test section. at the oiVwater monitors and
at the reference meters and mixing sections.
A programmable logic controller (PLC) controls operation and data acquisition from all the
field instruments. During the evaluation all relevant data were written to a test log file which
was time- and date- stamped for cross-reference to the SMFM 1000 log file. All
measurements are automatically corrected for temperature and pressure, the reference flow
rates are also compensated for the cany-over in the process streams. All the instruments are
regularly calibrated to local secondary standards, and density curves for the liquid phases are
measured before each evaluation commences.
The oil used during this evaluation was a mixture of stabilised Forties crude (flashpoint 60°C)
and Exxsol D80 in a 70130 ratio. The viscosity of the oil was 12.6 cSt at 18°C. and its density
was 860. 7 kg/m3 at 20°C. Brine was simulated by the additional of magnesium sulphate
(MgS04) salt to de·ionised water at concentrations of25g/l and 50gll. The conductivities of
these solutions at 25°C were 0.64 Sim and 1.05 Sim respectively, while the densities at 20°C
was 1010.45 and 1018.4 kg/m3 respectively. The gas phase was nitrogen.
Qliq GVF
(Vs) 0% 20% 50% 80%
11.1 x x x x
16.67 x x x x
25 x x x
34.72 x x x
For each evaluation the initial condition was set for approximately 20 min in order to enable
the SMFM 1000 to achieve thermal equilibrium with the test fluids. After this period the test
conditions were set for five minutes to allow the conditions to stabilise, before data recording
started. The reference flow rate measurements were recorded at five-second intervals for one
minute, and the SMFM 1000 readings were continuously averaged and written to file every
minute. The PC system clocks were synchronised prior to each evaluation in order to permit
cross-reference and comparison of the readings.
An additional repeatability test was performed continuously over an eight-hour period. The
condition was set to 161/s and 50% Gas Void Fraction (GVF) and the water cut was cycled at
30 minute intervals between 30 and 70%. The SMFM 1000 data were recorded continuously
during the test, the reference data were stored at one-minute intervals over a twenty-minute
period at each stabilised condition.
Figure 9. shows separate graphs for oil flowrates, gas flowrates and water flowrates for all test
points run during the extensive test series. Oil- and water-continuous test points have been
plotted together, but with different legends. During the test, new improved ·software was
installed. Test points from these tests have been given their own legend.
All test points have been plotted against the reference values from the test-rig. A deviation
band of± 20% has been dotted into all graphs.
In figure 10, the test results have been plotted against the claimed specifications given by
Fluenta for oil- and water-continuous flow. In the watercut versus gas fraction graph, both
claimed specifications for oil- and water-continuous flow have been plotted, since these are
not the same. The dotted lines represent specifications for water continuous flow, while the
continuous lines represent oil-continuous flow. Vutually all test points were inside the limits
specified, although the gas flowrate was generally underestimated. Most of these
underestimated test points occurred when the GVF was lower then 20% (then only the venturi
· was used for velocity measurements), when flowrates were low and at the lower limits of the
dP transmitter' s operating range in the venturi.
At high GVF there was a great improvement with the new software. These test points have
been given their own legend.
' »~~~~+-~~~-+-~~~-+-____..__~'::>'t'lir--~~~..:;._~~--;~~~--;
I
I! 11+--~~~+-~~~-+-~_,,.,.;;......::>f"':..._~"7"'""+~~~-t~~~---1r-~~--j
J
8 10-1-~~~~f--~--.~"*-::--::--...:::;~+-~~~~t--~~~-t~~~~-r~~~--;
0 10 °'15_ _ _20.,..
IO
,,,-
IO
.,,,·
/ /"
/
~
-- """
~
-
~-
~
/
--
/ - 0
-
/
, / ·- __ ,,
/
- /
/
_,,,
--
/ -~ _..Jr"
a
/•/ . ,,,,.,,""""' 4
a
...... -
/
·~...... ........
10
0
a
~
~ --·
,....
10
/
____ IO
.,.. IO 111
D
--
• ...,,__
Olll-
---#-2"1
,_
IO
,_
•r-~~~.....,~~~~-r~~~~.-~~~-,.~~~~"'T""~~-_,,,
-;;-''"T~~~""'::>"I'
/
/
/
(25+-~~~~1--~~~--1,_..~~~--t~---:a:--~-+.,...~~-.:of'"--~~~;-:::>"'"'-~~~
I
i 20
i u~~~--11--~~-+---....,....a--.....,.:;....a.___,~-+-~~--11--~~-+~~~~
J
I 10 i----+-~~"""----1---+-~-----+-----i
0...-....--~~-+~~~~~~~~~~-1-~~~~__,~~~~~ .....~~~~--1~~~~---1
0 5 10 15
- . - .... -11111
Figure 9
NSFMW 24-26 Oct. '95
(K.0)
211
•
-
•
·- -·--------
10
IO a
•... • •
• •'
·-c ~
CD
•
•
··•11.
•.,
e
D D
D ~ •
---- ---
D
--·
·11
1·°'-
a--.
·20
0 10 IO to TO 80 100
0•-"')
Liquid flow rates vs. gas fraction
• 20
,.
10 .. -
----- -----,,.,.--- a
.....-
ii) a-
•n -
,_
_H ~.
• a
•
--- ----- ---
"
D
-
-10
_,.
0
--""
IO
•
,..,_ '
0--
1·0l-.1
1CID
»i-~~-+-~~-+~~-+~~--1~~~1--~~-1--~~+-~~+i1 :::;::--
:IO
....... L
1·--·1
iI 10 •
---- __ "'1 __ _
- "
I o • '
1
.s
· 10 .. "
····~;··..
..
•
- .. __ ....
y
-~---
·- . •
"'
•
-
/ aDt~ .....
-211
\a
' ,,i ..
• • D
·...... _. ~_
.......
.a·· ···~-
-40
...,
--"1
..
•
0 10 40 IO eo 7D ID 1111
The gas flowrate was consistently under-estimated across the range of GVF tested, and the
magnitude of the error increased with GVF. During the limited range of tests in which the new
software was used, gas flowrate measurements were much improved, particularly at high
GVF.
This trial demonstrated that there may be scope for improvement in the software model; this
could have a beneficial impact on the meter's performance, particularly at high GVF.
The repeatability of all measurements was seen to be very good when cycling from oil- to
water-continuous regimes.
On the basis of these excellent results, Amerada Hess found the meter acceptable for
installation subsea at the South Scott field.
South Scott is an extension of the main Scott reservoir and contains recoverable oil reserves of
approximately 60 million barrels. The South Scott manifold is located S km south of the Scott
platform. The field has been developed with a manifold for four wells and a production line
tied .into the Scott platform. No test line has been installed. Savings obtained by Amerada Hess
. in using this SMFM I 000 instead of a traditional test line solution are expected to be around
GBP 8 -9 million. Ref /4/.
The SMFM 1000 has now been installed on the seabed as part of the South Scott manifold
and after recent successful commissioning and checks of communication and static readings
we are currently waiting for "first oil" which is due before the end of September.
3. G. High, Kva!mer FSSL Ltd; K.H. Frantzen, Fluenta als; M Marshall, Amerada Hess
(UK) Ltd: "On-Line Subsea Multiphase Flow Measurement", Houston March 1995
4. Subsea Engineering News vol 12, no. 1; 23. March 1995 "Subsea multiphase meter to
save millions at South Scott"