P & I Design Ltd

Process Instrumentation Consultancy & Design

2 Reed Street, Gladstone Industrial Estate,
Thornaby, TS17 7AF, United Kingdom.
Tel. +44 (0) 1642 617444 Fax. +44 (0) 1642 616447
Web Site: www.pidesign.co.uk

EMERGENCY RELIEF SYSTEM (ERS)

SIZING SOFTWARE
METHODS & PRACTICE
by John E. Edwards
jee@pidesign.co.uk

MNL043A 04/04
PAGE 1 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

Contents
1.0
2.0

3.0

4.0

5.0

Introduction
Relief Sizing Fundamentals
2.1 Vessel Models
2.2 Vent Flow Models
Definitions and Assumptions
3.1 % overpressure
3.2 Friction factors
3.3 Relief Device Discharge Coefficients
3.4 Equivalent Lengths
3.5 Discharge Pipe Considerations
3.6 Inlet / Outlet Pipe Sizing Rules
3.7 Environment F Factor
3.8 API Valve Selection
3.9 Exposed Area Considerations
3.10 Methods for Evaluating Gas Density
Methods Basis
4.1 Relief System Sizing General
4.2 Relief System Sizing Flow
4.3 Heat Models
4.3.1 External Fire
4.3.2 Chemical Reaction
4.3.3 Abnormal Operation
4.4 Sizing Methods for Single Phase Flow
4.4.1 Single Phase Liquid
4.4.2 Single Phase Gas or Vapour
4.4.3 Dry Steam
4.5 Sizing Methods for Two Phase Vapour Liquid Flow
4.5.1 Method for Predicting Onset of Two Phase Vapour Liquid Flow
4.5.2 Coupling Equation and Derivations
4.5.3 Omega Method for Two Phase Vapour Liquid Flow
4.5.4 Henry Fauskes Homogeneous Non-equilibrium HNE Model
4.5.5 Equilibrium Rate ERM Model
Nomenclature
Diagrams
Figure 1

ERS Sizing Strategy

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 2 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.

J. Wilday and J. Etchells, Workbook for Chemical Reactor Relief System Sizing
HSE Contract Research Report 136/1998.
Emergency Relief System Design Using DIERS Technology
The Design Institute for Emergency Relief Systems Project Manual, 1992.
Sizing Selection and Installation of Pressure Relieving Devices in Refineries
API 520, 7th Edition, January 2000.
Guide for Pressure Relieving and Depressuring Systems
API 521, 4th Edition, March 1997.
Venting Atmospheric and Low-Pressure Storage Tanks
API 2000, 5th Edition, April 1998.
D.K.Das, Specialist Consultant, Chemstations Private Communications.
H.K.Fauske, Pressure Relief and Venting
IChemE Symposium Series No 102, 1987.
L.K.Spink, Principles and Practice of Flowmeter Engineering
The Foxboro Company, The Plimpton Press, 1967.
Flow of Fluids Crane Company, Publication 410M, 1998.
J.C. Leung, Simplified Vent Sizing for ERS in Reactors and Storage Vessels
AIChemE Journal, Vol 32, No 10, October 1986.
Chemstations Emergency Relief System Sizing Software Validation
P&I Design Ltd, MNL 044, J.E.Edwards, 2002.

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 3 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

1.0

Introduction
Emergency Relief System (ERS) design has been covered extensively by the Design
Institute for Emergency Relief Systems (DIERS) (2) of the American Institute of Chemical
Engineers (AIChemE), the American Petroleum Institute Standards (API) (3,4,5), the UK
Health and Safety Executive (HSE) (1) and work by Leung (10), Fauske (7), Huff, Etchells,
Wilday and many others, with research still ongoing.
The practising engineer is required to provide safe, practical and timely solutions to ERS
design problems without having the opportunity to study the topic in depth. This technical
note attempts to provide a comprehensive review of ERS fundamentals, definitions and
design principles to allow the solution of cases encountered in normal practice. For more
complex cases or when in doubt reference should be made to an appropriate authority or
specialist in the field.
Chemstations have developed a menu driven Emergency Relief Sizing program supported
by a component data base with comprehensive thermodynamic options. The design
fundamentals presented in Sections 2, 3 and 4 have been used to validate the software by
spreadsheet methods in Report MNL 044 (11). The extensive features of the Chemstations
sizing program cover most situations encountered in practice and are summarised below

Batch or continuous process applications

Reactive or non reactive systems
Relief device: conventional relief valve, balanced relief valve, bursting disc or relief
valve and bursting disc in combination
Relief device: set pressure and allowable overpressure
Design or rating for pressure vessel or atmospheric vessel
Vessel model options: churn turbulent, bubbly or homogeneous equilibrium
Single phase phase vent flow: liquid, gas / vapour or steam flow
Prediction of two phase flow onset with relief sizing using Leungs method
Two phase vent flow model options: homogeneous equilibrium (HEM), equilibrium
rate (ERM) and Henry Fauskes homogeneous non-equilibrium (HNE)
Heat model code options: API 520/521, NFPA 30, OSHA 1919.106 or API 2000
Heat model options: specify heat rate, tempered runaway reaction or specify additional
heat input
Vessel geometry: cylindrical/spherical, vertical/horizontal
Vessel end type: ellipsoidal, spherical, conical or flat
Vessel parameters: diameter, tan-tan height and design pressure
Vessel operating liquid level
Vessel void volume calculated from geometry and liquid level or specify
Height of bottom tan line from grade where fire can be sustained
Options for fire fighting protection available and insulation factor
Relief device stagnant or flowing input stream composition and condition set from
CHEMCAD component database with appropriate thermodynamic options applied
Relief device inlet and outlet diameter and length specified and pressure drop
calculated and checked for design code compliance

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 4 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

