Gas Pipe Blowdown Investigation

Citation:
Rajiwate, Farhan Liyakat Husain. 2011. Investigation of compressible fluid behaviour in a vent pipe during
blowdown. M.Phil. Curtin University, Department of Chemical Engineering.
Permanent Link:
http://espace.library.curtin.edu.au/R?func=dbin-jump-full&local_base=gen01-era02&object_id=186160
The attached document may provide the author's accepted version of a published work.
See Citation for details of the published work.
Department of Chemical Engineering
Investigation of Compressible Fluid Behaviour in a Vent Pipe during

Blowdown
Farhan Liyakat Husain Rajiwate
This thesis is presented for the Degree of

Master of Philosophy (Chemical Engineering)
of
Curtin University
August 2011
Declaration
To the best of my knowledge and belief this thesis contains no material previously published by
any other person except where due acknowledgment has been made.
This thesis contains no material which has been accepted for the award of any other degree or
diploma in any university.
Signature:
Date:
...
Dedications
To Mum and Dad
Never regard study as a duty, but as the enviable opportunity to learn to know the liberating
influence of beauty in the realm of the spirit for your personal joy and to the profit of the
community to which your later work belongs
Albert Einstein
Abstract
In the process industry, upset conditions can result in the release of fluids to the atmosphere.
Such a release process is known as Blowdown. Accurate modeling and prediction of the
blowdown process is important in determining the consequences of venting operations and the
design conditions required for vent and flare systems. The predicted information such as the rate
at which the fluids are released, the total quantity of fluids released and the physical state of the
fluid is valuable and helps in evaluating the new process designs, process improvements and
improves the safety of the existing processes.
Blowdown events, amongst other transient processes, are the subject of particular interest to the
chemical, oil/gas, and power industries. In the process plants, particularly in the hydrocarbon
industry, there are many large vessels and pipelines operating under pressure and containing
hydrocarbon mixture. Depressurization of such equipments is frequently necessary during
maintenance, and in an emergency it may have to be rapid. Hazards arise because of the very
low temperatures generated within the fluid during the process and also from the large total
efflux and high efflux rates that arise from the large inventory of the long pipelines and high
pressure vessels. This inevitably leads to a reduction in the temperature of the vessel / pipeline
and associated vent system, possibly to a temperature below the ductile-brittle transition
temperature of the material from which the vessel, pipeline or piping is fabricated. To date, a
number of blowdown models and simulation codes related to pressure vessels and pipelines
have been developed to estimate the blowdown conditions in pressure vessels and pipelines.
There is no general model developed specifically for analyzing the conditions developed in a
vent pipe.
The scope of this work encompasses investigating the behavior of compressible gas in a vent
pipe, during venting, by developing a vent pipe model. A fluid dynamic and thermodynamic
approach is used in developing the model. The investigation is focused on the pressure,
temperature and flow rates of flowing gas and pipe wall temperatures. The model is validated
with experimental data generated by performing steady-state venting runs using compressed air.
The model is also validated by comparing the simulations performed in Aspen Hysys for single
component gases such as air, carbon dioxide, methane and multicomponent gases which are in
very close agreement.
II
Acknowledgement
It is great pleasure for me to acknowledge all the people who helped me to accomplish this
dissertation. First of all I would like to express my deepest gratitude to my supervisor Dr. Hari
B. Vuthaluru for giving me the opportunity to pursue my Masters Degree and the incredible
support provided during this study. I would like to thank my co-supervisors Mr. Clinton Smith,
Principal Process Engineer at Atkins Global and Mr. Dennis Kirk-Burnnand, Principal
Consultant at GHD Pty Ltd, for their timely support. Their in-depth advice, encouragement,
easy accessibility and freedom for work helped me to explore new ideas and to complete the
work in time. I would also like to thank Professor Moses Tad, Dean of Engineering and
Professor Ming H. Ang, Chairman of thesis committee, for evaluating my work.
I would like to thank all members of the administrations at Chemical Engineering Department as
well as library staff for their help during this course. Timely help from the secretarial staff, Mrs.
Naomi Mockford is highly appreciated. The support given by the technical staff, notably Mr.
John Murray, Ms. Karen Haynes, Mr. Jason Wright, Mr. Carl Lewis, Mr. Ashley Hughes and
Mr. Pierre Bastouil is worth mentioning. I am also thankful to my colleagues at Independent
Metallurgical Operations Pty Ltd, especially Mr. David Symons, Managing Director, Mr. Steve
McGhee, Director, Mr. Dennis Boska, Project Development Manager and Mrs. Sharon
OReilly, Recruitment Consultant for all the support they have provided during this thesis. I
would also like to thank my friends Faizaan, Jawad and Riyadh for all their help in nontechnical matters.
Finally, all the credentials of this degree go to my beloved parents. This course would not have
been possible without the financial support given by my father Mr. Liyakat Husain
Gulmohamad Rajiwate. I am deeply indebted to my parents, my brother Fahim and sister Faiza,
who have been with me at every step before and during this project completion. Last but not the
least my wife, Simins support before and during this thesis writing cannot be expressed in
words. I also acknowledge everyone who has assisted me directly or indirectly in the completion
of this work. Their assistance is invaluable and shall always be held in high regards.
III
Brief Biography of Author

Author of this thesis Mr. Farhan Rajiwate has completed Bachelor of Chemical Engineering in
year 2007 from Thadomal Shahani Engineering College, Mumbai, India, standing with First
Class Distinction. He worked as a graduate for Bharat Petroleum Corporation Ltd, India where
he was responsible for simulation of MTBE plant. He joined Curtin University, Perth, Australia
in February 2008 with the enrolment in Postgraduate Diploma in Chemical Engineering. On
successfully completing PG Diploma, he enrolled in Master of Philosophy (Chemical
Engineering) to work on the industry related research offered to him by GHD Pty Ltd. During
his time as a researcher, he was employed as a Process Engineer for Independent Metallurgical
Operations Pty Ltd (IMO) and has worked on a number of mineral processing projects related to
copper, gold, iron-ore and manganese with experience in plant commissioning, plant operation,
and designing. Prior to joining IMO, he worked for Nuplex Industries, a polyester resin
company, Perth and SGS Australia, Perth as a vacation-work student. Farhan holds interest in
process engineering related to oil & gas as well as mineral processing, specifically in the field of
process modeling / simulations, process control and plant operations.
He has written the following paper in the support of this thesis:
Investigations into Compressible Gas Behavior in a Vent Pipe during Blowdown, a

paper to be submitted for PSE-2012 conference, Singapore.
IV
Thesis Contents
Dedications.I
Abstract.II
Acknowledgement...III
Brief Biography of Author.IV
List of Figures...VIII
List of Tables..XV
Nomenclature....XVI
Chapter 1
Introduction and Objectives ................................................................................................... 1
1.1
Objectives ...................................................................................................................... 2
1.2
Thesis Outline ................................................................................................................ 4
Chapter 2
Literature Review .................................................................................................................... 5
2.1
Blowdown ...................................................................................................................... 5
2.2
Blowdown Process ......................................................................................................... 6
2.2.1
Blowdown of Pressure Vessel................................................................................ 7
2.2.2
Blowdown of Pipeline ............................................................................................ 8
2.3
Blowdown or Emergency Depressurising Systems ....................................................... 9

V
2.3.1
Pressure Safety Valve .......................................................................................... 10
2.4
Thermophysical Property ............................................................................................. 11
2.5
Blowdown Effects ........................................................................................................ 12
2.6
Investigations into Developed Simulation Codes and Models..................................... 13
2.7
Summary of Literature Review .................................................................................... 24
Chapter 3
Model Development ............................................................................................................... 26
3.1
Development of Mathematical Models ........................................................................ 27
3.1.1
Basic Conservation Equations ............................................................................. 27
3.1.2
Theoretical Aspects Related to Compressible Fluid Flow Behavior ................... 31
3.1.3
Model Assumptions ............................................................................................. 32
3.1.4
Development of Adiabatic Frictional Model ....................................................... 34
3.2
Modeling Approach ..................................................................................................... 46
3.2.1
Simulation Object ................................................................................................ 46
3.2.2
Method of Solution .............................................................................................. 46
3.3
Computations ............................................................................................................... 51
Chapter 4
Results and Discussion ........................................................................................................... 53
4.1
Experimental Design .................................................................................................... 53
4.2
Instrumentation and Data Collection ........................................................................... 55
VI
4.3
Experimental Data Noise Reduction ............................................................................ 56
4.4
Experimental Analysis ................................................................................................. 61
4.5
Experimental Validation .............................................................................................. 62
4.5.1
Experiment VPM-1 .............................................................................................. 63
4.5.2
Experiment VPM-4 .............................................................................................. 67
4.5.3
Experiment VPM-7 .............................................................................................. 70
4.6
Validation with Hysys.................................................................................................. 75
4.6.1
Comparison with Hysys Simulation for Air......................................................... 75
4.6.2
Comparison with Hysys Simulation for Carbon-dioxide ..................................... 87
4.6.3
Comparison with Hysys Simulation for Methane ................................................ 96
4.6.4
Comparison with Hysys Simulation for DBNGP Mixture: ............................... 104
Chapter 5
Conclusions and Recommendations for Future Work ..................................................... 112
5.1
Conclusions ................................................................................................................ 112
5.1.1
Comparison of Vent Pipe Model Predictions with Experimental Analysis ....... 112
5.1.2
Comparison of Vent Pipe Model Predictions with Aspen Hysys ...................... 113
5.2
Recommendations ...................................................................................................... 114
Appendices.116
References..265
VII
List of Figures
Figure 2-1: An air receiver with automatic isolation and blowdown valve installation (Spirax
Sarco Limited 2011) ...................................................................................................................... 7
Figure 2-2: Gas pipeline with a sectionalizing valve in centre, a typical shop fabricated
blowdown riser and valve on the left, and a typical field fabricated riser on the right (Gradle
1984) .............................................................................................................................................. 8
Figure 2-3: Typical layout of vent testing facility (Skouloudis 1992) ........................................... 9
Figure 2-4: Spring loaded PRV (American Petroleum Institute 2008) ........................................ 10
Figure 2-5: Typical inversion curve - Data for nitrogen gas (Wisniak and Avraham 1996) ....... 12
Figure 2-6: Brittle fracture developed in a pressure vessel (Keenan 2009) ................................. 13
Figure 3-1: Variation of fluid properties with friction ................................................................. 36
Figure 3-2: Condition for maximum mass flux ........................................................................... 39
Figure 3-3: Fanno curve (Enthalpy - Entropy diagram) Adapted from (Saad 1993) ................... 41
Figure 3-4: Representation of equation 3-27 ............................................................................... 49
Figure 3-5: Algorithm for vent pipe model .................................................................................. 52
Figure 4-1: Schematic representation of the vent pipe assembly ................................................. 54
Figure 4-2: Noise reduction for pressure transducers P1, P2, P3 ................................................... 59
Figure 4-3: Noise reduction for temperature sensors T1, T2, T3 ................................................... 60
Figure 4-4: Noise reduction for flow-meter ................................................................................. 61
Figure 4-5: Comparison of model predicted stagnation temperatures with experimental
stagnation temperatures for VPM-1 ............................................................................................. 64
Figure 4-6: Comparison of model predicted wall temperatures with experimental wall
temperatures for VPM-1 .............................................................................................................. 64
Figure 4-7: Comparison of model predicted standard volumetric flow rate with experimental
standard volumetric flow rate for VPM-1 .................................................................................... 65
Figure 4-8: Experimental pressure measurements for VPM-1 .................................................... 65
VIII
Figure 4-9: Comparison of model predicted stagnation temperature with experimental stagnation
temperature for VPM-4 ................................................................................................................ 68
Figure 4-10: Comparison of model predicted wall temperature with experimental wall
Figure 4-11: Comparison of model predicted standard volumetric flowrate with experimental
standard volumetric flow for VPM-4 ........................................................................................... 69
Figure 4-12: Experimental pressure for VPM-4 .......................................................................... 70
Figure 4-13: Comparison of model predicted stagnation temperature with experiment stagnation
Figure 4-15: Comparison of model predicted standard volumetric flowrate with experiment
flowrate for VPM-7...................................................................................................................... 72
Figure 4-16: Experiment pressure measurements for VPM-7 ..................................................... 73
Figure 4-17: Aspen Hysys Simulation Flowsheet........................................................................ 74
Figure 4-18: Predicted pressure comparison of vent pipe model with Hysys simulation for air in
the pressure range 100-500 KPa gauge ........................................................................................ 77
Figure 4-19: Predicted temperature comparison of vent pipe model with Hysys simulation for air
in the pressure range 100-500 KPa gauge .................................................................................... 77
Figure 4-20: Predicted mach no. comparison of vent pipe model with Hysys simulation for air in
pressure range 100-500 KPa gauge .............................................................................................. 78
Figure 4-21: Predicted density comparison of vent pipe model with Hysys simulation for air in
Figure 4-22: Predicted velocity comparison of vent pipe model with Hysys simulation for air in
Figure 4-23: Predicted Enthalpy-Entropy (Fanno curve) of vent pipe model for air in pressure
range 100-500 KPa gauge ............................................................................................................ 79
Figure 4-24: Predicted pressure comparison of vent pipe model with Hysys for air in the
pressure range 600-1000 KPa gauge ............................................................................................ 84
IX
Figure 4-25: Predicted temperature comparison of vent pipe model with Hysys simulation for air
in the pressure range 600-1000 KPa gauge .................................................................................. 84
Figure 4-26: Predicted mach no. comparison of vent pipe model with Hysys simulation for air in
the pressure range 600-1000 KPa gauge ...................................................................................... 85
Figure 4-27: Predicted density comparison of vent pipe model with Hysys simulation for air in
the pressure range 600-1000 KPa gauge ...................................................................................... 85
Figure 4-28: Predicted velocity comparison of vent pipe model with Hysys simulation for air in
the pressure range 600-1000 KPa gauge. ..................................................................................... 86
Figure 4-29: Predicted Enthalpy-Entropy (Fanno curve) of vent pipe model for air in the
Figure 4-30: Predicted pressure comparison of vent pipe model with Hysys simulation for CO2
in pressure range 100-500 KPa gauge.......................................................................................... 88
Figure 4-31: Predicted temperature comparison of vent pipe model with Hysys simulation for
CO2 in pressure range 100-500 KPa gauge .................................................................................. 88
Figure 4-32: Predicted Mach no. comparison of vent pipe model with Hysys simulation for CO2
Figure 4-33: Predicted density comparison of vent pipe model with Hysys simulation for CO2 in
Figure 4-34: Predicted velocity comparison of vent pipe model with Hysys simulation for CO2
Figure 4-35: Predicted Enthalpy-Entropy (Fanno curve) of vent pipe model for CO2 in the
Figure 4-36: Predicted pressure comparison of vent pipe model with Hysys simulation for CO2
in pressure range 600-1000 KPa gauge........................................................................................ 93
CO2 in pressure range 600-1000 KPa gauge ................................................................................ 93
Figure 4-38: Predicted Mach no. comparison of vent pipe model with Hysys simulation for CO2
Figure 4-39: Predicted density comparison of vent pipe model with Hysys simulation for CO2 in
X
Figure 4-40: Predicted velocity comparison of vent pipe model with Hysys simulation for CO2
Figure 4-41: Predicted Enthalpy-Entropy (Fanno curve) of vent pipe model for CO2 in the
Figure 4-42: Predicted pressure comparison of vent pipe with Hysys simulation for methane in
Figure 4-43: Predicted temperature comparison of vent pipe with Hysys simulation for methane
Figure 4-44: Predicted Mach no. comparison of vent pipe model with Hysys simulation for
methane in pressure range 100-500 KPa gauge ........................................................................... 98
Figure 4-45: Predicted density comparison of vent pipe model with Hysys simulation for
Figure 4-46: Predicted velocity comparison of vent pipe model with Hysys simulation for
Figure 4-47: Predicted Enthalpy-Entropy (Fanno curve) of vent pipe model for methane in
Figure 4-48: Predicted pressure comparison of vent pipe model with Hysys simulation for
methane in pressure range 600-1000 KPa gauge ....................................................................... 100
Figure 4-53: Predicted Enthalpy-Entropy (Fanno curve) of vent pipe model for methane in
pressure range 600-1000 KPa gauge .......................................................................................... 102
DBNGP in pressure range 100-500 KPa gauge ......................................................................... 106
XI
Figure 4-59: Predicted Enthalpy-Entropy (Fanno curve) of vent pipe model for DBNGP in
DBNGP in pressure range 600-1000 KPa gauge ....................................................................... 109
Figure 4-65: Predicted Enthalpy-Entropy (Fanno curve) of vent pipe model for DBNGP in
pressure range 600-1000 KPa gauge .......................................................................................... 111
Figure B 7-1: Mechanical drawings for vent pipe assembly...................................................... 119
Figure B 7-2: Photographs of test rig ......................................................................................... 124
Figure F 11-1: Vent pipe model predictions for air gas at 100 KPa gauge ................................ 146
XII
Figure F 11-8: Vent pipe model predictions for air gas at sonic conditions .............................. 153
Figure F 11-9: Vent pipe model predictions for methane gas at 100 KPa gauge ...................... 154
Figure F 11-10: Vent pipe model predictions for methane gas at 200 KPa gauge .................... 155
Figure F 11-16: Vent pipe model predictions for methane gas at sonic conditions ................... 161
Figure F 11-17: Vent pipe model predictions for carbon-dioxide gas at 100 KPa gauge .......... 162
Figure F 11-24: Vent pipe model predictions for carbon-dioxide gas at sonic conditions ........ 169
Figure F 11-25: Vent pipe model predictions for DBNGP gas mixture at 100 KPa gauge ....... 170
XIII
Figure F 11-32: Vent pipe model predictions for DBNGP gas mixture at sonic conditions ..... 177
Figure H 13-1: Vent pipe model user specification sheet .......................................................... 188
Figure H 13-2: Newtons Iteration method for nonlinear equations ........................................... 190
XIV
List of Tables
Table 2-1: Various release categories (Nolan 1996) ...................................................................... 6
Table 3-1: Pertinent equations related to frictional flow in constant area vent pipe .................... 34
Table 3-2: Static property relations for adiabatic flow in constant area vent pipe....................... 35
Table 3-3: Stagnation property relation ....................................................................................... 37
Table 3-4: Explicit approximation for Colebrook-White friction factor equation ....................... 43
Table 3-5: Overall average relative errors of fanning friction factor values obtained from
different explicit equations compared with those from the CW equation (Ouyang and Aziz 1995;
Swamee and Jain 1976; Romeo, Royo, and Monzon 2002; Sonnad and Goudar 2006) ............. 44
Table 3-6: Property relations in terms of Mach number .............................................................. 47
Table G 12-1: Comparison of vent pipe model predictions with Hysys simulation in pressure
range 100 - 500 KPa gauge for air ............................................................................................. 179
range 600 - 1000 KPa gauge for air ........................................................................................... 180
range 100 - 500 KPa gauge for carbon-dioxide ......................................................................... 181
range 600 - 1000 KPa gauge for carbon-dioxide ....................................................................... 182
range 100 - 500 KPa gauge for methane .................................................................................... 183
range 600 - 1000 KPa gauge for methane .................................................................................. 184
range 100 - 500 KPa gauge for DBNGP gas mixture ................................................................ 185
range 600 - 1000 KPa gauge for DBNGP gas mixture .............................................................. 186
XV
Nomenclature
(m2)
Area
Cp
Specific heat at constant pressure
(kJ/kg)
Cv
Specific heat at constant volume
(kJ/kg)
Diameter
DH
Hydraulic diameter or pipe diameter
Friction factor
Mass flux
Gravitational constant (m/s2)
Enthalpy
ho
Stagnation enthalpy
Length
L*
Critical length (m)
Mach number (-)
Mw
Molar mass
Mass flow rate (kg/s)
Pressure
Po
Stagnation pressure
(kPa)
P*
Critical pressure
(kPa)
Pr
Prandtl number
(-)
Universal gas constant
Re
Reynolds number
(m)
(m)
(-)
(kg/m2-s)
(kJ/kg)
(kJ/kg)
(m)
(kg/kmole)
(kPa)
(kJ/kmol.K)
(-)
XVI
Recovery factor
(-)
Entropy
(kJ/kg-K)
Temperature
(C)
To
Stagnation temperature
T*
Critical temperature
Taw
Adiabatic wall temperature
Velocity
V*
Critical velocity
Compressibility factor (-)
Density
Stagnation density
(kg/m3)
Critical density
(kg/m3)
Roughness
Specific heat ratio
Viscosity
(C)
(C)
(C)
(m/s)
(m/sec)
(kg/m3)
(m)
(-)
(Pa.s)
XVII
1 Chapter 1
Introduction and Objectives
Designing sustainable processes is one of the key challenges of the chemical industry. This is by
no means a trivial task as it requires translating the theoretical principles of chemical
engineering into design practice. Process design is central to chemical engineering and can be
considered to be the summit of chemical engineering, bringing together all of the components of
that field. Properly designed, constructed, operated and maintained equipment will not fail
provided that its design conditions are not exceeded. Risk reduction is another challenging task.
Safety in process plants starts at the design stage and is followed by series of steps in order to
reduce the risk completely.
In process plants, particularly in hydrocarbon industry, there are a large number of vessels and
process piping which contain / carry large amounts of flammable inventories of hydrocarbons.
Thus, the like hood of an occurrence of an incident or risk associated in such industry is high.
Such incidents can be significantly reduced by performing safety assessments and appropriate
safety precautions. Despite many safety precautions within the hydrocarbon industry, equipment
failures or operator errors may cause upset in process conditions beyond safe levels. If these
conditions rise too high, they may exceed the maximum strength of process vessels and process
piping systems. This can result in the rupturing of process vessels or piping, causing major
releases of toxic or flammable hydrocarbons. Such a sudden release process is called
Blowdown. Blowdown events, amongst other transient processes, are the subject of particular
interest to the chemical, oil/gas, and power industries. Blowdown can be an unexpected process
as seen on ruptured pipelines/process vessels or can be planned during maintenance of the
process equipments. Accurate modeling and prediction of the blowdown process is important in
determining the consequences of venting operations and the design conditions required for vent
and flare systems. The primary purpose for blowdown is to reduce pressure and remove
inventory in the least amount of time possible. Hazards mainly arise due to the changes in
equipment process conditions taking place during the blowdown process especially high efflux
rates. This inevitably leads to a reduction in the temperature of the vessel / pipeline and
associated vent piping system, possibly to a temperature below the ductile-brittle transition
temperature of the material from which the vessel, pipeline or piping system is fabricated. At a
temperature below the ductile-brittle transition temperature, the equipment material has a much
greater tendency to shatter on impact instead bending or deforming. It is under these
1
circumstances that the lowest wall temperatures will often be observed. In such cases, prior
estimations of the resulting temperature drop in the fluids and the equipment involved are of
primary importance. Such estimations can be predicted by developing models for performing
simulations of blowdown operations.
Today, safety is equal in importance to production and has developed into a scientific discipline
which includes many highly technical and complex theories and practices. More complex
processes require more complex safety technology. Examples of the technology of safety
include hydrodynamic modeling of flow through relief systems, developing mathematical
techniques to determine various ways that processes can fail and the probability of its failure etc.
Many blowdown models related to pressure vessels and pipelines have been developed till date
but each one has their own pros and cons. There is no general model developed specifically for
analyzing the fluid conditions developed in a vent pipe. A simple model for analyzing a gas
blowdown in vent pipe is required. The main objective of this work is to investigate the effects
of changes in gas flow conditions at different pressures and develop a simple steady-state vent
pipe model and validate the developed model by performing simulations in Aspen Hysys.Plant
and experimental analysis.
1.1 Objectives
As mentioned above, accurate prediction of blowdown conditions is of primary importance. A
number of blowdown models have been developed but no specific model is available for
predicting the blowdown conditions in a vent pipe. A thorough investigation of gas behavior in a
vent pipe during blowdown is required. Therefore, this study aims at developing a simple model
for a vent pipe by performing steady-state calculations in MS Excel simultaneously utilizing
Visual Basic code. Thermophysical properties for the gases are extracted from the REFPROP
software by writing a Visual Basic code. The REFPROP software calculated the thermophysical
properties using the GERG 2004 equation of state. This equation of state has been proved to be
better than AGA8-DC92, Peng-Robinson and other cubic equation of state. Pressure and
temperature variations of gas, temperature distribution on the vent pipe wall and the mass flow
through the vent pipe are the key parameters which are to be predicted by modeling. These
parameters govern the entire steady-state venting process. Hence, to investigate this venting
process of the gas in a vent pipe and to validate the model, a 24m long test rig is designed and
constructed. Experiments related to compressible gases such as air are conducted. The literature
available on venting through vent pipes is very scarce. Modeling of vent pipes associated with
pressure vessels and long pipelines is mentioned in literature but these models are based on
2
hypothetical assumptions (no validation). Few validated models do exists but are not available
on commercial scale. This investigation will be a significant contribution to the field of
blowdown operations.
The main aim of this research is to achieve the following objectives:
1.
Design and construction of a test rig. A combination of knowledge of related processes and
application of chemical and mechanical engineering first principles will be used to
satisfactorily design and fabricate the test rig.
2.
Investigation of the behavior of fluids during blowdown using fluid dynamics and
thermodynamic approach.
3.
Development of a vent pipe model into Microsoft Excel Visual Basic in order to predict the
pressures and temperatures of the inventory (gas) and the vent pipe wall temperatures
experienced during venting.
4.
Analyzing the results obtained from blowing down the test rig with air gas and providing a
brief discussion with respect to thermodynamic theories.
5.
Validating the developed model with the results obtained from the test rig blowdown and
Aspen Hysys.Plant.
1.2 Thesis Outline

This thesis comprises of the following in detail as shown below in the form of a flowchart:
INTRODUCTION AND OBJECTIVES

Need for Development of Vent Pipe Model during Blowdown
Clearly Stated Objectives
LITERATURE REVIEW
Reviewing Established Theories on Blowdown Operation
Reviewing Developed Blowdown Models
MODEL DEVELOPMENT
Fluid Dynamics and Thermodynamic Approach
Development of Mathematical Models
Modeling Approach
Computations
RESULTS AND DISCUSSIONS

Validation of Developed Model with Experimental Data
Validation of Developed Model with Aspen Hysys
CONCLUSION AND RECOMMENDATION

Brief Summary on Model Development and Validation
Future Work
.
4
2 Chapter 2
Literature Review
A brief literature review related to the blowdown of pressure vessels / pipelines, accentuating
the development of a simple steady-state gas flow model in a vent pipe, provides research
progress to date. An extensive literature on blowdown modeling and experimentation related to
pressure vessels and pipelines exists and is discussed in this section of the thesis.
The first section of the literature review explains comprehensively the purpose of blowing down
a pressure vessel / pipeline followed by a brief description of blowdown process in pressure
vessels and pipelines. Different release cases are tabulated and the need to design Emergency
Depressurization System is highlighted in this section. The second section emphasizes on the
current industrial practices in designing Emergency Design Systems (EDS) and operation of a
typical pressure relief valve. The third section involves reviewing of thermodynamic JouleThomson phenomenon taking place during the blowdown. This part will also provide insights
into the responsible parameters for causing changes in process conditions during blowdown. The
fourth section provides details of hazards related to depressurizing a pressure vessel / pipeline.
This part also provides an insight of the brittlement theory related to metals. The next section is
introduced here, which gives an extensive review on blowdown process modeling from safety
perspective and provides detailed investigations performed by various researchers on blowdown
modeling. A quick summarization of the available literature review is provided towards the end
of this chapter with an objective to focus on a simple gas flow vent model.
2.1 Blowdown
In the last few decades, oil & gas industries have shown excellent developing trends with respect
to production and technology. National Petroleum Council (NPC) of United States evaluated the
future demand and supply in oil and gas. This showed a growth by 50-60% in the demand for
energy by 2030 (Holditch and Chianelli 2008). Growing demand for energy produced from
natural resources such as oil and gas, coal, nuclear energy etc. calls for a strengthening in
exploration and development. However, it is evident from the fact that growing demand for
energy will pose a greater risk for the hazards that may arise in process industries. Processes
involved in the production of oil and gas facilities are always associated with risks and should be
recognized for probable hazards. One such process operation is the risk associated with
Blowdown. According to API (American Petroleum Institute 2007), the depressurization of a

plant or part of a plant, or equipment is known as Blowdown
During emergency situations in gas processing plants or on oil or gas platforms, the pressure of
the vessels / pipelines containing inventory must be reduced to avoid possible accidents. This is
mostly done by discharging the inventory to a flare or vent system (Evanger et al. 1995) and it is
called depressurization. The purpose of blowing down or depressurising a pressure vessel or a
pipeline filled with inventory is to prevent the vessel or the pipeline from rupturing against
overpressure caused mainly due to process upsets or during a major fire exposure so that the
resulting impacts to the vessel or pipeline are minimal. Sometimes, leaks due to abrupt rupture
of process vessels or pipelines can result in emergency depressurization of the system. There are
different instances during which blowdown of pressure vessels or pipelines section becomes
necessary. These may be for a planned maintenance schedule especially during the shut-down or
to protect the process equipment (vessel/pipeline) from over pressurization or in emergency
situations which arises in the vicinity of fire. Thus, the entire blowdown process can be
characterized to be a rapid release process. (Nolan 1996) has categorized such releases as
described in table 2-1.
Table 2-1: Various release categories (Nolan 1996)
Catastrophic Failure
Long Rupture
A vessel or pipeline opens completely immediately releasing its

contents.
A section of pipeline is removed being vented to atmosphere
whose CS areas are equal to the CS area of the pipe.
Open Pipe
The end of a pipe is fully opened exposing the CS area of the

pipe.
Vents, PRV
Smaller diameter piping or valves may be opened or fail which

release vapours or liquids to the environment unexpectedly.
Normal
Releases
Operation
Process storage or sewer vents, relief valve outlets, tank seals,

which are considered normal and acceptable practices that
release to the atmosphere.
2.2 Blowdown Process

The physical phenomenon that occurs during depressurization or blowdown begins with an
inventory filled vessel / pipeline reaching a trip pressure and the vessel / pipeline being isolated
(Marian, Vuthaluru, and Ghantala). Such an arrangement is shown in figure 2-1 and figure 2-2
where an air receiver with automatic isolation and blowdown valve is installed and a gas
pipeline with a sectionalizing valve in centre, a typical shop fabricated blowdown riser and
6
valve on the left, and a typical field fabricated riser on the right (Gradle 1984). As cited in the
literature (Richardson and Saville 1991; Haque, Richardson, and Saville 1992; Norris III, Exxon
Production Research Co, and R.C. Puls 1993; Fairuzov 1998), there exists a significant
difference in the blowdown process occurring inside a vessel and long pipeline. Spatial
uniformity of pressure distinguishes a vessel from a pipeline (Haque, Richardson, and Saville
1992).
2.2.1 Blowdown of Pressure Vessel

Skouloudis and Haque et al. have well explained the detailed phenomenon of the blowdown
process from pressure vessels during top venting transients (Skouloudis 1992; Haque et al.
1989) and is addressed here. The initial process is actuated by opening of pressure relief valve
(PRV) in both cases. The pressure vessel filled with inventory comprises of gas zone at the top,
liquid zone at the bottom and sometimes a third zone for free water formed from condensation
below the liquid zone. As soon as the relief valve opens, vapour contained in the freeboard
volume of the pressure vessel will be released and the pressure falls rapidly inside the vessel.
The liquid phase cannot follow this rapid change of pressure with a prompt change in
temperature
and
the
liquid
becomes
superheated. This leads to thermodynamic

disequilibrium between the phases which are
re-established after a short time by vigorous
re-evaporation of the liquid. During this
period the high depressurization rate is
reduced followed by a marked pressure
recovery which might occur when the vapour
Figure 2-1: An air receiver with automatic volume produced by evaporation exceeds the
isolation and blowdown valve installation volume of the mixture which flows out of the
(Spirax Sarco Limited 2011)
vessel. Vapour still discharges through the

vent line together with some droplets
entrained from the interface separating the predominantly liquid and the predominantly vapour
regions of the vessel. As soon as this level reaches the vent line a distinct two phase mixture is
discharged with large liquid content. Nevertheless, the evaporation processes continue and the
thermodynamic disequilibrium is reduced. The interface level gradually collapses so that the
vent line is no longer blocked. Then a predominantly vapour mixture again leaves the vessel
with several liquid droplets entrained. During this process the pressure in the vessel falls
continuously until a new state of equilibrium has been established with the surroundings.
2.2.2 Blowdown of Pipeline

Richardson et al. and Fairuzov have well explained the detailed phenomenon of the blowdown
process from pipeline (Richardson and Saville 1991; Fairuzov 1998) as is addressed here. The
process of pipeline depressurization can be divided into three stages: depressurization wave
propagation, choked critical flow from the
line and unchoked critical flow from the
line. Upon opening the blowdown valve,
an expansion wave travels from the
ruptured or open end of the line to the
intact end of the line. The pressure at the
intact end is unchanged from the initial
pressure. The flow is choked at the
Figure 2-2: Gas pipeline with a sectionalizing valve
in centre, a typical shop fabricated blowdown riser
and valve on the left, and a typical field fabricated
riser on the right (Gradle 1984)
ruptured or open end. After the expansion

wave has reached the intact end of the
pipeline, the fluid pressure inside the
pipeline is very close to the saturation
pressure corresponding to the fluid temperature. The pressure at the intact end starts to fall.
However, this does not affect the flow condition at the open end of the pipeline and is still
choked. The main contribution to the pressure drop in the line arises because of the friction at
the wall. When the pressure in the line starts decreasing sufficiently, the flow from the ruptured
end ceases to be choked. The main contribution to the pressure drop in the pipeline is again
caused due to friction at the pipe wall. In case of flashing liquids, flashing occurs within the
whole pipeline. The flashing process causes constant changes in the flow pattern. The fluid
temperature decreases due to the drop in the fluid pressure. The pipe wall is cooled by the fluid
flowing through the pipeline.
In both cases as the inventory passes through the choke Joule-Thomson expansion takes place.
Rapid cooling takes place due to isenthalpic expansion of the high pressure gas through the
throttling process. Due to Joule-Thomson expansion, the contained inventory cools and draws
heat from the vessel / pipeline walls, thus producing an auto-refrigeration effect or cooling the
vessel / pipeline walls. In case of gaseous phase expansion will take place. If liquid inventory is
present, flashing takes place soon after its pressure reaches the saturation pressure corresponding
8
to the fluid temperature (Fairuzov 1998) and the composition of the inventory changes with
decrease in pressure (Nageshwar 2003). Under this instance the mass flow rate of inventory
depends on the supply pressure and will decrease as the supply pressure decreases (Hong et al.
2004). Such a rapid release process accentuates the designing of emergency depressurizing
system and is discussed in the next section.
2.3 Blowdown or Emergency Depressurising Systems

In process industry, especially in hydrocarbon processing facilities, severe risks with respect to
fire, explosions and vessel ruptures are always associated. Designing safe technology has always
been a challenge for chemical engineers. Among the prime methods to prevent and limit the loss
potential from such incidents are the provisions of
hydrocarbon inventory isolation and removal
system (Nolan 1996). These systems are referred to
as Emergency Depressurizing Systems (Nolan
1996). For the emergency system relief in a
chemical plant several types of venting device are
installed such as nozzles, long pipes with or
without bends, orifice plates or other safety relief
valves. A typical layout of vent testing facility is
shown in figure 2-3. Currently, industry tends to
use American petroleum Institutes Recommended
Practices 520 (American Petroleum Institute 2008)
and American petroleum Institutes Recommended
Practices 521 (American Petroleum Institute 2007)
Figure 2-3: Typical layout of vent testing for specification and designing of emergency
facility (Skouloudis 1992)
depressurization systems (Roberts et al. 2004). API
RP 521 defines a vapour depressurizing system as a protective arrangement of valves and piping
intended to provide for rapid reduction of pressure in equipment by releasing vapours. API RP
521 defines a pressure-relieving system as an arrangement of a pressure-relieving device, piping
and a means of disposal intended for the safe relief, conveyance and the disposal of the fluids in
a vapour, liquid or gaseous state. Such a relieving system may consists of only one pressure
relief valve or rupture disc, either with or without discharge pipe, on a single vessel or line. The
function of blowdown facilities is to provide a means of venting the high pressure gas to the
atmosphere in a relatively short period of time (Gradle 1984). To relieve the overpressure build9
up in the vessel or pipeline, the pressure vessels / pipelines are installed with blowdown valves
or pressure relief valve (PRV) or pressure safety valve (PSV). These valves sense the
overpressure and are actuated automatically or manually to relieve the overpressure by reducing
the inventory and pressure within the isolated process vessel or pipeline section. The relieved
inventory is routed to a safe location e.g. to a blowdown or knockout drum and then to a flare or
a vent system to safely remove the vapours from the area and dispose without impact to the
environment.
2.3.1 Pressure Safety Valve

Typically, hydrocarbon pressure vessels are provided with a pressure safety valve (PSV), to
relieve internal pressure that develops above its designed
working pressure. The purpose of the PSV is to protect the
vessel from rupturing due to overpressure generated from
process condition or exposure to fire heat loads that
generate additional vaporization pressure inside the vessel.
A blowdown valve is a pressure relief valve which is
designed to open at a predetermined pressure in order to
protect a vessel or system from excess pressure by
removing or relieving fluid from that vessel or system. A
typical arrangement of a spring loaded PRV is shown in
figure 2-4. Although, different types of PRVs are
available all differ from each other with respect to their
operating function. As described in (American Petroleum
Figure 2-4: Spring loaded PRV Institute 2008), a spring loaded PRV consists of an inlet
(American Petroleum Institute nozzle which is connected to the vessel or the system to be
protected against overpressure, a movable disc which rests
2008)
on the nozzle head under normal operating conditions and a spring which controls the position
of the disc. The movable disc controls the flow through the nozzle. The spring loaded PRV
works on the principle of force balance which acts on the movable disc on the nozzle. The
spring load is preset to equal a force exerted on the movable disc (closed position) to equal the
force exerted on the closed disc by the inlet fluid through the nozzle. Under normal operating
conditions, the disc is seated on the nozzle head until the pressure exceeds the set pressure. Once
the inlet pressure exceeds the set pressure, the pressure force overcomes the spring force and the
valve opens. When the inlet pressure is reduced to the closing pressure, the valve re-closes. The
10
valve reseats when the inlet pressure or vessel pressure has dropped sufficiently below the set
pressure and this pressure at which the valve reseats is called the closing pressure. The gas then
passes through a vent system to the flare or vent header. A number of thermodynamic changes
take place in the gas properties while releasing the gas through the vent pipe into the
atmosphere. These changes in gas properties can have an impact on the vent pipe, thus, affecting
the material of construction of the metal wall, especially when low temperatures are experienced
in the process. To understand this phenomenon, the thermodynamic physical properties should
be well understood.
2.4 Thermophysical Property

When a gas expands through a restriction from a high pressure to low pressure changes in
temperature takes place. This process occurs under conditions of constant enthalpy and is known
as Joule-Thomson expansion (Shoemaker, Garland, and Nibler 1996). Joule-Thomson expansion
is a thermodynamic physical property which is experienced during blowdown. The temperature
change is related to pressure change and is characterized by the Joule-Thomson coefficient. The
temperature drop increases with increase of pressure drop and is proportional to the JouleThomson coefficient (Maric 2005). Joule-Thomson expansion takes place under adiabatic
conditions such as well insulated vessel or pipeline. In case of an uninsulated vessel or pipeline
the pressure change is rapid or the velocity of flow is high such that no heat transfer takes place.
The Joule-Thomson expansion phenomenon can be well understood by passing a gas through a
restriction while the fluid is allowed to expand adiabatically. During this process, no work is
done and the changes in potential and kinetic energy are negligible. It has been proved that the
gas flow through the restriction results in an isenthalpic (constant enthalpy) process (Jones and
Hawkins 1986). Thus, the gas escaping through the choke from vessel or pipeline into the vent
system will follow an isenthalpic path. At the same time, the gas flow pattern is affected because
of the entrance valve port area and frictional resistance in the vent pipe (Gradle 1984).
According to (Gradle 1984), an increase in the valve port area will increase the mass flow rate
through the vent system resulting in choked flow condition. At the same time an increase in
pressure drop results by an equivalent amount to valve port area opening. The pipe frictional
effects will equally contribute to the pressure drop and will tend to increase the flow path
resistance, thus, reducing the flow rate through the valve. Nonetheless, due to Joule-Thomson
expansion the bold inventory contained in the vessel or pipeline cools and draws heat from the
vessel / pipeline walls thus producing an auto-refrigeration effect. Generally, when JouleThomson expansion takes place, one of the two effects may take place- Joule-Thomson Cooling
11
effect and Joule-Thomson Inversion (heating) effect

(Wisniak
and Avraham 1996). Various authors
(Wisniak and Avraham 1996; Maric 2005) have

investigated
and
modelled
the
Joule-Thomson
coefficients and inversion curves. An inversion line is a

curve
formed
by
passing
through
maximum
temperature points for a given constant enthalpy line.