2.0
2.1

Relief Sizing Fundamentals

Vessel Models(1, 2 p 8)
In relief system design involving chemical reactions there are three main types of process
system:
Vapour pressure systems
The pressure generated by a runaway reaction is entirely due to the vapour pressure of the
reacting mixture which rises as the temperature of the mixture increases during a thermal
runaway. Vapour pressure systems are tempered in that the temperature and reaction rate
are controlled during relief due to latent heat removal.
Gassy systems
The pressure generated by a runaway reaction is entirely due to a permanent gas which is
evolved by the chemical reaction. Gassy systems are untempered in that pressure relief
does not control the temperature or reaction rate.
Hybrid systems
The pressure is due to both the evolution of a permanent gas and increasing vapour
pressure with increasing temperature.
Vessel flow models estimate the liquid swell (degree of vapour-liquid disengagement) as a
function of vapour throughput. The vessel flow models are coupled with vent flow capacity
models to determine the vapour mass fraction (onset of two phase-vapour-liquid venting)
and the total mass flow rate entering the vent line. There are three flow models considered:
Homogeneous
Assumes zero vapour-liquid disengagement in the vessel where vapour bubble rise velocity
relative to the liquid is zero. The vapour void fraction entering the vent line is the same as
the vapour void fraction in the vessel. Typical of extremely viscous or foamy conditions or
for when venting time is too short for an appreciable level rise.
Bubbly
Assumes uniform vapour generation throughout the liquid with limited vapour-liquid
disengagement in the vessel. The liquid phase is continuous with discrete bubbles.
Churn-turbulent
Assumes uniform vapour generation throughout the liquid with considerable vapour-liquid
disengagement in the vessel. The liquid phase is continuous with coalesced vapour regions
of increased size relative to the vessel bubble model.
Two phase-vapour-liquid venting is typical of gassy(foamy) and hybrid reaction systems
represented by the homogeneous and bubbly vessel models.
Reactor System Types

VAPOUR PRESSURE

REACTION SYSTEM

HYBRID

TEMPERED

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

GASSY

UNTEMPERED

MNL043A 04/04
PAGE 5 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

2.2

Vent Flow Models (1, 2)

Vent flow models for ERS sizing are based on single-phase or two-phase flow.
Single phase models considered are for Liquid, Vapor or Gas and Steam applications
which are covered rigorously in the standards API 520 (3 ) , API 521(4) and API 2000 (5).
ERS sizing involving chemical reactions invariably involves two-phase flow and three
models are considered, namely: Homogeneous Equilibrium Model (HEM) (1 p79)
Applicable to both flashing and non-flashing two-phase flow, assumes uniform mixing of
the phases across the pipe diameter with no phase slip(mechanical equilibrium), thermal
equilibrium between the phases, and complete vapour/liquid equilibrium.
Henry-Fauskes Homogeneous Non-equilibrium(HNE) Model (1 p165, 2 p64)
Basis for family of models with varying extent of non-equilibrium behaviour, the rate of
flashing at the choke point being a specified fraction (N) of the equilibrium value. Twophase mixture treated as a single pseudo-component and vapour/liquid equilibrium is
maintained in the reactor during the relief process. Assumes ideal vapour/liquid
equilibrium, vapour phase is an ideal gas, frictionless nozzle, turbulent flow, no heat gain
or loss and overpressure in the range 10-30% of absolute relief pressure.
Equilibrium Rate Model (ERM) (2 p68)
Special case of HNE family. Applicable to flashing two-phase flow only (vapour pressure
systems), assumes saturated liquid entering the vent, no flashing (non-equilibrium flow)
until the choke point and then flashing at equilibrium rate (N=1) at the choke point and the
vapour phase is treated as an ideal gas.
The following assumptions are considered conservative but may result in oversizing and an
impractical vent system design requiring further rationalisation and optimisation.
For tempered systems
Two-phase rather than single phase vapour relief.
Homogeneous vessel behaviour.
Homogeneous equilibrium model (HEM) for relief system flow.
For untempered systems
Two phase relief at the point of maximum gas generation rate.
Churn turbulent level swell vessel behaviour which minimises early loss of reactants.
Homogeneous equilibrium model (HEM) for relief system flow.
Vent Flow Models

VENT FLOW

HIGH VISCOSITY

INHERENTLY FOAMY

NON FOAMY

HOMOGENEOUS

BUBBLY

CHURN TURBULENT

VESSEL MODEL

VESSEL MODEL

VESSEL MODEL

TWO PHASE RELIEF

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

SINGLE PHASE RELIEF

MNL043A 04/04
PAGE 6 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

3.0

Definitions and Assumptions

3.1

% Overpressure
P CF P RG
% overpressure = mG
100
P RG

(3.1)

where CF = contingency factor

= 1.1 for non-fire contingency
= 1.21 for fire contingency
Note if PRG is set at the maximum allowable working pressure(MAWP)which is often the
design pressure then the allowable % overpressure is 10% for the fire contingency case and
21% for the non fire contingency case.
Note if the overpressure is stated based on absolute pressure then this must be corrected to
gauge pressure basis for entry into CHEMCAD as follows:

% overpressure = Pm P R 100
P RG

(3.2)

The relief device set pressure is normally specified in gauge units and CHEMCAD
requires the set pressure in absolute units which is found as follows
Pm = (P R Patm ) (1 + % overpressure 100) + Patm

3.2

(3.3)

Friction Factors (6 - Memo 5/ 96)

A generalised Fanning friction factor is used for establishing line loss in the inlet to the
relief device, with 0.005 being the default value. The following parameters are calculated
at the relief device inlet:
X0 v 0
+
X0 v 0 (1 X0 ) vf
Two-phase(subscript TP) viscosity TP = 0 g + (1 0 )f
0 =

Void fraction is given by

Re =

Reynolds Number

D G
TP

(3.4)
(3.5)
(3.6)

The following correlations are now used to calculate the Fanning friction factor based on
the user specified roughness factor R(ft) and all units in lb, ft, hr.
Re<=
<= 2100

f = 16 Re

3.24
7 0.9 2
R

Re > 2000 , f = 0.25 0.86859 ln
+

Re
D

Material
Steel
Cast Iron
Plastic
Glass

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

(3.7)
(3.8)

Roughness Factor R (ft)

0.00015
0.00085
0.000017
0.000005

MNL043A 04/04
PAGE 7 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

3.3

Relief Device Discharge Coefficients (6 - Memo 12/ 96)

Cd =
=
=
=

Device factor = Cd K c
0.975 for relief valve
0.953 for relief valve and rupture combination (6, memo 12/16/96)
line factor for rupture disc alone with 0.625 as default
0.625*0.6 = 0.375 for rupture disc with vacuum support

(3.9)

Kc = 1 for relief valve alone

= 1 for rupture disc alone
= 0.9 for safety valve certified per the rules of ASME (3)
= 0.9 for relief valve and rupture disc combination
N = 4f (L D) + 1.6

(3.10)

Where 1.6 is resistance coefficient of rupture disc(1.1) + entrance(0.5)

And f is the Fanning friction factor, assume 0.005 as first trial.
3.4
3.4.1

3.4.2

3.4.3

Equivalent Lengths (6 - Memo 10/ 96)

Valves
Valve Type
Gate Valve
Ball Valve
Plug Valve
Swing Tilting Check Valve
Vertical Lift Check Valve
Angled Lift Check Valve
Fittings
Fitting Type
Std 90 degree elbow
Std 45 degree elbow
Long radius elbow
180 degree return
Std Tee (Run)
45 degree angled Tee (Run)
Std Tee (Branch)
45 degree angled Tee (Branch)

L/D
13
3
18
100
600
55
L/D
30
16
20
50
20
10
60
30

Pipe Parameters
Pipe Parameters
Entrance
Exit
Expansion
Contraction

K
0.78
1.0

(1 )
0.5 (1 )
2

Where is the ratio of the smaller pipe diameter to the larger pipe diameter

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 8 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

3.5

Discharge Pipe Considerations (9)

All equivalent lengths can be referred to the same size by the resistance coefficient K
Relationship (9- p2-10 Equ 2-5) as shown in Equation 3.11 below
d
a
=
Ka Kb
d
b

where a defines K and d with reference to the pipe size to which all resistances are to be
expressed, and where K is defined (9- p2-8 Equ 2-4) as shown in Equation 3.12 below
L
K = 4f
D

The Fanning friction factor f is assumed identical in both pipes giving Equation 3.13 below
L

D a

3.6

L da

D b db

Inlet / Outlet Pipe Sizing Rules (6 - Memo 5/96)

For compressible flow, including two phase vapour-liquid, the relief device inlet line
pressure drop due to friction should not be more than 3% of the differential set pressure of
the relief device protecting a vessel under the jurisdiction of ASME.
If the safety device is not oversized the nominal diameter of the inlet line should be at least
equal to the nominal diameter of the inlet of the safety device.
If the safety device is a combination of rupture disc and relief valve, include 75D as the
equivalent length of pipe contributed by the rupture disc for pressure drop calculation.
The following is a summary of the general rules:
Conventional Spring Loaded Safety Relief Valves
Differential set pressure
Inlet pressure drop
Outlet pressure drop

PS = Set pressure Constant Back Pressure

PI = 3% PS
PO = 10% PS

Balanced Bellows Spring Loaded Safety Valve

PS = set pressure
PI = 3% PS
PO= 30%PS

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 9 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

3.7

Environment F Factor (3, 4 p17)

For ERS sizing based on fire exposure heat models the environment factor F is
determined from the relevant standards whose key recommendations are shown below:
Environment
Bare Vessel
Insulated Vessel
Sprinkler System
Both Insulation and Sprinklers

F Factor
1.0
0.3
0.3(Note 1)
0.15

Note 1: API 2000 does not recommend any allowance for sprinkler systems on bare
vessels (facilities assumed failed) F = 1.0 is recommended.
For insulated vessels the following additional requirements are to be considered:
The insulation conductance in
Btu/hr.ft2.F
4
2
1
0.67
0.5
0.4
0.33
3.8

F Factor
0.3
0.15
0.075
0.05
0.0376
0.03
0.026

API Valve Selection (6-memo)

Certified valves are open within 10% of set pressure for ASME Section VIII and 3% for
ASME Section I with certified flows being 90% of the observed flow when operating at
the previously stated overpressure.
The flow capacity of the valve body, at an acceptable back pressure, should be greater than
the flow capacity of the nozzle. It is not unusual to have an outlet pressure rating lower
than the inlet pressure rating of a relief valve.
Standard API valve type designations and nozzle sizes follow, where the first number
indicates the inlet nominal size(in) and the number following the nozzle letter code is the
outlet size (in).
Valve Type
1xDx2
1xEx2
1 x F x 2
1 x G x 2
1 x H x 3
2xJx3
3xKx4

Orifice Area(in2)
>0.06 & <=0.110
0.196
0.307
0.503
0.785
1.287
1.838

Valve Type
3xLx4
4xMx6
4xNx6
4xPx6
6xQx8
6xRx8
8 x T x 10

Orifice Area(in2)
2.853
3.6
4.34
6.38
11.05
16.0
26.0

If vent area is <= 0.06 the relief valve is based on x 1 size body.

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 10 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

3.9

Exposed Area Considerations

The heat transfer area to be used is determined by the agency governing the equipment
operation and the criteria for selection is detailed in the following table.
AGENCY GOVERNING THE EQUIPMENT OPERATION
EQUIPMENT

NFPA-30
OSHA 1919.106

API-520(3) & API-521(4)

Operating
Pressure
> 15 psig
Area up to the maximum
horizontal diameter or up
to the height of 25 ft.,
whichever is greater.

API-2000(5)
Operating
Pressure
< 15 psig
55% of total exposed area.
or surface area to a height
of 30ft above grade
whichever is greater.

1. Sphere

55% of total exposed area.

2. Horizontal Tank

75% of total exposed area.

If under 200 ft2, use 100%
of total exposed area.

Area equivalent to the

average inventory level up
to the height of 25ft.

75% of total exposed area

or the surface area to a
height of 30ft above grade,
whichever is greater.

3. Vertical Tank

100% of total exposed

area for the first 30 ft.
Exclude bottom area if the
bottom
is
flat
and
supported on ground.