As shown in the figure 2-5 the inversion curve divides
the pressure-temperature plane for nitrogen gas into two
Figure 2-5: Typical inversion curve - zones. In the zone inside the inversion curve the
Data for nitrogen gas (Wisniak and adiabatic Joule-Thomson effect is positive, so that
Avraham 1996)
decreasing the pressure leads to a decrease in
temperature whereas outside the inversion curve the adiabatic Joule-Thomson effect is negative
and a decrease in pressure leads to an increase in temperature. It is understood that an expansion
that begins from the inversion pressure leads to the highest cooling effect (Wisniak and
Avraham 1996).
2.5 Blowdown Effects

The problem related to the blowdown of pressure vessels / pipelines containing mixtures of
hydrocarbons are well known amongst industries involved in plant designing and hydrocarbon
extraction (Speranza and Terenzi 2005). As discussed earlier, during blowdown /
depressurisation of a pressure vessel or pipeline the most common effect encountered is the
Joule-Thomson Cooling effect. The primary hazard associated during this process is the
occurrence of brittle fracture in the vessel / piping material due to sudden decrease in
temperature.
Generally, steel type such as carbon steel and other ferritic steels which form the material of
construction for most pressure vessels / pipelines become susceptible to brittle fracture with
decrease in temperature (Khazrai, Haghighi, and Kordabadi 2001). The susceptibility of steel
such as carbon steel to brittle fracture is related to temperature. As the temperature decreases,
the susceptibility to brittle fracture increases (King 2006). If the temperature reaches close to or
below the ductile-brittle transition temperature of the vessel / pipeline material of construction,
the equipment will be prone to failure(Mahgerefteh and Wong 1999). The Joule-Thomson
cooling effect provides the mechanism for low temperature exposure.
12
Another
key
factor
which
increases
the
probability of metal brittle fracture is the

minimum level of applied stress to propagate a
brittle fracture. When the temperature of a body
is raised, or lowered, the material expands or
contracts. If this expansion or contraction is
wholly or partially resisted, stresses are set up in
the body (Case, Chilver, and Ross 1999). For
the crack to propagate through the material of
construction, it must have sufficient energy
Figure 2-6: Brittle fracture developed in a
pressure vessel (Keenan 2009)
which is available in the form of overpressure.

At lower temperatures the yield strength is
greater and the fracture is more brittle in nature.
The reason for this could be atomic vibrations (Shackelford 2005). As the temperature of
material decreases, atomic vibrations decreases and the atoms do not slip to new locations in the
material. As the stress increases, the atoms break their bonds and do not form new ones. This
decrease in slippage causes little plastic deformation before fracture. Thus, brittle fracture occurs
with rapid crack propagation and results in a catastrophic failure of a material with little or no
plastic deformation (King 2006). Figure 2-6 shows a pressure vessel under brittle fracture
caused by cold water for a hydrostatic pressure test and then pressurizing the vessel. The
temperature of the water caused the metal to become brittle.
A secondary hazard arises if there is a significant liquid. During complete blowdown of pressure
vessel, the gas-liquid interface reaches the top of the vessel choke. This results in a significant
liquid carryover with the gas into the vent or flare system. Carryover of a significant quantity of
liquid can present considerable operational difficulties to a flare or vent system designed to
handle gas alone(Haque, Richardson, and Saville 1992).
2.6 Investigations into Developed Simulation Codes and Models

The depressurization process is not amendable to simple analysis due to its highly transient
unsteady-state nature. There are several parameters that characterize the venting processes and
are classified according to their significance as geometrical, operational and physicochemical
during depressurization. The influence of these parameters during depressurization is well
studied (Skouloudis 1992). Nonetheless, the resulting effects from blowing down a pressure
13
vessel / pipeline can pose a significant safety hazard (Cumber 2001). Therefore, a fundamental
study of the blowdown process is crucial in the assessment of safety practices and procedures to
prevent or minimise the consequences of controlled or uncontrolled releases (Chen, Richardson,
and Saville 1995a). Predicting the conditions occurring during blowdown has always been a
challenge to chemical engineers (Mahgerefteh and Wong 1999). Consequently, in recent years
there have been a number of theoretical and experimental studies relating to blowdown
simulation with varying degree of sophistication (Mahgerefteh, Saha, and Economou 1999) and
several empirical correlations have been proposed (Weiss, Botros, and Jungowski 1988). These
models / coded programs developed are distinct from each other (very limited) in the range of
applicability.
Several numerical codes are available for monitoring some or all of the parameters which are
directly related to the depressurization of vessels or pipelines. These codes have been developed
for different types of application and although in principle solve similar sets of conservation
equations for the mass, momentum and energy. Despite based on the same principles, these
codes / programs differ significantly from each other in context to describing the
phenomenology of the transient, the method of solving the pertinent equations, homogeneity /
non-homogeneity and thermodynamic equilibrium / disequilibrium assumptions for multiple
phases. A number of benchmark exercises were conducted (Skouloudis 1992) which
concentrated on the hydrodynamic aspects of venting of vessels containing fluids (water /
refrigerant R114) under high pressure, identification of parameters characterizing the emergency
relief as well as the problems associated with the theoretical modeling of such processes with
four American codes namely RELAP, SAFIRE, RELIEF and DEERS
RELAP and its derivatives codes RELAP4/MOD6, RELAP5-EUR/MF (Worth, Staedtke, and
Franchello 1993) were developed to describe the transient single and two phase flows in
complex networks on the basis of a one dimensional approach. Correlations for single phase
natural and forced convection, sub-cooled and saturated nucleate boiling, critical heat flux,
transition boiling, minimum heat flux, annular and dispersed film boiling and calculations for
friction factors are included in the code. RELIEF (Nijsing and Brinkhof 1996) and DEERS
(Skouloudis 1992) codes also use a one dimensional mass, momentum, and energy conservation
equations. RELIEF code discretizes the vessel into several control volumes but a single control
volume for a vent line. DEERS code can be used in the venting of a large variety of systems.
However, the use of a single two phase model throughout the whole transient restricts the
accuracy of its predictions. CHARME-01 (Stoop, Bogaard, and Koning 1986), a thermo14
hydraulic computer program developed in the late 1970s provided more accurate computational
results in comparison with other numerical solution techniques in the calculation of transient
thermo-hydraulic phenomenon. CHARME-01 code based on the Method of Characteristics
(MOC) and includes proper treatment of the shock wave phenomenon. A comparison of
CHARME-01 and RELAP4/RELAP5 was demonstrated by (Stoop, Bogaard, and Koning 1985)
while describing the thermo-hydraulic loading condition of the reactor pressure vessel vent line
in the event of hydrogen being released from the reactor vessel into the vent line. All these codes
consisted of specific models for predicting the different conditions taking place during
blowdown / depressurization of reactor vessels.
The DIERS computer program SAFIRE (System Analysis for Integrated Relief Evaluation) was
developed primarily for vent-sizing calculations and for the interpreting the results of the largescale chemical reacting fluids. SAFIRE code is written in ANSI Standard FORTRAN-77
comprising of 9000 lines of FORTRAN with 66 subroutines (Tilley and Shaw 1990). The main
feature of the SAFIRE is its ability to handle up to 10 simultaneous chemical reactions with 10
components. The program solves one dimensional mass, momentum, and energy conservation
equations in the vent line and can also solve these pertinent equations for vessels; however, it
assumes a single control volume for describing the vessel. The code can model many different
aspects of emergency relief situations such as (Tilley and Shaw 1990)
x
Complex runaway reactions with or without gas generation
External heat loads (eg. Fire)
Venting of gases or mixtures of liquids and gas
Vapor-liquid disengagement in the vessel being vented
Non-idealities in vapor-liquid equilibria and in gas compressibility
Various vessels and vent line geometric combinations
SAFIRE has a wide range of vent flow calculation routines implemented as subroutines.
Example: Compressible gas flow through a nozzle is handled by subroutine GASN using
conventional gas dynamic relationships. Similar subroutine GASLT can also be used to solve
the compressible flow through a nozzle. While there are many vent flow models available in
SAFIRE, not all can be used in all situations. The choice of the most appropriate model for a
particular scenario requires the user to have a detailed knowledge of the range of application of
each model (Cumber 2001). The friction factor for vent line required for calculating the
frictional pressure drop has to be user defined. The two phase friction factor is calculated based
15
on the single phase relationship which is based on the liquid phase viscosity (only). The physical
properties for the components must be provided by the user in the input data in terms of the
coefficients to the correlations included in SAFIRE. The use of several input options in
characterizing the venting process makes the code user-dependent. An improper specification of
a flow model may lead to gross under-sizing of vent system with catastrophic consequences,
thus making the code very versatile. SAFIRE is not an appropriate tool for the inexperienced
user (Tilley and Shaw 1990). The model assumes a Homogenous Equilibrium Model (HEM)
and thermodynamic equilibrium for two phase system (Skouloudis 1992). A further difficulty
with the application of SAFIRE is that model robustness has been found to be a problem
(Cumber 2001).
For long gas pipelines in hydrocarbon service, the most impressive study was found by Botros et
al. (Botros, Jungowski, and Weiss 1989). In this study, a very mechanistic analysis that included
pipeline friction drop was supported by a full scale gas pipeline blowdown. Two physical
models were described one which takes into account the main pipeline as the volume model
(without frictional losses) with stagnation conditions inside the main pipe and the other as the
pipe model (with frictional losses) with velocity increasing towards the exit. Solutions for the
relevant model equations were obtained analytically and real gas properties for the gas (natural
gas) were obtained numerically. Blowdown time was calculated and the results were compared
with those obtained using the graphs (Gradle 1984) and own field measurements of a straight
pipe section and a compressor station yard piping. Effects of stack entrance and friction losses
and discharge coefficient were also evaluated. The study relates only to the main pipeline
section and effects of stack entrance and friction losses upstream of the blowdown valve (throat
area) are evaluated at which point sonic flow discharge results. Depending on the pressure in the
main pipeline, a subsonic or supersonic flow will result downstream of the blowdown valve. The
piping downstream of the blowdown valve or throat is neglected to provide simplicity in
modeling approach.
We agree to the fact that the physical processes taking place during blowdown are a complicated
mixture of several phenomena typically comprising of fluid mechanics, heat transfer, and phase
equilibrium. To investigate into these phenomenon, a programme of experimental work was
carried out (Haque et al. 1989). The experimental work was focused on depressurization related
to pressure vessels which varied from 5 to 110 cm in diameter, with a length to diameter ratio of
10 to 3 respectively. Depressurization experiments were conducted with nitrogen, 70-30%
mixture of nitrogen and natural gas/propane mixtures. Measurements were taken which included
16
the pressure, temperatures at a large number of positions both within the fluid phases and on the
wall of the vessel, and composition, all as a function of time which helped in the understanding
of the blowdown process. Based on the investigations performed and experimental data
available a mathematical model called BLOWDOWN program was developed. The objective
of this model is to be able to simulate all physically significant effects. Initial development
(Haque et al. 1989) of BLOWDOWN incorporated the presence of only two zones: the top
zone contains only vapor together with any suspended liquid-phase droplets; and the bottom
zone containing all liquid phase. The developed model provided a good understanding of the
physical processes occurring during the blowdown, even for multi-component multiphase
systems. However, it should be noted that there might be a possibility of free water formation
settled below zone2. With this in mind, Haque et al. extended the above work and incorporated
zone3 for free water (including dissolved hydrocarbons) in the BLOWDOWN program
(Haque, Richardson, and Saville 1992). This program was validated again (Haque et al. 1992)
with a number of experiments performed on pressure vessels and case studies. The measurement
results and predictions were found to be in good agreement.
The distinction between the blowdown of vessel and blowdown of a pipeline is that there is a
significant pressure difference within the latter but not within the former. This significant
pressure difference is mainly due to frictional effects encountered at the wall of the pipeline.
Also, in case of blowdown of pipelines it becomes necessary in predicting high efflux rates that
arise when the very large inventories are involved. With this in mind, an extension of the
BLOWDOWN program which can simulate the depressurization of a pipeline was undertaken
(Richardson and Saville 1991). Richardson and Saville divided the pipeline into a number of
elements and performed mass, momentum and energy balances for each element with variability
in elemental size to satisfy a number of requirements (Richardson and Saville 1991). Pertinent
equations involved in blowdown of gas line and condensate is well described and the developed
model is validated with two case studies one for the blowdown of the gas line between Piper
and MCP-01 and the other is for the full-bore blowdown of a typical condensate line. A
comparison of BLOWDOWN predictions with the measurements made during eight of the tests
using LPG carried out by Shell and BP on the Isle of Grain in 1985 (Richardson and Saville
1996). Four of the tests were for full-bore depressurizations and four for depressurizations with
orifices at the open ends of the lines. In all cases mentioned above, the BLOWDOWN
predictions were found to be in good agreement.
17
Although mentioned the use of BLOWDOWN program in simulating vessel / pipeline and
associated vent / piping system (Haque, Richardson, and Saville 1992; Richardson and Saville
1991, 1996), no thorough calculation procedures or computer algorithms have been described.
Also, the thermodynamic, phase and transport properties for BLOWDOWN are calculated using
PREPROP, which is a computer package developed to calculate thermo-physical properties of
multi-component mixtures by an extended principle of corresponding states which as well as
introducing uncertainties associated with its accuracy (Jones and Hawkins 1986), makes the
simulation computationally demanding (Mahgerefteh and Wong 1999).
A simple mechanistic model FRICRUP coded in FORTRAN program for predicting the
blowdown process of vessels and pipelines for both single phase and multiphase flow was
developed (Norris III, Exxon Production Research Co, and R.C. Puls 1993). A homogenous
equilibrium model and thermodynamic equilibrium model assumption, along with no relative
velocities between vapor and a liquid phase is assumed. The fact of steady-state hydrodynamic
conditions prevails in the vented pipe after the vessel is presented. Experiments are conducted
incorporating gases such as air, carbon dioxide and carbonated water for the validation of
FRICRUP code. The results of experiment and predictions by model are in good agreement. The
importance of pipe friction during the blowdown process is well highlighted. This factor
confirmed that the modeling of pipelines as vessels can be easily seriously inadequate. Despite
of its sophistication, the model does not agree very well for multiphase flow as can be seen from
experiments performed with carbon-dioxide which could be because of the assumption of
thermal equilibrium. Further experiments were carried out using several hydrocarbon gases
including both methane and heavier mixtures (Norris III and Exxon Production Research Co
1994). The pronounced difference in the blowdown behavior between pipelines and vessels
noted in the non-hydrocarbon experiments was confirmed for the hydrocarbon gases tested. The
basic assumptions for the model remained the same and similar results were obtained as
obtained when dealing with non-hydrocarbon gases.
Investigations into the blowdown of carbon dioxide from initially supercritical conditions have
been performed (Gebbeken and Eggers 1995). The supercritical condition selected for the
blowdown process was such that on pressure release flashing occurs after saturation condition
has reached. Experiments were accomplished for initial conditions that varied in temperature,
pressure, and minimum diameter of the venting line. Results showed that by enlarging the cross
sectional area of the venting line the outgoing mass flow rate from the vessel is increased.
18
Thermo-hydraulic phenomenon were discussed, particularly the pressure transients, the axial
temperature profile, and the axial void fraction profiles.
In order to evaluate the temperature effects of depressurization on the outside surface of the steel
wall, a full scale depressurization tests on parts of the topside piping on a riser platform in
operation was conducted (Evanger et al. 1995).The experimental results generated were
compared to the simulation CFD (Computational Fluid Dynamics) code PIA, developed at
NTH/SINTEF division. A one-dimensional and two dimensional analysis is performed by the
code PIA and incorporates a finite difference technique for numerical calculation of general heat
and mass transfer both in fluid and solid material. A brief description on the experimental set-up
is given and the calculations performed for the outer steel pipe wall temperatures are in good
agreement with the measurements. However, it seems to be that PIA gives satisfactory results
for gas systems with not too much liquid present in the inventory.
Guersts variational principle for bubbly flow was extended to generalized multi-component two
phase dispersions, and formulated a two fluid model for single and multi-component vaporliquid mixtures (Chen, Richardson, and Saville 1995a). In particular focus was on the
development of the energy conservation equation and equations of motion for compressible
single or multi-component vapor-liquid mixtures using a thermodynamic equilibrium
assumption. As described (Chen, Richardson, and Saville 1995a), the Guersts variational
principle allows both phases to be compressible in deriving the momentum equations which
contradicts the definition of compressible flow. In the second part of the article, a simplified
numerical method for solving two phase, multi-component flow equations was proposed and a
detailed study of the blowdown from pipelines containing one and two component flashing
mixtures was presented (Chen, Richardson, and Saville 1995b).
A mathematical model for simulating the blowdown of a pipeline conveying flashing
multicomponent mixtures was developed (Fairuzov 1998). The major features of the model
comprise of hydrodynamic model, break-flow model and heat transfer model are well explained.
Fairuzov suggested that a large amount of heat is transferred from the pipe wall into the fluid
during the blowdown process and hence the adiabatic assumption for simulating the blowdown
process is not valid. Based on this assumption, the effect of thermal capacitance was
incorporated into the model by employing a new approach in the formulation of energy
conservation equation for the fluid flow in the pipeline. The study revealed that the thermal
capacitance of the pipe wall has a significant influence on the two-phase flow behavior and
19
should not be neglected in the analysis of blowdown of long pipelines containing flashing
liquids. The model was compared to experimental data of and the model predictions hold in
good agreement to the experimental data. The effects of friction on the blowdown time were
assessed.
Further development of BLOWDOWN model, based on cubic equation of state, for blowdown
of vessels containing high pressure hydrocarbons was carried out (Mahgerefteh and Wong
1999). The model, termed as BLOWSIM incorporates the Soave Redlich Kwong EOS, Peng
Robinson EOS and the newly developed TCC cubic EOS for simulating vapour space
blowdown of vessels containing multicomponent hydrocarbon mixtures. BLOWSIM model
takes into account the non-equilibrium effects between phases, heat transfer between each fluid
phase and their corresponding sections of vessel wall, interphase fluxes due to evaporation and
condensation, and the effects of sonic flow at the orifice. BLOWSIM predicts the discharge
rates, pressure as well as the fluid and wall temperatures with time. The fluid phase material
balances depending on the zones formed inside the vessel, thermodynamic trajectories for fluid
phases, heat transfer between vessel wall-fluid phases, discharge calculations and calculation of
thermophysical properties are well explained. The performance of BLOWSIM is evaluated by
comparing the predictions generated to the predictions generated from BLOWDOWN as well as
to the published field data for high pressure blowdown of a full size vessel containing a
condensable hydrocarbon mixture. The model accurately predicts the vessel pressures as a
function of time and is in close agreement with BLOWDOWN. The minimum average bulk gas
temperature is predicted to within 2 K, the unwetted wall temperature is overestimated by ~4 K
and the wetted wall temperature is underestimated by ~5 K when compared to measured data.
The authors have provided reasoning for this over-estimation and under-estimation. The
instantaneous formation of liquid phase at the start of depressurization is predicted much earlier
by the BLOWSIM model then when compared to BLOWDOWN program.
An efficient numerical simulation (CNGS-MOC), based on the method of characteristics for
simulating full bore rupture of long pipelines containing two phase hydrocarbon is developed
(Mahgerefteh, Saha, and Economou 1999). The long CPU time has been largely addressed, and
this has been synonymous so far with such types of simulations by using curved characteristics
in conjunction with Compound Nested Grid System (CNGS). Curved characteristics are used as
they can afford the use of much larger discretization grids; while at the same time improve the
global accuracy. The method of characteristics is adopted to simulate the full bore rupture or
blowdown of long pipelines containing condensable or two phase hydrocarbon mixtures. This
20
technique is employed as opposed to Finite Difference method and Finite Element Method as
both have difficulty in handling the choking condition at the ruptured end. The MOC handles
choked flow intrinsically via the Mach line characteristics and is more accurate then the FDM
and FEM. The field data were from pipeline depressurization tests carried out in the Isle of
Grain (Richardson and Saville 1996) as well as those recorded during the night of Piper Alpha
tragedy. The performance of MOC in simulating Full bore rupture throughout the discharge
process is compared to other solution techniques including META-HEM (Chen, Richardson,
and Saville 1995a, 1995b), MSM-CS (Chen, Richardson, and Saville 1995a, 1995b),
BLOWDOWN(Haque, Richardson, and Saville 1992; Haque et al. 1992) and PLAC(Hall,
Butcher, and The 1993). The simulations were performed on the basis of a homogenous
equilibrium model (HEM) in which all phases are assumed to be at thermal and phase
equilibrium. Due to the absence of any theoretical and experimentally justified data for unsteady
friction factor in rough pipes, this parameter was ignored in the model and steady-state friction
factor estimated using the Moody approximation to Colebrooks equation. It is the most accurate
expression available. Two phase mixtures are simply handled by replacing single phase
properties by two-phase mixture properties. The simulations performed consider only rupture in
straight, horizontal well anchored pipelines in which the fluid compressibility is by far smaller
than pipe wall elasticity. Fluid structure interaction can effectively be ignored. Comparison
showed that CNGS-MOC, META-HEM and BLOWDOWN gave very similar predictions with
MSM doing less well and PLAC performing very poorly.
A model for predicting of outflow from high pressure vessels and associated vent pipe during
accidental failure was developed (Cumber 2001). The model was developed with a view of
incorporating its use in the safety assessments of industrial plant used to process or store
flammable material which in turn will provide source conditions for the mathematical models of
gas dispersion or accumulation and fires. For predicting the outflow, Cumber has sub divided
the model into 3 sub models a sub model for the vessel, a sub model for the vent conditions
and a library of physical property data such that thermodynamic and phase information
properties can be calculated as required. Model for a transient blowdown is described. The
model is based on homogeneity of two phase flow and thermodynamic equilibrium assumption
for both vessels and vent pipes. The system of ordinary differential equations is solved using the
fourth order Runga Kutta method. The developed model was compared for validation purposes
with the experimental data (Hervieu 1991), (Gebbeken and Eggers 1995) and (Haque et al.
1992) and the following was concluded. The vessel pressure and mass flowrate prediction is
21
well predicted. The vessel temperature is under-predicted, although this does not have a
significant effect on the predicted mass flowrate. To ensure the robustness of the model, nonlinear system solvers Powells hybrid method and the Simplex method of unconstrained
optimization is incorporated into the model. However, the outflow model can fail when the
phase of the vessel contents changes. This is because the non-linear systems of equations
describing mass and energy conservation is degenerate at the critical point. The mass flow rate
for the gas phase release through a hole is calculated by a variant of the isentropic nozzle flow
equations. The gas phase density is evaluated using the cubic EOS rather than the ideal equation
of state significantly improves the accuracy of the vent model. Liquid phase release is modeled
by the application of Bernoullis equation, including a liquid head contribution where
appropriate. To calculate the mass flow rate for two phase flow through an orifice, the
homogenous equilibrium model has been implemented. The two phase mixture is treated as a
single phase fluid, and the two phases are taken to be in equilibrium with equal velocities and
temperatures. Gas outflow from a vent pipe is calculated by taking the flow of gas from the
vessel to the pipe entrance to be isentropic, and the flow of gas along the vent pipe to be
isenthalpic with friction effects included. The model of liquid flow through a pipe is a direct
extension of Bernoullis equation with friction and entrance losses included. Two phase flow
through a vent pipe is calculated by solving an equation for the conservation of momentum
under the homogenous equilibrium assumption for two phase flow.
A model for the simulation of blowdown of pressure vessels containing two-phase (gas-liquid)
hydrocarbon fluids was proposed (Speranza and Terenzi 2005). Their model is based on a global
mass and energy balance between the phases, gas and occasionally liquid, present in the vessel,
at a very stage of blowdown. The model takes into account the heat transfer taking place with
the external environment, the presence of many components in the vessel and the possibility of
situations in which the phase equilibrium is not appropriate. The model takes into account the
strong cooling effect taking place between the wall of the vessel and the liquid in contact with it
which helps in avoiding cracks in the vessel wall. The model takes into account the
compositional approach, allowing for the presence of many different hydrocarbons within the
vessel, as well as non-equilibrium conditions between the phases. The model was validated by
performing 2 experiments 100% Nitrogen (I1) and a mixture of hydrocarbons (S9). The
predicted conditions during blowdown by the model are in close agreement with the
experimental results. It was suggested that before gas escapes through the choke a rapid motion
is induced by the acceleration of the gas far upstream, and we can imagine it to get mixed and
22
homogenized at all the time, especially in the early stages of the blowdown, while pressure is
dropping steeply. However, the model does not provide any facts related to modeling of gas in
the vent pipe after the choke is mentioned. The model focuses on the average quantities rather
than local variations for homogeneity of fluid, pressure drop and temperature.
Several other authors have analyzed the behavior of blowdown of vessel / pipelines and
associated vent piping system. Analysis and experiment data on the discharge from carbondioxide filled vessels is published in literature (Eggers and Green 1990). Goh has described
simplified pipeline method employing quasi-ideal gas thermodynamics and has shown limited
experimental validation (Goh 1989). Here experiments were performed with air from which the
flow rate for natural gas was estimated. Integrated safety relief valve inlet piping design for
compressible gas flow from an overpressurised pressure vessel was performed (Westman 1997).
The design was based on ideal gas adiabatic flow principles which involved simultaneous
solution of parametric equations derived from these principles. Effects of SRV inlet line
pressure loss and the use of pipe bends is highlighted. Mass flow rates calculations for the inlet
line and nozzle based on isentropic flow are performed and illustrated; however, its use is
restricted only to ideal gas assumption. A simple and practical method for sizing pipelines
incorporating the theories of adiabatic and isothermal frictional flow was investigated (Cochran
1996). However, no validations were provided. Based on the concept of critical length,
calculations relating to compressible fluid flow incorporating non-linear equations were
analyzed (Farina 1997).
23
2.7 Summary of Literature Review

An extensive literature review related to blowdown of high pressure vessels and pipelines has
been carried out. In the last few decades, a number of theoretical and experimental studies
relating to blowdown simulations with varying degree of sophistication were conducted based
on which different blowdown models were developed. Most of these developed models were in
good agreement with the experimental analyses and hence were validated whereas few models
did not provide any validations. Moreover, a number of validated models were not well
documented properly and are not commercially available. The use of such models is beyond the
reach of scientific community. Although detailed investigations were conducted on the fluid
behavior inside the pressure vessel and pipelines, very few investigations related to fluid
behavior within the associated vent piping system were conducted. As addressed in literature
review, very few models provided an insight into the vent pipe modeling. It should be noted that
the impact of piping systems on process plant economics is so great that the initial investment in
piping systems for new installations has been estimated to range from 18 to 61% of the
equipment costs and from 7 to 15% of the total cost of the installed plant (Cochran 1996).
Currently, industry tends to use API Recommended Practices 520 and 521 for specification of
pressure relieving systems. However, these practices are more relevant to the case of fire
scenarios. A thorough investigation of compressible fluids in a vent pipe is therefore required.
The vital and foremost step to tackle this issue is to have more detailed knowledge of events
occurring prior to fluid entering the vent pipe through the relief valve. As cited in literature
(Norris III, Exxon Production Research Co, and R.C. Puls 1993; Norris III and Exxon
Production Research Co 1994), the vent pipe associated with pressure vessels and pipelines for
venting purposes contains no mass or momentum storage. As a result a steady-state
hydrodynamics can be adopted in vent pipe analysis. Second step will be identification of
parameters which bring about changes in the fluid flow conditions along the vent pipe. To date,
a number of blowdown models and simulation codes related to pressure vessels and pipelines
have been developed based on the same pertinent equations (continuity equation, energy
equation and momentum equation) and differ from each other in methods of solving these
pertinent equations. There is no general model developed specifically for analyzing the
conditions developed in a vent pipe.
Due to unavailability of analysis and data applicable to the simulation of a vent pipe, a
combined analytical and experimental program was initiated. The goal was to develop a steadystate adiabatic vent pipe model for a single phase (gas only) compressible gases. The model was
24
programmed into Visual basic in conjunction with MS Excel spreadsheet because of its
simplicity and easy to use user interface. Investigation to be performed will involve determining
the thermodynamic fluid properties, pressure drop, temperature drop and mass flow along the
vent pipe. The vent pipe model will incorporate the newly developed GERG 2004 equation of
state which has proved to be more suitable than other cubic equation of state developed. This
will help the model in predicting more accurately the thermophysical properties during the
venting process. The developed model will be validated with experimental data obtained for air
gas from the test rig designed and constructed in Curtin Universitys facility. The developed
model will be compared to Aspen Hysys.Plant version 7.1, Process Engineering software for
single phase single component gases such as air, carbon-dioxide, methane and single phase
multicomponent gas mixtures, thus providing additional validation.
25
3 Chapter 3
Model Development
Pressure vessels and pipelines, with many more utilization in process industry, nuclear industry,
marine and space industry, operating under extreme of high and low temperatures and high
pressures, are becoming highly sophisticated (Mackerle 1999). Their operations are often
subjected to interference from accidents, corrosion, and human error, etc. A potential of risk is
always associated with such equipments and safe operations is an important issue for operators
worldwide. A safety assessment must be performed on these equipments and a quantitative risk
assessment of their operation should be conducted.
The problems related to blowdown of pressure vessels / pipelines containing compressible gases
are well known among process industries. Process modeling and computer simulation have
proved to be an extremely successful engineering tool for design and optimization of such
processes (Ramirez 1998). The use of simulation has expanded rapidly during the past few
decades because of the availability of high speed computers and computer workstations. A
number of factors which influence the blowdown of pressure vessels / pipelines were discussed
in the literature review. These factors have led to the modeling of blowdown of pressure vessel /
pipeline and associated vent piping system. Development of such models has progressed in the
last few decades which use the same pertinent equations and differ from each other in the
method of solution approach. Despite availability of blowdown models, very few models are
available for determining the compressible fluid flow conditions, specifically, in vent piping
associated with pressure vessels and pipelines. Robustness and efficiency of these available vent
models have been proved to be a problem. Keeping this in mind, we develop a vent pipe model
for predicting the pressure and temperature of flowing compressible fluid (gas), surface
temperature of the vent pipe wall, and the mass flow rate which can be passed through the vent
pipe during blowdown.
Since a simple model for predicting the compressible fluid conditions in a vent pipe is desired,
every approach has been made to characterize the model as mechanistic as possible. The user
must understand that the developed model will provide a very close estimate of the compressible
fluid flow properties which bring about the changes in the vent pipe flow conditions.
Assumptions are clearly stated when developing the pertinent equations in order to ensure a
better understanding prevails. A well-defined strategy was adopted in developing our vent pipe
26
model consisting of a series of logical steps. These steps involved problem definition,
development of mathematical models for the process, method of solution, computation and
interpretation of the results. Problem definition was very precisely stated in chapter 1. The need
for a vent pipe model for predicting the compressible fluid flow conditions in vent pipe
associated with pressure vessels / pipelines was highlighted.
3.1 Development of Mathematical Models

Compressible flows are limited to low viscosity fluids such as single phase gases and multiphase
fluids containing mostly gases. In current work, our investigations are related to single phase
gases only. The model will not encounter any phase change. Compressible fluid flow is a
complex process, the interpretation of which can be analyzed by a combination of several other
physical factors. These factors which impact the compressible fluid behavior will be examined
in order to provide a better insight into compressible flow and are discussed in the concept of
compressible fluid behavior. Before proceeding with the modeling theory for the vent pipe
model, basic concept of fluid dynamics involved with the compressible fluid flow is described.
This will help us in better understanding of our vent pipe model theory.
3.1.1
Basic Conservation Equations
All analyses concerning the motion of compressible fluids must necessarily begin, either directly
or indirectly, with the statements of the four basic physical laws governing such motions. These
physical laws are independent of the nature of the particular fluid and are as follows:
x
Law of Conservation of Mass The Continuity Principle
Momentum Principle
The First Law of Thermodynamics
The Second Law of Thermodynamics
3.1.1.1
Law of Conservation of Mass The Continuity Equation
The Principle of Conservation of Mass, when referred to a system of fixed identity, simply states
that the mass of the system under consideration is constant. This statement is a concise summary
of experimental observation, relativity and nuclear effects being absent (Shapiro 1954). Under
unsteady state conditions, both density and velocity are functions of space and time. Thus,
applying the continuity equation for a fixed identity occupying the control volume is
27
3-1
Where mc.v - Instantaneous mass within the control volume; dw - Mass rate of flow entering and
leaving the control volume
Thus, it can be stated that the rate of accumulation of mass within the control volume is equal to
the excess of the incoming rate of flow over the outgoing rate of flow. Under steady state
conditions, the total mass remains constant, thus, there will be no mass accumulation. For a
control volume at any instant, the mass rate of flow is a function of element of control volume
and the local mass density. Thus for a steady state, the continuity equation can be expressed as

3-2
In general form,

3-3
Where m Mass flow rate of the compressible fluid; Instantaneous mass density of the fluid
corresponding to the inlet and outlet area; Vn - Instantaneous velocity of the fluid corresponding
to the inlet and outlet area
3.1.1.2
Momentum Principle
When the net external force acting on a system is zero, the linear momentum of the system in
the direction of the force is conserved in both magnitude and direction. This is the principle of
conservation of linear momentum. When there is a net external force, however, the linear
momentum is no longer conserved. The resultant behavior is described by Newtons second law
of motion, which is more general than the momentum principle.
According to Newtons second law of motion, the resultant of forces applied to a particle, which
may be at rest or in motion, is equal to the rate of change of momentum of the particle in the
direction of the resultant force. Newtons second law of motion yields:

3-4
Where F Sum of the forces acting on the particle in any one direction; (mV) Kinetic
momentum in the same direction
The rate of change of momentum of a fixed-mass system can be related to the rate of change of
momentum of a control volume in accordance to the following equation
28
3-5
Under steady state conditions, the rate of change of momentum within the control surface is
zero, thus the above momentum equation reduces to

3-6
It should be noted that even if frictional forces or non-equilibrium regions exists within the
control volume, the momentum equation is still valid. This allows the momentum principle to be
used in evaluating the forces generated by the flow of fluid.
3.1.1.3
The First Law of Thermodynamics
The First Law of Thermodynamics or Law of Conservation of Energy states that energy can
neither be created nor destroyed but can be converted from one form to another. The total
energy is always conserved. From the first law of thermodynamics or law of conservation of
energy we can conclude that for any system, open or closed, there is an energy balance as

Mathematically the first law can be represented as

3-7
Where Qsys Net amount of heat associated with the system; Wsys Net amount of work
associated with the system; Usys Net amount of energy stored inside the system
Thus for a steady-state steady flow system we have,

3-8
Where Q Net amount of heat associated with the system or control volume; W Net amount
of work associated with the control volume and is different from system work; The integral term
represents the shaft or expansion work, or flow work; h Enthalpy of the system
3.1.1.4
The Second Law of Thermodynamics:
The Second Law of Thermodynamics is far-reaching principle of nature that has been stated in
many forms. One of the following two forms mentioned in (Jones and Hawkins 1986;
Nageshwar 2003) are usually the most valuable:
29
The Clausius Statement: It is impossible for any device to operate in such a manner that it
produces no effect other than the transfer of heat from one body to another body at a higher
temperature
The Kelvin-Planck Statement: It is impossible for any device to operate in a cycle and produce
work while exchanging heat only with the bodies at a single fixed temperature
These two statements of the second law and many other statements are entirely equivalent in
their consequences. The first law of thermodynamics introduces the internal energy property and
the second law of thermodynamics introduces the entropy property. The property entropy often
provides a means of determining if a process is reversible, irreversible, or even possible. This
application of entropy is based on the principle of the increase of entropy, which states that the
entropy of an isolated system always increases or, in the limiting case of a reversible process,
remains constant with respect to time.
Thus in mathematical form we have,
3-9
With the understanding that time is the independent variable, this statement is usually written