Area equivalent to the

average inventory level up
to the height of 25 ft.

As in OSHA.

4. Process Vessel

Area equivalent to the

average inventory level up
to the height of 25 ft.
-

5. Fractionating Column

Area equivalent to liquid

level in bottom, and
reboiler if part of the
column, plus liquid hold
up from all trays up to a
height of 25 ft.

For vertical tanks the area of the top head is excluded if the top head is more than 30ft above the
ground for NFPA, OSHA and API 2000 codes.
However the top head area is included if there is two phase flow conditions. (6, memo 4/97)

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 11 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

3.10

Methods for Evaluating Gas Density (8)

The gas density at relief device inlet conditions is calculated using the following general
relationships. Non ideal behaviour is considered by determining the compressibility factor
Z which can be estimated using the Compressibility Charts of Nelson and Obert (8,p376)
provided the pseudocritical absolute temperature Tc and pressure Pc are known by
evaluating the ordinates from the following
TR =

Tf and
Pf
PR =
Tc
Pc

(3.14)

The gas density is calculated from the following

G =

273
M W Pf
MW
P
3
f
kg / m 3
lb / ft or =
10.73 T f Z
22.415
Z
Tf

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

(3.15)

MNL043A 04/04
PAGE 12 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

4.0
4.1

Sizing Methods Basis

Relief System Sizing General
Figure 1 provides details of the general strategy used for ERS design and sizing.
This procedure can be summarised as follows

If relief from vessel decide if churn-turbulent, bubbly or homogeneous (2.1)

Decide if system is vapour pressure, gassy or hybrid (2.1)
Determine if single or two phase venting by checking onset conditions (2.2)
Determine required relief rate W (mass flow/sec) based on worst case (4.2)
Calculate proposed relief system capacity G (mass flow/unit area) (4.3)
Calculated required relief device area A from A=W/G

The conditions for the onset of two phase flow (4.5.1) depend on the vessel geometry, the
charge volume, the vessel flow model selected and the relief system vapour generation rate
(mass flow/sec).
The phase split or quality of the venting stream is the key parameter, as defined by the
mass fraction of gas/vapour X or the volume (void) fraction , which depends on the flow
regime in the venting vessel and are related by the following
X =

g
g + (1 )f

(4.1)

(X

X g

g + ((1 X

) f )

(4.2)

If single phase flow has been used for initial sizing it should be confirmed that the onset
conditions have not been established thus negating this assumption.
The required relief rate W is determined from the heat model selection (4.3) and whether we
have single phase or two phase relief. The heat rate is determined from the relevant design
codes for heat input QH, the heat generated by the reaction QT or equipment failure modes.
The single phase gas flow capacity G is determined from the relevant sizing equation in
the applicable standard and using relief device manufacturers data for sizing coefficients.
For bursting discs if friction is significant (LE > 40) the Omega (4.5.3) method is used.
The two phase gas flow capacity G depends on the vent flow model selected and can be
determined from the Omega method which is applicable to vapour pressure, gassy and
hybrid systems.

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 13 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

4.2

Relief System Sizing Flow

The required relief rate W can be determined from the following relationships depending
on whether we have single or two phase vapour venting
For single phase vapour pressure systems, with no runaway reaction, we have
W =

QH
hfg

(vapourisation rate)

(4.3)

For runaway reactions the adiabatic temperature rise (dT/dt) and the peak rate of
permanent gas evolution QG are determined experimentally. For single phase vapour
venting the following apply (1, A6.2.2)
Tempered system W =

dT m Cf
dt

Untempered W = QG G +

hfg

+ QG G (vapourisation rate + gas evolution rate) (4.4)

m Cf dT
hfg dt

(gas evolution+corresponding vapour rate) (4.5)

max

For two phase vapour pressure systems Leungs method (1, A5.7) can be used where
W =

mR qav
0.5

V
h
fg

+ (Cf T
mR vfg

)0.5

(4.6)

The average value of the difference between the vapour and liquid specific volumes vfg at
the set and relief pressures can be determined from
vfg =

1
1

g f

(4.7)

At constant pressure there is no temperature rise and equation (4.6) simplifies to

2

W =

mR qav vfg

QT m R vfg

V hfg

(4.8)

V hfg

For the external heating case Leungs method equation (4.6) can be used with a modified
value of heat release rate per unit mass q as follows (1 A5.7)
qmod = qav +

2QH

(4.9)

mR

Alternatively the modified form of Leungs equation (10) is used where we have
Tm Ts =

ln
GA Cv
QT

m0 QT vfg

1 + V hfg
V GA hfg
m0 Cv vfg

(4.10)

It is recommended that average values of the physical properties and heat input rate,
between the vent opening pressure and accumulated pressure, be used. Liquid specific heat
at constant pressure Cf can be used in place of specific heat at constant volume Cv .
For single phase liquid venting due to thermal expansion refer to section 4.4.1

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 14 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

4.3
4.3.1

Heat Models
External Fire (2)
Heat input rates as a result of external fires have received extensive investigation by
several organisations including API, NFPA and OSHA.
The operating pressure determines the applicable standard as follows:
API 520 (3) / API 521 (4)
API 2000 (5)

operating pressure > 15 psig

operating pressure < 15 psig

In API 520/API 521 the heat input QH is determined from:With adequate drainage and fire fighting equipment
Without adequate drainage and fire fighting equipment

QH = 21000 FA0.82
QH = 34500 FA0.82

(4.11)
(4.12)

In API 2000 for low pressure storage tanks the heat input QH is determined from:In the range 0.4 x 106 < QH < 4.0 x 106 Btu/hr
6
6
In the range 4.0 x 10 < QH < 9.95 x 10 Btu/hr
Where,
QH = total heat absorption
A = total wetted surface
F = environmental factor

QH = 20000 A
QH = 199300 A0.566

(4.13)
(4.14)

Btu/hr
ft2
(API 521 Table 5)

The exposed area to be used is determined by the agency governing the equipment
operation and the criteria for selection is detailed in previous section 3.9.
Single phase vapour venting design basis for non-foamy systems is usually considered the
appropriate vent flow model for external fire cases unless the vessel is completely liquid
filled or the entrainment velocity (7) UE is exceeded where
0.25

g
f

U E 3 .0
2
g

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

(4.15)

MNL043A 04/04
PAGE 15 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

4.3.2

Chemical Reaction (1) (2)

For relief device sizing involving runaway exothermic reactions experimental data is
required from reaction screening techniques to establish the vessel model from reaction
behaviour (vapour pressure, gassy or hybrid) and the vent flow model from the reaction
type (tempered or untempered).
For most exothermic runaway reactions, the reaction rate which determines the heat release
rate, increases exponentially with temperature. Hence for tempered systems the relief
pressure should be set as low as practicable consistent with operational requirements for
inerting procedures and pressure transfers.
For tempered runaway reactions (vapour pressure or hybrid) calorimetry can determine the
tempering temperature at the relief pressure and if it can be established that the reaction
will continue to temper until completion the Omega method can be used based on a
homogeneous two phase system (vessel and vent flow model).
CHEMCAD process modelling and simulation software can be used to predict the
tempering temperature at the relief pressure and temperature at the accumulated pressure.
Alternatively if the rate of temperature rise at the relief device set pressure (dT dt )s and
the maximum rate of temperature rise at the maximum design overpressure (dT dt )m are
known the average value of heat release rate per unit mass of reacting mixture qav can be
calculated from
dT dT
qav = 0.5 Cp
+

dt s dt m

(4.16)

where Cp is the average liquid specific heat (kJ/kg K)

Leungs method for vapour pressure systems can now be used based on a homogeneous
two phase system. The ERM vent model may also be used if friction can be ignored such
as in the case for safety valves.
For untempered runaway reactions (gas evolution) the reaction rate depends primarily on
the temperature and not on the pressure. The operation of the relief system cannot control
the rate of runaway reaction but simply acts to remove material from the reactor. The
vessel model with the most vapour/liquid disengagement ie churn turbulent, will normally
provide the worst case. Relief device sizing is not based on a heat model, as such, but uses
calorimetry to establish the peak gas evolution rate QG (m3/s) by determining the peak rate
of pressure rise (dP dt )e , the corresponding temperature Te and the rate of temperature rise
(dT dt )e suitably adjusted for thermal inertia, where subscript e refers to experimental
parameters and Tc the containment vessel temperature.

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 16 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

For closed tests the following relationship is used where mR is the vessel reactant mass
V dP V dT T m

e R
QG =

Tc me

dt
dt
P
T

e

(4.17)

For open tests the above can be simplified to

V dP T m
e R
QG =

P dt e Tc me

(4.18)

If the containment vessel temperature Tc is not given it can be approximated by

Tc = 0.5 (Te + Tamb )

(4.19)

The required relief rate W (kg/s) can now be determined from

W = QG

mR
V

(4.20)

This relief rate is now used in the Heat Model option field Specify Vent Flowrate
In the absence of any calorimetric data the heat of reaction should be established from the
components heats of formation using the following relationship
HPr oducts = HRe ac tan ts

(4.21)

This heat of reaction is now used in the Heat Model option field Specify Heat Rate.
It should be noted that a facility is available to add an additional Heat Input which could be
as in the case of a services failure or external fire.
4.3.3

Abnormal Operation
Heat input rates can be user defined from a knowledge of the thermal characteristics of the
reaction system and associated jacket services. This is particularly useful when considering
operational and control system failure modes such as in the case of maximum heat input
from vessel jacket services systems.
Control valve or pressure regulator failure is a common design case for determining the
relief rate with the flowrate being based on the maximum differential pressure attainable
across the valve seat.
Another common application is to protect equipment against overpressure due to thermal
expansion of a trapped liquid.

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 17 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

4.4
4.4.1

Sizing Methods for Single Phase Flow

Single Phase Liquid (3)
The sizing method presented is based on API 520 7th edition which should be referred to
for more specific details and the relevant capacity factor correction graphs.
When capacity certification is required per ASME Section VIII, Div L
A =

Qgpm
38 Cd K w K c K v

SG
p i po

in US customary units (4.22)

Where
A
= required effective discharge area (in2)
Qgpm = flow rate (US gpm)
Cd = manufacturers rated coefficient of discharge, 0.65 as default
Kw = back pressure correction factor, 1.0 if atmospheric, otherwise refer (3)
Kc
= combination correction factor, 1.0 if no rupture disc or 0.9 with rupture disc

Kv

2.878
342.75 1
+
= viscosity correction factor from 0.9935 +
0.5
1.5
Re

SG
pi
po

Re

(4.23)

= specific gravity of liquid at flowing temperature to water at standard conditions

= upstream relieving pressure (psig)
= back pressure (psig)

When capacity certification is not required the method assumes a Cd of 0.62 and an
overpressure of 25%.
A =

Qgpm
38 Cd K w K c K v K p

SG
1.25p po

(4.24)

Where
Kp = over pressure correction factor, 1.0 at 25% overpressure otherwise refer (3)
A common application is to protect equipment against overpressure due to thermal
expansion of a trapped liquid. Expansion rates can be determined (4) from the following:
Qgpm =

B QH

(4.25)

500 G C

Where
B = cubical expansion coefficient per F at expected temperature
QH= total heat transfer rate Btu/h
G = specific gravity reference water=1.00 at 60F
C = specific heat trapped fluid Btu/lbF
If the blocked in liquid has a vapour pressure higher than the relief pressure then ERS
design is based on vapour generation rate.