3-10
Thus, based upon the above basic physical laws, the following conditions should exist under
steady state conditions
x
The mass flow rate is constant. This means that the mass flow rate at the entrance is the
same as at the exit and that the mass contained within the volume neither increases nor
diminishes at any time.
The rate of change of momentum within the control volume is zero.
No change in properties or in energy level of fluid occurs at the entrance, at the exit, or
at any point within a control volume
The rate at which energy, in the form of heat or work, crosses the boundaries of the
control volume is constant.
The entropy of an adiabatic closed system always increases
30
3.1.2
Theoretical Aspects Related to Compressible Fluid Flow Behavior
The flow of compressible fluids during blowdown from large pressure vessels or pipelines into
vent systems is influenced by a number of factors (Skouloudis 1992). These factors could be
classified according to their significance as geometrical, operational and thermophysical
parameters. The geometrical parameters which influence the venting process rely to a certain
extent on the size, type of material of construction and orientation of the vent piping associated
with the pressure vessels / pipelines. The operational factors which influence the flow of fluid
into the vent pipe system are the vessel / pipeline conditions present prior to blowdown and the
changes taking place in the gas behavior (heat transfer) inside the pressure vessel or pipeline
during blowdown. The thermophysical factors include the physical and transport properties of
the fluids contained in the vessels / pipeline. These thermophysical factors affect the flow
regimes of compressible fluid in vent systems, thus determination of these properties along the
vent pipe is central to this investigation.
The changes taking place in the properties of the compressible fluid enforces the thermodynamic
behavior of the fluids to be taken into account. These changes taking place during expansion or
compression in the vent pipe are brought about by two processes: isothermal process and
adiabatic process (Bansal 2005). When compression or expansion of gas takes place under
constant temperature conditions, the resulting process is an isothermal process. In such a
process, heat transfer takes between the system carrying the compressible gas and the
surrounding. On the other hand, in an adiabatic process expansion or compression takes place
with no heat transfer between the system and the surrounding. Such a process occurs if the
system is well insulated. The use of these two models depends on the situation encountered. It
has been cited in literature (Cochran 1996; Shapiro 1954; Saad 1993; Yuhu et al. 2002) that
isothermal models best describe the flow of compressible gases taking place through long
uninsulated pipelines while the adiabatic model is more appropriate for shorter and insulated
pipings such as the vent systems. The solution obtained by incorporating the isothermal model
yields higher pressure drop at the same mass flow rate and provides a more conservative
estimate for the pipe diameter sizing. On the other hand, an adiabatic model at constant pressure
drop predicts higher efflux rates and so is frequently the choice for conservative design of
emergency depressurization system. Moreover, the velocity of flowing gas in a short pipe is fast
enough so that no time is provided for heat transfer to take place and hence the flow can be
modeled as adiabatic.
31
Frictional effects, heat transfer effects and changes in cross sectional area contribute to the
changes of compressible fluid behavior taking place in the vent pipe. As we adopt an adiabatic
approach to develop our model, the heat transfer effects can be neglected. The vent pipe is a
constant cross-sectional area pipe; hence area changes are not relevant to our model. Thus we
consider pipe wall friction to be the chief factor bringing about the changes in compressible
fluid properties. In vent pipe subjected to compressible flow, the losses encountered due to
friction are of two types: skin friction and form friction. The skin frictional losses are
encountered due to internal surface roughness of the pipe present between the flowing fluid and
the pipe material. Form frictional losses are due to obstructions present in the piping system
such as bend pipe fittings, control valve or anything that changes the course of motion of the
flowing fluid. Thus, change in properties of fluid taking place inside the vent pipe is due to
frictional effects generated at the wall surface. This is because the behavior of flowing fluid
depends strongly on whether the fluid is under the influence of solid boundaries. The effect of
solid boundary on the flow is confined to a layer of the fluid immediately adjacent to the solid
wall where shear stress is confined (McCabe, Smith, and Harriott 2001). The effects of friction
on compressible fluid flow parameters are explained in detail using the Fanno curves in the later
part of this chapter.
Based on the above theoretical aspects related to compressible fluid flow behavior, we
understand that the behavior of compressible fluid in the vent pipe associated with emergency
blowdown facilities should follow an adiabatic path in which the changes in fluid flow
properties are brought about due to frictional effects. Thus, we progress with the development of
a vent pipe model based on adiabatic and frictional approach.
3.1.3
3.1.3.1
Model Assumptions
Steady State Analysis
The geometry visualized in the development of model comprises of a source and vent pipe
arrangement. The source can be visualized to be a pressure vessel / pipeline which has all the
mass storage of the system at isobaric and isothermal conditions throughout its volume. The
source delivers the supply of compressible gas to the vent pipe arrangement through a nozzle at
subsonic conditions. Norris et al. have developed their pipeline model based on this approach
and have provided validated results (Norris III, Exxon Production Research Co, and R.C. Puls
1993; Norris III and Exxon Production Research Co 1994). The vent pipe arrangement contains
no mass or momentum storage. As a result, steady-state hydrodynamics are used in the vent pipe
32
analysis. These steady-state hydrodynamics do, however, contain all frictional pressures drops in
the system. The pressure, temperature, and fluid properties are considered continuous across
both the source-vent pipe boundaries.
3.1.3.2
One Dimensional Approach
As discussed earlier a number of factors influence the behavior of compressible fluid in a vent
pipe which results in complexity of the process. Because of the complicated nature of the
problem, it will be assumed that the flow is one-dimensional, i.e. that all properties are uniform
over each cross section or a flow in which the rate of change of fluid properties normal to the
streamline direction is negligibly small compared with the rate of change along the streamline.
The assumption of one dimensional flow is justified largely by the great simplifications it makes
possible (Shapiro and Hawthorne 1947). According to (Shapiro and Hawthorne 1947; Shapiro
1954; Parker 1989) one-dimensional treatment introduces no significant errors especially when
changes in stream properties in the direction of flow are much larger than in the direction normal
to flow and when changes in properties in the direction normal to flow are the same in all
planes, that is, the velocity, temperature, and density profiles are unchanged. An additional
assumption is inherent in the one-dimensional analysis, namely, that the effect of turbulence on
the computation of the mean properties is small.
3.1.3.3
x
Clearly Stated Assumptions

The flow is considered to be steady and one dimensional for single-phase gas through a
constant cross sectional area vent pipe
No mechanical work done or heat exchange on or by the fluid during the flow
Differences in elevation produce negligible changes compared with the frictional effects
and hence neglected
Specific heats are constant across a particular cross sectional area for a given segment or
vent pipe length
Friction is restricted to wall shear
Velocity gradients within a cross section are neglected
33
3.1.4
Development of Adiabatic Frictional Model
The flowing compressible fluid at a short distance above the vent pipe wall possesses some
momentum, whereas the fluid immediately adjacent to the pipe wall, where the fluid velocity is
zero, has no momentum. The flowing compressible fluid must therefore acquire momentum
from faster flowing layer above it, which in turn receives momentum from the next layer up and
so on (McCabe, Smith, and Harriott 2001). Each layer is dragged along by the layer above it
except the wall where all the momentum is delivered as shear force. Momentum is thus
transferred from a region of high fluid velocity to low fluid velocity. The rate of momentum
transfer is governed by velocity gradient which acts as the driving force. Our purpose is to find
in analytical form the variations in all stream properties along the vent pipe profile of constant
area. As discussed earlier, the change of fluid properties is brought about by frictional force and
will depend upon the amount of frictional force. In order to evaluate this frictional force
generated by the flow of compressible fluid, we apply the momentum principle and obtain a
differential form of relation between the fluid properties and friction (Saad 1993; Shapiro 1954).

3-11
Where f - Fanning friction factor; DH - Hydraulic diameter or the diameter of the vent pipe;
- Density of the compressible fluid; V-Velocity of the flowing stream
3.1.4.1
Static Property Relations
The physical phenomenon that causes changes in fluid is viscous friction and is measured by the
term 4f/DH in equation 3-11. Relevant equations discussed earlier necessary to the solution of
the problems pertaining to frictional flow in constant area vent pipe are the continuity equation,
energy equation and the increase in entropy principle by second law of thermodynamics.
Additional equations include the real gas equation and the equation for Mach number. All
equations are summarized in the table 3-1 below:
Table 3-1: Pertinent equations related to frictional flow in constant area vent pipe
Real Gas Law
(a)
Continuity Equation
(b)
Energy Equation
(c)
Definition of Mach number
Increase in Entropy principle by Second Law of Thermodynamics
(d)
(e)
34
Equation 3-11 and above five equations incorporate seven different fluid parameters which can
be or are inter-related to each other. These property equations can be related to each other by
defining a single independent variable, the value of which can be changed following which the
other dependent variables can be calculated. By defining a single parameter we easily determine
the corresponding values of these compressible fluid properties. Since the effect of friction on
the changes encountered in compressible fluid parameters is desired we define the independent
variable as 4f/DH. The entire derivation for relating the compressible fluid parameters to the
independent variable is given in (Saad 1993; Shapiro 1954). It should be noted that the
derivation given in (Saad 1993; Shapiro 1954) incorporates the perfect or ideal gas law. We
incorporate a compressibility factor, Z, to deviate the behavior to real gas. However, when
deriving the real gas relation, the compressibility factor cancels off and results in same equations
as of for ideal gas behavior (refer Appendix E for derivation). The table 3-2 below summarizes
the various static property relations for the compressible fluid.
Table 3-2: Static property relations for adiabatic flow in constant area vent pipe
Friction and Mach number relation
(a)
Frictional effects on velocity
(b)
Frictional effects on density
(c)
Frictional effects on pressure
(d)
Frictional effects on temperature
(e)
Where f - Fanning friction factor; M - Mach number; - Density of compressible fluid; P Static pressure of flowing fluid; T - Static temperature of flowing fluid; DH - Hydraulic diameter
of vent pipe; dx - Differential vent pipe length
35
3.1.4.1.1
Static Property Deviations with Friction
5
Effect of friction on Mach number
Effect of friction on Velocity
Effect of Friction on Density
Effect of Friction on Pressure
Effect of Friction on Temperature
Fluid Property
4
3
Subsonic Supersonic
2
1
0
0.0
0.5
1.0
1.5
2.0
Mach Number
Figure 3-1: Variation of fluid properties with friction

Figure 3-1 gives a better understanding of the variation of compressible fluid properties due to
friction. Equations involved in plotting the property relations are tabulated in table 3-2. It is seen
that frictional flow in a vent pipe causing changes in compressible fluid properties is always
decreasing in a subsonic or supersonic flow and becomes negligible at Mach unity. Figure 3-1
can be well explained by taking into consideration the inlet flow conditions in the vent pipe. One
should understand that continuous transitions from subsonic to supersonic flow or from
supersonic to subsonic flow, are impossible (McCabe, Smith, and Harriott 2001) until and unless
the flow is mechanically altered. We restrict our vent pipe model to subsonic region. By
referring to figure 3-1 we can say that with decreasing frictional effects the velocity of the fluid
is increasing along with increasing Mach number. Pressure, temperature and density are found
to be of decreasing order in subsonic region. The compressible fluid entering the vent pipe at
subsonic condition (M < 1) will attain Mach number less than 1 or approach unity at the exit of
the vent pipe. At Mach unity, choked flow results at the exit of the vent pipe. Relevant
adjustments are made to overcome the choking condition. Overall it can be said that friction has
the net effect of accelerating a subsonic stream.
Although not incorporated in our model, we give the reader an idea of what changes are caused
due to friction in supersonic flow. The compressible fluid entering the vent pipe at supersonic
condition (M > 1) will always try to approach Mach unity at the exit of the vent pipe. This is
because the frictional effects at the exit of the vent pipe are at minimum. In the supersonic
region, velocity is of decreasing order. When Mach of unity is attained for initial supersonic
condition, choking takes place which involves the appearance of shock waves. Adjustments are
made by increasing the vent pipe length to overcome choking condition. Overall, it can be said
36
that friction has the net effect of decelerating a supersonic stream. A better understanding of the
effects of friction on fluid properties is provided when discussing the Fanno curves in the latter
section.
3.1.4.2
Stagnation Property Relations
Now that the static properties for flowing compressible fluid are defined we define the
stagnation properties for these compressible fluid. In a steady state adiabatic process, when the
fluid is decelerated to zero velocity provided that no work interaction occurs the resulting
properties of the fluid are called stagnation properties. Stagnation properties are developed by
taking into account the process to be adiabatic and frictionless, that is, isentropic process. Such a
process is encountered in variable cross sectional area where the frictional effects are minimal.
Stagnation properties provide a convenient reference state in analyzing the flowing compressible
fluid properties, that is, static properties. Stagnation properties are more related to the source
conditions. Although valid for variable cross sectional area, (Shapiro 1954; Shapiro and
Hawthorne 1947) have suggested that these isentropic stagnation properties are valid for
adiabatic frictional constant area vent pipe. These properties are defined by (Saad 1993; Shapiro
1954; Bansal 2005) are represented in table 3-3:
Table 3-3: Stagnation property relation
Stagnation and Static Pressure Relation
(a)
Stagnation and Static Temperature Relation
(b)
Stagnation and Static Density Relation
(c)
Where Po Stagnation Pressure; To Stagnation Temperature; M: Mach number; - Specific

heat ratio. Stagnation enthalpy and stagnation temperature are considered to be a constant
throughout the process whereas stagnation pressure is not. Compressible fluid when brought to
rest adiabatically, the static enthalpy of the fluid is equal to the stagnation enthalpy, and the
static temperature is equal to the stagnation temperature. However, the pressure is equal to the
initial stagnation pressure only if the fluid is brought to rest both adiabatically and reversibly,
that is, isentropically. Adiabatic frictional process is considered to be an irreversible process.
According to (Saad 1993; Shapiro 1954)
3-12
37
Where s - Entropy; Po - Stagnation pressure. The above equation provides a better understanding
of relationship between entropy and stagnation pressure. For an increase in entropy, there will
always be a decrease in the stagnation pressure. The relative change in stagnation pressure
therefore provides an indication of degree of irreversibility of the process. Friction present in the
vent pipe causes an increase in the entropy and therefore stagnation pressure decreases. The
property relations in table 3 have been derived by (Saad 1993; Shapiro 1954; Shapiro and
Hawthorne 1947; Bansal 2005).
3.1.4.3
Estimation of Mass Flow
The function of blowdown facilities on pressure vessels / pipelines is to provide a means of

venting the high pressure inventory to atmosphere in a very short period of time (Gradle 1984).
The short blowdown time is always associated with high velocities and high mass flow-rates.
Flow of compressible fluid such as natural gas and other gas mixtures is dependent upon
Reynolds number, friction factor, pipe roughness, pipe diameter, pipe length, temperature,
pressure, pressure drop and gas properties (Ouyang and Aziz 1995). The prediction of mass
efflux from pressure vessels / pipelines through vent system is a central step in the design of
emergency depressurization system. Accurate predictions are required for optimum design. This
is analyzed in our vent pipe model. The relevant equations adopted for analyzing the flow in
vent pipes depend on the basic physical law of fluid mechanics, that is, the Continuity Equation.
For a constant area flow, mass flux is independent of length. The mass velocity can be evaluated
at any point inside the entrance of the vent pipe. The process of blowdown of pressure vessels /
pipelines is characterized to be an unsteady process where the properties of compressible fluid
are functions of space and time causing the flow to change throughout the flow path. However,
we discussed earlier, as per the geometry visualized in the model analysis steady state
hydrodynamics prevail in vent pipe. Thus, for steady state conditions, the mass rate of flow
across two different sections of the vent pipe can be expressed by continuity equation as

Where - Mass flow rate; - Density, A - Cross sectional area; V - Velocity of flowing fluid
The mass flow per unit area or the mass flux, G, can then be written as
3-13
For vessels or pipelines of commercial interest, the pressure to be released almost always results
in sonic velocity at some restriction, and choked flow results (Norris III, Exxon Production
Research Co, and R.C. Puls 1993). Choked flow is the condition wherein the mass flow rate
becomes independent of the downstream conditions i.e. that point at which further reduction in
38
downstream pressure does not result in change of the mass flow rate (Haque, Richardson, and
Saville 1992). Basically, a limit occurs because acoustic signals can no longer propagate
upstream. This limit occurs when the fluid velocity just equals the propagation velocity. Such a
condition is seen at Mach unity. Thus it is advisable to relate the gas flow relation in form of
dimensionless Mach number. The mass flux in terms of static pressure and static temperature
can be expressed as

3-14
The above equation for mass flux in terms of stagnation properties can be expressed as
3-15
Where Po Stagnation Pressure; To Stagnation Temperature; M: Mach number; - Specific

heat ratio; Z Compressibility factor.
According to equation 3-15, for a given Mach number, the flow is proportional to the stagnation
pressure and inversely proportional to the square root of stagnation temperature. For a given
geometry,
stagnation
and
downstream
pressures, and assumed friction factor, these

equations 3-14 & 3-15 define the flow
(Parker 1985). If choking condition is
attained at the exit of the vent pipe, the rate
of flow through the system increases and the
flow is choked by the vent pipe. The mass
rate of flow can be increased only by
decreasing the stagnation temperature and /or
increasing the stagnation pressure. For this
reason, flow test data for many applications
over wide range of pressure and temperature
Figure 3-2: Condition for maximum mass flux
levels, are plotted with as the flow
variable (Shapiro 1954). The condition at which maximum flow can be achieved occurs at Mach
unity. This condition is plotted in figure 3-2.
3.1.4.4
Estimation of Adiabatic Wall Temperature
During blowdown of pressure vessels / pipelines, the time required to reduce the overpressure
build-up and inventory is influenced by high efflux rates. This inevitably leads to a reduction in
39
the temperature of the vessel / pipeline and associated vent pipe system, possibly to a
temperature below the ductile-brittle transition temperature of the material from which the
vessel / pipeline and associated vent piping is fabricated (Haque, Richardson, and Saville 1992;
Haque et al. 1989; Marian, Vuthaluru, and Ghantala). At this temperature, the probability of
failure of equipment material is high. The temperature of flowing gas in the vent pipe along with
high speed velocities will influence the temperature of the vent pipe wall. Due to high velocities
encountered viscous stresses set-up which do shearing work on the fluid particles which results
in an increase in internal energy as well as the temperature of fluid very close to the wall (Saad
1993). This work is dissipated in form of viscous heating. At high velocities, dissipation is
largest close to the wall. The flow is not locally adiabatic and a difference will exist between the
wall temperature and the stagnation gas temperature (Prandtl 2004). Also, the adiabatic wall
temperature will be realistically higher than the flowing gas temperature.
The adiabatic wall temperature has been well studied in the boundary layer flow on a flat plate
and is usually correlated with the recovery factor (Shi et al. 2001). It has become common
knowledge that for laminar flow recovery factor, r, is Pr1/2 while for turbulent flow recovery
factor is Pr1/3 These equations neglect the fact that the recovery factors are also influenced by
Mach number (Kaye 1953) given by the expression:

3-16
Where r - Recovery factor; Pr - Prandtl number; N - Reciprocal of the exponent of the turbulent
boundary-layer velocity profile approximated by power law. This relation holds for Prandtl
numbers greater than 0.65 and less than 0.75. Equation 3-16 is not validated.
The investigations related to adiabatic wall temperature are very few (Shi et al. 2001). Although
many of these approximations are valid for flat plates, these can be applied to circular pipes.
(McAdams, Nicolai, and Keenan 1946) have performed investigation related to adiabatic wall
temperature for the subsonic turbulent flow in a pipe and have defined the recovery factor as:
3-17
Where T : Bulk mean gas temperature; TO : Stagnation gas temperature; Taw : Adiabatic wall
temperature; r: Recovery factor. A number of approximation and typical ranges for recovery
factor are provided with no proper validations (Kaye 1953). (Shi et al. 2001)Shi et al. have
defined the recovery factor as a function of Prandtl number and Knudsen number. The recovery
factor for continuous flow is always equal to Prandtl number and will increase above Prandtl
40
number as Knudsen number increases (Shi et al. 2001). Validations are been provided by Shi et
al. for the proposed method of determining the recovery factor. Hence, we equate the recovery
factor to Prandtl number and calculate the adiabatic wall temperature using the relation by
McAdams et al. into our model. The use of recovery factor relation for predicting the outlet pipe
wall temperature will be confirmed with validation of the model.
3.1.4.5
Effects of Friction - Fanno Process

The effects of friction on the flow parameters in a vent
pipe during blowdown can be well explained by
means of Enthalpy-Entropy diagram. The curve
formed on such a plane is defined by Continuity
Equation and Energy Equation is known as the Fanno
curve. The Fanno process is one steady, adiabatic flow
with friction in a duct in which the cross sectional area
does not change along its length (Chan and Woods
1992). The friction leads to a force on the fluid in the
Figure 3-3: Fanno curve (Enthalpy - opposite direction to the flow. In Fanno flow, the
Entropy diagram) Adapted from (Saad stagnation enthalpy and mass flux are constant in all
sections of the vent pipe. The continuity equation and
1993)
energy equation, describes the Fanno process in the

plane of thermodynamic properties, enthalpy and density as (nomenclature remains the same as
defined earlier)

3-18
The above equation indicates that when the flow of gas is accelerating in velocity, the enthalpy
is decreasing by a corresponding amount, and when the gas is decelerating the enthalpy
increases. As enthalpy is a function of temperature, it is valid that similar results will be seen in
the temperature profiles.
The gradient of the Fanno curve is given by (Chan and Woods 1992) expressed as
3-19
Where the subscript FANNO indicates that the differentiation is taken while keeping stagnation
enthalpy and mass flux unchanged. The slope of the Fanno curve in the enthalpy-entropy plane
is given by (Saad 1993; Shapiro 1954; Chan and Woods 1992)
41
3-20
Equation 3-20 expresses enthalpy as a function of temperature, Mach number and entropy and
implies that the effect of friction in a Fanno flow is to drive the flow towards Mach unity, with
enthalpy and pressure decreasing in the subsonic branch and increasing in the supersonic branch.
This is represented in figure 3-3. The upper part of the curve represents the subsonic condition
whereas the lower portion represents the supersonic condition. Since the flow is adiabatic with
friction, the second law of thermodynamic tells us that entropy may increase but may not
decrease. Thus the path of states along any one of the Fanno curves must be towards the right.
Thus a subsonic flow may therefore never become supersonic and a supersonic flow may never
become subsonic, unless a discontinuity is present. Frictional effects present in the vent pipe
alone cannot change subsonic flow into supersonic flow or vice versa because part of such
processes will involve decrease in entropy, thus, violating the increasing entropy principle by
Second Law of Thermodynamics. Emphasis is on frictional effects taking place in the subsonic
region. In subsonic flow, frictional effects increase the internal energy with a corresponding
reduction in the density of the fluid. The mass flow rate per unit area or mass flux must remain
constant in the vent pipe during subsonic flow condition. In order to achieve this, constant mass
flow rate condition, there must be an increase in velocity leading to expansion of compressible
fluid. Friction has no effects on stagnation temperature or on stagnation enthalpy; however,
friction reduces stagnation pressure in both subsonic and supersonic flow.
3.1.4.6
Estimation of Friction factor
Friction is the chief factor bringing about changes in fluid properties. The drag of a fluid at the
contact between the fluid and the pipe is caused by friction factor (Ellenberger 2010). As cited
in (Bansal 2005; Ellenberger 2010; Ouyang and Aziz 1995), there are two major friction factors
available in fluid mechanics which are used to determine the pressure loss due to friction in
pipes: the Fanning friction factor and Darcy-Weisbach or Moody friction factor. The Darcy
friction factor is four times larger than the Fanning friction factor. The variation of the friction
factor with Reynolds number and pipe roughness for circular pipes can be divided into different
regimes (Govier and Aziz 1972): laminar flow, smooth wall turbulent flow, partially rough wall
turbulent flow and fully rough wall turbulent flow. Partially rough wall turbulent flow and fully
rough wall turbulent flow are also named as partially developed turbulent flow and fully
developed turbulent flow (Ouyang and Aziz 1995). For Laminar flow, the friction factor can be
shown to be a simple function of Reynolds number (Bansal 2005):
42
3-21
Where f - Fanning friction factor and Re - Reynolds number. The friction factor is only a
function of Reynolds number for smooth wall turbulent flow, and a function of relative pipe
roughness for fully rough wall turbulent flow, whereas it depends upon both the Reynolds
number and relative pipe roughness in partially rough wall turbulent flow.
Table 3-4: Explicit approximation for Colebrook-White friction factor equation
(a)
(Moody 1947)
(Wood 1966)
(b)
(Jain 1976)
(c)
(Churchill 1977)
(d)
(Chen 1979)
(e)
(Zigrang and Slyvester 1982)
(f)
; ;
(Serghides 1984)
(Swamee and Jain 1976)
(Romeo, Royo, and Monzon 2002)
(Sonnad and Goudar 2006)
(g)
(h)
(i)
(j)
In practical situations, the flow of compressible fluid (gas) is turbulent. A number of different
approximations are been reported to analyze the friction on turbulent flow regime. These
methods can be classified as smooth pipe correlations and rough pipe correlations. Our
investigations are only related to rough pipes hence we do not consider smooth pipe correlations
into our vent pipe model. Difficulty of solving turbulent flow problems in rough pipes lies in the
fact that hydraulic friction factor is a complex function of relative surface roughness and
Reynolds number (Brkic` 2011). The equation for computing the friction factor in the DarcyWeisbach pipe friction loss equation, as presented by Colebrook and White (Colebrook 1939),
has been preferred because of its presumed superior accuracy and sound theoretical basis
(Bernuth 1990).
The Colebrook and White (CW) equation which related to pipe roughness and Reynolds
number, is customarily given by (Franzini, Finnemore, and Daugherty 1997)
3-22
Where f - fanning friction factor; (/D) - Relative pipe roughness; Re - Reynolds number.
Colebrook equation is transcendental which means that it cannot be solved by using only
43
elementary functions and basic arithmetic operation in definitive form (Brkic` 2011). Clearly,
the above Colebrook and White equations are implicit in the friction factor estimation, and
requires either an iterative numerical scheme or by graphical representation for solution. An
alternative solution to iterative methods is the direct use of an explicit equation which is precise
enough to calculate the value of friction factor.
Numerous researches (Moody 1947; Wood 1966; Jain 1976; Churchill 1977; Chen 1979;
Zigrang and Slyvester 1982; Serghides 1984; Swamee and Jain 1976; Romeo, Royo, and
Monzon 2002; Sonnad and Goudar 2006) have been conducted in this area. The most widely
used explicit approximations for Colebrook-White equation postulated since the end of 1940s
are synthesized in table 3-4. These approximations differ from each other in degree of accuracy.
Average percentage errors generated by these approximations when compared to ColebrookWhite equation have been indicated in table 3-5. Referring to the accuracy table 3-5, we can say
that the deviation of Serghides approximation (Serghides 1984) table 3-4 equation (g) from the
Colebrook-White equation for rough pipe results in a very low average error compared to any
other approximation listed in table 3-5. Hence we apply Serghides approximation for
Colebrook-White equation into our model for determining the friction factor in the transitional
and turbulent flow (Re > 2100) at any relative roughness (/D). The Serghides approximation for
Colebrook-White equation is derived by applying Steffensons accelerated convergence
technique to an iterative, numerical solution of Colebrook-White equation. The constants A, B
and C are approximations of Colebrook-White equation obtained by three iterations of direct
substitution method (Serghides 1984).
Table 3-5: Overall average relative errors of fanning friction factor values obtained from
different explicit equations compared with those from the CW equation (Ouyang and Aziz 1995;
Swamee and Jain 1976; Romeo, Royo, and Monzon 2002; Sonnad and Goudar 2006)
Average
Error
3.1.4.7
Serghides
Chen
Z-S
Jain
Romeo
Sonnad
Swamee
Churchill
Wood
Moody
0.00037
0.137
0.234
0.929
1.04
1.09
1.34
4.092
5.107
6.276
Estimation of Thermophysical Properties
The accurate knowledge of thermodynamic properties of gases such as natural gases and other
gas mixtures is of indispensable importance for the basic engineering and performance of
technical processes (Kunz et al. 2007). These properties can significantly affect the flow regimes
occurring during the venting process, thus introducing unexpected variations in the
depressurization mechanism (Skouloudis 1992).
44
The thermodynamic properties of mixtures can be calculated in a very convenient way from the
equations of state. The advantage of employing equation of state in determining these properties
is because it does not rely on activity coefficient concepts. A number of equations of state are
available which serve his purpose. AGA8-DC92 equation of state is currently an internationally
accepted standard only for P--T relation in homogenous gas region of natural gases. Aside from
the restriction to the homogenous gas phase, the AGA8-DC92 equation of state shows
significant weaknesses in the description of natural gas properties and covers only a limited
temperature, pressure and composition range (Kunz et al. 2007). Cubic equation of states such
as Soave-Redlich Kwong (Soave 1972) and Peng Robinson (Peng and Robinson 1976) are
widely used in many technical applications due to their simple mathematical structure.
Technical applications which demand high accuracy of the calculated mixture properties, the
cubic equation of state show major weaknesses with respect to representation of thermal
properties in the liquid phase, speed of sound (thus impacting density, velocity profiles) and the
description of caloric properties (Soave 1995; Kunz et al. 2007; Won, Smith, and Zeininger
2005). As a result there are inconsistencies in calculations when moving from one region to
another. Experimental evidence has also shown that it is most important to model the
thermodynamics of depressurization accurately since failure to do so can lead to trajectories
through phase (pressure-temperature-composition) space which are grossly in error (Richardson
and Saville 1996). For this reason, thermodynamic, phase and transport properties of single
phase multi-component fluids involved in the vent pipe model are calculated using a
thermophysical computer package called REFPROP (Lemmon, Huber, and McLinden 2009).
This program has been developed by National Institute of Standards and Technology (NIST)
and provides the thermodynamic and transport properties of industrially important fluids and
their mixtures
REFPROP is based on the most accurate pure fluid and mixture models. The program
implements three models for the thermodynamic properties of pure fluids: the GERG-2004
equation of state explicit in Helmholtz energy (Kunz et al. 2007). Mixture calculations employ a
model that applies mixing rules to the Helmholtz energy of the mixture components; it uses a
departure function to account for the departure from ideal mixing (Lemmon, Huber, and
McLinden 2009). The GERG-2004 (Kunz et al. 2007) equation of state is a fundamental
equation explicit in the Helmholtz free energy as a function of density, temperature, and
composition. The GERG-2004 equation of state is developed with a view to overcome the
weaknesses and limitations of the previous equations of state. The development and evaluation
of GERG-2004 mixture model is based on more than 100,000 experimental data for multiple
45
thermodynamic properties in different fluid regions (Kunz et al. 2007). The GERG-2004
formulation is able to represent the most accurate experimental binary and multi-component data
for gas phase and gas-like supercritical densities, speed of sound, and enthalpy differences
mostly to within their low experimental uncertainties. The normal range of validity covers
temperatures from 90 K to 450 K and pressure up to 35 MPa for natural gases and other single
or gaseous mixture consisting of the 18 components methane, nitrogen, carbon-dioxide, ethane,
propane, n-butane, isobutene, n-pentane, iso-pentane, n-hexane, n-heptane, n-octane, hydrogen,
oxygen, carbon monoxide, water, helium, and argon (Lemmon, Huber, and McLinden 2009).
The uncertainties in gas phase density and speed of sound for a broad variety of natural gases
and related mixtures are less than 0.1% over the temperature range 250 K to 450 K at pressures
up to 35 MPa (Kunz et al. 2007). Thus, the utilization of REFPROP into the vent pipe model in
determining the thermophysical properties will improve the accuracy of predictions of
compressible fluid flow properties and make the simulation in the vent pipe model competent.
3.2 Modeling Approach

3.2.1
Simulation Object
As discussed earlier the geometry visualized in model development consists of a source and vent
pipe arrangement. The conditions in vent pipe have been proved to be at steady state. In order to
perform simulation using vent model, we chose 8 NB schedule 80 stainless steel straight pipe of
length 12 m. The roughness of the pipe is assumed to be as 0.00015 m. No fittings are involved
hence we neglect the form friction. The vent pipe predictions which we need to calculate are
pressure and temperature of the flowing gas, adiabatic wall temperature, Mach number, density,
velocity, enthalpy, entropy, friction factor, mass flow, standard volumetric flow, stagnation
properties and critical properties. These properties are calculated for each and every segment
along the vent profile for specified inlet static pressure and temperature and gas composition.
Venting is to atmosphere hence the outlet static pressure is 1 bar atm.
3.2.2
Method of Solution
A number of equations are involved in determining compressible fluid flow properties in a vent
pipe during blowdown. One of the methods of applying these is Multi-Step or Segmented
Design Method (Ouyang and Aziz 1995). Multi-Step or Segmented Design Method require that
calculations be performed over very small segments of the vent pipe and that iterations be
employed to obtain the change in pressure, temperature and other thermophysical properties
over each segment. The vent pipe length can be equally divided or chosen in such a way that
46
their sum of the segment lengths is exactly equal to the total vent pipe length. The procedure can
be applied from upstream to downstream end of the vent pipe. The method, however, becomes
bi-directional for calculating properties at sonic condition.
Table 3-6: Property relations in terms of Mach number
PART A
PART B
(a)
(f)
(b)
(g)
(c)
(h)
(d)
(i)
(e)
(j)
Properties of a fluid determined in table 3-2 at any section of a vent pipe can be related to the
properties at any other section of the vent pipe. It is always advisable to relate the property
relations in table 3-2 in form of dimensionless Mach number. In order to achieve this, property
relations in table 3-2 are integrated between the inlet and exit conditions of the vent pipe. The
inlet conditions are represented by subscript 1 and exit conditions by subscript 2. The integrated
expressions are tabulated in table 3-6 Part A. A problem develops when the speed of gas is
approaching sonic velocity. Obtaining results of table 3-6 Part A will sometimes result in
solutions of subsonic to supersonic flow. This will affect the calculation procedures and will
result in an error. Hence in order to overcome these situations, we restrict the properties in table
3-6 Part A to approach those characteristic of Mach unity. Properties of a fluid when the gas is
flowing at Mach unity are called critical properties and are identified by means of an asterisk
(*). These equations are represented in table 3-6 Part B.
Mass flow rate is a requirement in predicting the properties along the vent pipe segments. Thus,
initially we consider the entire length of vent pipe and compute the mass flow rate. In order to
calculate the constant mass flow rate, inlet value of Mach number is required. This is achieved
47
by solving equations (a) & (e) of table 3-6 and equation 23 which relates the inlet and outlet
Mach numbers for a segment or for the entire length of the vent pipe.
3-23
This result in two non-linear equations 24 & 25 with three unknown variable: friction factor,
Mach number, and outlet temperature.
3-24
3-25
Where a = (-1)/2; : Specific heat ratio; M1 Inlet Mach number for the segment or vent pipe;
M2 Outlet Mach number for the segment or vent pipe; P1 Inlet static pressure for the segment
or vent pipe; P2 Outlet static pressure for the segment or vent pipe; T1 Inlet static
temperature for the segment or vent pipe; T2 Outlet static temperature for the segment or vent
pipe. We have two dependent variables and one independent variable in equations 24 & 25.
Since we interested in the effects of friction in the vent pipe, we choose friction factor, f, as
independent variable.
Frictional resistance between the moving gas and pipe wall is quantified using Darcy friction
factor, f. For fully turbulent flow, friction factor is independent of Reynolds number and is
determined using the Von Karman equation customarily given by (Cochran 1996):

3-26
For flow regimes other than fully turbulent, the friction factor is dependent on Reynolds number.
However, the above Von-Karman equation can be conveniently used as an initial estimate of
friction factor. On estimating the initial friction factor, the non-linear equation 24 & 25 can be
solved by applying Newtons Iteration method for multi-variable non-linear equations. We
incorporate the use of Jacobian matrix in calculating our dependent variables. The procedure for
Newtons Iteration method can be cited in a (Franz and Melching ND; Bellman 1970; Ortega
and Rheinboldt 1970). One of the serious difficulties associated with the use of the Newtons
technique is calculation of the Jacobian matrix and its inversion at each step which sometimes
results in errors (Bellman 1970). This difficulty is overcome by solving the matrix on excel
spreadsheet.
48
An initial estimate for Mach number is assumed to be 0.01 and for outlet temperature is assumed
to be as inlet temperature. In subsonic flow the exit temperature is always decreasing. Hence the
inlet temperature will provide a good estimate in calculating the outlet temperature. A minimum
of 400 iterations are performed to calculate the final friction factor incorporating a number of
nested iterations. After approximating the friction factor value on first iteration, Serghides
approximation (Serghides 1984) to the Colebrook-White equation 3-22 is used to estimate the
friction factor up to final iteration. Iterations are performed until the friction factor convergence
is of the order 10-16. This will improve the accuracy of vent pipe model predictions. The inlet
Mach number and outlet static temperature of the flowing gas are calculated from iterations for
the final friction factor. The inlet Mach number is used to calculate the steady state mass flowrate which remains constant for all segments of the vent pipe.
Figure 3-4: Representation of equation 3-27