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 18 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

4.4.2

Single Phase Gas or Vapour (3)

The sizing method presented is based on API 520 7th edition which should be referred to
for more specific details and the relevant capacity factor correction graphs.
Sizing equation for critical flow in US customary units
A =

T Z

C K d P1 K b K c

MW

(4.26)

k +1

C = 520

Where

k =

A
W
Cp
Kd
P1

Kb

Cp
Cv

2 k 1
k

k +1

(4.27)

M W Cp
MW Cp 1.99

(4.28)

required effective discharge area (in2)

flow rate (lb/hr)
specific heat (Btu/lb F) If k cannot be estimated use C = 315 or refer Table 8 (3)
effective coefficient of discharge. For preliminary sizing use the following
0.975 for pressure relief device with or without rupture disc
0.62 for rupture disc
upstream relieving pressure (psia)
Set pressure(psig) x (1+ accumulation) + atmospheric pressure
Accumulation = 0.1 for unfired emergency relief
= 0.21 for fire condition
= 0.33 for piping
= back pressure correction factor from manufacturers data for balanced bellows
valves only. Conventional and pilot operated valves Kb = 1.0.
=
=
=
=
=
=
=
=

Refer Figure 35 (3) or estimate using

Kc

T
Z
MW

Kb =

1 r

735 F

(4.29)

= combination correction factor,

= 1.0 if no rupture disc
= 0.9 when rupture disc installed in combination with pressure relief valve
= relieving temperature of inlet gas or vapour (R = F + 460)
= compressibility factor of gas or vapour at inlet relieving conditions
= molecular weight of gas or vapour at inlet relieving conditions

Sizing equation for subcritical flow in US customary units with above nomenclature
A =

T Z

(4.30)

735 F K d K c M W P1 (P1 P2)

k
F =
k 1

(r

)2 k 1

r (k 1) k

1r

P2
r =
P1

(4.31)

P2 = back pressure (psia)

Balanced bellows relief valves operating in the subcritical region may be sized using
the critical flow formulae.

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 19 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

4.4.3

Dry Steam (3)

The sizing method presented is based on API 520 7th edition which should be referred to
for more specific details and the relevant capacity factor correction graphs.
Sizing equation for steam service at critical flow in US customary units
A =

(4.32)

51.5 P1 K d K b K c K N K SH

Where
KN = correction factor for Napier equation
= 1.0 where P1 1500 psia
=

0.1906 P1 1000
where P1 > 1500 psia and 3200 psia
0.2292 P1 1061

KSH = superheat steam correction factor from Table 9 (3)

= 1.0 for saturated steam at any pressure

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 20 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

4.5
4.5.1

Sizing Methods for Two Phase Vapour Liquid Flow (2, p 27)
Method for Predicting Onset of Two Phase Vapour Liquid Flow
Determine the vapour capacity F (lb/h) of the relief device at the desired set pressure.
The superficial vapour velocity jg is then calculated using the following relationship:
jg =

F
3600 g AR

(4.33)

where g is the vapour density (lb/ft3) and AR is the vessel cross sectional area (ft2)
The bubble rise velocity U (ft/s) is now calculated using the following relationship:

U =

) (

K 32.174 2.2046E 03 f g

)] 0.25

(4.34)

0.5

Where K is 1.53 for churn turbulent fluid and 1.18 for bubbly fluid
The dimensionless superficial gas/vapour velocity due to flow is given by:
F =

jgx

(4.35)

The dimensionless superficial vapour velocity at which two phase-vapour-liquid flow

commences is now determined from the following:
Churn turbulent fluid

Bubbly fluid

2
1 C0

(4.36)

(1 )(1 C )
3

(4.37)

Where C0 the correlating parameter is given by:

Vessel Fluid Model
Churn-turbulent
Bubbly
Homogeneous

C0 Best Estimate
1.5
1.2

Note that

C0 Conservative
1.0
1.0
1.0

V Vf

(4.38)

Where V is the total reactor volume and Vf is the volume of liquid in the reactor.
In summary we can now predict the onset of two phase-vapour-liquid flow from:
Two phase-vapour-liquid venting is predicted if

F >=

All vapour venting is predicted if F <

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 21 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

4.5.2

Coupling Equation and Derivations (2, 6 - Memo 1/96)

The relief of a two phase-vapour-liquid fluid is to be achieved in accordance with the
overpressure requirements of the relevant design code or standard.
When the onset of two phase vapour liquid flow is predicted the coupling equation (2, p5) is
solved to determine the flowing quality of vapour at the vent inlet.
The coupling equation is a vapour material balance at the vent inlet as follows:
Total weight rate of vapour entering the vent =
Vapour disengaging the liquid surface + Vapour in the aerated liquid passing to the vent

'
,
X0 A G = jg g ,s A R + Xm A G jg g ,s A R

(4.39)

Where:
j'g = superficial vapour velocity at liquid surface necessary to just swell the liquid to the
top of the vessel (ft/s)
3
g ,s = vapour density at relief pressure and corresponding saturation temperature (lb/ft )
Xm = stagnation quality at liquid surface, where Xm =

m g ,s

(4.40)

m g ,s + (1 m )f

m = void fraction at the liquid surface

For Churn-Turbulent Vessel Model the condition at the relief device inlet, denoted by
subscript 1, is used where
21 =

(1

2a
+ C0 a

( )

v 21 = vf 0 + X1 vfg

(4.41)

(4.42)

1 1

g f

vfG =

(4.43)

a = average void fraction in the swelled liquid for Churn-Turbulent Vessel Model
Co = Correlation parameter
For Homogeneous Vessel Model it can be shown
0 =

X 0 vg 0

X0 =

(4.44)

v 20

0 g 0
0 g 0
=
20
0 g 0 + (1 0 )f 0

v 20 = X 0 vg 0 + (1 X 0 ) vf 0 =

V
m0

1
20

(4.45)
(4.46)

Alternatively in the terms of initial vessel conditions with vapour volume Vg , vessel
gross volume V and reactant mass m0 we have
0 =

Vg

V
= 1 Vf =
V m0
V

the average vessel void fraction

(4.47)

The vent mass flux G (lb/ft2s) can be determined from frictionless flow:

G = 2gc 21 (P PB )144

1/ 2

where vent two phase density 21 =

G A
20

Two phase flow vent rate (ft3/s)

Q 20 =

Gas generation rate (ft3/s)

QG = jg AR

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

X 0 g ,s + (1 X0 ) f

(4.48)
(4.49)
(4.50)

MNL043A 04/04
PAGE 22 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

4.5.3

Omega Method for Two Phase Vapour Liquid Flow (2, 6 - Memo 1/ 96)
Omega is a correlating parameter in the following equation of state
v

1 =
v0

p0

1
p

(4.51)

Where < 1 gassy two phase, = 1 gas only single phase and > 1 flashing two phase
The method is applicable when the following conditions are satisfied
Homogeneous two phase turbulent flow
If two phase flashing mixture, vapour liquid equilibrium is maintained and
expansion process is isenthalpic
If two phase non flashing mixture thermal equilibrium is maintained between
phases and expansion process is isentropic
Friction factor constant
Conditions of applicability T TTc < 0.9
and
P PTc < 0.5
Flowing quality X0 (vapour wt flow/total wt flow) is determined by solving the coupling
equation (4.4.1) where subscript 0 refers to stagnation conditions in the upstream vessel
The Omega method is used to calculate the dimensionless mass flux G*c and the critical
pressure ratio c using Figure A8.2 (1) or by curve fitting
For vapour pressure and hybrid systems
vfg 0 2
0 (

+ 1 0 )f 0 Cf T0 Pv 0
=
hfg 0
k

Vapour volume vg 0 = X 0
0 =

vg 0
vg 0 + vf 0

(4.55)

Liquid volume vf 0 =

(4.53)

g 0

vfg 0 =

(4.52)

1
1

g 0 f 0

(4.56)

(1

X0 )
f 0

0 = vf0 + X0 v20

(4.54)
(4.57)

Note that the term based on stagnation liquid conditions can also be represented by inlet
two phase density or specific volume as follows

(1 0 )f 0 = 21 =
For gassy systems the above reduces to

(4.58)

v 21

0
k

(4.59)

c = 0.6055 + 0.1356 log 0.0131 (ln

Gc =
*

If < 4 then G*c is calculated from

0.66

G*c =

If >= 4 then G*c is calculated from

0.39

)2

(4.60)
(4.61)
(4.62)

Calculate choked nozzle flow Gc from G c = G*c P0 (4.63) Relief area A=W/G (4.64)
v0

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 23 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

4.5.4

Henry-Fauskes Homogeneous Non-equilibrium HNE Model (1 A53 , 2 p64 )

This method is applicable to two-phase mixtures treated as a single pseudo-component
with vapour/liquid equilibrium being maintained in the reactor during the relief process. It
assumes ideal vapour/liquid equilibrium, vapour phase is an ideal gas, frictionless nozzle,
turbulent flow, no heat gain or loss and overpressure in the range 10-30% of absolute relief
pressure.
No vapour disengagement ie system inherently foamy
HNE model uses the following relationship for relief system sizing where all pseudocomponent properties are at the relief conditions.
A =

1 m R dT dt R
2 F (Pm P R )

CfR
TR

(4.65)

where CfR is liquid specific heat at the relief pressure and (Pm PR ) is the maximum
accumulated pressure and relief pressure difference.
The frictional correction factor F values are shown below as a function of vent line
equivalent length
LE/D
0
50
100
200
400
600

F
1
0.87
0.78
0.68
0.57
0.5

Vapour / liquid disengagement

The area calculated from the equation above is multiplied by the disengagement factor
D R
1

(4.66)

where R is void fraction in reactor at relief pressure and D is void fraction at

disengagement. Note application of this correction will result in a reduced relief area.

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 24 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

Equilibrium Rate ERM Model (1

9.4.2 , 2 p68 )

Special case of HNE critical flow model obtained by setting parameter N equal to unity.
The model is only applicable to flashing two phase flow with vapour phase treated as an
ideal gas ie tempered vapour pressure systems. It assumes saturated liquid entering the
vent, no flashing (non-equilibrium flow) until the choke point and then flashing at
equilibrium rate (N=1) at the choke point.
The simplified ERM correlation is given by
dP
G =
dT

T0 0.5

Cf 0
0

(4.67)

CHEMCAD component data base can be used to evaluate the slope of the vapour pressure
vs temperature curve at stagnation condition.
For single component systems or treating as a single pseudo-component mixture with a
narrow boiling range and using the isenthalpic rate of vaporization at the critical condition
we obtain the simplified ERM non-equilibrium correlation which uses stagnation
conditions subscript 0 in the reactor.
G =

hfg 0

(4.68)

vfg 0 (Cf 0 T0 )

0.5

The method finds its main application in the sizing of relief valves on vapour pressure
systems with an appropriate discharge coefficient applied.
The mass velocity in an ideal nozzle will depend on the model selected and the flowing
quality as shown by Diers.(p 64)

Nozzle Ideal Flow vs Inlet Quality

Ref Diers Fig II-2
12
Mass velocity G
kg/m2s x1000

4.5.5

10
8
6
4
2
0
0.01

0.1

10

100

Inlet Vapor Quality % ww

HEM FLOW

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

ERM FLOW

HNE FLOW

MNL043A 04/04
PAGE 25 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

5.0

Nomenclature
Definition of symbols used unless defined otherwise in main text for specific equations.
SI units are the preferred but consistent units should be used as appropriate.
Some correlation constants require units to be used as defined in main text.
A
AR
B
Cd
Cf
C0
Cp
Cv
D
F
f
G
Gc
gc
hfg
jg