The predicted compressible fluid properties for each segment of vent pipe depend on the friction
term 4fL/D. The accuracy of particular model or method of solution is greatly dependent on the
4fL/D term of the pipe section under blowdown (Botros, Jungowski, and Weiss 1989). For a
given segment length, L1-2, the term 4fL1-2/D is estimated from the following equation 3-27
(Parker 1989; Saad 1993; Shapiro 1954) and is represented in figure 3-4:
3-27
Where L* - Maximum length of vent pipe which does not cause choking; (L*)
M1:
Vent pipe
length associated with M1; (L ) M2: Vent pipe length associated with M2; L1-2 - Vent pipe length
between the section 1 corresponding to Mach M1, and section 2 corresponding to Mach M2
Upon calculating the friction term, the corresponding compressible fluid properties are
calculated using the critical property relations in table 3-6 Part B. The resulting venting
conditions are then calculated for each segment of the pipe. The exit conditions calculated for a
segment becomes the inlet condition for the next segment of the vent pipe. The balances
obtained for each segment of the vent pipe are then linked together to satisfy the boundary
conditions. The boundary conditions for the vent pipe are specified gas static pressure, gas static
49
temperature and gas composition at the inlet of the vent pipe and complete pressure drop to
atmosphere, that is, a pressure of 1 bar atm or choking pressure (at which flow becomes choked)
at the exit of the vent pipe.
If choking condition is attained, the user has following 3 optionsx
Decreasing the inlet static pressure At sonic condition by decreasing the inlet static
pressure, the mass flow rate will be reduced. Thus, the flow will enter the subsonic region
with a shift in Fanno curve
Increasing the vent pipe length - When a compressible gas flows through a constant area
vent pipe, the flow characteristics in the vent pipe are affected by the length of the vent
pipe. If the flow entering the vent pipe is at subsonic condition, the gas will accelerate in
the vent pipe owing to friction, approaching sonic conditions at the exit. At the same time,
the static pressure as well as stagnation pressure decreases in the direction of the flow.
Stagnation temperature and stagnation enthalpy will remain constant. If choking condition
is attained at the vent pipe exit, the mass flow rate through the vent pipe is at its maximum
(refer figure 2) and the flow is choked at the exit. If a further increase in mass flow rate is
desired it can be achieved by decreasing the stagnation temperature and or increasing the
stagnation pressure at the inlet of the vent pipe (as per equation 15). However, the velocity
at the exit of the vent pipe would still be sonic, but the exit pressure would be higher.
According to the Fanno process, friction present in adiabatic flow will cause changes in the
compressible fluid properties and increases the gas velocity so that the sonic velocity is
approached at the pipe exit. Apart from friction factor, the term, 4fL/D, also incorporates
the length and diameter of the vent pipe. Precisely, the mass flow rate achieved in the vent
pipe depends on friction resistance (Brkic` 2011). Hence, the length of the pipe can directly
affect the mass flow through the vent pipe. If the term 4fL/D is as large as the maximum
value appropriate for the Mach number at the entrance to the vent pipe, then the gas flow at
the pipe exit is at Mach 1 and the length of the pipe is at its maximum. Thus when choking
occurs, the Mach number at the inlet of the pipe depends on the length of the pipe and
decreases as the length is increased. When the flow is choked, an increase in pipe length
produces a reduction in the mass flow, so that the operating point is shifted to a different
Fanno line.
Increasing or decreasing the exit gas static pressure - When a compressible gas flows
through a constant area vent pipe, the flow characteristics in the vent pipe are affected by
the back pressure at the vent pipe exit. For a constant vent pipe length, an increase or
decrease in exit gas pressure will result in a sonic condition depending on the back pressure
50
applied. In subsonic flow, gas accelerates continuously such that the exit pressure is equal
to back pressure. If the back pressure is reduced, the exit pressure of the gas will reduce
such that sonic conditions are approached at the exit of the vent pipe. At this point the mass
flow will be at its maximum through the vent pipe and the exit Mach number will be at
unity. Any further reduction in back pressure will have no effect on the mass flow. In our
case venting is straight to atmosphere. On achieving choking pressure, an increase or
decrease in exit pressure from back pressure of 1 bar will cause a decrease in mass flow and
the Mach number will be less than unity.
This procedure gives the complete state of the line for specified upstream conditions at all points
along the vent pipe. Compressible gases used in performing simulations using the vent pipe
model were air, methane, carbon dioxide and DBNGP gas mixture. Simulations were performed
in the pressure range from 100 KPa gauge up to choking pressure condition. All results are
presented in Appendix F.
3.3 Computations
For obtaining solutions to process simulations, several levels of computation are available
ranging from solution by inspection to analytical and high speed computer solution (Ramirez
1998). Because of the complexity and non-linearity of process simulation problems, most
solution require high speed computer. All computations related to the vent pipe model were
carried out on Core 2 Duo 3.00 GHz computer with 2 GB RAM provided by Curtin University.
The vent pipe models programming functions were scripted in Visual Basic in conjunction with
Microsoft Excel which will act as a user interface for data input, model prediction results, and
report generations. The algorithm adopted in computing the predictions of the vent pipe model is
presented in figure 3-5. The MS Excel and Visual Basic program functions for the vent pipe
model and for obtaining the thermophysical properties are presented in Appendix H.
51
Figure 3-5: Algorithm for vent pipe model
52
4 Chapter 4
Results and Discussion
As discussed in the literature review, the ranges of transient and steady-state vent pipe flow
experiments are limited. Venting experiments have been conducted with working fluids such as
water and refrigerant R114 with an operating pressure up to 7.2 MPa (Skouloudis 1992). These
experiments, which encounter liquid only, are more related to the reactor cooling system for
nuclear industry and were conducted for validating models SAFIRE (Tilley and Shaw 1990),
RELAP (Worth, Staedtke, and Franchello 1993), RELIEF (Nijsing and Brinkhof 1996) and
DEERS (Skouloudis 1992). A wide range of experiments related to blowdown of single phase
gas/liquid or multiphase mixtures from pressure vessels and pipelines are been conducted
(Evanger et al. 1995; Gebbeken and Eggers 1995; Norris III, Exxon Production Research Co,
and R.C. Puls 1993; Haque et al. 1992). A rough idea to model the vent pipe-work associated
with vessels and pipelines is mentioned (Haque, Richardson, and Saville 1992). In order to
evaluate the performance of the developed vent pipe model and provide experimental data for
future development of models, a small facility was constructed to perform venting experiments.
4.1 Experimental Design

The experimental test rig was developed and designed based on first principles of chemical
engineering. The test rig was designed for a maximum design pressure of 1500 KPa G. Relevant
standard/codes were employed in mechanical designing of the experimental test rig. The design
was confirmed and signed for construction by Dr. Hari Vuthaluru, Associate Prof. Department
of Chemical Engineering, Curtin University, Mr. Clinton Smith, Principal Process Engineer,
Atkins Global and Dennis Kirk-Burnnand, Principal Consultant, GHD Pty Ltd. The construction
of the experimental test rig was carried out in Curtin Universitys Mechanical Workshop by Carl
Lewis, Senior Technician.
The experimental test rig consists of a vent pipe and an accumulator pipe arrangement. The vent
pipe is a 12m long 8NB schedule 80 stainless steel type 316 pipe. The entire test rig arrangement
is positioned horizontally on 90 brackets mounted in the wall. In order to achieve a steady-state
flow condition, an 11.6 m long 50 NB schedule 40 seamless carbon steel ASTM A106 GR B
3000 pipe was positioned in parallel with the vent pipe. The purpose of the carbon steel pipe was
to act as an accumulator, a steady supply source of gas through the 12 m vent pipe section in
case if the supply from the compressor and gas bottles was observed depleting.
53
Figure 4-1: Schematic representation of the vent pipe assembly
54
In order to comply with safety, the vent pipe and accumulator pipe arrangements were
hydrostatically tested with water at a pressure of 600 psi for a period of 30 min. No leaks were
found and a pressure test certificate (refer Appendix A) was provided. Pressure reliefs were also
installed in case of pressure built up in the accumulator. Due to high velocity noise produced at
the end of the test section, a noise controller was designed and attached to end flange of the vent
pipe. The noise controller is a 150NB SS pipe with 80NB schedule 10s SS pipe inside both
welded to a flange. The 80NB pipe has inch perforated holes along its length. The gas exiting
at the end of the vent pipe is absorbed by the acoustic foam packed inside the 150NB pipe. The
entire arrangement and mechanical drawings can be seen in figure 4-1 and Appendix B. In order
to have no heat transfer with the surroundings prevailing adiabatic process assumptions, the
entire vent pipe arrangement was insulated with glass wool. Extensive safety and operational
controls were instituted to prevent the ingress of unauthorized personnel into the facility during
the gas blowdown.
4.2 Instrumentation and Data Collection

The vent pipe tests were performed by measuring the stagnation temperature of the gas at the
inlet and exit, temperature of the vent pipe wall at the exit, the temperature of the vent pipe wall
at every 1m section of the pipe, pressure and the flow of the gas through the vent pipe. The
instruments are selected based on the engineering design parameters which could sustain the
maximum design pressure. Pressure transducers were used to measure the pressure of the gas at
three different positions: before ball valve (bv2), inlet and exit of vent pipe. The 130C Ceramic
Pressure Transducer made of Wheatstone bridge circuit transmitting an analog output of 420mA was used to sense the pressure. The pressure range of the transducer is 0-20 bar at an
accuracy of 0.1 bar. The pressure transducers were calibrated by the vendor. A digital pressure
gauge was also used to give direct measurements. This digital pressure gauge was provided by
BOC Gases. RTDs were used to sense the temperature of the gas at the inlet and exit ends.
Model RTD-PT100 output was used with initial calibration performed by the vendor. In order to
cross check the accuracy, RTDs were immersed in the ice / water bath and corresponding
temperatures were recorded. All RTD readings were found to be in close agreement (refer
Appendix C for Commissioning and Testing report). The temperatures of the outside of the pipe
wall at the vent pipe exit were obtained using Adjustable Ring T-Type thermocouple. These
thermocouples make direct contact with the pipe for maximum performance and have grounded
junctions. The operating range for these thermocouples is -100C to 400C. The temperature
sensed by the temperature sensors was confirmed by a Non-Contact Thermometer with Dual
55
Laser Targeting (temperature gun). The contact thermocouples temperatures were checked for
accuracy in the ice / water bath and measurements were found to be in close agreement. The
temperature range was -50C to 650C with an accuracy of 1%. The outside surface pipe wall
temperatures after every 1m section of the vent pipe were obtained using temperature gun. The
flowrate of the gas through the vent pipe was monitored and obtained using an IFM Effector 300
Flow Sensor Model SD6000. This flow sensor is been developed especially for compressed air
with integrated pipe length. The flow sensor measurement is based on the calorimetric principle
transmitting an analogue signal of 4-20mA proportional to the standard volumetric flow. The
compressed air meter detects the standard volume flow (to ISO 2533) directly, eliminating the
need to correct for temperature and pressure variation. The high measurement dynamics of the
system enables reliable detection of minute quantities. The range of the flow meter is 0-75
Nm3/hr at an accuracy of 3%. High accuracy and repeatability are ensured by the integration of
the measurement sensors key elements into a defined pipe length.
All data was telemetered to a NI CompactDAQ data acquisition system developed by National
Instruments. Model NI cDAQ-9172 is an eight-slot NI CompactDAQ chassis that can hold up to
eight I/O modules and is capable of measuring a broad range of analog and digital I/O signals
and sensors using a Hi-Speed USB 2.0 interface. The analog signals from the pressure
transducer, temperature RTD, temperature thermocouple and flow meter are transmitted to NI
input modules NI9203, NI9217 and NI9211 via 2pr screened dekron cable. The data collection
is controlled by the NI LabView Signal Express software version 3.5 from where trend data is
exported to Microsoft Excel for further analysis. The advantage of using LabView Signal
Express is that it provides instant interactive measurements that require no programming, thus,
making it easier to use. Although the accuracy of the instruments and modules is found to be
agreeable there exists a potential for signal noise caused primarily due to power supply
fluctuations, signal transmission etc. This difficulty was solved by adopting signal noise
reduction technique and is described in next section.
4.3 Experimental Data Noise Reduction

Noise is a high-frequency variation in the process measurement that is not associated with the
true process measurement i.e. it is the variation in the sensor reading that does not correspond to
changes in the process and can be by background electrical interference, mechanical vibrations
and process fluctuations (Riggs and Karim 2006). These signal noises are equivalent to errors
which inevitably corrupt the process measurement and render the steady-state performance
56
during the measurement processing and transmission of signal. Hence it is therefore important to
reduce, if not completely eliminate, the effect of noise or errors.
The total error in a measurement, which is the difference between the measured value and the
true value of the variable, can be conveniently represented as the sum of the contributions from
two types of errors random errors and gross errors (Narasimhan and Jordache 2000). Random
error (Nagy 1992; Narasimhan and Jordache 2000) implies that neither the magnitude nor the
sign of the error can be predicted with certainty. In other words, if the measurement is repeated
with the same instrument under identical process conditions, a different value may be obtained
depending on the outcome of the random error. Gross errors imply that at any given time they
have a certain magnitude and sign which may be unknown. Thus, if the measurement is repeated
with the same instrument under identical conditions, the contributions of the systematic gross
error to the measured value will be the same. Random errors can be caused by a number of
different sources such as power supply fluctuations, network transmission and signal conversion
noise, analog input filtering, changes in ambient conditions whereas gross errors are caused by
nonrandom events such as instrument malfunctioning, miscalibration, wear or corrosion of
sensors, and solids deposits. Such gross errors do not apply to our measurement process as no
malfunctioning, miscalibration, wear or corrosion exists with our sensing instruments. The
instruments purchased from relevant vendors are new which certify calibration performed on
them. The temperature instruments have been tested from time to time in ice / water bath to
ensure its accuracy. The instruments are well fitted by qualified Mechanical Technicians. The
only error relevant in our case is the random errors on measurements as additive contributions.
An abundant literature exists on measurement error and its calculation (Lloyd and Lipow 1962).
Characteristics of random error can be described using statistical properties. Hence its mean or
expected value is usually the DC voltage we trying to measure, to which noise are added and its
variance is the standard deviation of the noise. As recommended by (National Instruments
2006), we assume an identical distribution of each of the samples. Specifically, the means of all
the samples are the same, as are the standard deviations. This assumption is convenient because
in calculations we can now use the same statistics to describe each of the samples.
Characterizing the sample as independent is not a good assumption because the character of
noise is often time varying. The standard deviation is a measure of the magnitude of the energy
of whatever AC signal is present (just noise, we hope, in case of a DC measurement) and is
independent of whatever DC signal is present. Since the true standard deviation is never known,
57
an estimate of the standard deviation can be obtained by using the following equation
recommended (National Instruments 2006)

4-1
Where - standard deviation (just noise in case of DC measurement); Xi - sample of noise in

question; E (.) - Expectation (average value) of the quantity inside the brackets.
An important requirement for estimating the standard deviation of a measurement error from a
sample of measurements is that all the measurements of the variable should be drawn from the
same statistical population. We apply this logic to our initial start-up measurements for which
the mean or expected value is fixed at 4mA. This makes the task trivial. Now, the standard
errors calculated are subtracted from the measured values to obtain a true measured value. It
should be noted that the random error generated will not be entirely eliminated (Narasimhan and
Jordache 2000). A second type of redundancy, called temporal redundancy exists as we generate
more data continually from CompactDAQ to determine a steady-state. Temporal redundancy
can be exploited by simple averaging the calculated measurements. This task is accomplished by
using a digital filter. Different digital filters such as exponential filter, Moving Average filter,
polynomial filters and hybrid filters exists. Each filter type has its own advantages. Moving
is a well known low-pass filter defined, for discrete signals, by (Alessio et al.
average
2002)
4-2
The moving average is a finite impulse response (FIR) filter which means that the effect of any
input lasts only for N steps. The equal weight moving average cancels out periodic noise. The
moving average is easy to tune for steady-state or quasi steady-state signals, requiring only the
adjustment of the number of input values used to calculate the average. The moving average
does not overshoot and reaches correct steady-state after a step change. The moving average is
also easy to implement and fast to compute. These calculations are not trivial and are performed
in Excel Visual Basic Program. An Excel Visual Basic program is written to accomplish this
task of reducing the effects of errors on pressure transducer; temperature sensors and flow meter
(refer Appendix D).
58
4.20
NoiseReductionforP1 Transducer
4.20
4.10
4.10
4.00
4.00
3.90
3.90
3.80
4.20
ExpectedValue
mAP1withnoise
Poly.(mAP1withnoisereduction)
mAP1withnoisereduction
Poly.(ExpectedValue)
Poly.(mAP1withnoise)
3.80
4.20
4.10
4.10
4.00
4.00
3.90
3.90
3.80
4.20
ExpectedValue
mAP2withnoise
4.00
3.80
3.80
4.20
4.00
ExpectedValue
mAP3withnoise
3.80
Figure 4-2: Noise reduction for pressure transducers P1, P2, P3
59
NoiseReductionforT1 RTD
19.10
19.10
19.05
19.05
19.00
19.00
18.95
18.95
18.90
18.90
18.85
18.85
18.80
18.80
0 100 200 300 400 500 600 700 800 90010001100120013001400150016001700180019002000

ExpectedValue
T1withnoisereduction
T1withnoise
NoiseReductionforT2 RTD
19.50
19.50
19.00
19.00
18.50
18.50
18.00
18.00
17.50
17.50
17.00
17.00
16.50
16.50
16.00
16.00
15.50
15.50
15.00
15.00
19.50
100
200
ExpectedValue
300
400
500
600
700
800
900
1000
T2withnoise
NoiseReductionforT3 Thermocouple
19.50
19.00
19.00
18.50
18.50
18.00
18.00
17.50
17.50
17.00
17.00
0 100 200 300 400 500 600 700 800 90010001100120013001400150016001700180019002000

ExpectedValue
T3withnoise
Figure 4-3: Noise reduction for temperature sensors T1, T2, T3
60
NoiseReductionforFlowTransducer
4.10
4.10
4.08
4.08
4.06
4.06
4.04
4.04
4.02
4.02
4.00
4.00
3.98
3.98
3.96
3.96
3.94
3.94
3.92
3.92
3.90
3.90
mAFlowratewithnoisereduction
0 100ExpectedValue
200 300 400 500 600 700 800 9001000110012001300
1400150016001700180019002000
mAFlowratewithnoise
Poly.(mAFlowratewithnoisereduction)
Poly.(mAFlowratewithnoise)
Figure 4-4: Noise reduction for flow-meter

The procedure explained above for noise reduction works well and its implementation can be
confirmed by running the program on a set of measurements obtained during 200 KPa gauge
test. The graphs are summarized for signals obtained from pressure transducers, flow
transducers and temperature sensors. A clear reduction in the effect of random error can be seen
in Figure 4-2, figure 4-3, and figure 4-4. Referring to these graphs it can be said that the noise or
random error produced during the signal measurement is reduced, thus, attaining the true value
of the measurement.
4.4 Experimental Analysis

A total of 9 venting experiments were carried out in the fluid flow laboratory which were
designated from VPM-1 to VPM-9 (VPM Vent Pipe Model). These experiments were divided
into three sets each containing 3 experiments. A set differs with respect to the initial pressure.
Set-1 experiments were performed at an initial pressure of 200 KPa G, Set-2 at an initial
pressure of 300 KPa G and Set-3 at an initial pressure of 400 KPa G. Maintaining a steady-state
pressure above 400 KPa G into the vent pipe was not possible due to restricted flow supply from
the laboratory air compressor. In all cases, venting was to atmosphere so the back pressure was 0
KPa G. Experiments were repeated in order to ensure reproducibility. Not all experiments were
carried out for the same time period. Compressed Air from laboratory air compressor and an
instrument graded air in G-size cylinder (single) was used in the experiments. Cylindrical gas
bottles were provided by BOC Gases. The air gas composed of 78.12% Nitrogen, 20.96%
61
Oxygen and 0.92% Argon by mole (BOC Gases 2006). The reason for utilizing air is because of
its simplicity and cheapness (Glushkov, Selyanskaya, and Kas'yanov 2003). Air has only a
single phase over the pressures and temperatures encountered in our experiment, and departures
from ideal gas behavior are small. Also, the restriction of blowing down any supercritical or
hydrocarbon gases into the atmosphere on the Curtin University premises favored air only.
A simple procedure was adopted to ensure steady-state conditions are achieved into the vent
pipe. A pressure regulator and a bleed valve arrangement was installed initially which did not
prove to be effective and was discarded. Two 20 NB full bore ball valves were used in order to
achieve steady-state flow conditions. One ball valve (bv1) was placed after the accumulator and
other (bv2) before the gas enters the vent pipe. Valve (bv2) was used as the open/close valve
whereas the valve (bv1) was used to function as a regulator to achieve the steady-state
conditions. Air was supplied by a rubber air hose to the accumulator. Initially, on start-up the
valve (bv2) was kept in closed position and valve (bv1) was opened slowly. Pressure gauge
installed on the accumulator line was used to observe the pressure required. Once the required
air pressure is achieved valve (bv2) was opened slowly and steady-state conditions were
achieved with valve (bv1). Pressure, temperature and standard flow readings were recorded as
analog values and the entire process was monitored using LabView Signal Express software.
Not all experimental readings / logs could be stored as the software was only a demo version
provided with Compact DAQ. The results obtained from the experiments are compared with the
model predictions and interpreted in the latter section.
4.5 Experimental Validation

Predictions made using the vent pipe model have been conducted with all of the validatory
experiments VPM-1 to VPM-9. Three selected representative comparisons namely VPM-1,
VPM-4 and VPM-7 are given in what follows. In all experiments conducted, air is always in
gaseous state. No condensation or formation of two-phase is likely to take place due to low
pressures involved. One point that must always be borne in mind while comparing the
experimental results and vent pipe model predictions is that the model contains no disposable
parameters. Thus there can be no adjustment of parameters in order to ensure good agreement
between the experimental measurements and the predictions. The vent pipe model is completely
predictive.
The test section is well insulated with glass wool so that the entire arrangement is considered to
be an adiabatic process. The RTDs are not fitted exactly in the streamline of the flowing gas.
62
This is because of area restrictions present with the geometry of RTD tube and the inside
diameter of the vent pipe. If RTDs are fitted in such a way that the tip of the RTD is immersed
half way into the flowing air stream then this will cause restriction to flow inside the duct. Care
is taken to ensure the tip of RTD is not causing any restriction in the flow. Along the duct length
the velocity of air is always accelerating towards the exit of the vent pipe. As the air flows inside
the duct, RTD senses the temperature of air at the point of contact. At this point, the velocity of
air is likely to be decelerated. According to (Saad 1993), when a fluid is decelerated to zero
velocity in a steady-flow adiabatic process, the resulting properties of the fluid are called
stagnation properties, provided that no work interactions occurs and also gravitational, magnetic,
electric and capillary effects are absent. According to this definition, the measured temperature
of air will be the stagnation temperature and not static temperature i.e. the temperature measured
will not be the actual temperature of the flowing air gas.
4.5.1
Experiment VPM-1
Stagnation temperature measurements were taken at the entry and exit of the vent pipe.
Measuring the stagnation temperature along the entire length of the vent pipe was not possible
due to difficulty in getting the instruments fitted along the vent pipe. The system was allowed to
attain steady-state condition by controlling the flow. The final steady-state stagnation
temperature measurements were recorded. After performing noise reductions on the recorded
measurements, these were summarized in figure 4-5. The experimental values were plotted for a
steady-state period only. The disturbance occurring prior to achieving steady-state condition was
not plotted. The stagnation temperature predictions by the vent model were compared to the
experimental values. It was seen that the exit stagnation temperature achieved a steady-state
value quicker than the inlet stagnation temperature. The final steady-state value for the inlet
stagnation temperature was 19.04C whereas the exit stagnation temperature value achieved was
18.92C. The inlet stagnation temperature value obtained from experimental analysis was
inputted into the vent model to predict the exit stagnation temperature value. The stagnation
temperature values along the vent profile were also predicted and are summarized in figure 4-5.
The predicted steady-state exit stagnation temperature was 18.97 C. This predicted value when
compared to the exit experimental value results in a percent difference of 0.26% which
equivalent to 0.05C and is relatively very small. There is clearly excellent agreement between
the predicted stagnation temperature and experimental stagnation temperatures.
63
VentPipeLength(m)
1
10
11
12
20.0
20.0
ExperimentTemperatureT1
19.6
PredictedStagnation
Temperatures
19.4
TemperatureC
19.8
ExperimentTemperatureT2
19.6
19.2
19.4
19.2
19.0
19.0
18.8
18.8
18.6
18.6
18.4
18.4
18.2
18.2
18.0
TemperatureC
19.8
18.0
900
920
940
960
980
1000
1020
1040
1060
Figure 4-5: Comparison of model predicted stagnation temperatures with experimental

stagnation temperatures for VPM-1
VentPipeLength(m)
2
10
11
12
19.0
18.0
18.0
17.0
17.0
16.0
16.0
15.0
15.0
ThermocoupleOutsideWall
Temperature
PredictedAdiabaticWall
Temperatures
TemperatureGunSurface
TemperatureMeasurements
PredictedGasTemperatures
14.0
13.0
12.0
11.0
14.0
13.0
TemperatureC
TemperatureC
1
19.0
12.0
11.0
10.0
10.0
900
920
940
960
980
1000
1020
1040
1060
Figure 4-6: Comparison of model predicted wall temperatures with experimental wall
temperatures for VPM-1
64
25
24
24
23
23
22
22
21
21
20
20
19
19
18
18
17
ExperimentalFlowrate
17
16
PredictedFlowrate
16
15
15
900
920
940
960
980
1000
Time (sec)
1020
1040
StandardVolumetricFlowrate(Nm3/hr)
25
1060
Figure 4-7: Comparison of model predicted standard volumetric flow rate with experimental
250
250
200
200
150
150
100
PressureP1
100
PressureP2
50
PressureP3
Pressure(KPaG)
Pressure(KPaG)
standard volumetric flow rate for VPM-1
50
900 910 920 930 940 950 960 970 980 990 1000 1010 1020 1030 1040 1050 1060 1070
Time (sec)
Figure 4-8: Experimental pressure measurements for VPM-1

Actual gas temperature measurement is not a trivial task and requires very accurate measuring
instruments. One way of estimating the actual gas temperature is by determining the pipe wall
temperature during the steady-state flow process and calculating the gas temperature assuming a
recovery factor (McAdams, Nicolai, and Keenan 1946). In order to evaluate this, the process has
to be adiabatic so that no heat transfer takes place with the surrounding. A T-type thermocouple
was attached to the pipe wall at the exit of the vent pipe. The thermocouple was attached using
an adjustable ring so that a firm contact exists between the pipe wall and thermocouple. The
temperature readings were recorded at a nominal sampling rate. After performing noise
reductions on these temperature readings, these measurements were summarized in figure 4-6.
The final temperature recorded on achieving steady-state was 16.56 C. The surface pipe wall
temperatures at every 1m surface were measured by a non-contact dual laser thermometer. The
65
temperatures recorded by temperature gun are also summarized in figure 4-6. The predicted gas
temperature values for the predicted stagnation temperatures were used in determining the
adiabatic wall temperature. The predicted adiabatic wall temperatures are represented in figure
4-6 and compared to the measured surface pipe wall temperatures. The percentage differences
between predicted and temperature gun measurements were calculated to be 0.0% at the inlet
and -4.53% (equivalent to -0.73C) at the exit of the vent pipe. The percent difference between
predicted and thermocouple temperature measurement was calculated to be 2.26% which is
equivalent to -0.37C. Clearly, there is also a good agreement between the measured and
predicted pipe wall temperatures. In particular, the minimum wall temperature at the exit of the
vent pipe, which is of significance to the materials of construction of the pipe itself, is underpredicted within 2.26%. Thus, there exists a very close agreement between the vent pipe model
temperature predictions and experimental analysis.
Now, that we have a close agreement between the predicted wall temperatures and the
experimental values, we can say that the predicted static temperature values and actual
temperature values must be in close agreement as well and are summarized in figure 4-6. A gas
temperature drop of 6.53 C was predicted for an inlet pressure of 200 KPa gauge.
The standard volumetric flow rate for the gas was recorded using IFM Effector 300 flow sensor.
The flow measurements were recorded for the steady-state pressure of 200 KPa gauge for the
same time period as for temperatures. After performing noise reductions on these readings, these
values are summarized in figure 4-7. However, it was difficult to maintain a steady-state
condition in the vent pipe due to supply issues from the laboratory air compressor. This resulted
in slight variations in the pressure and flow rate readings. In order to have a close comparison
between the predictions and experimental values, it was decided to predict the flow rates for the
corresponding experimental pressure readings. The predicted flow rates were compared to the
experimental flow rate results. An average standard volumetric flow of 22.89 Nm3/hr was
attained on achieving steady-state during the experiment whereas an average standard
volumetric flow of 20.46 Nm3/hr was predicted by the vent pipe model. The comparison result
tells us that there exists a percentage difference of -10.6% which is equivalent to 2.43 Nm3/hr.
Hence the flow rate is under-predicted by the vent pipe model. However, the calculated
difference is not very high and is acceptable.
Pressure measurements were recorded at the entry and exit of the vent pipe using pressure
transducers. After performing noise reductions these values are summarized in figure 4-8.
66
However, it was not possible to determine the pressure along the vent pipe and hence the
pressure profile is discussed in more detail in Hysys validation. A similar approach to
experiment VPM-1 was adopted in comparing the vent pipe model predictions to experimental
analysis at 300 and 400 KPa inlet gauge pressures.
4.5.2
Experiment VPM-4
Stagnation temperature measurements were taken at the entry and exit of the vent pipe. The
system was allowed to attain steady-state condition by controlling the flow. The final steadystate stagnation temperature measurements were recorded. After performing noise reductions on
the recorded measurements, these were summarized in figure 4-9. The experimental values were
plotted for a steady-state period only. The disturbance occurring prior to achieving steady-state
condition was not plotted. The stagnation temperature predictions by the vent model were
compared to the experimental values. It was seen that the exit stagnation temperature achieved a
steady-state value quicker than the inlet stagnation temperature. The final steady-state value for
the inlet stagnation temperature was 18.59C whereas the exit stagnation temperature value
achieved was 18.35C. The inlet stagnation temperature value obtained from experimental
analysis was inputted into the vent model to predict the exit stagnation temperature value. The
stagnation temperature values along the vent profile were also predicted and are summarized in
figure 4-9. The predicted steady-state exit stagnation temperature was 18.41C. This predicted
value when compared to the exit experimental value results in a percent difference of 0.33%
which is equivalent to 0.06C and is very small. There is clearly excellent agreement between
the predicted stagnation temperature and experimental stagnation temperatures. Actual gas
temperature measurement was estimated in a similar manner to experiment VPM-1. After
performing noise reductions on these temperature readings, these measurements were
summarized in figure 4-10. The final temperature recorded on achieving steady-state was
15.33C. The surface pipe wall temperatures at every 1m surface were measured by a noncontact dual laser thermometer. The temperatures recorded by temperature gun are also
summarized in figure 4-10. The predicted gas temperature values for the predicted stagnation
temperatures were used in determining the adiabatic wall temperature. The predicted adiabatic
wall temperatures are represented in figure 4-10 and compared to the measured surface pipe wall
temperatures. The percentage differences between predicted and temperature gun measurements
were calculated to be 0.19% at the inlet and 3.88% (equivalent to 0.55C) at the exit of the vent
pipe. The percent difference between predicted and thermocouple temperature measurement was
calculated to be -3.78% which is equivalent to -0.58C.
67
VentPipeLength(m)
1
10
11
12
19.5
19.5
ExperimentStagnation
TemperatureT1
ExperimentStagnation
TemperatureT2
PredictedStagnation
Temperatures
19.1
TemperatureC
18.9
19.3
19.1
18.9
18.7
18.7
18.5
18.5
18.3
18.3
18.1
18.1
17.9
17.9
17.7
17.7
TemperatureC
19.3
17.5
17.5
80
180
280
380
480
580
680
Figure 4-9: Comparison of model predicted stagnation temperature with experimental stagnation
temperature for VPM-4
VentPipeLength(m)
2
10
11
12
19.0
19.0
17.0
17.0
15.0
15.0
13.0
13.0
Temperature
11.0
9.0
Temperatures
9.0
7.0
7.0
11.0
5.0
80
180
TemperatureC
TemperatureC
5.0
280
380
480
580
680
Figure 4-10: Comparison of model predicted wall temperature with experimental wall
Clearly, there is also a good agreement between the measured and predicted pipe wall
temperatures. In particular, the minimum wall temperature at the exit of the vent pipe, which is
of significance to the materials of construction of the pipe itself, is under-predicted within 3.78% which is equivalent to -0.58C. The percent difference is relatively small and is
acceptable. Once again there is a good agreement between the vent pipe model temperature
68
predictions and experimental analysis. Now, that we have a close agreement between the
predicted wall temperatures and the experimental values, we can say that the predicted static
temperature values and actual temperature values must be in close agreement as well and are
summarized in figure 4-9. A gas temperature drop of 12.53 C was predicted for a pressure drop
of 300 KPa gauge.
The standard volumetric flow measurements were recorded for the steady-state pressure of 300
KPa gauge for the same time period as for temperatures. After performing noise reductions on
these readings, these values are summarized in figure 4-11. It was difficult to maintain a steadystate condition in the vent pipe due to supply issues from the laboratory air compressor. This
35
35
PredictedFlowrate
33
33
31
31
29
29
27
27
25
25
80
180
280
380 (sec)
Time
480
580
resulted in slight variations in the pressure and flow rate readings.
680
Figure 4-11: Comparison of model predicted standard volumetric flowrate with experimental
standard volumetric flow for VPM-4
In order to have a close comparison between the predictions and experimental values, it was
decided to predict the flowrates for the corresponding experimental pressure readings. The
predicted flowrates were compared to the experimental flowrate results. An average standard
volumetric flow of 29.86 Nm3/hr was attained on achieving steady-state during the experiment
whereas an average standard volumetric flow of 27.98 Nm3/hr was predicted by the vent pipe
model. The comparison result tells us that there exists a percentage difference of -6.3% which is
equivalent to 1.88 Nm3/hr. Hence the flowrate is under-predicted by the vent pipe model.
However, the calculated difference is not very high and is acceptable. Pressure measurements
were recorded at the entry and exit of the vent pipe using pressure transducers. After performing
noise reductions these values are summarized in figure 4-12. However, it was not possible to
determine the pressure along the vent pipe and hence the pressure profile is discussed in more
detail in Hysys validation.
69
350
300
300
250
PressureP1
250
200
PressureP2
200
PressureP3
150
150
100
100
50
50
Pressure(KpaG)
Pressure(KpaG)
350
0
80
180
280
Time380
(sec)
480
580
680
Figure 4-12: Experimental pressure for VPM-4
4.5.3
Experiment VPM-7
Stagnation temperature measurements were taken at the entry and exit of the vent pipe. The
system was allowed to attain steady-state condition by controlling the flow. The final steadystate stagnation temperature measurements were recorded. After performing noise reductions on
the recorded measurements, these were summarized in figure 4-13. The experimental values
were plotted for a steady-state period only. The disturbance occurring prior to achieving steadystate condition was not plotted. The stagnation temperature predictions by the vent model were
compared to the experimental values. It was seen that the exit stagnation temperature achieved a
steady-state value quicker than the inlet stagnation temperature. The final steady-state value for
the inlet stagnation temperature was 18.55C whereas the exit stagnation temperature value
achieved was 18.16C. The inlet stagnation temperature value obtained from experimental
analysis was inputted into the vent model to predict the exit stagnation temperature value. The
stagnation temperature values along the vent profile were also predicted and are summarized in
figure 4-13. The predicted steady-state exit stagnation temperature was 18.27C. This predicted
value when compared to the exit experimental value results in a percent difference of 0.6%
which is equivalent to 0.11C and is very small. There is clearly excellent agreement between
the predicted stagnation temperature and experimental stagnation temperatures. Actual gas
temperature measurement was estimated in a similar manner to experiment VPM-1 and VPM-4.
After performing noise reductions on these temperature readings, these measurements were
summarized in figure 4-14. The final temperature recorded on achieving steady-state was
12.39C. The surface pipe wall temperatures at every 1m surface were measured by a noncontact dual laser thermometer. The temperatures recorded by temperature gun are also
summarized in figure 4-14. The predicted gas temperature values for the predicted stagnation
temperatures were used in determining the adiabatic wall temperature.
70
VentPipeLength(m)
1
10
11
12
19.5
19.5
ExperimentalStagnation
TemperatureT1
ExperimentalStagnation
TemperatureT2
PredictedStagnation
Temperatures
19.1
TemperatureC
18.9
18.7
19.3
19.1
18.9
18.7
18.5
18.5
18.3
18.3
18.1
18.1
17.9
17.9
17.7
17.7
TemperatureC
19.3
17.5
17.5
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
170
180
190
200
The predicted adiabatic wall temperatures are represented in figure 4-14 and compared to the
measured surface pipe wall temperatures. The percentage differences between predicted and
temperature gun measurements were calculated to be 0.49% at the inlet and 9.79% (equivalent
to 1.13C) at the exit of the vent pipe. The percent difference between predicted and
thermocouple temperature measurement was calculated to be 1.91% which is equivalent to
0.24C.
VentPipeLength(m)
2
10
11
12
18
18
16
16
14
14
12
12
10
10
Temperature
Temperatures
8
6
4
2
TemperatureC
TemperatureC
6
4
2
0
0
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
170
180
190
200
Again, there is good agreement between the measured and predicted pipe wall temperatures. In
particular, the minimum wall temperature at the exit of the vent pipe, which is of significance to
71
the materials of construction of the pipe itself, is over-predicted within -1.91% which is
equivalent to 0.24C. Thus, there exists a good agreement between the vent pipe model
temperature predictions and experimental analysis.
Now, that we have a close agreement between the predicted wall temperatures and the
experimental values, we can say that the predicted static temperature values and actual
temperature values must be in close agreement as well and are summarized in figure 4-14. A gas
40
40
39
PredictedFlowrate
39
38
38
37
37
36
36
35
35
34
34
33
33
32
32
31
31
temperature drop of 17.35 C was predicted for a pressure drop of 400 KPa gauge.
30
30
20
30
40
50
60
70
80
90
100
110 (sec)
120
Time
130
140
150
160
170
180
190
200
Figure 4-15: Comparison of model predicted standard volumetric flowrate with experiment
flowrate for VPM-7
The standard volumetric flow measurements were recorded for the steady-state pressure of 400
KPa gauge for the same time period as for temperatures. After performing noise reductions on
these readings, these values are summarized in figure 4-15. It was difficult to maintain a steadystate condition in the vent pipe due to supply issues from the laboratory air compressor as the
pressure vent on increasing. This resulted in slight variations in the pressure and flowrate
readings. In order to have a close comparison between the predictions and experimental values,
it was decided to predict the flowrates for the corresponding experimental pressure readings.
The predicted flowrates were compared to the experimental flowrate results. An average
standard volumetric flow of 35.44 Nm3/hr was attained on achieving steady-state during the
experiment whereas an average standard volumetric flow of 35.40 Nm3/hr was predicted by the
vent pipe model. The comparison result tells us that there exists a percentage difference of 0.11%. Hence, an excellent agreement exists between the predicted and experimental flow rates
at 400 KPa gauge pressure. In particular, maximum flow which is of significance in determining
72
the choke conditions for designing of flare systems and velocities is accurately predicted by the
450
450
400
400
350
350
300
300
250
250
200
200
PressureP1
150
150
PressureP2
100
Pressure(KpaG)
Pressure(KpaG)
vent pipe model within 0.11% which is equivalent to 0.04Nm3/hr.
100
PressureP3
50
50
0
20
30
40
50
60
70
80
90
100Time(sec)
110 120
130
140
150
160
170
180
190
200
Figure 4-16: Experiment pressure measurements for VPM-7