Kc
KN
KSH
Kp
Kv
Kw
L
MW
me
m0
mR
Patm
Pf
Pm
PmG
PR
PRG
QH
QG
QT
qav
qmod
Re
Tamb
Tc

cross-sectional area of relief device (m2)

cross-sectional area of vessel (m2)
cubical expansion coefficient
relief device discharge coefficient
liquid specific heat capacity (J/kgK)
correlating parameter two phase flow
average liquid specific heat at constant pressure (J/kgK)
liquid specific heat at constant volume (J/kgK)
relief system (or pipe) diameter (m)
environmental factor
Fanning friction factor
two phase mass flowrate per unit flow area (kg/m2s)
choked nozzle flow (kg/m2s)
gravitational constant consistent units
latent heat of vaporisation (J/kg)
superficial vapour velocity consistent units
relief device combination factor
correction factor for Napier equation
correction factor for superheat steam
overpressure correction factor
viscosity correction factor
back pressure correction factor
line length (m)
vapour or gas molecular weight
mass in experimental calorimeter (kg)
initial mass in the reactor (kg)
mass in reactor at relief pressure (kg)
atmospheric pressure (N/m2 abs)
flowing pressure (N/m2 abs)
maximum accumulated pressure (N/m2 abs)
maximum accumulated pressure (N/m2 gauge)
relief pressure (N/m2 abs)
relief pressure (N/m2 gauge)
relief system external heat input (J/s)
peak rate of permanent gas evolution (m3/s)
total heat release rate of reacting mixture (J/s)
average heat release rate per unit mass reacting mixture (J/kg s)
average heat release rate per unit mass reacting mixture modified for external
heating (J/kg s)
Reynolds number
ambient temperature (K)
containment vessel temperature (K)

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 26 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

5.0

Nomenclature

Tf
UE
V
Vf
Vg
v0
vf
vfg
W
X
Xm
X0
Z

m
0

R
G
g
f
SG
21
TP
g
f

H
PS
PI
PO
(dT/dt)
(dP/dt)e
(dT/dt)s
(dT/dt)m

flowing temperature (K)

entrainment velocity (m/s)
volume of reactor (m3)
volume of liquid in reactor (m3)
volume of vapour in reactor (m3)
specific volume at stagnation conditions at inlet to relief system (m3/kg)
liquid specific volume (m3/kg)
difference between vapour and liquid specific volumes (m3/kg)
mass flowrate (kg/s)
vapour or gas mass fraction
stagnation quality at liquid surface
mass fraction of gas/vapour in a two phase mixture at stagnation conditions
vapour or gas compressibility
vapour or gas volume(void) fraction
void fraction at liquid surface
initial void fraction in vessel
ratio of small pipe diameter to large pipe diameter
roughness factor (ft)
gas density (kg/m3)
gas or vapour density (kg/m3)
liquid density (kg/m3)
specific gravity of flowing fluid
two phase flowing density at relief device inlet (kg/m3)
two phase dynamic viscosity (Ns/m2)
gas/vapour dynamic viscosity (Ns/m2)
liquid dynamic viscosity (Ns/m2)
liquid surface tension (N/m)
superficial vapour velocity (dimensionless)
correlating parameter
heat of formation in consistent units
relief device set pressure minus back pressure (N/m2 abs)
inlet pipe pressure drop (N/m2 abs)
outlet pipe pressure drop (N/m2 abs)
adiabatic temperature rise (K/s)
peak rate of pressure rise in experimental calorimeter (N/m2s)
temperature rise at set pressure (K/s)
temperature rise at maximum design overpressure (K/s)

Subscripts
This is not a comprehensive list but is included for clarification only
0
condition at initial or stagnation
1
condition at inlet
2
two phase condition
21
two phase condition at inlet
R
relief condition
f
liquid
g
gas or vapour

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 27 OF 28

Emergency Relief System (ERS)

Sizing Software Methods and Practice

FIGURE 1
ERS SIZING STRATEGY

P & I Design Ltd

2 Reed Street, Thornaby, UK, TS17 7AF
Tel: + 44 (0)1642 617444
Fax: + 44 (0)1642 616447
www.pidesign.co.uk

MNL043A 04/04
PAGE 28 OF 28

AREA from
API 520
3.6.3
4.29

SUBCRITICAL FLOW TEST

API 520
Pout > Pc P 3.6.1

(1) HEM vessel

model 2PH flow same
vessel void fraction
(X=m/V)
(2) Untempered hybrid
refer HSE A6.2.2
(3) All gassy reactions
are untempered. Single
phase refer HSE
A6.2.2
(4) Tempered single
phase systems refer
HSE A6.2.1
(5) See Technical
Paper references 1&2

NOTES

PID/JEE/02

FIGURE 1

ERS
SIZING
STRATEGY

AREA from
API 520
3.6.2
4.25

Pout Pc

CRITICAL

GAS or VAPOUR

OSHA 1910.106
NFPA 30

PROCESS
TANKS
>15 PSI
API 520

STORAGE
TANKS
VAC to 15PSI
API 2000

AREA from
API 520
3.7
4.31

AREA from
API 520
3.8.1
4.22

ASME VIII

CAPACITY
CERTIFY

2PH
CONDITION
TEST
I-B7

AREA from
API 520
3.8.2
4.23

NONE

TWO PHASE
GAS LIQUID

LEUNGS
METHOD
DIERS VI-A7
HSE 8.3

HENRY-FAUSKES HNE
HOMOGENEOUS
NON-EQUILIBRIUM
3.2.1.1 & HSE A5.3

AREA from
FLOW W
MASS FLUX G
A=W/G

LEUNGS
METHOD
HEM

LEUNGS
METHOD
HEM
HSE 6.3.2

EQUILIBRIUM
-RATE
ERM
DIERS 3.2.1.2

OMEGA
METHOD
HEM

OMEGA
METHOD

G
GASSY
FLOW
MODEL
I-A3
TEMPERED
HYBRID (2)

CALCULATE
FLOWING
QUALITY X
DIERS I-C2

OMEGA
METHOD
HEM
HSE A8.4

TEMPERED

TEMPERED

LEUNGS
METHOD
DIERS VI-A7
HSE 6.3.2

VOLATILE (4)

VOLATILE (4)

TEMPERED
HYBRID (2)

TYPE
OF
PROCESS

CONDITION

NON FLASHING LIQUID

SINGLE PHASE

PROCESS

N
o (1)
HEM

(5)

GASSY (3)

PEAK GAS
EVOLUTION
HSE 7.3
HSE A5.10

EXTERNAL HEAT

STEAM

PHASE
TYPE

W
RELIEF
FLOW
II-3

BUBBLY

VESSEL
MODEL
I-A2

PROCESS WITH

MODIFIED
LEUNGS
METHOD
HSE A5.7

EXTERNAL
HEAT

CHURN TURBULENT

Note