Pressure measurements were recorded at the entry and exit of the vent pipe using pressure
transducers. After performing noise reductions these values are summarized in figure 4-16.
However, it was not possible to determine the pressure along the vent pipe and hence the
pressure profile is discussed in more detail in Hysys validation.
73
Figure 4-17: Aspen Hysys Simulation Flowsheet
74
4.6 Validation with Hysys

In order to assess the predicted performance of the developed vent pipe model, complete process
simulations were performed. Despite some expected differences between a process simulation
and real-life operation, process simulators are commonly used to provide reliable information on
process operation, owing to their vast component libraries, comprehensive thermodynamic
packages and advanced computational methods (West, Posarac, and Ellis 2008). Predictions
made using the vent pipe model have been compared with simulations performed for air,
carbon-dioxide, methane and a multicomponent mixture of hydrocarbons (Kirk-Burnnand
2009). HYSYS was selected as a process simulator for both its simulation capabilities and its
ability to incorporate calculations using the spreadsheet tool. It differs from other process
simulators such as ASPEN PLUS in two respects: interactive interpretation of the
commands/units as they entered and bi-directional information flow (Pareek 2008). Steady-state
simulations were performed in Hysys version 7.1. The first step in developing the process
simulation was selecting the chemical components for the process, as well as a thermodynamic
model. Additionally, the unit components and input conditions for the venting process must be
selected and specified. The unit operations and input conditions were selected based on the vent
pipe model to ensure that the venting process simulated in HYSYS could be compared in a
consistent manner. Since no polar compounds are present, Peng-Robinson model was selected as
the property package for the simulation because of its simplicity and accuracy (Peng and
Robinson 1976). A number of cubic equations of states are available but Peng-Robinson
thermodynamic model is selected because of its wide use in development of different
mathematical models. Another equation known for its accuracy is the Soave-Redlich-Kwong
(Soave 1972). However, the performance of Peng-Robinson equation is better than SoaveRedlich-Kwong equation in all cases tested and shows its greatest advantage in the prediction of
vapor pressure of pure substances, liquid phase densities and equilibrium ratios of mixtures. In
regions where engineering calculations are frequently required the Peng-Robinson equation
gives better agreement between predictions and experimental PVT data (Peng and Robinson
1976). No heat transfer approach with the surrounding was considered. The Hysys process flowsheet for the vent pipe model is represented in figure 4-17 where the CGP-100 is the vent pipe
section.
4.6.1
Comparison with Hysys Simulation for Air
The vent pipe models predicted results for compressible gas such as air are compared with the
simulated results of Aspen Hysys. The predicted mass flow rates, pressure profile, temperature
75
profile, Mach number profile, density profile and velocity profile for air from the vent pipe
model and Hysys are analyzed on an excel spreadsheet and various graphs are plotted.
Remember that we are specifically interested in the exit conditions of the vent pipe. However,
the different parameter profiles are discussed as well. The vent pipe design specifications used
in the vent model and Aspen Hysys model puts restriction on the flow and results in a choking
condition at the end of the vent pipe with sonic conditions. This sonic condition for air was
calculated at ~750 K Pa gauge pressure. Due to limitations imposed in Aspen Hysys, the vent
pipe flow sheet did not converge which resulted in increasing the back pressure. The following
two cases are evaluated here: Air Case 1- Pressure range of 100-500 K Pa gauge (atmospheric
blowdown) and Air Case 2- Pressure range of 600-1000 K Pa gauge (back pressure). The
comparison percentage differences are calculated in both cases for Hysys simulations and Vent
pipe model. Enthalpy and Entropy along the vent profile are also assessed which helped in
understanding the fanno line.
4.6.1.1
Air Case 1: Pressure range 100-500 KPa gauge
The predicted and simulated results for mass flow rates at steady-state conditions in the pressure
range 100-500 KPa gauge are in close agreement. The comparison percentage difference
calculated in table G 11-1 on mass flow rates, predicted by the vent model, are at minimal. The
minimum percentage difference calculated for mass flow rate was 0.25% and maximum
percentage difference was calculated at 0.568%. Figure 4-18 shows the pressure profile for the
vent model predictions and Aspen Hysys simulated results for air at steady-state mass flow
conditions in the pressure range 100-500 K Pa gauge. The pressures at the inlet and exit
conditions were calculated in both cases and were found to be in close agreement. Obviously,
the percentage comparison difference at the inlet of the vent pipe was 0% whereas that
calculated at the exit of vent was 0.025%. This could be due to minor calculation discrepancy.
Overall an excellent agreement in the mass flow rate and exit pressure prevails. The pressure
profile along the vent pipe was analyzed. The vent pipe was divided into twelve sections and the
pressures at entry / exit of each section was calculated. The predicted results were compared
with Hysys simulated results. Initially, the pressure profile follows a linear path and then
decreases exponentially attaining exit conditions (atmospheric). The mass flowrate in all cases
(from 100-500 K Pa gauge vent pipe profiles) should be constant prevailing steady-state
conditions. The predicted pressure readings along the vent length compares well with Hysys and
are within 0.6% of the simulated Hysys values for the first ten sections.
76
1.2
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
Vent Pipe Lenght, m
10.2
10.8
11.4
12
0.6
3.6
4.2
Vent Model Prediction @ 200 KPaG
Aspen HYSYS @ 200 KpaG
2.4
Vent Model Prediction @ 100 Kpa G
1.8
1.2
4.8
6.6
Aspen HYSYS @ 300 KPaG
Vent Pipe Length, m
5.4
7.2
8.4
7.8
9.6
10.8
11.4
10.2
Figure 4-19: Predicted temperature comparison of vent pipe model with Hysys simulation for air in the pressure range 100-500 KPa gauge
20
18
16
14
12
10
8
6
4
2
0
-2 0
-4
-6
-8
-10
20
18
16
14
12
10
8
6
4
2
0
12 -2
-4
-6
-8
-10
101.3
0.6
101.3
201.3
201.3
401.3
501.3
601.3
301.3
301.3
401.3
501.3
601.3
Figure 4-18: Predicted pressure comparison of vent pipe model with Hysys simulation for air in the pressure range 100-500 KPa gauge
Aspen HYSYS Simulation: Pressure K Pa
Aspen HYSYS Simulation: Temperature C
Vent Pipe Model Predictions: Pressure K Pa
77
Vent Pipe Model Predictions: Temperature C
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.2
10.8
11.4
12
Vent Pipe Length, m
0.6
1.2
1.8
2.4
3.6
4.2
4.8
6.6
Vent Pipe Lenght, m
5.4
7.2
7.8
8.4
9.6
10.2
10.8
11.4
Figure 4-21: Predicted density comparison of vent pipe model with Hysys simulation for air in the pressure range 100-500 KPa gauge
7.5
7.0
6.5
6.0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
12
7.5
7.0
6.5
6.0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0.0
1.2
0.1
0.0
0.6
0.2
0.1
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Figure 4-20: Predicted mach no. comparison of vent pipe model with Hysys simulation for air in pressure range 100-500 KPa gauge
Aspen HYSYS Simulation: Mach Number
Aspen HYSYS Simulation: Density kg/m3
Vent Pipe Model Predictions: Mach Number
78
Vent Pipe Model Predictions: Density kg/m3
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.2
10.8
11.4
12
Vent Pipe Lenght, m
260
6.20
265
270
275
280
285
290
295
6.30
Fanno Curve @ 500 Kpa G
6.40
6.50
Entropy, KJ/kg-K
6.60
6.70
6.80
6.90
1.2
0.6
50
50
100
150
200
250
300
100
150
200
250

Figure 4-23: Predicted Enthalpy-Entropy (Fanno curve) of vent pipe model for air in pressure range 100-500 KPa gauge
Enthalpy, KJ/kg
Figure 4-22: Predicted velocity comparison of vent pipe model with Hysys simulation for air in the pressure range 100-500 KPa gauge
Aspen HYSYS Simulation: Velocity m/sec
300
79
Vent Pipe Model Predictions: Velocity m/sec
The mystical cases are the last two sections. Although, the exit condition (atmospheric
pressure) is obtained at the end of the vent pipe, the pressure drop in the 11th section of the
vent pipe is significantly high in Hysys simulation then predicted by the model. The reason
for existence of such a pressure profile in the last two sections is unclear at this stage.
The model predictions and Hysys simulated temperatures along the vent pipe were plotted
and percent comparison differences were calculated. Figure 4-19 shows the temperature
profile for the vent model predictions and Aspen Hysys simulated results for air at steadystate mass flow conditions in the pressure range 100-500 KPa gauge. Once again looking at
the Figure 4-19, it can be said that the exit temperature readings in all cases are in close
agreement with a minimum difference of -0.72% and a maximum difference of 1.67%. The
temperature profile along the vent pipe again follows a linear profile initially and then
decreases exponentially to attain a final exit temperature. The model predicted temperatures
in the first ten sections of the vent pipe match with the Hysys simulated temperature readings
and are within 1.67%. A similar temperature profile as in case of pressure is obtained in the
last two sections of the vent pipe. The reason for existence of such a temperature profile is
unclear at this stage.
Mach number, density and velocity along the vent pipe were assessed and plotted. Figure 420 shows the Mach number profile for the vent model predictions and Aspen Hysys
simulated results for air at steady-state mass flow conditions in the pressure range 100-500
KPa gauge. Based on the assessment performed, it can be said that the Mach number along
the vent profile is increasing towards the exit of the vent pipe and approaching towards sonic
velocity. The exit Mach number readings in first ten sections are in close agreement with a
minimum difference of 0.13% and a maximum difference of 0.63%.
Figure 4-21 shows the density profile for the vent model predictions and Aspen Hysys
KPa gauge. A fall in density of fluid is noticed along the vent profile. The graph is very
similar to pressure Figure 4-18 and temperature Figure 4-19 which shows decreasing
linearity and an exponential fall. The exit density readings in first ten sections are in close
agreement with a minimum difference of -0.2% and a maximum difference of 0.02%.
Figure 4-22 shows the velocity profile for the vent model predictions and Aspen Hysys
80
KPa gauge. Once again, it can be said that the exit velocity readings in all cases are in close
agreement with a minimum difference of 0.4% and a maximum difference of 0.61%. The
velocity profile developed is completely opposite to temperature profile and very similar to
Mach number profile. Overall the results were found to be in very close agreement for all
parameters in case 1. The percent difference between the vent model and Hysys was
comparatively high in the 11th section of the vent pipe.
Figure 4-23 represents the enthalpy and entropy along the vent profile for the pressure range
100-500 K Pa gauge. The various curves formed are known as Fanno Curve or Fanno
Line. As can be seen from the Figure 4-23, the enthalpy is decreasing along the vent profile
and a simultaneous increase in entropy is noticed. As discussed in model development,
friction is an important parameter which brings about the changes in the flow conditions. To
define a flow in a region or duct, the effects of friction must be monitored. In our case,
friction is causing an increase in the velocity and Mach number with a simultaneous
decrease in enthalpy and pressure. The fanno line represents the effects of friction on the
flow parameters. In Figure 3-3, the maximum Mach number which could be obtained at the
end of the vent will be unity representing a case of adiabatic sonic flow. At this point flow is
choked. On comparison of Figure 4-23 with Figure 3-3, it can be said that the fanno curve in
Figure 4-23 represents the upper part of the general fanno curve. This region represents the
subsonic flow region. Thus for case 1, the qualitative character of the flow is markedly
influenced by subsonic flow conditions.
4.6.1.2
Air Case 2: Pressure range 600-1000 KPa gauge
Simulations performed in Aspen Hysys at pressures > 600 K Pa gauge did not converge to
achieve atmospheric pressure at the exit of the vent pipe. This could be a restriction in
Hysys. However, no further investigations were performed on this matter. In order to
simulate the Hysys model, the back pressure (pressure at the exit of the vent) was increased
by a relative amount such that the vent exit pressure equals to the back pressure. This was
done by adjusting the steady-state mass flow condition. Trial and error methods were
performed in order to solve the Hysys model at minimal back pressure (above atmosphere).
The exit pressures obtained from the Hysys simulations were inputted in the vent pipe model
and relevant predictions were calculated. The results obtained were tabulated in table G 11-2
and pressure profile, temperature profile, Mach number profiles, density profile and velocity
81
profile along the vent pipe were plotted. Similar results were obtained as in case1 and are
discussed here.
The vent model predictions and Aspen Hysys simulated results for mass flow in the pressure
range 600-1000 K Pa gauge are found to be in very close agreement. A minimum of 0.08%
and a maximum of 0.22% of comparison difference were calculated. Figure 4-24 shows the
pressure profile for the vent model predictions and Aspen Hysys simulated results for air at
steady-state mass flow conditions in the pressure range 600-1000 K Pa gauge. The pressures
at the inlet and exit conditions were calculated and were found to be in close agreement. The
pressure profile along the vent pipe was analyzed in the same way as in case-1. Similar
results representative to case-1 were obtained. The percentage comparison difference was
well within limits for the first ten sections and was calculated to be 0.47% better than case1. Figure 4-25 shows the temperature profile for the vent model predictions and Aspen
Hysys simulated results for air at steady-state mass flow conditions in the pressure range
600-1000 KPa gauge. The temperature profile developed along the vent pipe was of the
same pattern as case-1 representing a decreasing linearity followed by an exponential
decrease to attain exit conditions. The percentage comparison difference calculated was
1.71% for the first ten sections of the vent pipe. Mach number, density and velocity profile
were also plotted and are represented in Figure 4-26, Figure 4-27 and Figure 4-28. The
density profile developed is very similar to that of the pressure drop profile which confirms
the existence of relationship between them. The profiles developed represent a linear and
exponential increase in Mach number and velocity and are opposite to the temperature
profile. Enthalpy-Entropy plots representing the fanno curve for the pressure range 600-1000
K Pa gauge are plotted in Figure 4-29. Once again the flow is characterized to be as subsonic
with a decrease in enthalpy and an increase in entropy proving the irreversibility of the
process.
The mystical condition developed in the last two sections of the vent pipe in case 1 was also
seen in case 2. This condition can be explained by the fanno line equation stated in model
development. As discussed in the previous section of model development, a decrease in
density is always registered according to the fanno equation. The mass flow per unit area
must remain constant and in order to compensate for this the velocity in this region
increases. As can be seen in the 11th section, the difference in the density of air is high when
compared to the other sections (refer table 4-1 and table 4-2). However, this does not solve
82
our problem. After a careful consideration, it was concluded that this discrepancy could be
generated because of calculations performed with different equation of states used in the
vent model and Aspen Hysys. The GERG-2004 (Kunz et al. 2007) equation of state was
used in calculating the thermophysical properties in vent model whereas Peng-Robinson
(Peng and Robinson 1976) equation of state was used in calculating the thermophysical
properties in Aspen Hysys simulation. The reason for the profile difference must be a result
of limitations imposed by the equations of state. According to (Setzmann and Wagner 1991),
the density values calculated from the Peng-Robinson equation of state deviate from the
reference equation of state by up to +5% at pressures below 30MPa. However, this research
was conducted for Methane gas. (Kunz et al. 2007) mentioned that the calculated values for
the speed of sound show deviations of more than 10% in the same temperature and pressure
ranges. This can affect our density, velocity and Mach number profiles. It was also reported
that the suitability of the Peng-Robinson equation of state for use in technical applications
which require high accuracy predictions of the properties of natural gases quickly revealed
serious deficiencies. The GERG-2004 equation of state was developed with a view to
overcome such difficulties. Hence the vent model predictions can be proved to be more
accurate than the simulated results from Aspen Hysys.
Overall it can be concluded that the vent pipe models predictions for air are in very close
agreement with Aspen Hysys simulated results.
In order to investigate that the vent model predictions hold true not only for compressible
gas such as air but also for other gases, it was decided to perform simulations in Aspen
Hysys incorporating supercritical and hydrocarbon gases such as carbon-dioxide and
methane. The predicted and simulated pressure, temperature, Mach number, density and
velocity profiles were assessed. The percentage comparison difference is calculated in all
cases. The results for these gases are discussed here.
83
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
Vent Pipe Lenght, m
10.2
10.8
12
7.2
7.8
8.4
9.6
10.2
11.4
10.8
12
-12
-8
-4
Figure 4-25: Predicted temperature comparison of vent pipe model with Hysys simulation for air in the pressure range 600-1000 KPa gauge
6.6
-20
Vent Pipe Length, m
5.4
-16
4.8
-20
4.2
-16
-12
-8
3.6
0
3
0
2.4
4
1.8
1.2
12
12
0.6
16
16
-4
20
20
11.4
101.3
2.4
101.3
1.8
201.3
1.2
301.3
201.3
0.6
401.3
301.3
501.3
601.3
701.3
801.3
901.3
1001.3
1101.3
401.3
501.3
601.3
701.3
801.3
901.3
1,001.3
1,101.3
Figure 4-24: Predicted pressure comparison of vent pipe model with Hysys for air in the pressure range 600-1000 KPa gauge
84
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.2
10.8
11.4
12
Vent Pipe Length, m
0.6
1.2
1.8
2.4
3.6
4.2
4.8
6.6
Vent Pipe Lenght, m
5.4
7.2
7.8
8.4
9.6
10.2
10.8
11.4
12
Figure 4-27: Predicted density comparison of vent pipe model with Hysys simulation for air in the pressure range 600-1000 KPa gauge
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
0.0
1.2
0.1
0.0
0.6
0.2
0.1
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Figure 4-26: Predicted mach no. comparison of vent pipe model with Hysys simulation for air in the pressure range 600-1000 KPa gauge
85

0.6
1.2
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.2
10.8
11.4
12
Vent Pipe Lenght, m
100
150
200
250
300
50
350
50
100
150
200
250
300
250
6.10
255
260
265
270
275
280
285
290
295
6.20
6.30
6.40
Entropy, KJ/kg-K
6.50
6.60
6.70
Figure 4-29: Predicted Enthalpy-Entropy (Fanno curve) of vent pipe model for air in the pressure range 600-1000 KPa gauge
Enthalpy, KJ/kg
6.80
Figure 4-28: Predicted velocity comparison of vent pipe model with Hysys simulation for air in the pressure range 600-1000 KPa gauge.
350
86
4.6.2 Comparison with Hysys Simulation for Carbon-dioxide

The results from vent pipe model predictions and Aspen Hysys simulations for carbondioxide gas are explained here. The vent pipe Hysys flow sheet did not converge to complete
atmospheric pressure after 600 K Pa gauge inlet pressure. Suspected reason for this could be
the limitations with Hysys. Hence it was decided to increase the back pressure in order to
solve the Hysys flow sheet. The following two cases are evaluated here: CO2 Case 1Pressure range of 100-500 K Pa gauge (atmospheric venting); CO2 Case 2-Pressure range of
600-1000 K Pa gauge (back pressure).
4.6.2.1
CO2 Case 1: Pressure range 100-500 KPa gauge
The predicted and simulated results for mass flow rates at steady-state conditions in the
pressure range 100-500 KPa gauge are in close agreement. The comparison percentage
difference calculated in table G 11-3 on mass flow rates, predicted by the vent model, are at
minimal. The minimum comparison percentage difference calculated for mass flow rate was
-0.16% and maximum percentage difference was calculated at -0.92%. Figure 4-30 shows
the pressure profile for the vent model predictions and Aspen Hysys simulated results for
carbon-dioxide at steady-state mass flow conditions in the pressure range 100-500 K Pa
gauge. The percentage comparison difference at the inlet and exit of the vent pipe was 0%
whereas at the exit of vent was 0.03%. This could be due to minor calculation discrepancy.
Overall an excellent agreement in the mass flow rate and exit pressure prevails. The pressure
profile along the vent pipe for carbon-dioxide was analyzed in a similar manner as analyzed
for air. The predicted and simulated pressure results for carbon-dioxide at the entry & exit of
each section of vent pipe were compared. It was found that the pressure profile developed
was very similar to that developed in case of air. An initial decreasing linearity followed by
an exponential fall to attain exit pressure condition (atmospheric) was established in the vent
pipe. The predicted pressure readings along the vent length compares well and are within
0.6% of the simulated Hysys values for the first ten sections. High percent differences in
the pressure drop are seen in the last two sections of the vent pipe. The reason for existence
of such a pressure profile in the last two sections is unclear at this stage.
Figure 4-31 shows the temperature profile for the vent model predictions and Aspen Hysys
simulated results for carbon-dioxide at steady-state mass flow conditions in the pressure
range 100-500 KPa gauge. A similar temperature profile pattern as seen with air was
recognized in case of carbon-dioxide.
87
0.6
1.2
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
Vent Pipe Lenght, m
10.2
10.8
11.4
12
0.6
3.6
4.2
4.8
2.4
1.8
1.2
6.6
Vent Pipe Length, m
7.2
5.4
8.4
7.8
9.6
10.8
11.4
10.2
Figure 4-31: Predicted temperature comparison of vent pipe model with Hysys simulation for CO2 in pressure range 100-500 KPa gauge
20
18
16
14
12
10
8
6
4
2
0
-2 0
-4
-6
-8
-10
20
18
16
14
12
10
8
6
4
2
0
12 -2
-4
-6
-8
-10
101.3
101.3
301.3
401.3
501.3
601.3
201.3
201.3
301.3
401.3
501.3
601.3
Figure 4-30: Predicted pressure comparison of vent pipe model with Hysys simulation for CO2 in pressure range 100-500 KPa gauge
88
Vent Pipe Model Predictions: Temperature

C
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.2
10.8
11.4
12
Vent Pipe Length, m
0.6
1.2
1.8
2.4
3.6
4.2
4.8
6.6
Vent Pipe Lenght, m
5.4
7.2
7.8
8.4
9.6
10.2
10.8
11.4
Figure 4-33: Predicted density comparison of vent pipe model with Hysys simulation for CO2 in pressure range 100-500 KPa gauge
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
12
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
0.0
1.2
0.1
0.0
0.6
0.2
0.1
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Figure 4-32: Predicted Mach no. comparison of vent pipe model with Hysys simulation for CO2 in pressure range 100-500 KPa gauge
89
Vent Pipe Model Predictions: Mach

Number
0.6
1.2
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.2
10.8
11.4
12
Vent Pipe Lenght, m
2.40
2.45
2.50
2.55
Entropy, KJ/kg-K
2.60
2.65
2.70
Figure 4-35: Predicted Enthalpy-Entropy (Fanno curve) of vent pipe model for CO2 in the pressure range 100-500 KPa gauge
475
2.35
480
485
490
495
500
505
2.75
100
150
Figure 4-34: Predicted velocity comparison of vent pipe model with Hysys simulation for CO2 in pressure range 100-500 KPa gauge
Enthalpy, KJ/kg
200
50
50
100
150
200
90
However, the temperature percentage differences were found to be of slight variant at the
exit of the vent pipe. The calculated temperature differences increased with the increase in
pressure. The model predicted temperatures in the first ten sections of the vent pipe match
with Hysys simulations and are in close agreement. The percentage temperature differences
in the 11th section of the pipe have been decreased when compared to the temperature profile
of air. This decrease in temperature in the 11th section could have been compensated into the
exit temperature, thus, increasing our final comparison percentage difference. A temperature
percentage difference of -30.24% at 400 K Pa gauge and 24.65% at 500 K Pa gauge was
calculated. Percentage differences below 400 K Pa gauge are in close agreement.
Mach number, density and velocity profiles along the vent pipe were assessed and plotted.
Figure 4-32 shows the Mach number profile for the vent model predictions and Aspen Hysys
range 100-500 KPa gauge. Based on the assessment performed, it can be said that the Mach
number along the vent profile is increasing towards the exit of the vent pipe and approaching
towards sonic velocity. The exit Mach number readings in all cases are in close agreement
with a minimum difference of -0.97% and a maximum difference of -2.41%.
Figure 4-33 shows the density profile for the vent model predictions and Aspen Hysys
range 100-500 KPa gauge. A fall in density of fluid is noticed along the vent profile. The
graph is very similar to pressure Figure 4-30 and temperature Figure 4-31 which shows
decreasing linearity and an exponential fall. The exit density readings in all cases are in close
agreement with a minimum difference of -0.32% and a maximum difference of -1.24%.
Figure 4-34 shows the velocity profile for the vent model predictions and Aspen Hysys
range 100-500 KPa gauge. The minimum and maximum comparison difference calculated
for the exit velocity was 0.15% and 0.35%. The velocity profile pattern developed is a
horizontal mirror image of temperature profile and very similar to Mach number profile.
Overall the results were found to be in very close agreement for all parameters in case 1.
Figure 4-35 shows the enthalpy-entropy diagram representing the fanno curve for the vent
pipe model. The friction in the pipe results in a decrease in enthalpy with a simultaneous
increase in entropy towards the exit of the vent pipe is indicated, thus, defining the flow as
subsonic.
91
4.6.2.2
CO2 Case 2: Pressure range 600-1000 KPa gauge
In the case of air, simulations performed in Aspen Hysys at pressures > 600 K Pa gauge did
not converge to achieve atmospheric pressure at the exit of the vent pipe. An identical
situation was seen when simulating the vent pipe for carbon-dioxide gas in Hysys. In order
to simulate the Hysys model, the back pressure (pressure at the exit of the vent) was
increased by a relative amount such that the vent exit pressure equals to the back pressure.
This was performed by adjusting the steady-state mass flow condition. Trial and error
methods were performed in order to solve the Hysys model at minimal back pressure (above
atmosphere). The exit pressures obtained from the Hysys simulations were inputted in the
vent pipe model and relevant predictions were calculated. The results obtained were
tabulated in table G 11-4 and pressure profile, temperature profile, Mach number profiles,
density profile and velocity profile along the vent pipe were plotted.
The vent model predictions and Aspen Hysys simulated results for mass flow in the pressure
range 600-1000 K Pa gauge are found to be in close agreement with a minimum and
maximum difference of -1.13% and -2.02%. Figure 4-36 shows the pressure profile for the
vent model predictions and Aspen Hysys simulated results for carbon-dioxide at steady-state
mass flow conditions in the pressure range 600-1000 K Pa gauge. The predicted pressures at
the inlet and exit conditions were found to be in close agreement with minimal difference.
This minimal difference could be because of minor calculation error. A difference of 0.44%
was calculated on analyzing the pressure profile for the first ten sections along the vent
length. The predicted and simulated temperature profile for the first ten sections agreed
closely and is represented in figure 4-37. The predicted and simulated Mach number, density
and velocity profiles along the vent pipe were plotted in figure 4-38, figure 4-39 and figure
4-40. Similar results to CO2 case-1 were obtained. Enthalpy-entropy curve were plotted in
figure 4-41 which defined the flow in the vent to be subsonic.
92
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
Vent Pipe Lenght, m
10.2
10.8
11.4
12
6.6
8.4
9.6
10.8
11.4
10.2
Figure 4-37: Predicted temperature comparison of vent pipe model with Hysys simulation for CO2 in pressure range 600-1000 KPa gauge
7.8
-8
-4
-20
7.2
-16
Vent Pipe Length, m
5.4
-20
4.8
-12
4.2
-16
3.6
-12
-8
0
2.4
0
1.8
1.2
0.6
12
12
16
16
-4
20
20
12
101.3
1.8
201.3
101.3
1.2
301.3
201.3
0.6
401.3
301.3
501.3
601.3
701.3
801.3
901.3
1001.3
1101.3
401.3
501.3
601.3
701.3
801.3
901.3
1,001.3
1,101.3
Figure 4-36: Predicted pressure comparison of vent pipe model with Hysys simulation for CO2 in pressure range 600-1000 KPa gauge
93
1.2
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.2
10.8
11.4
12
Vent Pipe Length, m
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.8
11.4
Figure 4-39: Predicted density comparison of vent pipe model with Hysys simulation for CO2 in pressure range 600-1000 KPa gauge
Vent Pipe Lenght, m
10.2
12
0
2.4
0
1.8
1.2
0.6
10
12
14
16
18
20
10
12
14
16
18
20
22
22
0.0
0.6
0.1
0.0
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Figure 4-38: Predicted Mach no. comparison of vent pipe model with Hysys simulation for CO2 in pressure range 600-1000 KPa gauge
94
0.6
1.2
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.2
10.8
11.4
12
Vent Pipe Lenght, m
475
2.20
480
485
490
495
500
2.25
2.30
2.35
2.40
Entropy, KJ/kg-K
2.45
2.50
2.55
2.60
2.65
100
150
50
200
50
100
150
Figure 4-41: Predicted Enthalpy-Entropy (Fanno curve) of vent pipe model for CO2 in the pressure range 600-1000 KPa gauge
Enthalpy, KJ/kg
Figure 4-40: Predicted velocity comparison of vent pipe model with Hysys simulation for CO2 in pressure range 600-1000 KPa gauge
200
95
4.6.3 Comparison with Hysys Simulation for Methane

The vent pipe model predictions for hydrocarbon gas such as methane were also evaluated.
The results were compared to the simulated results from Aspen Hysys. Previous results for
air and carbon-dioxide gases were evaluated on case by case basis due to convergence issue
present in Hysys simulation. A similar situation was encountered here. The following cases
were evaluated: CH4 Case-1: Pressure ranges 100-500 K Pa gauge and CH4 case-2: Pressure
range 600-1000 K Pa gauge.
4.6.3.1
CH4 Case 1: Pressure range 100-500 KPa gauge
The predicted and simulated results for methane mass flow rates in the pressure range 100500 KPa gauge are found to be in close agreement, thus maintaining steady-state conditions.
The mass flow rate comparison percentage differences calculated in table G 11-5 are at
minimal. The minimum comparison percentage difference calculated for mass flow rate was
-0.02% and maximum percentage difference was calculated to be 0.3%. Figure 4-42 shows
the pressure profile for the vent model predictions and Aspen Hysys simulated results for
methane at steady-state mass flow conditions in the pressure range 100-500 K Pa gauge. On
comparing the exit pressure values from the vent model predictions, a minimal percentage
difference of 0.03% is calculated. This could be due to minor calculation discrepancy. The
predicted pressure values in the first ten sections of the vent pipe compares well with the
simulated results and are within 0.64%. The pressure drop in the 11th section of the vent
pipe is high (10.92%) in case of Hysys resulting in a high percentage difference. The reason
for existence of such a pressure profile in the last two sections is unclear at this stage.
The model predictions and Hysys simulated temperatures along the vent pipe were plotted
and percent comparison differences were calculated. Figure 4-43 shows the temperature
profile for the vent model predictions and Aspen Hysys simulated results for methane at
steady-state mass flow conditions in the pressure range 100-500 KPa gauge. The predicted
exit temperature comparison percentage difference for methane is less when compared to
carbon-dioxide. A minimum difference of -0.16% and a maximum difference of -9.14%
were calculated for the exit temperature. The graph follows a linear decrease with an
exponential fall pattern to attain the final exit temperature is seen. The model predicted
temperatures in the first ten sections of the vent pipe match with the Hysys simulated
temperature readings and are within 1.15% better than air and carbon-dioxide.
96
1.2
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
Vent Pipe Lenght, m
10.2
10.8
11.4
12
101.3
0.6
101.3
201.3
201.3
401.3
501.3
301.3
601.3
301.3
401.3
501.3
Aspen HYSYS Simulation: Temperature

C

4.2
3.6
4.8
6.6
Vent Pipe Length, m
5.4
7.2
8.4
7.8
9.6
10.8
11.4
10.2
Figure 4-43: Predicted temperature comparison of vent pipe with Hysys simulation for methane in pressure range 100-500 KPa gauge
-10
-6
-10
-6
12 -2
2
2.4
1.8
10
10
1.2
14
14
0.6
18
18
-2 0
22
22
Figure 4-42: Predicted pressure comparison of vent pipe with Hysys simulation for methane in pressure range 100-500 KPa gauge
601.3
97
Vent Pipe Model Predictions:

Temperature C
1.2
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.2
10.8
11.4
12
Vent Pipe Length, m
0.0
0.6
0.1
0.0
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.6
1.2
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.8
11.4
12
Figure 4-45: Predicted density comparison of vent pipe model with Hysys simulation for methane in pressure range 100-500 KPa gauge
Vent Pipe Lenght, m
10.2
Figure 4-44: Predicted Mach no. comparison of vent pipe model with Hysys simulation for methane in pressure range 100-500 KPa gauge
98

0.6
1.2
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.2
10.8
11.4
12
Vent Pipe Lenght, m
100
150
200
250
300
5.80
6.00
6.20
Entropy, KJ/kg-K
6.40
6.60
Figure 4-47: Predicted Enthalpy-Entropy (Fanno curve) of vent pipe model for methane in pressure range 100-500 KPa gauge
840
5.60
850
860
870
880
890
900
6.80
Figure 4-46: Predicted velocity comparison of vent pipe model with Hysys simulation for methane in pressure range 100-500 KPa gauge
Enthalpy, KJ/kg
350
50
50
100
150
200
250
300
350
99
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
Vent Pipe Lenght, m
10.2
10.8
11.4
12
101.3
1.8
201.3
101.3
1.2
301.3
201.3
0.6
401.3
301.3
501.3
601.3
701.3
801.3
901.3
1001.3
1101.3
401.3
501.3
601.3
701.3
801.3
901.3
1,001.3
6.6
8.4
9.6
10.8
11.4
10.2
-8

-20
7.8
-16
7.2
-20
Vent Pipe Length, m
5.4
-12
4.8
-16
-8
4.2
-12
3.6
0
12 -4
0
2.4
4
1.8
1.2
12
12
0.6
16
16
-4 0
20
20
100
Figure 4-49: Predicted temperature comparison of vent pipe model with Hysys simulation for methane in pressure range 600-1000 KPa gauge

C
Figure 4-48: Predicted pressure comparison of vent pipe model with Hysys simulation for methane in pressure range 600-1000 KPa gauge
1,101.3

Temperature C
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.2
10.8
11.4
12
Vent Pipe Length, m
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.8
11.4
12
Figure 4-51: Predicted density comparison of vent pipe model with Hysys simulation for methane in pressure range 600-1000 KPa gauge
Vent Pipe Lenght, m
10.2
0
1.8
1.2
0.6
Figure 4-50: Predicted Mach no. comparison of vent pipe model with Hysys simulation for methane in pressure range 600-1000 KPa gauge
1.2
0.0
0.6
0.1
0.0
0.2
0.3
0.4
0.5
0.6
0.7
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0.9
0.8
1.0
1.0
101
1.2
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.2
10.8
11.4
12
Vent Pipe Lenght, m
5.40
5.60
5.80
Entropy, KJ/kg-K
6.00
6.20
6.40
6.60
Figure 4-53: Predicted Enthalpy-Entropy (Fanno curve) of vent pipe model for methane in pressure range 600-1000 KPa gauge
830
5.20
840
850
860
870
880
890
900
102
0.6
50
150
200
250
300
50
Figure 4-52: Predicted velocity comparison of vent pipe model with Hysys simulation for methane in pressure range 600-1000 KPa gauge
Enthalpy, KJ/kg
350
100
100
150
200
250
300
350
A maximum percentage difference of 136.33% is calculated in the 11th section of the vent
pipe. The reason for existence of such a temperature profile is unclear at this stage.
Mach number, density and velocity along the vent pipe were assessed and plotted. Figure 444, Figure 4-45 and Figure 4-46 show the Mach number, density and velocity profiles for the
vent model predictions and Aspen Hysys simulated results for methane at steady-state mass
flow conditions in the pressure range 100-500 KPa gauge. Referring to the relevant graphs, it
can be said that the density along the vent profile is decreasing towards the exit of the vent
pipe and the graph pattern resembles the same as the pressure graph. The Mach number and
velocity profiles attain the exit conditions exponentially preceded by a linear rise. The
predicted results are in close agreement for Mach number, velocity and density profile of the
Hysys simulated results. Enthalpy-Entropy diagram (fanno curve) was plotted to explain the
effects of flow conditions developed in the vent pipe and to define the flow in the vent pipe.
Figure 4-47 represents the fanno curves in the pressure ranges 100-500 K Pa gauge. A
decrease in enthalpy and pressure with a simultaneous increase in entropy defines the flow to
be subsonic for the predicted results.
4.6.3.2
CH4 Case 2: Pressure range 600-1000 KPa gauge
The vent model predictions and Hysys simulated results with comparison differences are
tabulated in table G 11-6. Mass flow predictions agreed well with the Hysys simulated
results for the pressure range 600-1000 K Pa gauge. The minimum and maximum
comparison percentage difference calculated was -0.2% and 0.58% for the mass flow. The
predicted and simulated results of pressure profile for the vent pipe flowing with methane
were plotted and are represented in Figure 4-48. A close agreement in the pressure results is
seen at the exit and in the first ten sections of the vent pipe. The percentage difference in the
11th section is high compared to the other sections of the vent pipe and is ~11.6%. The
pressure profile pattern developed for methane is very similar to air and carbon-dioxide
gases evaluated before. The temperature predictions and simulated results for methane were
evaluated and the pattern developed was very similar to air and carbon-dioxide gas. This is
represented in Figure 4-49. A minimum percentage comparison difference of 4.08% and a
maximum of 6.15% were calculated at the exit of the vent pipe. The temperature profile
developed along the vent pipe was of the same pattern as CH4 case-1 representing a
decreasing linearity followed by an exponential decrease to attain exit conditions. Predicted
and simulated Mach numbers, density and velocity were also plotted in Figure 4-50, Figure
103
4-51 and Figure 4-52 for methane and were found to be in close agreement with each other.
Enthalpy-Entropy diagram (Figure 4-53) characterized the flow to be subsonic. A decrease
in enthalpy and increase in entropy was noticed.
4.6.4 Comparison with Hysys Simulation for DBNGP Mixture:

The evaluations performed for compressible gases such as air, carbon-dioxide and methane
indicate that the vent model predictions and the Aspen Hysys simulated results are in good
agreement for single component gas phase steady-state adiabatic conditions. In order to test
the performance of the vent pipe model flowing with a multi-component gas, an evaluation
was performed on multi-component DBNGP (Kirk-Burnnand 2009) gas mixture. The
evaluations were performed in the pressure range 100-1000 KPa gauge on a similar case by
case basis as performed with air, carbon-dioxide and methane.
4.6.4.1
DBNGPmixture Case 1: Pressure range 100-500 KPa gauge
The vent model predictions and Hysys simulated results with comparison differences are
tabulated in table G 11-7. The mass flow rate predictions are in excellent agreement with
Hysys simulated results. A minimum percentage difference of 0.04% and a maximum
difference of -0.23% were calculated. These percentage differences are under acceptable
limits. The pressure and temperature predictions hold in good agreement too. The pressure
and temperature profile are plotted in figure 4-54 and figure 4-55. The predicted pressures
for the first ten sections along the vent are within 0.6% of the comparison difference
whereas the predicted temperatures are within 1.09% of the comparison difference in the
same sections. The predicted exit conditions for pressure are in good agreement. An increase
in temperature difference is noticed at pressure 400 and 500 KPa gauge. This calculated
difference of is still acceptable. Predicted Mach number, densities and velocities along the
vent pipe were compared with Hysys results and are plotted in Figure 4-56, Figure 4-57 and
Figure 4-58. Except for the 11th section of the vent, the results are in good agreement for the
first ten sections of the vent pipe and at the exit conditions. The percentage difference
increases in section 11 and the reason for this is discussed when performing evaluations with
air. Fanno lines were plotted in Figure 4-59 for the pressure ranging between 100-500 KPa
gauge and flow was characterized to be subsonic. A decrease in enthalpy with a
simultaneous increase in entropy was seen.
104
4.6.4.2
DBNGPmixture Case 2: Pressure range 600-1000 KPa gauge
Hysys simulations performed at pressures > 600 KPa gauge did not converge the flow sheet
with DBNGP gas mixture which resulted in an increase in back pressure. The vent model
predictions and Hysys simulated results with comparison difference are tabulated in table G
11-8. Overall the predicted results were in close agreement with the Hysys simulated results.
The minimum and maximum difference calculated when comparing the mass flow was 0.33% and -0.72%. The pressure and temperature profiles plotted for predicted and Hysys
simulated results in Figure 4-60 and Figure 4-61 resemble very closely to the profile patterns
developed for single component gases such as air, carbon-dioxide and methane. The
percentage differences are within 0.35% for pressure and 1.08% for temperature in the
first ten sections of the vent pipe. The exit pressure predictions are very closely agreeable
with Hysys simulations. However, differences for exit temperatures are slightly higher than
within the profile but are within acceptable ranges. Figure 4-62, Figure 4-63 and Figure 4-64
which represent the Mach number, density and velocity profiles for model predicted and
Hysys simulated results show similar resemblance to air, carbon-dioxide and methane. Flow
was characterized to be subsonic as per the fanno curve plotted in Figure 4-65.
105
1.2
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
Vent Pipe Lenght, m
10.2
10.8
11.4
12
101.3
0.6
101.3
201.3
201.3
401.3
501.3
601.3
301.3
301.3
401.3
501.3
601.3
4.8
4.2
3.6
6.6
7.2
Vent Pipe Length, m
5.4
7.8
9.6
8.4
10.2
11.4
-10
-6
12 -2
10.8
106
Figure 4-55: Predicted temperature comparison of vent pipe model with Hysys simulation for DBNGP in pressure range 100-500 KPa gauge
-10
-6
2
2.4
1.8
10
10
1.2
14
14
0.6
18
18
-2 0
22
22
Figure 4-54: Predicted pressure comparison of vent pipe model with Hysys simulation for DBNGP in pressure range 100-500 KPa gauge

C

Vent Pipe Model Predictions: Temperature
C
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.2
10.8
11.4
12
Vent Pipe Length, m
1.2
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.8
11.4
12
Figure 4-57: Predicted density comparison of vent pipe model with Hysys simulation for DBNGP in pressure range 100-500 KPa gauge
Vent Pipe Lenght, m
10.2
0.6
Figure 4-56: Predicted Mach no. comparison of vent pipe model with Hysys simulation for DBNGP in pressure range 100-500 KPa gauge
1.2
0.0
0.6
0.1
0.0
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Vent Pipe Model Predictions: Mach

Number
107
0.6
1.2
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.2
10.8
11.4
12
Vent Pipe Lenght, m
100
150
200
250
300
5.40
5.50
5.60
5.70
5.80
Entropy, KJ/kg-K
5.90
6.00
6.10
6.20
Figure 4-59: Predicted Enthalpy-Entropy (Fanno curve) of vent pipe model for DBNGP in pressure range 100-500 KPa gauge
770
5.30
780
790
800
810
820
830
6.30
Figure 4-58: Predicted velocity comparison of vent pipe model with Hysys simulation for DBNGP in pressure range 100-500 KPa gauge
Enthalpy, KJ/kg
350
50
50
100
150
200
250
300
350
108
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.8
11.4
12
-8
6.6
7.2
7.8
8.4
9.6
10.2
11.4
10.8
-12
-8
-20
Vent Pipe Length, m
5.4
-16
4.8
-20
4.2
-16
-12
3.6
0
12 -4
0
2.4
4
1.8
1.2
12
12
0.6
16
16
-4 0
20
20
109
Figure 4-61: Predicted temperature comparison of vent pipe model with Hysys simulation for DBNGP in pressure range 600-1000 KPa gauge

C
Figure 4-60: Predicted pressure comparison of vent pipe model with Hysys simulation for DBNGP in pressure range 600-1000 KPa gauge
Vent Pipe Lenght, m
10.2
101.3
1.8
201.3
101.3
1.2
301.3
201.3
0.6
401.3
301.3
501.3
601.3
701.3
801.3
901.3
1001.3
1101.3
401.3

501.3
601.3
701.3
801.3
901.3
1,001.3
1,101.3

Temperature C
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.2
10.8
11.4
12
Vent Pipe Length, m
0.0
1.2
0.1
0.0
0.6
0.2
0.1
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.8
11.4
12
Figure 4-63: Predicted density comparison of vent pipe model with Hysys simulation for DBNGP in pressure range 600-1000 KPa gauge
Vent Pipe Lenght, m
10.2
0
1.8
1.2
0.6
10
10
Figure 4-62: Predicted Mach no. comparison of vent pipe model with Hysys simulation for DBNGP in pressure range 600-1000 KPa gauge
110
1.8
2.4
3.6
4.2
4.8
5.4
6.6
7.2
7.8
8.4
9.6
10.2
10.8
Vent Pipe Lenght, m
760
5.00
770
780
790
800
810
820
830
5.10
5.20
5.30
5.40
5.60
Entropy, KJ/kg-K
5.50
5.70
5.80
5.90
6.00
Figure 4-65: Predicted Enthalpy-Entropy (Fanno curve) of vent pipe model for DBNGP in pressure range 600-1000 KPa gauge
Enthalpy, KJ/kg
11.4
12
6.10
Figure 4-64: Predicted velocity comparison of vent pipe model with Hysys simulation for DBNGP in pressure range 600-1000 KPa gauge
1.2
111
0.6
50
50
150
200
250
300
350
100
100
150
200
250
300
350
5 Chapter 5
Conclusions and Recommendations for Future Work
5.1 Conclusions
A thorough investigation has been conducted into compressible fluid (single-phase gas)
behavior taking place in a vent pipe. The factors affecting the compressible fluid behavior
and their influence on the compressible fluid parameters have been discussed. Friction is
found to be the chief factor bringing about the changes in compressible fluid flow properties.
This has been well explained by Fanno process. Based on the investigations performed and
to satisfy the need of a model for venting through associated vent piping with pressure
vessels / pipelines, a steady-state vent pipe model to predict the compressible fluid flow
conditions during blowdown of pressure vessels / pipelines was developed. A fluid dynamic
and thermodynamic approach was used in developing the model. The vent pipe model is
described best as a model encountering adiabatic frictional flow conditions. The vent pipe
model predicts the flowing gas properties such as pressure, temperature, mass flow /
standard volumetric flow, temperature of the pipe wall at the exit along with stagnation
properties and critical properties. The use of REFPROP, which incorporates the GERG 2004
equation of state, makes the simulation with the vent pipe model highly competent. All
thermophysical properties are determined using REFPROP. The vent pipe model has been
validated by comparing its predictions to experimental analysis and process simulation
software, Aspen Hysys. Overall, it can be stated that the vent pipe models predictions are in
good agreement with experimental and Aspen Hysys results. The vent pipe model contains
no disposable parameters and no adjustments have been made during validation to ensure the
good agreement.
5.1.1 Comparison of Vent Pipe Model Predictions with Experimental Analysis

A test rig was designed and constructed for experimental analysis which incorporated the use
of fast acting pressure, temperature and flowrate instruments. Steady state experiments were
conducted with air in the pressure range of 200 400 KPa gauge and the results have been
discussed. It follows from the three sets of comparisons (VPM-1, VPM-4 and VPM-7)
reported here that the vent pipe model predicts results in very close agreement with the
112
experimental measurements. This agreement permits confidence to be placed in the

predictions made using the vent pipe model. One of the most important parameters when
designing venting systems is the temperature of the vent pipe. This parameter helps in
estimating the minimum temperature that could be attained at the exit of the pipe wall during
venting. The adiabatic wall temperature was predicted based on the recovery factor approach
where the recovery factor was taken to be same as the Prandtl number. Comparison of the
predicted wall temperature using recovery factor approach has provided a very close
agreement. Based on experimental comparison, the standard volumetric flow rates, the
stagnation temperatures and minimum pipe wall temperatures can be predicted using the
vent pipe model with an estimated uncertainty of 2.5 Nm3/hr, 0.15C and 0.6C.
However, there are certain other validatory comparisons which would increase confidence in
the program and are discussed in the recommendation section.
5.1.2 Comparison of Vent Pipe Model Predictions with Aspen Hysys

The vent pipe predictions were compared to Aspen Hysys simulated results for single
component compressible gases such as air, carbon-dioxide and methane and multicomponent gases such as DBNGP mixtures in the pressure range of 100-1000 KPa gauge.
The predicted results were found to be in close agreement for all parameters involved. The
Mach number and velocity profile pattern developed were similar in all cases forming a
horizontal mirror image to the temperature profiles for the flowing gases. Pressure greater
than 600 K Pa gauge were evaluated by increasing the exit gas pressure at the end of the vent
pipe due to convergence issues with Hysys. A high percentage comparison difference was
seen in the 11th section of the vent pipe in all the parameters plotted for all gases. The reason
for this high difference was explained during the evaluation of air and holds for all other
gases evaluated here. Fanno curves were plotted. The flow was characterized to be subsonic
and irreversibility of the process was confirmed. The vent pipe model predictions compares
well with Hysys simulations with very small percentage differences at the exit. Based on
Hysys comparison results, the two important parameters - the minimum temperature of the
flowing gas at the exit and the maximum mass flow are predicted using the vent pipe model
with an estimated uncertainty of at most 0.25C with air, 1.65C with carbon dioxide,
0.40 C with methane. 0.42 C with DBNGP gas mixture for minimum temperature of
flowing gas and 0.15 kg/hr with air, 2.5 kg/hr with carbon dioxide, 0.42 with methane,
0.56 kg/hr with DBNGP gas mixture for maximum mass flux. Moreover, the vent pipe
model predictions are calculated based on the GERG-2004 equation of state which is proved
113
to better than AGA8-DC92, Peng-Robinson and other equations of state (Kunz et al.
2007)(Kunz et al. 2007). Thus, it can be said that the developed vent pipe model can be
successfully employed for predicting the single phase steady-state adiabatic vent pipe
performance for single and multi-component gas mixtures.
Overall, a very close agreement exists between the predictions of the vent pipe model and
experimental / Aspen Hysys process simulations. Based on these results we can conclude
that the vent pipe model can be used in designing the vent piping systems associated with
pressure vessels / pipelines. The vent pipe model can become much more robust when
certain gaps in the experimental validation, in particular for higher pressure venting
conditions, are filled.
5.2 Recommendations
There are certain other validatory comparisons which would increase the robustness of the
vent pipe model and can be undertaken as a future scope of work.
Some of the
recommendations are as follows:

x
The existing experimental evidence was performed in the pressure range from 200
400 KPa gauge due to issues related to laboratory compressor air supply. There is a
need for further experiments to be performed at pressures higher than 400 KPa
gauge. The existing experimental evidence was performed with air gas only. Due to
constraints and other restrictions in fluid flow laboratory hydrocarbon or other
supercritical gases could not be vented. There is a need for further experiments with
hydrocarbon gases and supercritical gases.
The vent pipe model is developed based on an adiabatic approach and will be
employed mostly to short pipes. However, as discussed in this thesis, the actual
behavior of the gas lies somewhere between the isothermal and adiabatic conditions.
Hence, the development of an isothermal model with heat transfer will be an added
advantage for accurate prediction of real gases.
The vent pipe model is developed in visual basic in conjunction with Microsoft
Excel spreadsheet. The convergence of results is delayed due to time taken by
processor for performing calculations. This problem can be efficiently solved by
scripting the program in FORTRAN language. The FORTRAN language is designed
114
for scientific usage and also has excellent logical capabilities. Also, FORTRAN is
used heavily by experienced process engineers.
x
The vent pipe model has been developed for single phase gases. A new model for
multi-component multiphase gases can be developed in conjunction with the current
single phase model.
115
Appendices
6 Appendix A
6.1 Pressure Testing Certificate
116
117
7 Appendix B
7.1 Test Rig Representation and Mechanical Drawings
118
Figure B 7-2: Photographs of test rig
124
8 Appendix C
8.1 Commissioning and Testing Report
125
126
127
128
129
130
9 Appendix D
9.1 Visual Basic Program for Noise Reduction
131
9.2 General Program for Eliminating Standard Deviation in

Pressure, Flowrate and Temperature
Public Function NewmAVAR(mAVAR, ExpectedValue, N) As Double
Dim SQmAVAR, SQstddev, stddev, stddevAvg As Double
'Computing square of the mean (measured value - mA current)
SQmAVAR = WorksheetFunction.Power(mAVAR, 2)
SQstddev = SQmAVAR - WorksheetFunction.Power(ExpectedValue, 2)
'Computing Standard deviation as the noise
If SQstddev < 0 Then
SQstddev = SQstddev * -1
'Calculating standard deviation of a single sample
stddev = WorksheetFunction.Power(SQstddev, 0.5)
'Calculating the average standard deviation over a N samples
stddevAvg = stddev / WorksheetFunction.Power(N, 0.5)
Else
stddev = WorksheetFunction.Power(SQstddev, 0.5)
stddevAvg = stddev / WorksheetFunction.Power(N, 0.5)
End If
If mAVAR < 4 Then
NewmAVAR = mAVAR + stddevAvg
Else
NewmAVAR = mAVAR - stddevAvg
End If
End Function
132
Public Sub NewVAR()

Dim ws, ws1 As String
Dim rangemAVAR1, rangeClear As Range
Dim AP1, mAP1, count As Variant
Dim missing As Variant
Sheet2.Activate
ws = Worksheets.Application.ActiveSheet.Name
ws1 = "VAR to Current conversion"
Set
rangemAVAR1
=
Range("StartmAVAR1").End(xlDown))
Worksheets(ws).Range(Range("StartmAVAR1"),
Set rangeClear = Worksheets(ws).Range("startp1clear:endp1clear")

rangeClear.ClearContents
missing = ""
count = 0
Application.StatusBar = "Converting Start-up Amp to mA"
For Each AP1 In rangemAVAR1
If AP1.Offset(0, -8).Value > 0 Then
AP1.Value = AP1.Offset(0, -8).Value * 1000
AP1.Offset(0, 1).Value = AP1.Offset(0, -7).Value * 1000
AP1.Offset(0, 2).Value = AP1.Offset(0, -6).Value * 1000
Else
AP1.Value = missing
AP1.Offset(0, 1).Value = missing
AP1.Offset(0, 2).Value = missing
End If
Next AP1
133
Application.StatusBar = "Amp to mA conversion complete"

Application.StatusBar = "Progressing with Standard Deviation Calculations"
'Performing check if values present
For Each mAP1 In rangemAVAR1
If mAP1.Offset(0, -8).Value = missing Then
Application.StatusBar = "Zero value encountered @" & count
Exit For
Else
'VAR 1
mAP1.Offset(0,3).Value=NewmAVAR(mAP1.Value,Range("ExpectedValue").Value,
Range("NumberofSamples").Value)
mAP1.Offset(0,4).Value=NewmAVAR(mAP1.Offset(0,1).Value,Range("ExpectedValue").
Value, Range("NumberofSamples").Value)
End If
count = count + 1
Application.StatusBar = "Reducing the mA data point to low Standard Deviation. Currently
at " & count
Next mAP1
Application.StatusBar = "All mA data points reduced to low Standard Deviation"
End Sub
134
9.3 General Program for Performing Moving Average on Pressure,

Flowrate and Temperature measurements
Public Sub CalculateMovingAverage()
Application.Calculation = xlCalculationManual
Dim ws As String
Dim MAve, DInterval, a, b, c, d, e, f, count As Variant
Dim rangeMAveCurrent1, rangeMAveCurrent2, rangeMAveCurrent3, rangeMAverage1,
rangeMAverage2, rangeMAverage3 As Range
ws = "VAR Noise Reduction"
DInterval = Worksheets(ws).Range("Interval").Value
Set
rangeMAveCurrent1=Worksheets(ws).Range("StartMovingAverage:EndMovingAverage")
Set
rangeMAveCurrent2=Worksheets(ws).Range("StartMovingAverage:EndMovingAverage").
Offset(0, 1)
Set
rangeMAveCurrent3
=
Worksheets(ws).Range("StartMovingAverage:EndMovingAverage").Offset(0, 2)
rangeMAveCurrent1.ClearContents
count = 0
For Each MAve In rangeMAveCurrent1
If MAve.Offset(DInterval - 1, -7).Value > 0 Then
'current 1
a = MAve.Offset(0, -4).Address
b = MAve.Offset(DInterval - 1, -4).Address
'current 2
c = MAve.Offset(0, -3).Address
d = MAve.Offset(DInterval - 1, -3).Address
'current 3
e = MAve.Offset(0, -2).Address
f = MAve.Offset(DInterval - 1, -2).Address
135
Set rangeMAverage1 = Worksheets(ws).Range(a, Worksheets(ws).Range(b))

Set rangeMAverage2 = Worksheets(ws).Range(c, Worksheets(ws).Range(d))
Set rangeMAverage3 = Worksheets(ws).Range(e, Worksheets(ws).Range(f))
'Starting from the top
'MAve.Value = Application.WorksheetFunction.Average(rangeMAverage1)
'Starting after the interval point
MAve.Offset(DInterval-1,0).Value
Application.WorksheetFunction.Average(rangeMAverage1)
MAve.Offset(DInterval
1,
1).Value
MAve.Offset(DInterval
1,
2).Value
=
=
=
count = count + 1
Application.StatusBar = "Performing Noise Reduction. Currently at " & count
Else
Application.StatusBar = "VAR Noise Reduction Calculation Complete"
Application.Calculation = xlCalculationAutomatic
Exit Sub
End If
Next MAve
Application.StatusBar = "VAR Noise Reduction Calculation Complete"
Application.Calculation = xlCalculationAutomatic
MsgBox "VAR Noise Reduction Calculation Complete"
End Sub
136
10 Appendix E
10.1 Adiabatic frictional flow derivation
137
Consider a steady one dimensional flow of a real gas with constant specific heats across a
control surface, as shown in the figure below,
Control volume
V+dV
Gas flow
Shear stress
Above figure adapted from (Saad 1993)

Momentum equation is expressed as
Ap A( p dp ) WZ.P.dx
UAv (v dv v )
Ap Ap Adp WZ.P.dx
UAvdv
Adp Ww.P.dx UAvdv
Now, the friction factor is related to the shear stress in the flow direction in the following
way:
WZ
1 2
Uv
2
Where f : Friction factor; W Z : Shear stress; P : Wetted perimeter
The wetted perimeter of the duct P in terms of hydraulic diameter is given as:
4A
DH
Where DH = hydraulic diameter
For circular ducts DH=D, diameter of circular duct
? Adp
? dp
fUv 2 4 A
UAv 2 dv 2
dx 2
.
D
2
2
v
Uv 2 dv 2
4 f Uv 2
dx
Du2
2 v2
0 1
138
Other equations necessary for the solution

Real Gas Equation
ZURT
Continuity Equation
UvA Constant
Energy Equation:
'H
v2 2
Mach number
M2
v 2 zJRT
Second law of thermodynamics
ds t 0
Dividing equation 1 by p
dp 4 fUv 2
Uv 2 dv 2
dx

p
D.2 p
2 p v2
Now,
Uv 2
v2
u zJRT
zJRT
UM 2 u zJRT
JM 2 u URT
JM 2 P
Therefore, above equation becomes
dp 4 fJM 2
JM 2 dv 2
dx

p
2D
2 v2
From equation 4
v2
2
'H
?H
v2
2
Constant
139
? dh
2vdv
2
? dh vdv
? dh
vdv
dv 2
2
? Cp.dT
Dividing above equation by Cp.T
dT
T
Cp
Now,
dv 2
2CpT
dz
JR
(TREQ
z REQ )
J 1
dT
The term (TREQ
dz
dT
M2
&
v2
zJRT
) is small and hence can be neglected.

REQ
dv 2
JR
zT
2
J 1
dT
T
dT
T
2
J 1 dv

2 zJRT
dT
T
2
J 1 dv

2
2 v
M2
dT
T
2
J 1 2 dv

M
v2
2
From equation 5, we have
M2
v2
zJRT
? Taking "n on both sides
140
? "nM 2
"nv 2 "nT "nz
dM 2
M2
dv 2 dT dz

T
z
v2
dv 2 dM 2 dz

z
M2
v2
dT
T
10
From equations 8 & 9, we get
dv 2 dM 2

M2
v2
J 1 M 2 dv 2
2
v2
dv 2
v2
J 1 2
1 2 M
dM 2
M2
dv 2
v2
1
J 1 2
1
M
2
dM 2
2
M
10
From equation 2 & 3
dp
p
&
dU
dU
U

dT dz

T
z
dv
v
dp
p
dv dT dz

v
T
z
dp
p
dv 2
v2
1 J 1 2
2 2 M
11
Substitute equation 11 in equation 7
dv 2
v2
?
2
JM 2 dv 2
1 J 1 2 4 fJM
2 2 M 2 D dx 2 v 2
dv 2
v2
1 J 1 2 JM 2 4 fJM 2
M
dx

2
2
2D
2
141
dv 2
? 2 1 M 2
v
4 fJM
dx
D
2
4f
dx
D
4f
dx
D
4 fJM 2
dx
D
12
1
dM
1 M 2
2
J 1 2 M
1 2 M
2
1
1 M 2 dM
J 1 2 2
M2

M
M
1
J
2 1 M 2
dM
J 1 2 M
JM 2 1
M
2
13
From equation 10
dv 2
v2
dv
v
1
dM 2
J 1 2 M 2
1
M
2
dM
1
J 1 2 M
1
M
2
14
Now,
dv
v
dU
dM
1
J 1 2 M
1
M
2
15
From equation 11
dp
p
dv 2
v2
1 J 1 2
2 2 M
2dv 1 J 1 2
M

v 2
2
142
dp
?
p
>
dv 2
1 J 1M 2
2
v
dp
p
dM
1
1 J 1M 2

1
J
M
2

1 2 M
dp
p
1 J 1M 2 dM
J 1 2 M
1 2 M
>
16
From equation 8
dT
T
2
J 1 2 dv

M
v2
2
J 1 2 2dv

M
v
2
J 1M 2 dM

J 1 2 M
1 2 M
dT
T
J 1M
dM

J 1 2
1 2 M
17
Properties of a fluid at any section of a vent pipe may be related to properties at any other
section. Equation 13 represents the changes in Mach number with displacement along the
vent pipe. By integrating equation 13 within the limits M=M1 to M=M2 and x=0 to x=L
(Maximum vent pipe length at which Mach number is unity)
4f
dx
D
M2
2
M JM 2
1
1 M
2
J 1 2
1
M
2
dM
M
143
M2
1
M JM 4
1
1 M
2
J 1 2
1
M
2
dM 2
The solution of the above equation can be obtained by method of partial fractions which
results in
4 fL
J 1 2
1
M 2
1 1
1 J 1 2

"n

J M 12 M 22 2J J 1 2
M 1
1
18
Friction is the chief parameter which causes the properties of any flow, whether subsonic or
supersonic to approach these Mach unity characteristics. Hence, M1=M; M2 =1and L=L*
4 fL *
2
1 M J 1
(J 1) M
ln
JM 2 2J J 1 2

21
M
19
144
11 Appendix F
11.1 Vent Pipe Model Simulations Results for Air, Carbon Dioxide,
Methane and DBNGP Gas Mixture
145
Figure F 11-1: Vent pipe model predictions for air gas at 100 KPa gauge
146
147
148
149
150
151
152
Figure F 11-8: Vent pipe model predictions for air gas at sonic conditions
153
Figure F 11-9: Vent pipe model predictions for methane gas at 100 KPa gauge
154
155
156
157
158
159
160
Figure F 11-16: Vent pipe model predictions for methane gas at sonic conditions
161
Figure F 11-17: Vent pipe model predictions for carbon-dioxide gas at 100 KPa gauge
162
163
164
165
166
167
168
Figure F 11-24: Vent pipe model predictions for carbon-dioxide gas at sonic conditions
169
Figure F 11-25: Vent pipe model predictions for DBNGP gas mixture at 100 KPa gauge
170
171
172
173
174
175
176
Figure F 11-32: Vent pipe model predictions for DBNGP gas mixture at sonic conditions
177
12 Appendix G
12.1 Vent Pipe Model Comparison with Hysys Simulation
178
Table G 12-1: Comparison of vent pipe model predictions with Hysys simulation in pressure range 100 - 500 KPa gauge for air
179
Table G 12-2: Comparison of vent pipe model predictions with Hysys simulation in pressure range 600 - 1000 KPa gauge for air
180
Table G 12-3: Comparison of vent pipe model predictions with Hysys simulation in pressure range 100 - 500 KPa gauge for carbon-dioxide
181
182
Table G 12-4: Comparison of vent pipe model predictions with Hysys simulation in pressure range 600 - 1000 KPa gauge for carbon-dioxide
Table G 12-5: Comparison of vent pipe model predictions with Hysys simulation in pressure range 100 - 500 KPa gauge for methane
183
Table G 12-6: Comparison of vent pipe model predictions with Hysys simulation in pressure range 600 - 1000 KPa gauge for methane
184
mixture
185
Table G 12-7: Comparison of vent pipe model predictions with Hysys simulation in pressure range 100 - 500 KPa gauge for DBNGP gas
mixture
186
Table G 12-8: Comparison of vent pipe model predictions with Hysys simulation in pressure range 600 - 1000 KPa gauge for DBNGP gas
13 Appendix H
13.1 Vent Pipe Model Program
187
Figure H 13-1: Vent pipe model user specification sheet
188
189
Figure H 13-2: Newtons Iteration method for nonlinear equations
190
Interpolation of Property Relation

Static Function XYinterpolate(xyarray As Variant, x, y As Single) As Single
Dim n1, M1 As Integer
Dim x1, x2, y1, y2, Ry1x1, Ry1x2, Ry1x1x2, Ry2x1, Ry2x2, Ry2x1x2 As Single
x1 = Application.HLookup(x, xyarray, 1)
n1 = Application.Match(x1, xyarray.Rows(1), 0)
x2 = xyarray.Cells(1, n1 + 1).Value
y1 = Application.VLookup(y, xyarray, 1)
M1 = Application.Match(y1, xyarray.Columns(1), 0)
y2 = xyarray.Cells(M1 + 1, 1).Value
Ry1x1 = xyarray.Cells(M1, n1)
Ry1x2 = xyarray.Cells(M1, n1 + 1)
Ry1x1x2 = (x - x1) / (x2 - x1) * (Ry1x2 - Ry1x1) + Ry1x1
Ry2x1 = xyarray.Cells(M1 + 1, n1)
Ry2x2 = xyarray.Cells(M1 + 1, n1 + 1)
Ry2x1x2 = (x - x1) / (x2 - x1) * (Ry2x2 - Ry2x1) + Ry2x1
XYinterpolate = (y - y1) / (y2 - y1) * (Ry2x1x2 - Ry1x1x2) + Ry1x1x2
End Function
Static Function Log10(x)
Log10 = Log(x) / Log(10)
End Function
Serghides approximation to Colebrook-White Equation for calculating friction factor:

Function MoodyFrictFactor(Nre, rel_rough)
If Nre > 2000 Then
a = -2 * Log10(rel_rough / 3.7 + 12 / Nre)
b = -2 * Log10(rel_rough / 3.7 + 2.51 * a / Nre)
c = -2 * Log10(rel_rough / 3.7 + 2.51 * b / Nre)
MoodyFrictFactor = (a - (b - a) ^ 2 / (c - 2 * b + a)) ^ -2
Else
MoodyFrictFactor = 64 / Nre
End If
End Function
191
Initial estimate of friction factor using Von-Karman equation

Function VonKarmanFrictFactor(rel_rough)
If rel_rough > 0 Then
a = 2 * Log10(rel_rough / 3.7)
b = -1 * a
c=1/b
VonKarmanFrictFactor = c ^ 2
Else
VonKarmanFrictFactor = "Pipe is smooth"
End If
End Function
Non-Linear equations for predicting T2 and M1

Function equation1(press1, press2, temp1, temp2, Compressibility, Gamma, constA, Mach1)
As Double
''Main Equation for T2
equation1 = (temp1 / temp2) - ((1 + (constA * Mach1 * Mach1 * ((press1 * press1) /
(press2 * press2)) * (temp2 / temp1))) / (1 + (constA * Mach1 * Mach1)))
End Function
Function equation2(press1, press2, temp1, temp2, Compressibility, Gamma, constA, Mach1,

fricf, Lenght, Diameter) As Double
'Main equation for M1
equation2 = (((((2 * constA * Compressibility) - (1 / Compressibility) - (3 *
constA)) * ((Application.ln((press1 * press1) / (press2 * press2))) +
(Application.ln(temp2 / temp1)) + (Application.ln((1 + (constA * Mach1 * Mach1))
/ (1 + (constA * ((press1 * press1) / (press2 * press2)) * (temp2 / temp1) * Mach1 *
Mach1)))))) + ((1 / (Mach1 * Mach1)) * (1 - (1 / (((press1 * press1) / (press2 *
press2)) * (temp2 / temp1)))))) * (Compressibility / Gamma)) - ((fricf * Lenght) /
Diameter)
End Function
192

As Double
'derivative of eqn 1 with respect to T2 which is treated as X1 in our case
equation3 = (-1) * ((temp1 / (temp2 * temp2)) + ((a * Mach1 * Mach1 * ((press1 * press1) /
(press2 * press2)) * (1 / temp1)) / (1 + (constA * Mach1 * Mach1))))
End Function

As Double
'derivative of eqn 1 with respect to mach1 which is treated as X2 in our case
equation4 = (-1) * ((((1 + (constA * Mach1 * Mach1)) * (constA) * ((press1 * press1) /
(press2 * press2)) * (temp2 / temp1) * (2 * Mach1)) - ((1 + (constA * Mach1 * Mach1 *
((press1 * press1) / (press2 * press2)) * (temp2 / temp1))) * (2 * constA * Mach1))) / ((1 +
(constA * Mach1 * Mach1)) * (1 + (constA * Mach1 * Mach1))))
End Function

As Double
'derivative of eqn 2 with respect to T2 which is treated as X1 in our case
equation5 = ((((2 * constA * Compressibility) - (1 / Compressibility) - (3 * constA)) * ((1 /
temp2) - (1 / ((1 / (constA * ((press1 * press1) / (press2 * press2)) * (Mach1 * Mach1) * (1 /
temp1))) + (temp2))))) + (1 / ((Mach1 * Mach1) * ((press1 * press1) / (press2 * press2)) *
(temp2 * temp2) * (1 / temp1)))) * (Compressibility / Gamma)
End Function

As Double
'derivative of eqn 2 with respect to mach1 which is treated as X2 in our case
equation6 = (((2 * constA * Mach1 * ((2 * constA * Compressibility) - (1 / Compressibility)
- (3 * constA))) * ((1 / (1 + (constA * Mach1 * Mach1))) - (1 / ((1 / ((press1 * press1 *
temp2) / (press2 * press2 * temp1))) + (constA * Mach1 * Mach1))))) - ((2 / (Mach1 *
Mach1 * Mach1)) * (1 - (1 / ((press1 * press1 * temp2) / (press2 * press2 * temp1)))))) *
(Compressibility / Gamma)
End Function
193
Calculations and Experimental flow comparisons

Option Explicit
Public Sub OutletPressureGoalSeek()
Dim ws As String
ws = "Controlling sheet"
Worksheets(ws).Range("AR10").GoalSeek _
Goal:=Range("outletpressureKPaG").Value + 101.325, _
ChangingCell:=Worksheets(ws).Range("Error")
End Sub
Public Sub experimentflowcal()
Dim ws, ws1 As String
Dim rangeIP, rangeOP, rangeMF, rangeIT, rangeOT As Range
Dim x, StartInletPressure, StartOutletPressure, bob As Variant
ws = "Controlling sheet"
ws1 = "Experiment Flow Calculation"
Set rangeIP = Worksheets(ws1).Range("StartIP:EndIP")
Set rangeOP = Worksheets(ws1).Range("StartOP:EndOP")
Set rangeMF = Worksheets(ws1).Range("StartMF:EndMF")
Set rangeIT = Worksheets(ws1).Range("StartIT:EndIT")
Set rangeOT = Worksheets(ws1).Range("StartOT:EndOT")
'StartInletPressure = Worksheets(ws).Range("InletPressureKPaG")
'StartOutletPressure = Worksheets(ws).Range("OutletPressureKPaG")
For Each x In rangeIP
If x.Value > 150 Then
If x.Value < x.Offset(0, 1).Value Then
x.Value = ""
x.Offset(0, 1).Value = ""
Else
Worksheets(ws).Range("InletPressureKPaG") = x.Value
Worksheets(ws).Range("OutletPressureKPaG") = x.Offset(0, 1).Value
Worksheets(ws).Range("InletTemperatureDegC") = x.Offset(0, 3).Value
'bob = x.Offset(0, 1).Value + 101.325
'Worksheets(ws).Range("AK2").GoalSeek , _
' Goal:=bob, _
' ChangingCell:=Worksheets(ws).Range("Y4")
Call OutletPressureGoalSeek
Call SnapShotOfControllingSheet
194
End If
End If
Next x
MsgBox "all done"
End Sub
195
Functions for Vent Pipe Model Program

Function Mass_Flux(Mach1, Density, Compressibility, Gamma, R, Temperature) As Double
Mass_Flux = Mach1 * Density * Application.WorksheetFunction.Power(Compressibility
* Gamma * R * Temperature, 0.5)
End Function
Function Mass_Flow(MassFlux, Diameter) As Double
Dim Area As Double
Area = (Application.WorksheetFunction.pi() / 4) *
Application.WorksheetFunction.Power(Diameter, 2)
Mass_Flow = MassFlux * Area
End Function
Function Velocity(MassFlux, Density)
Velocity = MassFlux / Density
End Function
Function Mach2(Mach1, Pressure1, Pressure2, Temperature1, Temperature2)
Mach2 = Mach1 * (Pressure1 / Pressure2) * Application.WorksheetFunction. Power
((Temperature2 / Temperature1), 0.5)
End Function
Function Reynolds(Diameter, Velocity, Density, Viscosity)
Reynolds = (Diameter * Velocity * Density) / Viscosity
End Function
Function Area(Diameter)
Area = (Application.WorksheetFunction.pi() / 4) *
Application.WorksheetFunction.Power(Diameter, 2)
End Function
196
Newtons Iteration Method

Public Sub Newton_Iteration_Method()
Dim ws As String
Dim a, b, c, d, e, f, Ro, A1, B1, C1, D1, G1, Gi, mu1, P1, P2, T1, Z1, A_Constant,
Pipe_Rel_Roughness, P_Length, P_Diameter As Variant
Dim T2i, M1i, M2i, mi, Vi, Rei, fi, Iterations As Variant
Dim i, j As Variant
On Error Resume Next
ws = "Mass Flow Rate"
P1 = Worksheets(ws).Range("Inlet_Pressure").Value
P2 = Worksheets(ws).Range("Outlet_Pressure").Value
T1 = Worksheets(ws).Range("Inlet_Temperature").Value
Z1 = Worksheets(ws).Range("Inlet_Compressibility").Value
G1 = Worksheets(ws).Range("Inlet_specific_heat_ratio").Value
D1 = Worksheets(ws).Range("Inlet_Density").Value
mu1 = Worksheets(ws).Range("Inlet_Viscosity").Value
Ro = Worksheets(ws).Range("Gas_Constant").Value
A_Constant = Worksheets(ws).Range("Inlet_A").Value
Pipe_Rel_Roughness
=
Worksheets(ws).Range("Pipe_Roughness").Value
Worksheets(ws).Range("Pipe_Diameter").Value
P_Length = Worksheets(ws).Range("Pipe_Length").Value
P_Diameter = Worksheets(ws).Range("Pipe_Diameter").Value
'Initial estimates
T2i = Worksheets(ws).Range("Inlet_Temperature").Value
M1i = 0.01
fi = VonKarmanFrictFactor(Pipe_Rel_Roughness)
Iterations = 10
For i = 1 To Iterations
For j = 1 To 20
a = equation3(P1, P2, T1, T2i, Z1, G1, A_Constant, M1i)
c = equation4(P1, P2, T1, T2i, Z1, G1, A_Constant, M1i)
b = equation5(P1, P2, T1, T2i, Z1, G1, A_Constant, M1i)
d = equation6(P1, P2, T1, T2i, Z1, G1, A_Constant, M1i)
e = equation1(P1, P2, T1, T2i, Z1, G1, A_Constant, M1i)
f = equation2(P1, P2, T1, T2i, Z1, G1, A_Constant, M1i, fi, P_Length, P_Diameter)
A1 = d * ((a * d) - (b * c))
B1 = b / ((b * c) - (a * d))
C1 = c / ((b * c) - (a * d))
D1 = a / ((a * d) - (b * c))
T2i = T2i - ((A1 * e) + (B1 * f))
M1i = M1i - ((B1 * e) + (D1 * f))
197
j=j+1
Next j
'New friction factor using moody
Gi = Mass_Flux(M1i, D1, Z1, G1, Ro, T2i)
mi = Mass_Flow(Gi, P_Diameter)
Vi = Velocity(Gi, D1)
Rei = Reynolds(P_Diameter, Vi, D1, mu1)
fi = MoodyFrictFactor(Rei, Pipe_Rel_Roughness)
Next i
M2i = Mach2(M1i, P1, P2, T1, T2i)
MsgBox M2i
End Sub
Snap-Shot Module
Public Sub SnapShotOfControllingSheet()
Dim ws, ws1, cell As String
ws = "Snap Shot"
ws1 = "Controlling sheet"
cell = Worksheets(ws).Range("A8").Address
'cell = "SnapShotDateTime"
On Error Resume Next
Err.Clear
Worksheets(ws).Range(cell).End(xlDown).Offset(1, 0).Value = Date
Worksheets(ws).Range(cell).End(xlDown).Offset(0, 0).NumberFormat = "dd-mmm' yy"
Worksheets(ws).Range(cell).End(xlDown).Offset(1, 0).NumberFormat = "dd-mmm' yy"
Worksheets(ws).Range(cell).End(xlDown).Offset(0, 1).Value = Range("PipeScheduleNo")
Worksheets(ws).Range(cell).End(xlDown).Offset(1, 0).NumberFormat = "general"
Worksheets(ws).Range(cell).End(xlDown).Offset(0, 2).Value = Range("PipeSizeInch")
Worksheets(ws).Range(cell).End(xlDown).Offset(0, 2).NumberFormat = "0.000"
Worksheets(ws).Range(cell).End(xlDown).Offset(0, 3).Value = Range("PipeLenghtm")
Worksheets(ws).Range(cell).End(xlDown).Offset(0,4).Value
Range("SurfaceRoughness").Value
Range("InletPressureKPaG").Value
Range("OutletPressureKPaG").Value
198

Range("PressureDrop").Value
Range("InletTemperatureDegC").Value
Range("OutletTemperatureDegC").Value
Worksheets(ws).Range(cell).End(xlDown).Offset(0, 10).Value = Range("InletMach").Value

Range("OutletMach").Value
Range("PipeMaxMassFlowrate").Value
Range("PipeMaxNormalFlowrate").Value
'MsgBox "Values entered on Snap Shot Spreadsheet"

End Sub
199
Program for determining thermodynamic properties from REFPROP (Lemmon,

Huber, and McLinden 2009). Formatted accordingly for vent pipe model by Farhan
Rajiwate
Option Explicit
Private Const FluidsDirectory As String = "fluids\"
Private Const MixturesDirectory As String = "mixtures\"
Private Const MaxComps As Integer = 20
Private Declare Sub SETUPdll Lib "REFPROP.DLL" (i As Long, ByVal hfld As String,
ByVal hfmix As String, ByVal hrf As String, ierr As Long, ByVal herr As String, ln1 As
Long, ln2 As Long, ln3 As Long, ln4 As Long)
Private Declare Sub SETREFdll Lib "REFPROP.DLL" (ByVal hrf As String, ixflag As
Long, x0 As Double, h0 As Double, s0 As Double, t0 As Double, p0 As Double, ierr As
Long, ByVal herr As String, ln1 As Long, ln2 As Long)
Private Declare Sub SETMIXdll Lib "REFPROP.DLL" (ByVal hmxnme As String, ByVal
hfmix As String, ByVal hrf As String, ncc As Long, ByVal hfile As String, x As Double,
ierr As Long, ByVal herr As String, ln1 As Long, ln2 As Long, ln3 As Long, ln4 As Long,
ln5 As Long)
Private Declare Sub SETMODdll Lib "REFPROP.DLL" (i As Long, ByVal htype As String,
ByVal hmix As String, ByVal hcomp As String, ierr As Long, ByVal herr As String, ln1 As
Long, ln2 As Long, ln3 As Long, ln4 As Long)
Private Declare Sub GERG04dll Lib "REFPROP.DLL" (nc As Long, iflag As Long, ierr As
Long, ByVal herr As String, ln1 As Long)
Private Declare Sub TPRHOdll Lib "REFPROP.DLL" (t As Double, p As Double, x As
Double, j As Long, i As Long, d As Double, ierr As Long, ByVal herr As String, ln As
Long)
Private Declare Sub THERM2dll Lib "REFPROP.DLL" (t As Double, d As Double, x As
Double, p As Double, e As Double, h As Double, s As Double, cv As Double, cp As Double,
w As Double, Z As Double, hjt As Double, aH As Double, G As Double, kappa As Double,
beta As Double, dPdD As Double, d2PdD2 As Double, dPdT As Double, dDdT As Double,
dDdP As Double, spare1 As Double, spare2 As Double, spare3 As Double, spare4 As
Double)
200

Double, kappa As Double, beta As Double, isenk As Double, kt As Double, betas As
Double, bs As Double, kkt As Double, thrott As Double, pi As Double, spht As Double)
Private Declare Sub THERMdll Lib "REFPROP.DLL" (t As Double, d As Double, x As
w As Double, hjt As Double)
w As Double, a As Double, G As Double)
Private Declare Sub ENTROdll Lib "REFPROP.DLL" (t As Double, d As Double, x As
Double, s As Double)
Private Declare Sub ENTHALdll Lib "REFPROP.DLL" (t As Double, d As Double, x As
Double, h As Double)
Private Declare Sub CVCPdll Lib "REFPROP.DLL" (t As Double, d As Double, x As
Double, cv As Double, cp As Double)
Private Declare Sub PRESSdll Lib "REFPROP.DLL" (t As Double, d As Double, x As
Double, p As Double)
Private Declare Sub AGdll Lib "REFPROP.DLL" (t As Double, d As Double, x As Double,
a As Double, G As Double)
Private Declare Sub DPDDdll Lib "REFPROP.DLL" (t As Double, rho As Double, x As
Double, dPdD As Double)
Private Declare Sub DPDD2dll Lib "REFPROP.DLL" (t As Double, rho As Double, x As
Double, d2PdD2 As Double)
Private Declare Sub DPDTdll Lib "REFPROP.DLL" (t As Double, rho As Double, x As
Double, dPdT As Double)
Private Declare Sub DDDPdll Lib "REFPROP.DLL" (t As Double, rho As Double, x As
Double, dDdP As Double)
Private Declare Sub DDDTdll Lib "REFPROP.DLL" (t As Double, rho As Double, x As
Double, dDdT As Double)
Private Declare Sub DHD1dll Lib "REFPROP.DLL" (t As Double, rho As Double, x As
Double, dHdT_D As Double, dHdT_P As Double, dHdD_T As Double, dHdD_P As
Double, dHdP_T As Double, dHdP_D As Double)
Private Declare Sub SATTdll Lib "REFPROP.DLL" (t As Double, x As Double, i As Long,
p As Double, Dl As Double, Dv As Double, xliq As Double, xvap As Double, ierr As Long,
ByVal herr As String, ln As Long)
201
Private Declare Sub SATPdll Lib "REFPROP.DLL" (p As Double, x As Double, i As Long,

t As Double, Dl As Double, Dv As Double, xliq As Double, xvap As Double, ierr As Long,
Private Declare Sub SATDdll Lib "REFPROP.DLL" (d As Double, x As Double, kph As
Long, kr As Long, t As Double, p As Double, Dl As Double, Dv As Double, xliq As Double,
xvap As Double, ierr As Long, ByVal herr As String, ln As Long)
Private Declare Sub SATHdll Lib "REFPROP.DLL" (h As Double, x As Double, kph As
Long, nroot As Long, k1 As Long, T1 As Double, P1 As Double, D1 As Double, k2 As
Long, T2 As Double, P2 As Double, d2 As Double, ierr As Long, ByVal herr As String, ln
As Long)
Private Declare Sub SATEdll Lib "REFPROP.DLL" (e As Double, x As Double, kph As
Long, T2 As Double, P2 As Double, d2 As Double, ierr As Long, ByVal herr As String, ln
As Long)
Private Declare Sub SATSdll Lib "REFPROP.DLL" (s As Double, x As Double, kph As
Long, T2 As Double, P2 As Double, d2 As Double, k3 As Long, t3 As Double, p3 As
Double, d3 As Double, ierr As Long, ByVal herr As String, ln As Long)
Private Declare Sub CV2PKdll Lib "REFPROP.DLL" (icomp As Long, t As Double, rho As
Double, cv2p As Double, csat As Double, ierr As Long, ByVal herr As String, ln As Long)
Private Declare Sub CSATKdll Lib "REFPROP.DLL" (icomp As Long, t As Double, kph
As Long, p As Double, rho As Double, csat As Double, ierr As Long, ByVal herr As String,
ln As Long)
Private Declare Sub DPTSATKdll Lib "REFPROP.DLL" (icomp As Long, t As Double, kph
As Long, p As Double, rho As Double, csat As Double, dpdtsat As Double, ierr As Long,
Private Declare Sub TPFLSHdll Lib "REFPROP.DLL" (t As Double, p As Double, x As
Double, d As Double, Dl As Double, Dv As Double, xliq As Double, xvap As Double, q As
Double, e As Double, h As Double, s As Double, cv As Double, cp As Double, w As
Double, ierr As Long, ByVal herr As String, ln As Long)
Private Declare Sub TDFLSHdll Lib "REFPROP.DLL" (t As Double, d As Double, x As
Double, p As Double, Dl As Double, Dv As Double, xliq As Double, xvap As Double, q As
Private Declare Sub PDFLSHdll Lib "REFPROP.DLL" (p As Double, d As Double, x As
Double, t As Double, Dl As Double, Dv As Double, xliq As Double, xvap As Double, q As
202
Private Declare Sub PHFLSHdll Lib "REFPROP.DLL" (p As Double, h As Double, x As

Double, t As Double, d As Double, Dl As Double, Dv As Double, xliq As Double, xvap As
Double, q As Double, e As Double, s As Double, cv As Double, cp As Double, w As
Private Declare Sub PSFLSHdll Lib "REFPROP.DLL" (p As Double, s As Double, x As
Double, q As Double, e As Double, h As Double, cv As Double, cp As Double, w As
Private Declare Sub PEFLSHdll Lib "REFPROP.DLL" (p As Double, e As Double, x As
Double, q As Double, h As Double, s As Double, cv As Double, cp As Double, w As
Private Declare Sub THFLSHdll Lib "REFPROP.DLL" (t As Double, h As Double, x As
Double, i As Long, p As Double, d As Double, Dl As Double, Dv As Double, xliq As
Double, xvap As Double, q As Double, e As Double, s As Double, cv As Double, cp As
Double, w As Double, ierr As Long, ByVal herr As String, ln As Long)
Private Declare Sub TSFLSHdll Lib "REFPROP.DLL" (t As Double, s As Double, x As
Double, xvap As Double, q As Double, e As Double, h As Double, cv As Double, cp As
Private Declare Sub TEFLSHdll Lib "REFPROP.DLL" (t As Double, e As Double, x As
Double, xvap As Double, q As Double, h As Double, s As Double, cv As Double, cp As
Private Declare Sub DHFLSHdll Lib "REFPROP.DLL" (d As Double, h As Double, x As
Double, t As Double, p As Double, Dl As Double, Dv As Double, xliq As Double, xvap As
Double, q As Double, e As Double, s As Double, cv As Double, cp As Double, w As
Private Declare Sub DSFLSHdll Lib "REFPROP.DLL" (d As Double, s As Double, x As
Double, q As Double, e As Double, h As Double, cv As Double, cp As Double, w As
Private Declare Sub DEFLSHdll Lib "REFPROP.DLL" (d As Double, e As Double, x As
Double, q As Double, h As Double, s As Double, cv As Double, cp As Double, w As
Private Declare Sub HSFLSHdll Lib "REFPROP.DLL" (h As Double, s As Double, Z As
Double, t As Double, p As Double, d As Double, Dl As Double, Dv As Double, xliq As
203
Double, xvap As Double, q As Double, e As Double, cv As Double, cp As Double, w As

Private Declare Sub ESFLSHdll Lib "REFPROP.DLL" (e As Double, s As Double, Z As
Double, t As Double, p As Double, d As Double, Dl As Double, Dv As Double, xliq As
Double, xvap As Double, q As Double, h As Double, cv As Double, cp As Double, w As
Private Declare Sub CCRITdll Lib "REFPROP.DLL" (t As Double, p As Double, V As
Double, x As Double, cs As Double, ts As Double, Ds As Double, ps As Double, ws As
Private Declare Sub FPVdll Lib "REFPROP.DLL" (t As Double, d As Double, p As Double,
x As Double, f As Double)
'private Declare Sub SPECGRdll Lib "REFPROP.DLL" (t As Double, d As Double, p As
Double, Gr As Double)
Private Declare Sub TQFLSHdll Lib "REFPROP.DLL" (t As Double, q As Double, x As
Double, kq As Long, p As Double, d As Double, Dl As Double, Dv As Double, xliq As
Double, xvap As Double, e As Double, h As Double, s As Double, cv As Double, cp As
Private Declare Sub PQFLSHdll Lib "REFPROP.DLL" (p As Double, q As Double, x As
Double, kq As Long, t As Double, d As Double, Dl As Double, Dv As Double, xliq As
Double, xvap As Double, e As Double, h As Double, s As Double, cv As Double, cp As
Private Declare Sub ABFL1dll Lib "REFPROP.DLL" (a As Double, b As Double, x As
Double, i As Long, ByVal ab As String, dmin As Double, dmax As Double, t As Double, p
As Double, d As Double, ierr As Long, ByVal herr As String, ln1 As Long, ln2 As Long)
Private Declare Sub ABFL2dll Lib "REFPROP.DLL" (a As Double, b As Double, x As
Double, kq As Long, ksat As Long, ByVal ab As String, tbub As Double, tdew As Double,
pbub As Double, pdew As Double, Dlbub As Double, Dvdew As Double, ybub As Double,
xdew As Double, t As Double, p As Double, Dl As Double, Dv As Double, x As Double, y
As Double, q As Double, ierr As Long, ByVal herr As String, ln As Long, ln2 As Long)
Private Declare Sub DBFL2dll Lib "REFPROP.DLL" (d As Double, b As Double, x As
Double, i As Long, ByVal ab As String, t As Double, p As Double, Dl As Double, Dv As
Double, xliq As Double, xvap As Double, q As Double, ierr As Long, ByVal herr As String,
ln As Long, ln2 As Long)
Private Declare Sub CRITPdll Lib "REFPROP.DLL" (x As Double, tc As Double, pc As
Double, dc As Double, ierr As Long, ByVal herr As String, ln As Long)
Private Declare Sub VIRBdll Lib "REFPROP.DLL" (t As Double, x As Double, b As
Double)
204
Private Declare Sub DBDTdll Lib "REFPROP.DLL" (t As Double, x As Double, dbt As

Double)
Private Declare Sub VIRCdll Lib "REFPROP.DLL" (t As Double, x As Double, c As
Double)
Private Declare Sub TRNPRPdll Lib "REFPROP.DLL" (t As Double, d As Double, x As
Double, eta As Double, tcx As Double, ierr As Long, ByVal herr As String, ln As Long)
Private Declare Sub FGCTYdll Lib "REFPROP.DLL" (t As Double, d As Double, x As
Double, f As Double)
Private Declare Sub DIELECdll Lib "REFPROP.DLL" (t As Double, d As Double, x As
Double, de As Double)
Private Declare Sub SURFTdll Lib "REFPROP.DLL" (t As Double, d As Double, x As
Double, sigma As Double, ierr As Long, ByVal herr As String, ln As Long)
Private Declare Sub SURTENdll Lib "REFPROP.DLL" (t As Double, rhol As Double, rhov
As Double, xl As Double, xv As Double, sigma As Double, ierr As Long, ByVal herr As
String, ln As Long)
Private Declare Sub MELTTdll Lib "REFPROP.DLL" (t As Double, x As Double, p As
Private Declare Sub MLTH2Odll Lib "REFPROP.DLL" (t As Double, P1 As Double, P2 As
Double)
Private Declare Sub MELTPdll Lib "REFPROP.DLL" (p As Double, x As Double, t As
Private Declare Sub SUBLTdll Lib "REFPROP.DLL" (t As Double, x As Double, p As
Private Declare Sub SUBLPdll Lib "REFPROP.DLL" (p As Double, x As Double, t As
Private Declare Sub WMOLdll Lib "REFPROP.DLL" (x As Double, wm As Double)
Private Declare Sub XMASSdll Lib "REFPROP.DLL" (xmol As Double, xkg As Double,
wmix As Double)
Private Declare Sub XMOLEdll Lib "REFPROP.DLL" (xkg As Double, xmol As Double,
wmix As Double)
Private Declare Sub QMASSdll Lib "REFPROP.DLL" (qmol As Double, xl As Double, xv
As Double, qkg As Double, xlkg As Double, xvkg As Double, wliq As Double, wvap As
205
Private Declare Sub QMOLEdll Lib "REFPROP.DLL" (qkg As Double, xlkg As Double,
xvkg As Double, qmol As Double, xl As Double, xv As Double, wliq As Double, wvap As
Private Declare Sub INFOdll Lib "REFPROP.DLL" (icomp As Long, wmm As Double, ttrp
As Double, tnbpt As Double, tc As Double, pc As Double, dc As Double, Zc As Double, acf
As Double, dip As Double, Rgas As Double)
Private Declare Sub LIMITXdll Lib "REFPROP.DLL" (ByVal htyp As String, t As Double,
d As Double, p As Double, x As Double, tmin As Double, tmax As Double, dmax As
Double, pmax As Double, ierr As Long, ByVal herr As String, ln1 As Long, ln2 As Long)
Private Declare Sub LIMITKdll Lib "REFPROP.DLL" (ByVal htyp As String, icomp As
Long, t As Double, d As Double, p As Double, tmin As Double, tmax As Double, dmax As
Double, pmax As Double, ierr As Long, ByVal herr As String, ln1 As Long, ln2 As Long)
Private Declare Sub SETKTVdll Lib "REFPROP.DLL" (icomp As Long, jcomp As Long,
ByVal hmodij As String, fij As Double, ByVal hfmix As String, ierr As Long, ByVal herr
As String, ln1 As Long, ln2 As Long, ln3 As Long)
Private Declare Sub GETKTVdll Lib "REFPROP.DLL" (icomp As Long, jcomp As Long,
ByVal hmodij As String, fij As Double, ByVal hfmix As String, ByVal hfij As String,
ByVal hbinp As String, ByVal hmxrul As String, ln1 As Long, ln2 As Long, ln3 As Long,
ln4 As Long, ln5 As Long)
Private Declare Sub GETFIJdll Lib "REFPROP.DLL" (ByVal hmodij As String, fij As
Double, ByVal hfij As String, ByVal hmxrul As String, ln1 As Long, ln2 As Long, ln3 As
Long)
Private Declare Sub PREOSdll Lib "REFPROP.DLL" (i As Long)
Private Declare Sub SETAGAdll Lib "REFPROP.DLL" (ierr As Long, ByVal herr As
String, ln1 As Long)
'Used to call Refprop:
Private herr As String * 255, herr2 As String * 255, hfmix As String * 255, hfmix2 As String
* 255, hrf As String * 3, htyp As String * 3, hmxnme As String * 255
Private hfld As String * 10000
Private nc As Long, phase As Long
Private x(1 To MaxComps) As Double, xliq(1 To MaxComps) As Double, xvap(1 To
MaxComps) As Double, xmm(1 To MaxComps) As Double, xkg(1 To MaxComps) As
Double, xmol(1 To MaxComps) As Double, wmix As Double
Private ierr As Long, ierr2 As Long, kq As Long, kr As Long
206
Private t As Double, p As Double, d As Double, Dl As Double, Dv As Double, q As Double,

wm As Double, tz As Double, pz As Double, dz As Double, dd As Double
Private e As Double, h As Double, s As Double, Cvcalc As Double, Cpcalc As Double, w
As Double
Private tmin As Double, tmax As Double, dmax As Double, pmax As Double
Private tc As Double, pc As Double, dc As Double
Private tbub As Double, tdew As Double, pbub As Double, pdew As Double, Dlbub As
Double, Dvdew As Double, ybub(1 To MaxComps) As Double, xdew(1 To MaxComps) As
Double
Private eta As Double, tcx As Double, sigma As Double, hjt As Double, de As Double
Private wmm As Double, ttrp As Double, tnbpt As Double, Zc As Double, acf As Double,
dip As Double, Rgas As Double
Private tUnits(10) As String, taUnits(10) As String, pUnits(10) As String, dUnits(10) As
String, vUnits(10) As String, hUnits(10) As String, sUnits(10) As String, wUnits(10) As
String, visUnits(10) As String, tcxUnits(10) As String, stUnits(10) As String
Private tUnits2 As String, taUnits2 As String, pUnits2 As String, dUnits2 As String, vUnits2
As String, hUnits2 As String, sUnits2 As String, wUnits2 As String, visUnits2 As String,
tcxUnits2 As String, stUnits2 As String
Private FldOld As String
Private Z As Double, aHelm As Double, Gibbs As Double, xkappa As Double, beta As
Double
Private dPdD As Double, d2PdD2 As Double, dPdT As Double, dDdT As Double, dDdP As
Double
Private spare1 As Double, spare2 As Double, spare3 As Double, spare4 As Double
Private Const CtoK = 273.15
'Exact conversion
Private Const FtoR = 459.67
'Exact conversion
Private Const RtoK = 5 / 9

Private Const HtoS = 3600
Private Const ATMtoMPa = 0.101325
Private Const BARtoMPA = 0.1
Private Const KGFtoN = 98.0665 / 10
Private Const INtoM = 0.0254
Private Const FTtoM = 12 * INtoM
'Exact conversion
'Exact conversion
'Exact conversion
'Exact conversion
'Exact conversion
'Exact conversion
'Exact conversion
207
Private Const LBMtoKG = 0.45359237

Private Const CALtoJ = 4.184
'Exact conversion
'Exact conversion (tc)
'private Const CALtoJ = 4.1868

pure water)
'Exact conversion (IT) (Use this one only with
Private Const MMHGtoMPA = ATMtoMPa / 760
'Exact conversion
Private Const INH2OtoMPA = 0.000249082
Private Const BTUtoKJ = CALtoJ * LBMtoKG * RtoK

Private Const LBFtoN = LBMtoKG * KGFtoN
Private Const IN3toM3 = INtoM * INtoM * INtoM
Private Const FT3toM3 = FTtoM * FTtoM * FTtoM
Private Const GALLONtoM3 = IN3toM3 * 231
Private Const PSIAtoMPA = LBMtoKG / INtoM / INtoM * KGFtoN / 1000000
Private Const FTLBFtoJ = FTtoM * LBFtoN
Private Const HPtoW = 550 * FTLBFtoJ
Private Const BTUtoW = BTUtoKJ * 1000
Private Const LBFTtoNM = LBFtoN / FTtoM
Private CompFlag As Integer
Function Setup(FluidName)
Dim a As String, ab As String, FluidNme As String, FlNme As String
Dim i As Integer, sum As Double, sc As Integer, ncc As Integer, nc2 As Long, mass As
Integer
Dim hRef As Double, sRef As Double, Tref As Double, pref As Double
Dim htype As String * 3, hmix As String * 3, hcomp As String * 60
Dim RPPrefix As String, FluidsPrefix As String, MixturesPrefix As String
Dim xtemp(1 To MaxComps) As Double
ierr = 0
herr = ""
FlNme = FluidName
208
If InStr(FluidName, "error") Then Exit Function

If InStr(FluidName, "Inputs are out of range") Then Exit Function
If FluidName = FldOld Then Exit Function
FldOld = ""
Call CheckName(FluidName)
RPPrefix = Environ("RPPrefix")
If RPPrefix = "" Then
FluidsPrefix = FluidsDirectory
MixturesPrefix = MixturesDirectory
Else
FluidsPrefix = RPPrefix & "\" & FluidsDirectory
MixturesPrefix = RPPrefix & "\" & MixturesDirectory
End If
hrf = "DEF"
hfmix = FluidsPrefix & "hmx.bnc"
On Error GoTo ErrorHandler:
ChDrive (Application.ActiveWorkbook.Path)
ChDir (Application.ActiveWorkbook.Path)
On Error GoTo 0
a = ""
For i = 1 To MaxComps: xtemp(i) = 0: Next
mass = 0
If InStr(UCase(FluidName), ".MIX") Then
'Open MixturesPrefix & FluidName For Input As #1
'Line Input #1, ab
'Line Input #1, ab
'Input #1, nc2
'For i = 1 To nc2
' Line Input #1, ab
209
' a = a & FluidsPrefix & ab & "|"

'Next
'For i = 1 To nc2
' Input #1, xtemp(i)
'Next
'Close 1
'hfld = a
hmxnme = MixturesPrefix & FluidName
Call SETMIXdll(hmxnme, hfmix, hrf, nc2, hfld, xtemp(1), ierr, herr, 255&, 255&, 3&,
10000&, 255&)
ElseIf InStr(FluidName, ",") Or InStr(FluidName, ";") Then
FluidNme = Trim(FluidName)
If InStr(FluidNme, ";") Then sc = 1 Else sc = 0
If UCase(Right(FluidNme, 4)) = "MASS"
Trim(Left(FluidNme, Len(FluidNme) - 4))
Then
mass
1:
FluidNme
nc2 = 0
Do
If sc = 0 Then i = InStr(FluidNme, ",") Else i = InStr(FluidNme, ";")
If i = 0 Then i = Len(FluidNme) + 1
nc2 = nc2 + 1
If nc2 > MaxComps Then ierr = 1: herr = Trim2("Too many components"): Exit
Function
ab = Trim(Left(FluidNme, i - 1))
Call CheckName(ab)
If InStr(LCase(ab), ".fld") = 0 Then ab = ab + ".fld"
a = a & FluidsPrefix & ab & "|"
FluidNme = Mid(FluidNme, i + 1)
If sc = 0 Then i = InStr(FluidNme, ",") Else i = InStr(FluidNme, ";")
xtemp(nc2) = CDbl(Left(FluidNme, i - 1))
210
FluidNme = Trim(Mid(FluidNme, i + 1))

Loop Until FluidNme = ""
sum = 0
For i = 1 To nc2: sum = sum + xtemp(i): Next
If sum <= 0 Then ierr = 1: herr = Trim2("Composition not set"): Exit Function
For i = 1 To nc2: xtemp(i) = xtemp(i) / sum: Next
hfld = a
If nc2 < 1 Then ierr = 1: herr = Trim2("Setup failed"): Exit Function
'To load the GERG-2004 pure fluid equations of state rather than the defaults
'that come with Refprop, call the GERG04dll routine with a 1 as the second input.
'Call GERG04dll(nc2, 1&, ierr, herr, 255&)
Call SETUPdll(nc2, hfld, hfmix, hrf, ierr, herr, 10000&, 255&, 3&, 255&)
ElseIf InStr(FluidName, "/") <> 0 And InStr(FluidName, "(") <> 0 Then
FluidNme = Trim(FluidName)
If UCase(Right(FluidNme, 4)) = "MASS"
Trim(Left(FluidNme, Len(FluidNme) - 4))
Then
mass
1:
FluidNme
nc2 = 0
Do
i = InStr(FluidNme, "/")
If InStr(FluidNme, "(") < i Then i = InStr(FluidNme, "(")
nc2 = nc2 + 1
If nc2 > MaxComps Then ierr = 1: herr = Trim2("Too many components"): Exit
Function
ab = Trim(Left(FluidNme, i - 1))
Call CheckName(ab)
If InStr(LCase(ab), ".fld") = 0 Then ab = ab + ".fld"
a = a & FluidsPrefix & ab & "|"
FluidNme = Trim(Mid(FluidNme, i))
If Left(FluidNme, 1) = "/" Then FluidNme = Trim(Mid(FluidNme, 2))
211
Loop Until Left(FluidNme, 1) = "("

FluidNme = Mid(FluidNme, 2)
If Right(FluidNme, 1) = ")" Then FluidNme = Trim(Left(FluidNme, Len(FluidNme) - 1))
ncc = 0
Do
i = InStr(FluidNme, "/")
ncc = ncc + 1
If ncc > MaxComps Then ierr = 1: herr = Trim2("Too many components"): Exit
Function
xtemp(ncc) = CDbl(Left(FluidNme, i - 1))
FluidNme = Mid(FluidNme, i + 1)
Loop Until FluidNme = ""
sum = 0
For i = 1 To nc2: sum = sum + xtemp(i): Next
If sum <= 0 Then ierr = 1: herr = Trim2("Composition not set"): Exit Function
For i = 1 To nc2: xtemp(i) = xtemp(i) / sum: Next
hfld = a
If nc2 < 1 Then ierr = 1: herr = Trim2("Setup failed"): Exit Function
'To load the GERG-2004 pure fluid equations of state rather than the defaults
'that come with Refprop, call the GERG04dll routine with a 1 as the second input.
'Call GERG04dll(nc2, 1&, ierr, herr, 255&)
Else
nc2 = 1
If InStr(LCase(FluidName), ".fld") = 0 And InStr(LCase(FluidName), ".ppf") = 0 Then
FluidName = FluidName + ".fld"
If InStr(FluidName, "\") Then
hfld = FluidName
Else
212
hfld = FluidsPrefix & FluidName

End If
'...Use call to SETMOD to change the equation of state for any of the
'.....pure components from the default (recommended) values.
'.....This should only be implemented by an experienced user.
'If InStr(LCase(hfld), "argon") <> 0 And nc2 = 1 Then
' hcomp = "FE1": htype = "EOS": hmix = hcomp
' Call SETMODdll(nc2, htype, hmix, hcomp, ierr, herr, 3&, 3&, 60&, 255&)
'End If
End If
If mass Then
For i = 1 To nc2
xkg(i) = xtemp(i)
Next
Call XMOLEdll(xkg(1), xtemp(1), wmix)
End If
If ierr <= 0 Then
nc = nc2
'If setup was successful, load new values of nc and x()
For i = 1 To nc
x(i) = xtemp(i)
Next
Setup = FluidName
FldOld = FlNme
'Use the following line to calculate enthalpies and entropies on a reference state
213
'based on the currently defined mixture, or to change to some other reference state.
'The routine does not have to be called, but doing so will cause calculations
'to be the same as those produced from the graphical interface for mixtures.
Call SETREFdll(hrf, 2&, x(1), hRef, sRef, Tref, pref, ierr, herr, 3&, 255&)
Else
Setup = Trim2(herr)
FldOld = ""
End If
Exit Function
ErrorHandler:
Resume Next
End Function
Sub CheckName(FluidName)
Restart:
If Left(FluidName, 1) = Chr(34) Then
FluidName = Mid(FluidName, 2): GoTo Restart
End If
If Right(FluidName, 1) = Chr(34) Then
FluidName = Left(FluidName, Len(FluidName) - 1): GoTo Restart
End If
If
UCase(FluidName)
=
"nitrogen;7812;argon;0092;oxygen;2096"
"AIR"
Then
FluidName
If UCase(FluidName) = "CARBON DIOXIDE" Then FluidName = "CO2"

If UCase(FluidName) = "CARBON MONOXIDE" Then FluidName = "CO"
If UCase(FluidName) = "CARBONYL SULFIDE" Then FluidName = "COS"
If UCase(FluidName) = "CYCLOHEXANE" Then FluidName = "CYCLOHEX"
If UCase(FluidName) = "CYCLOPROPANE" Then FluidName = "CYCLOPRO"
If UCase(FluidName) = "DEUTERIUM" Then FluidName = "D2"
214
If UCase(FluidName) = "HEAVY WATER" Then FluidName = "D2O"

If UCase(FluidName) = "HYDROGEN SULFIDE" Then FluidName = "H2S"
If UCase(FluidName) = "IBUTANE" Then FluidName = "ISOBUTAN"
If UCase(FluidName) = "ISOBUTANE" Then FluidName = "ISOBUTAN"
If UCase(FluidName) = "ISOPENTANE" Then FluidName = "IPENTANE"
If UCase(FluidName) = "NEOPENTANE" Then FluidName = "NEOPENTN"
If UCase(FluidName) = "ISOHEXANE" Then FluidName = "IHEXANE"
If UCase(FluidName) = "NITROUS OXIDE" Then FluidName = "N2O"
If UCase(FluidName) = "PARAHYDROGEN" Then FluidName = "PARAHYD"
If UCase(FluidName) = "PROPYLENE" Then FluidName = "PROPYLEN"
If UCase(FluidName) = "SULFUR HEXAFLUORIDE" Then FluidName = "SF6"
End Sub
Sub CalcSetup(FluidName, InpCode, Units, Prop1, Prop2)

Call Setup(FluidName)
If ierr > 0 Then Exit Sub
Call ConvertUnits(InpCode, Units, Prop1, Prop2)
herr = ""
q = 0: t = 0: p = 0: d = 0: Dl = 0: Dv = 0: e = 0: h = 0: s = 0: Cvcalc = 0: Cpcalc = 0: w = 0
End Sub
Sub CalcProp(FluidName, InpCode, ByVal Units, ByVal Prop1, ByVal Prop2)

Dim iflag1 As Integer, iflag2 As Integer
ThisWorkbook.Activate
q = 0: t = 0: p = 0: d = 0: Dl = 0: Dv = 0: e = 0: h = 0: s = 0: Cvcalc = 0: Cpcalc = 0: w = 0
If IsMissing(Prop1) Then iflag1 = 1
If iflag1 = 0 Then
If Len(Trim(Prop1)) = 0 Then iflag1 = 2
215
If iflag1 = 0 Then If CDbl(Prop1) = 0 And Prop1 <> "0" Then ierr = 1: herr =
Trim2("Invalid input: ") + Prop1: Exit Sub
End If
If IsMissing(Prop2) Then iflag2 = 1
If iflag2 = 0 Then
If Len(Trim(Prop2)) = 0 Then iflag2 = 2
If iflag2 = 0 Then If CDbl(Prop2) = 0 And Prop2 <> "0" Then ierr = 1: herr =
Trim2("Invalid input: ") + Prop2: Exit Sub
End If
If IsMissing(InpCode) Then InpCode = ""
Call CalcSetup(FluidName, InpCode, Units, Prop1, Prop2)
If UCase(Left(InpCode, 4)) = "CRIT" Then
Call CRITPdll(x(1), t, p, d, ierr, herr, 255&)
If ierr = 0 Then Call THERMdll(t, d, x(1), pc, e, h, s, Cvcalc, Cpcalc, w, hjt)
Exit Sub
ElseIf UCase(Left(InpCode, 4)) = "TRIP" Then
If nc <> 1 Then ierr = 1: herr = Trim2("Can only return triple point for a pure fluid"): Exit
Sub
Call INFOdll(1, wmm, t, tnbpt, tc, pc, dc, Zc, acf, dip, Rgas)
Call SATTdll(t, x(1), 1, p, d, Dv, xliq(1), xvap(1), ierr, herr, 255&)
If ierr = 0 Then Call THERMdll(t, d, x(1), pc, e, h, s, Cvcalc, Cpcalc, w, hjt)
Exit Sub
End If
If iflag1 Then ierr = 1: herr = Trim2("Inputs are missing"): Exit Sub

If ierr > 0 Then Exit Sub
If InpCode <> "" Then Call Calc(InpCode, Prop1, Prop2, iflag1, iflag2)
End Sub
Sub Calc(InputCode, Prop1, Prop2, iflag1, iflag2)
216
Dim a As String, Input1 As String, Input2 As String, InpCode, i As Integer, pp As Double

ierr = 0
herr = ""
InpCode = Trim(UCase(InputCode))
Input2 = ""
Input1 = Left(InpCode, 1)
If Len(InpCode) = 2 Then Input2 = Mid(InpCode, 2, 1)
If Len(InpCode) = 3 And Right(InpCode, 1) = "&" Then Input2 = Mid(InpCode, 2, 1)
If Left(InpCode, 2) = "TP" Or Left(InpCode, 2) = "PT" Then Input2 = Mid(InpCode, 2, 1)
If Input1 = "T" Then t = Prop1: If iflag1 >= 1 Then GoTo Error1

If Input1 = "P" Then p = Prop1: If iflag1 >= 1 Then GoTo Error1
If Input1 = "D" Then d = Prop1: If iflag1 >= 1 Then GoTo Error1
If Input1 = "V" And Prop1 <> 0 And Len(InpCode) = 2 Then d = 1 / Prop1: Mid(InpCode,
1, 1) = "D": If iflag1 >= 1 Then GoTo Error1
If Input1 = "E" Then e = Prop1: If iflag1 >= 1 Then GoTo Error1
If Input1 = "H" Then h = Prop1: If iflag1 >= 1 Then GoTo Error1
If Input1 = "S" Then s = Prop1: If iflag1 >= 1 Then GoTo Error1
If Input1 = "Q" Then q = Prop1: If iflag1 >= 1 Then GoTo Error1
If Input2 = "T" Then t = Prop2: If iflag2 >= 1 Then GoTo Error2
If Input2 = "P" Then p = Prop2: If iflag2 >= 1 Then GoTo Error2
If Input2 = "D" Then d = Prop2: If iflag2 >= 1 Then GoTo Error2
If Input2 = "V" And Prop2 <> 0 And Len(InpCode) = 2 Then d = 1 / Prop2: Mid(InpCode,
2, 1) = "D": If iflag2 >= 1 Then GoTo Error2
If Input2 = "E" Then e = Prop2: If iflag2 >= 1 Then GoTo Error2
If Input2 = "H" Then h = Prop2: If iflag2 >= 1 Then GoTo Error2
If Input2 = "S" Then s = Prop2: If iflag2 >= 1 Then GoTo Error2
If Input2 = "Q" Then q = Prop2: If iflag2 >= 1 Then GoTo Error2
phase = 2
217
If Len(InpCode) > 1 Then If UCase(Mid(InpCode, 2, 1)) = "L" Then phase = 1
For i = 1 To nc
xliq(i) = 0: xvap(i) = 0
Next
If Left(InpCode, 1) = "T" And t <= 0 Then herr = Trim2("Input temperature is zero"): Exit
Sub
'Calculate saturation values given temperature
If InpCode = "TL" Or InpCode = "TLIQ" Or InpCode = "TVAP" Then
Call SATTdll(t, x(1), phase, p, Dl, Dv, xliq(1), xvap(1), ierr, herr, 255&)
If (p = 0 Or Dl = 0) And ierr = 0 Then ierr = 1: herr = Trim2("Inputs are out of range"):
Exit Sub
d = Dl: q = 0
If phase = 2 Then d = Dv: q = 1
Call THERMdll(t, d, x(1), p, e, h, s, Cvcalc, Cpcalc, w, hjt)
'Calculate saturation values given pressure
ElseIf InpCode = "PL" Or InpCode = "PLIQ" Or InpCode = "PVAP" Then
Call SATPdll(p, x(1), phase, t, Dl, Dv, xliq(1), xvap(1), ierr, herr, 255&)
If (p = 0 Or Dl = 0) And ierr = 0 Then ierr = 1: herr = Trim2("Inputs are out of range"):
Exit Sub
d = Dl: q = 0
If phase = 2 Then d = Dv: q = 1
'Calculate saturation values given density
ElseIf InpCode = "DL" Or InpCode = "DLIQ" Or InpCode = "DVAP" Then
Call SATDdll(d, x(1), 1&, kr, t, p, Dl, Dv, xliq(1), xvap(1), ierr, herr, 255&)
q = kr - 1
ElseIf InpCode = "TPL" Or InpCode = "PTL" Then
Call TPRHOdll(t, p, x(1), 1&, 0&, d, ierr, herr, 255&)
218
Dl = d: Dv = d: q = 990
Call THERMdll(t, d, x(1), pp, e, h, s, Cvcalc, Cpcalc, w, hjt)
ElseIf InpCode = "TPV" Or InpCode = "PTV" Then
Dl = d: Dv = d: q = 990
ElseIf InpCode = "TP" Or InpCode = "PT" Then
Call TPFLSHdll(t, p, x(1), d, Dl, Dv, xliq(1), xvap(1), q, e, h, s, Cvcalc, Cpcalc, w, ierr,
herr, 255&)
ElseIf InpCode = "TD" Or InpCode = "DT" Then
Call TDFLSHdll(t, d, x(1), p, Dl, Dv, xliq(1), xvap(1), q, e, h, s, Cvcalc, Cpcalc, w, ierr,
herr, 255&)
ElseIf InpCode = "TD&" Or InpCode = "DT&" Then
'Do not perform any flash calculation here
q = 990
ElseIf InpCode = "TH" Or InpCode = "HT" Then
Call THFLSHdll(t, h, x(1), 2&, p, d, Dl, Dv, xliq(1), xvap(1), q, e, s, Cvcalc, Cpcalc, w,
ierr, herr, 255&)
ElseIf InpCode = "TS" Or InpCode = "ST" Then
Call TSFLSHdll(t, s, x(1), 1&, p, d, Dl, Dv, xliq(1), xvap(1), q, e, h, Cvcalc, Cpcalc, w,
ierr, herr, 255&)
ElseIf InpCode = "TE" Or InpCode = "ET" Then
Call TEFLSHdll(t, e, x(1), 2&, p, d, Dl, Dv, xliq(1), xvap(1), q, h, s, Cvcalc, Cpcalc, w,
ierr, herr, 255&)
ElseIf InpCode = "TQ" Or InpCode = "QT" Then
Call TQFLSHdll(t, q, x(1), 1&, p, d, Dl, Dv, xliq(1), xvap(1), e, h, s, Cvcalc, Cpcalc, w,
ierr, herr, 255&)
ElseIf InpCode = "PD" Or InpCode = "DP" Then
Call PDFLSHdll(p, d, x(1), t, Dl, Dv, xliq(1), xvap(1), q, e, h, s, Cvcalc, Cpcalc, w, ierr,
herr, 255&)
219
ElseIf InpCode = "PH" Or InpCode = "HP" Then

Call PHFLSHdll(p, h, x(1), t, d, Dl, Dv, xliq(1), xvap(1), q, e, s, Cvcalc, Cpcalc, w, ierr,
herr, 255&)
ElseIf InpCode = "PS" Or InpCode = "SP" Then
Call PSFLSHdll(p, s, x(1), t, d, Dl, Dv, xliq(1), xvap(1), q, e, h, Cvcalc, Cpcalc, w, ierr,
herr, 255&)
ElseIf InpCode = "PE" Or InpCode = "EP" Then
Call PEFLSHdll(p, e, x(1), t, d, Dl, Dv, xliq(1), xvap(1), q, h, s, Cvcalc, Cpcalc, w, ierr,
herr, 255&)
ElseIf InpCode = "PQ" Or InpCode = "QP" Then
Call PQFLSHdll(p, q, x(1), 1&, t, d, Dl, Dv, xliq(1), xvap(1), e, h, s, Cvcalc, Cpcalc, w,
ierr, herr, 255&)
ElseIf InpCode = "DH" Or InpCode = "HD" Then
Call DHFLSHdll(d, h, x(1), t, p, Dl, Dv, xliq(1), xvap(1), q, e, s, Cvcalc, Cpcalc, w, ierr,
herr, 255&)
ElseIf InpCode = "DS" Or InpCode = "SD" Then
Call DSFLSHdll(d, s, x(1), t, p, Dl, Dv, xliq(1), xvap(1), q, e, h, Cvcalc, Cpcalc, w, ierr,
herr, 255&)
ElseIf InpCode = "DE" Or InpCode = "ED" Then
Call DEFLSHdll(d, e, x(1), t, p, Dl, Dv, xliq(1), xvap(1), q, h, s, Cvcalc, Cpcalc, w, ierr,
herr, 255&)
ElseIf InpCode = "HS" Or InpCode = "SH" Then
Call HSFLSHdll(h, s, x(1), t, p, d, Dl, Dv, xliq(1), xvap(1), q, e, Cvcalc, Cpcalc, w, ierr,
herr, 255&)
ElseIf InpCode = "TMELT" Then
Call MELTTdll(t, x(1), p, ierr, herr, 255&)
If ierr = 0 Then Call TPFLSHdll(t, p, x(1), d, Dl, Dv, xliq(1), xvap(1), q, e, h, s, Cvcalc,
Cpcalc, w, ierr, herr, 255&)
ElseIf InpCode = "PMELT" Then
If p = 0 Then ierr = 1: herr = Trim2("Input pressure is zero"): Exit Sub
Call MELTPdll(p, x(1), t, ierr, herr, 255&)
220
If ierr = 0 Then Call TPFLSHdll(t, p, x(1), d, Dl, Dv, xliq(1), xvap(1), q, e, h, s, Cvcalc,
Cpcalc, w, ierr, herr, 255&)
ElseIf InpCode = "TSUBL" Then
Call SUBLTdll(t, x(1), p, ierr, herr, 255&)
If ierr = 0 And p = 0 Then ierr = 1: herr = Trim2("No sublimation line available")
If ierr = 0 Then
q=1
d = p / 8.314472 / t
End If
ElseIf InpCode = "PSUBL" Then
If p = 0 Then ierr = 1: herr = Trim2("Input pressure is zero"): Exit Sub
Call SUBLPdll(p, x(1), t, ierr, herr, 255&)
If ierr = 0 And t = 0 Then ierr = 1: herr = Trim2("No sublimation line available")
If ierr = 0 Then
q=1
d = p / 8.314472 / t
End If
Else
ierr = 1: herr = Trim2("Invalid input code")
End If
If (q <= 0.000001 Or q >= 0.999999) And Cvcalc = -9999980 Then Call THERMdll(t, d,
x(1), p, e, h, s, Cvcalc, Cpcalc, w, hjt)
Exit Sub
Error1:
ierr = 1: herr = Trim2("First property missing"): Exit Sub
Error2:
221
ierr = 1: herr = Trim2("Second property missing"): Exit Sub

End Sub
Function Temperature(FluidName, InpCode, Optional Units, Optional Prop1, Optional
Prop2)
Call CalcProp(FluidName, InpCode, Units, Prop1, Prop2)
Temperature = ConvertUnits("-T", Units, t, 0)
End Function
Function Pressure(FluidName, InpCode, Optional Units, Optional Prop1, Optional Prop2)
Pressure = ConvertUnits("-P", Units, p, 0)
End Function
Function Density(FluidName, InpCode, Optional Units, Optional Prop1, Optional Prop2)
Density = ConvertUnits("-D", Units, d, 0)
End Function
Function CompressibilityFactor(FluidName, InpCode, Optional Units, Optional Prop1,
Optional Prop2)
Call INFOdll(1, wmm, ttrp, tnbpt, tc, pc, dc, Zc, acf, dip, Rgas)
CompressibilityFactor = p / d / t / Rgas
End Function
Function LiquidDensity(FluidName, InpCode, Optional Units, Optional Prop1, Optional
Prop2)
If ierr <> 0 Then LiquidDensity = Trim2(herr): Exit Function
If q < 0 Or q > 1 Then
LiquidDensity = Trim2("Inputs are single phase")
Else
CompFlag = 1
LiquidDensity = ConvertUnits("-D", Units, Dl, 0)
222
CompFlag = 0
End If
End Function
Function VaporDensity(FluidName, InpCode, Optional Units, Optional Prop1, Optional
Prop2)
If ierr <> 0 Then VaporDensity = Trim2(herr): Exit Function
VaporDensity = Trim2("Inputs are single phase")
Else
CompFlag = 2
VaporDensity = ConvertUnits("-D", Units, Dv, 0)
CompFlag = 0
End If
End Function
Function Volume(FluidName, InpCode, Optional Units, Optional Prop1, Optional Prop2)
Dim V As Double
Volume = 0
If d <= 0 Then Volume = Trim2("Density is zero"): Exit Function
V=1/d
Volume = ConvertUnits("-V", Units, V, 0)
End Function
Function Energy(FluidName, InpCode, Optional Units, Optional Prop1, Optional Prop2)

Energy = ConvertUnits("-H", Units, e, 0)
End Function
223
Function Enthalpy(FluidName, InpCode, Optional Units, Optional Prop1, Optional Prop2)

Enthalpy = ConvertUnits("-H", Units, h, 0)
End Function
Function Entropy(FluidName, InpCode, Optional Units, Optional Prop1, Optional Prop2)

Entropy = ConvertUnits("-S", Units, s, 0)
End Function
Function IsochoricHeatCapacity(FluidName, InpCode, Optional Units, Optional Prop1,

Optional Prop2)
IsochoricHeatCapacity = ConvertUnits("-S", Units, Cvcalc, 0)
End Function
Function cv(FluidName, InpCode, Optional Units, Optional Prop1, Optional Prop2)

cv = ConvertUnits("-S", Units, Cvcalc, 0)
End Function
Function IsobaricHeatCapacity(FluidName, InpCode, Optional Units, Optional Prop1,

Optional Prop2)
IsobaricHeatCapacity = ConvertUnits("-S", Units, Cpcalc, 0)
End Function
Function cp(FluidName, InpCode, Optional Units, Optional Prop1, Optional Prop2)

cp = ConvertUnits("-S", Units, Cpcalc, 0)
224
End Function
Function SpeedOfSound(FluidName, InpCode, Optional Units, Optional Prop1, Optional

Prop2)
SpeedOfSound = ConvertUnits("-W", Units, w, 0)
End Function
Function Sound(FluidName, InpCode, Optional Units, Optional Prop1, Optional Prop2)

Sound = ConvertUnits("-W", Units, w, 0)
End Function
Function LatentHeat(FluidName, InpCode, Optional Units, Optional Prop1, Optional Prop2)

Dim hl As Double, hv As Double
InpCode = Trim(UCase(InpCode))
If Left(InpCode, 1) = "T" Then
Call CalcSetup(FluidName, "T", Units, Prop1, Prop2)
If ierr <> 0 Then LatentHeat = Trim2(herr): Exit Function
If nc <> 1 Then LatentHeat = Trim2("Can only be calculated for pure fluids"): Exit
Function
t = Prop1
If t <= 0 Then LatentHeat = Trim2("Input temperature is zero"): Exit Function
If t > tc Then LatentHeat = Trim2("Temperature is greater than the critical point
temperture"): Exit Function
Call SATTdll(t, x(1), 1&, p, Dl, Dv, xliq(1), xvap(1), ierr, herr, 255&)
If (p = 0 Or Dl = 0) And ierr = 0 Then ierr = 1: LatentHeat = Trim2("Inputs are out of
range"): Exit Function
ElseIf Left(InpCode, 1) = "P" Then
225
Call CalcSetup(FluidName, "P", Units, Prop1, Prop2)

If nc <> 1 Then LatentHeat = Trim2("Can only be calculated for pure fluids"): Exit
Function
p = Prop1
If p <= 0 Then LatentHeat = Trim2("Input pressure is zero"): Exit Function
If p > pc Then LatentHeat = Trim2("Pressure is greater than the critical point pressure"):
Exit Function
Call SATPdll(p, x(1), 1&, t, Dl, Dv, xliq(1), xvap(1), ierr, herr, 255&)
If (t = 0 Or Dl = 0) And ierr = 0 Then ierr = 1: LatentHeat = Trim2("Inputs are out of
range"): Exit Function
Else
LatentHeat = Trim2("Valid inputs are only 'T' or 'P'"): Exit Function
End If
Call THERMdll(t, Dl, x(1), p, e, hl, s, Cvcalc, Cpcalc, w, hjt)
Call THERMdll(t, Dv, x(1), p, e, hv, s, Cvcalc, Cpcalc, w, hjt)
LatentHeat = ConvertUnits("-H", Units, hv - hl, 0)
End Function
Function HeatOfVaporization(FluidName, InpCode, Optional Units, Optional Prop1,

Optional Prop2)
HeatOfVaporization = LatentHeat(FluidName, InpCode, Units, Prop1, Prop2)
End Function
Function JouleThompson(FluidName, InpCode, Optional Units, Optional Prop1, Optional

Prop2)
JouleThompson = ConvertUnits("-J", Units, hjt, 0)
226
End Function
Function IsentropicExpansionCoef(FluidName, InpCode, Optional Units, Optional Prop1,

Optional Prop2)
Call WMOLdll(x(1), wm)
If d = 0 Then
IsentropicExpansionCoef = w ^ 2 / Rgas / t * wm * 0.001
Else
IsentropicExpansionCoef = w ^ 2 * d / p * wm * 0.001
End If
End Function
Function IsothermalCompressibility(FluidName, InpCode, Optional Units, Optional Prop1,

Optional Prop2)
Call THERM2dll(t, d, x(1), p, e, h, s, Cvcalc, Cpcalc, w, Z, hjt, aHelm, Gibbs, xkappa,
beta, dPdD, d2PdD2, dPdT, dDdT, dDdP, spare1, spare2, spare3, spare4)
IsothermalCompressibility = Trim2("Infinite")
If d > 1E-20 And Not (xkappa = -9999990 Or xkappa > 1E+15) Then
IsothermalCompressibility = 1 / ConvertUnits("-P", Units, 1 / xkappa, 0)
End Function
Function VolumeExpansivity(FluidName, InpCode, Optional Units, Optional Prop1,

Optional Prop2)
VolumeExpansivity = 1 / ConvertUnits("-A", Units, 1 / beta, 0)
227
End Function
Function AdiabaticCompressibility(FluidName, InpCode, Optional Units, Optional Prop1,

Optional Prop2)
AdiabaticCompressibility = Trim2("Infinite")
If d > 1E-20 And w <> 0 Then AdiabaticCompressibility = 1 / ConvertUnits("-P", Units, 1 /
(1 / d / w ^ 2 / wm * 1000), 0)
End Function
Function AdiabaticBulkModulus(FluidName, InpCode, Optional Units, Optional Prop1,

Optional Prop2)
If p = 0 Then
AdiabaticBulkModulus = 0
Else
AdiabaticBulkModulus = ConvertUnits("-P", Units, w ^ 2 * d * wm * 0.001, 0)
End If
End Function
Function IsothermalExpansionCoef(FluidName, InpCode, Optional Units, Optional Prop1,

Optional Prop2)
228

If p = 0 Then
IsothermalExpansionCoef = 1
Else
IsothermalExpansionCoef = d / p * dPdD
End If
End Function
Function IsothermalBulkModulus(FluidName, InpCode, Optional Units, Optional Prop1,

Optional Prop2)
If p = 0 Then
IsothermalBulkModulus = 0
Else
IsothermalBulkModulus = ConvertUnits("-P", Units, d * dPdD, 0)
End If
End Function
Function Quality(FluidName, InpCode, Optional Units, Optional Prop1, Optional Prop2)

If ierr > 0 Then Quality = Trim2(herr): Exit Function
Quality = q
If q = 990 Then Quality = Trim2("Not calculated")
If q = 998 Then Quality = Trim2("Superheated vapor with T>Tc")
If q = 999 Then Quality = Trim2("Supercritical state (T>Tc, p>pc)")
If q = -998 Then Quality = Trim2("Subcooled liquid with p>pc")
229
End Function
Function LiquidMoleFraction(FluidName, Optional InpCode, Optional Units, Optional

Prop1, Optional Prop2, Optional i)
If ierr > 0 Then LiquidMoleFraction = Trim2(herr): Exit Function
If i < 1 Or i > nc Then LiquidMoleFraction = Trim2("Constituent number out of range"):
Exit Function
LiquidMoleFraction = x(i)
Else
LiquidMoleFraction = xliq(i)
End If
If nc = 1 Then LiquidMoleFraction = Trim2("Not applicable for a pure fluid")
End Function
Function VaporMoleFraction(FluidName, Optional InpCode, Optional Units, Optional

Prop1, Optional Prop2, Optional i)
If ierr > 0 Then VaporMoleFraction = Trim2(herr): Exit Function
If i < 1 Or i > nc Then VaporMoleFraction = Trim2("Constituent number out of range"):
Exit Function
VaporMoleFraction = x(i)
Else
VaporMoleFraction = xvap(i)
End If
If nc = 1 Then VaporMoleFraction = Trim2("Not applicable for a pure fluid")

End Function
230
Function Viscosity(FluidName, InpCode, Optional Units, Optional Prop1, Optional Prop2)

If ierr > 0 Then Viscosity = Trim2(herr): Exit Function
If t = 0 Or d = 0 Then Viscosity = Trim2("Inputs out of range"): Exit Function
Call TRNPRPdll(t, d, x(1), eta, tcx, ierr2, herr2, 255&)
If q > 0.000001 And q < 1 - 0.000001 Then eta = -9999999
Viscosity = ConvertUnits("-U", Units, eta, 0)
If eta = 0 Then Viscosity = Trim2("Unable to calculate property")
End Function
Function ThermalConductivity(FluidName, InpCode, Optional Units, Optional Prop1,

Optional Prop2)
If ierr > 0 Then ThermalConductivity = Trim2(herr): Exit Function
If t = 0 Or d = 0 Then ThermalConductivity = Trim2("Inputs out of range"): Exit Function
If q > 0.000001 And q < 1 - 0.000001 Then tcx = -9999999
ThermalConductivity = ConvertUnits("-K", Units, tcx, 0)
If tcx = 0 Then ThermalConductivity = Trim2("Unable to calculate property")
End Function
Function Prandtl(FluidName, InpCode, Optional Units, Optional Prop1, Optional Prop2)

If ierr > 0 Then Prandtl = Trim2(herr): Exit Function
If t = 0 Or d = 0 Then Prandtl = Trim2("Inputs out of range"): Exit Function
If q > 0.000001 And q < 1 - 0.000001 Then Prandtl = Trim2("Undefined"): Exit Function
If tcx = 0 Or eta = 0 Then Prandtl = Trim2("Unable to calculate property")
231

Prandtl = eta * Cpcalc / tcx / wm / 1000
End Function
Function SurfaceTension(FluidName, InpCode, Optional Units, Optional Prop1, Optional

Prop2)
If ierr > 0 Then SurfaceTension = Trim2(herr): Exit Function
If t = 0 Then SurfaceTension = Trim2("Input temperature is zero"): Exit Function
If q >= 0 And q <= 1 Then
Call SURFTdll(t, Dl, xliq(1), sigma, ierr2, herr2, 255&)
Else
Call SURFTdll(t, d, x(1), sigma, ierr2, herr2, 255&)
End If
SurfaceTension = ConvertUnits("-N", Units, sigma, 0)
If sigma = 0 Or ierr2 <> 0 Then SurfaceTension = Trim2("Inputs out of range")
End Function
Function DielectricConstant(FluidName, InpCode, Optional Units, Optional Prop1, Optional

Prop2)
If ierr > 0 Then DielectricConstant = Trim2(herr): Exit Function
If q > 0.000001 And q < 1 - 0.000001 Then DielectricConstant = Trim2("Undefined"): Exit
Function
If t = 0 Then DielectricConstant = Trim2("Inputs out of range"): Exit Function
Call DIELECdll(t, d, x(1), de)
DielectricConstant = de
End Function
232
Function MolarMass(FluidName, Optional InpCode, Optional Units, Optional Prop1,

Optional Prop2)
Call CalcSetup(FluidName, "", Units, Prop1, Prop2)
MolarMass = wm
End Function
Function MoleFraction(FluidName, i)
Call CalcProp(FluidName, "", "", 0, 0)
If ierr > 0 Then MoleFraction = Trim2(herr): Exit Function
If i < 1 Or i > nc Then MoleFraction = Trim2("Constituent number out of range"): Exit
Function
MoleFraction = x(i)
If nc = 1 Then MoleFraction = Trim2("Not applicable for a pure fluid")
End Function
Function MassFraction(FluidName, i)
Call CalcProp(FluidName, "", "", 0, 0)
If ierr > 0 Then MassFraction = Trim2(herr): Exit Function
If i < 1 Or i > nc Then MassFraction = Trim2("Constituent number out of range"): Exit
Function
Call XMASSdll(x(1), xmm(1), wm)
MassFraction = xmm(i)
If nc = 1 Then MassFraction = Trim2("Not applicable for a pure fluid")
End Function
'Change molar composition to mass composition

'Prop1 - Prop20 are the molar values for the components in the mixture.
'i specifies which component's mole fraction is returned. If zero, the molar mass is returned
233
Function Mole2Mass(FluidName, i, Prop1, Prop2, Optional Prop3, Optional Prop4, Optional

Prop5, Optional Prop6, Optional Prop7, Optional Prop8, Optional Prop9, Optional Prop10,
Optional Prop11, Optional Prop12, Optional Prop13, Optional Prop14, Optional Prop15,
Optional Prop16, Optional Prop17, Optional Prop18, Optional Prop19, Optional Prop20)
Dim j As Integer, xkg2(1 To MaxComps) As Double, xmol2(1 To MaxComps) As Double,
wmix2 As Double, sum As Double
For j = 1 To MaxComps: xmol2(j) = 0: Next
xmol2(1) = Prop1
xmol2(2) = Prop2
If Not IsMissing(Prop3) Then xmol2(3) = Prop3
Call CalcSetup(FluidName, "", "", 0, 0)
If ierr > 0 Then Mole2Mass = Trim2(herr): Exit Function
If i < 0 Or i > nc Then Mole2Mass = Trim2("Index out of Range (greater than number of
components in mixture)"): Exit Function
234
sum = 0
For j = 1 To nc
sum = sum + xmol2(j)
Next
If Abs(sum - 1) > 0.0001 Then Mole2Mass = Trim2("Composition does not sum to 1"): Exit
Function
Call XMASSdll(xmol2(1), xkg2(1), wmix2)
If i = 0 Then 'Molar mass of mixture
Mole2Mass = wmix2
Else
'Mass fraction
Mole2Mass = xkg2(i)
End If
End Function
'Change mass composition to molar composition

'Prop1 - Prop20 are the mass values for the components in the mixture.
'i specifies which component's mass fraction is returned. If zero, the molar mass is returned
Function Mass2Mole(FluidName, i, Prop1, Prop2, Optional Prop3, Optional Prop4, Optional
Prop5, Optional Prop6, Optional Prop7, Optional Prop8, Optional Prop9, Optional Prop10,
Optional Prop11, Optional Prop12, Optional Prop13, Optional Prop14, Optional Prop15,
Optional Prop16, Optional Prop17, Optional Prop18, Optional Prop19, Optional Prop20)
Dim j As Integer, xkg2(1 To MaxComps) As Double, xmol2(1 To MaxComps) As Double,
wmix2 As Double, sum As Double
For j = 1 To MaxComps: xkg2(j) = 0: Next
xkg2(1) = Prop1
xkg2(2) = Prop2
If Not IsMissing(Prop3) Then xkg2(3) = Prop3
235

Call CalcSetup(FluidName, "", "", 0, 0)
If ierr > 0 Then Mass2Mole = Trim2(herr): Exit Function
If i < 0 Or i > nc Then Mass2Mole = Trim2("Index out of Range (greater than number of
components in mixture)"): Exit Function
sum = 0
For j = 1 To nc
sum = sum + xkg2(j)
Next
If Abs(sum - 1) > 0.0001 Then Mass2Mole = Trim2("Composition does not sum to 1"): Exit
Function
Call XMOLEdll(xkg2(1), xmol2(1), wmix2)
If i = 0 Then 'Molar mass of mixture
Mass2Mole = wmix2
Else
'Mole fraction
Mass2Mole = xmol2(i)
End If
End Function
236
Function EOSMax(FluidName, Optional InpCode, Optional Units, Optional Prop1, Optional

Prop2)
If nc > 1 Then
Call LIMITXdll("EOS", 300#, 0#, 0#, x(1), tmin, tmax, dmax, pmax, ierr2, herr2, 3&,
255&)
Else
Call LIMITKdll("EOS", 1, 300#, 0#, 0#, tmin, tmax, dmax, pmax, ierr2, herr2, 3&, 255&)
End If
If InpCode = "P" Or InpCode = "p" Then
EOSMax = ConvertUnits("-P", Units, pmax, 0)
ElseIf InpCode = "D" Or InpCode = "d" Then
EOSMax = ConvertUnits("-D", Units, dmax, 0)
Else
EOSMax = ConvertUnits("-T", Units, tmax, 0)
End If
End Function
Function EOSMin(FluidName, Optional InpCode, Optional Units, Optional Prop1, Optional

Prop2)
If nc > 1 Then
Call LIMITXdll("EOS", 300#, 0#, 0#, x(1), tmin, tmax, dmax, pmax, ierr2, herr2, 3&,
255&)
Else
Call LIMITKdll("EOS", 1, 300#, 0#, 0#, tmin, tmax, dmax, pmax, ierr2, herr2, 3&, 255&)
End If
237
If InpCode = "P" Or InpCode = "p" Then

EOSMin = 0
ElseIf InpCode = "D" Or InpCode = "d" Then
EOSMin = 0
Else
EOSMin = ConvertUnits("-T", Units, tmin, 0)
End If
End Function
Function ErrorCode(InputCell)
ErrorCode = ierr
End Function
Function ErrorString(InputCell)
ErrorString = Trim2(herr)
End Function
Function Trim2(a)
'All error messages call this routine to add the pound sign (#) to the beginning of the line.
'If you do not want this error code, simply remove the ["#" +] piece below.
'It can also be changed to any other symbol(s) you desire.
If Left(a, 1) <> "#" Then
Trim2 = "#" + Trim(a)
Else
Trim2 = Trim(a)
End If
End Function
238
Function UnitConvert(InputValue, UnitType As String, OldUnits As String, NewUnits As

String)
'InputValue is the value to be converted from OldUnits to NewUnits

'UnitType is one of the following letters (one character only in most cases):
'UnitType
Unit name
SI units
' T
Temperature
' P
Pressure
' D
Density or specific volume
mol/m^3 or kg/m^3 (or m^3/mol or m^3/kg)
' H
Enthalpy or specific energy
J/mol or J/kg
' S
Entropy or heat capacity
Pa
J/mol-K or J/kg-K
' W
Speed of sound
m/s
' U
Viscosity
' K
Thermal conductivity
' JT
Joule Thompson
' L
Length
' A
Area
m^2
' V
Volume
' M
Mass
' F
Force
' E
Energy
' Q
Power
' N
Surface tension
Pa-s
W/m-K
K/Pa
m^3
kg
N
N/m
' Gage pressures can be used by adding "_g" to the unit, e.g., "MPa_g"
Dim Value As Double, Tpe As String, Unit1 As String, Unit2 As String

Dim Drct As Integer, Gage As Integer, Vacm As Integer
Dim MolWt As Double, Rgas As Double
239
If Not IsNumeric(InputValue) Then UnitConvert = 0: Exit Function

If NewUnits = "" Then UnitConvert = InputValue: Exit Function
Value = InputValue
Tpe = UCase(Trim(UnitType))
Unit1 = UCase(Trim(OldUnits))
Unit2 = UCase(Trim(NewUnits))
Rgas = 8.314472
If CompFlag = 1 Then Call WMOLdll(xliq(1), wm)
If CompFlag = 2 Then Call WMOLdll(xvap(1), wm)
MolWt = wm
For Drct = 1 To -1 Step -2

'----------------------------------------------------------------------' Temperature Conversion
'----------------------------------------------------------------------If Tpe = "T" Then
If Unit1 = "K" Then
ElseIf Unit1 = "C" Then
Value = Value + Drct * CtoK
ElseIf Unit1 = "R" Then
Value = Value * RtoK ^ Drct
ElseIf Unit1 = "F" Then
If Drct = 1 Then
'Value = RtoK * (Value + FtoR)
'Does not give exactly zero at 32 F
Value = (Value - 32) * RtoK + CtoK

Else
'Value = Value / RtoK - FtoR
'Does not give exactly 32 at 273.15 K
240
Value = (Value - CtoK) / RtoK + 32

End If
Else
UnitConvert = Trim2("Undefined input unit"): Exit Function
End If
'----------------------------------------------------------------------' Pressure Conversion

'----------------------------------------------------------------------ElseIf Tpe = "P" Then
Gage = InStr(Unit1, "GAGE")
Vacm = InStr(Unit1, "VACM")
If Gage = 0 Then Gage = InStr(Unit1, "_G")
If Vacm = 0 Then Vacm = InStr(Unit1, "_V")
If Gage <> 0 And Drct = -1 Then Value = Value - ATMtoMPa
If Vacm <> 0 And Drct = -1 Then Value = ATMtoMPa - Value
If Gage <> 0 Then Unit1 = Trim(Left(Unit1, Gage - 1))
If Vacm <> 0 Then Unit1 = Trim(Left(Unit1, Vacm - 1))
If Unit1 = "PA" Then
Value = Value / 1000000 ^ Drct
ElseIf Unit1 = "KPA" Then
ElseIf Unit1 = "MPA" Then
Value = Value
ElseIf Unit1 = "GPA" Then
Value = Value * 1000 ^ Drct
ElseIf Unit1 = "BAR" Then
Value = Value * BARtoMPA ^ Drct
ElseIf Unit1 = "KBAR" Then
241
Value = Value * (BARtoMPA * 1000) ^ Drct

ElseIf Unit1 = "ATM" Then
Value = Value * ATMtoMPa ^ Drct
ElseIf Unit1 = "KGF/CM^2" Or Unit1 = "KG/CM^2" Or Unit1 = "ATA" Or Unit1 =
"AT" Or Unit1 = "ATMA" Then
Value = Value * (KGFtoN / 100) ^ Drct
ElseIf Unit1 = "PSI" Or Unit1 = "PSIA" Then
Value = Value * PSIAtoMPA ^ Drct
ElseIf Unit1 = "PSF" Then
Value = Value * (PSIAtoMPA / 144) ^ Drct
ElseIf Unit1 = "MMHG" Or Unit1 = "TORR" Then
Value = Value * MMHGtoMPA ^ Drct
ElseIf Unit1 = "CMHG" Then
Value = Value * (MMHGtoMPA * 10) ^ Drct
ElseIf Unit1 = "INHG" Then
Value = Value * (MMHGtoMPA * INtoM * 1000) ^ Drct
ElseIf Unit1 = "INH2O" Then
Value = Value * INH2OtoMPA ^ Drct
ElseIf Unit1 = "PSIG" Then
If Drct = 1 Then
Value = PSIAtoMPA * Value + ATMtoMPa
Else
Value = (Value - ATMtoMPa) / PSIAtoMPA
End If
Else
End If
If Gage <> 0 And Drct = 1 Then Value = Value + ATMtoMPa
If Vacm <> 0 And Drct = 1 Then Value = ATMtoMPa - Value
242
'----------------------------------------------------------------------' Density Conversion

'----------------------------------------------------------------------ElseIf Tpe = "D" Then
If Value = 0 Then Value = 1E-50
If Unit1 = "MOL/DM^3" Or Unit1 = "MOL/L" Or Unit1 = "KMOL/M^3" Then
ElseIf Unit1 = "MOL/CM^3" Or Unit1 = "MOL/CC" Then
ElseIf Unit1 = "MOL/M^3" Then
ElseIf Unit1 = "KG/M^3" Then
Value = Value / MolWt ^ Drct
ElseIf Unit1 = "KG/DM^3" Or Unit1 = "KG/L" Then
Value = Value * (1000 / MolWt) ^ Drct
ElseIf Unit1 = "G/DM^3" Or Unit1 = "G/L" Then
ElseIf Unit1 = "G/CC" Or Unit1 = "G/CM^3" Or Unit1 = "G/ML" Then
ElseIf Unit1 = "G/DM^3" Then
ElseIf Unit1 = "LBM/FT^3" Or Unit1 = "LB/FT^3" Then
Value = Value * (LBMtoKG / FT3toM3 / MolWt) ^ Drct
ElseIf Unit1 = "LBMOL/FT^3" Then
Value = Value * (LBMtoKG / FT3toM3) ^ Drct
ElseIf Unit1 = "SLUG/FT^3" Then
Value = Value * (LBMtoKG / FT3toM3 / MolWt * KGFtoN / FTtoM) ^ Drct
ElseIf Unit1 = "LB/GAL" Then
Value = Value * (LBMtoKG / GALLONtoM3 / MolWt) ^ Drct
243
'----------------------------------------------------------------------' Specific Volume Conversion

'----------------------------------------------------------------------ElseIf Unit1 = "DM^3/MOL" Or Unit1 = "L/MOL" Or Unit1 = "M^3/KMOL" Then
Value = 1 / Value
ElseIf Unit1 = "CM^3/MOL" Or Unit1 = "CC/MOL" Or Unit1 = "ML/MOL" Then
Value = 1000 / Value
ElseIf Unit1 = "M^3/MOL" Then
Value = 1 / Value / 1000
ElseIf Unit1 = "M^3/KG" Then
Value = 1 / Value / MolWt
ElseIf Unit1 = "DM^3/KG" Or Unit1 = "L/KG" Then
ElseIf Unit1 = "CC/G" Or Unit1 = "CM^3/G" Or Unit1 = "ML/G" Then
ElseIf Unit1 = "DM^3/G" Then
ElseIf Unit1 = "FT^3/LBM" Or Unit1 = "FT^3/LB" Then
Value = 1 / Value * (LBMtoKG / FT3toM3 / MolWt)
ElseIf Unit1 = "FT^3/LBMOL" Then
Value = 1 / Value * (LBMtoKG / FT3toM3)
ElseIf Unit1 = "FT^3/SLUG" Then
Value = 1 / Value * (LBMtoKG / FT3toM3 / MolWt * KGFtoN / FTtoM)
Else
End If
If Abs(Value) < 1E-30 Then Value = 0
244
'----------------------------------------------------------------------' Specific Energy and Enthalpy Conversions

'----------------------------------------------------------------------ElseIf Tpe = "H" Then
If Unit1 = "J/MOL" Or Unit1 = "KJ/KMOL" Then
ElseIf Unit1 = "KJ/MOL" Then
ElseIf Unit1 = "MJ/MOL" Then
ElseIf Unit1 = "KJ/KG" Or Unit1 = "J/G" Then
Value = MolWt ^ Drct * Value
ElseIf Unit1 = "J/KG" Then
Value = (MolWt / 1000) ^ Drct * Value
ElseIf Unit1 = "M^2/S^2" Then
ElseIf Unit1 = "FT^2/S^2" Then
Value = (MolWt / 1000 * FTtoM ^ 2) ^ Drct * Value
ElseIf Unit1 = "CAL/MOL" Or Unit1 = "KCAL/KMOL" Then
Value = CALtoJ ^ Drct * Value
ElseIf Unit1 = "CAL/G" Or Unit1 = "KCAL/KG" Then
Value = (CALtoJ * MolWt) ^ Drct * Value
ElseIf Unit1 = "BTU/LBM" Or Unit1 = "BTU/LB" Then
Value = (BTUtoKJ / LBMtoKG * MolWt) ^ Drct * Value
ElseIf Unit1 = "BTU/LBMOL" Then
Value = (BTUtoKJ / LBMtoKG) ^ Drct * Value
Else
End If
245
'----------------------------------------------------------------------' Entropy and Heat Capacity Conversions

'----------------------------------------------------------------------ElseIf Tpe = "S" Then
If Unit1 = "J/MOL-K" Or Unit1 = "KJ/KMOL-K" Then
Value = Value
ElseIf Unit1 = "KJ/MOL-K" Then
ElseIf Unit1 = "KJ/KG-K" Or Unit1 = "J/G-K" Then
Value = MolWt ^ Drct * Value
ElseIf Unit1 = "J/KG-K" Then
ElseIf Unit1 = "BTU/LBM-R" Or Unit1 = "BTU/LB-R" Then
Value = (BTUtoKJ / LBMtoKG / RtoK * MolWt) ^ Drct * Value
ElseIf Unit1 = "BTU/LBMOL-R" Then
Value = (BTUtoKJ / LBMtoKG / RtoK) ^ Drct * Value
ElseIf Unit1 = "CAL/G-K" Or Unit1 = "CAL/G-C" Or Unit1 = "KCAL/KG-K" Or Unit1
= "KCAL/KG-C" Then
Value = (CALtoJ * MolWt) ^ Drct * Value
ElseIf Unit1 = "CAL/MOL-K" Or Unit1 = "CAL/MOL-C" Then
ElseIf Unit1 = "FT-LBF/LBMOL-R" Then
Value = (FTLBFtoJ / LBMtoKG / RtoK / 1000) ^ Drct * Value
ElseIf Unit1 = "CP/R" Then
Value = Rgas ^ Drct * Value * 1000
Else
End If
246
'----------------------------------------------------------------------' Speed of Sound Conversion

'----------------------------------------------------------------------ElseIf Tpe = "W" Then
If Unit1 = "M/S" Then
ElseIf Unit1 = "M^2/S^2" Then
Value = Sqr(Value)
ElseIf Unit1 = "CM/S" Then
ElseIf Unit1 = "KM/H" Then
Value = Value * (1000 / HtoS) ^ Drct
ElseIf Unit1 = "FT/S" Then
Value = Value * FTtoM ^ Drct
ElseIf Unit1 = "IN/S" Then
Value = Value * INtoM ^ Drct
ElseIf Unit1 = "MILE/H" Or Unit1 = "MPH" Then
Value = Value * (INtoM * 63360 / HtoS) ^ Drct
ElseIf Unit1 = "KNOT" Then
Value = Value * 0.5144444444 ^ Drct
ElseIf Unit1 = "MACH" Then
Value = Value * Sqr(1.4 * 298.15 * 8314.51 / 28.95853816) ^ Drct
Else
End If
'----------------------------------------------------------------------' Viscosity Conversion

'----------------------------------------------------------------------ElseIf Tpe = "U" Then
247
If Unit1 = "PA-S" Or Unit1 = "KG/M-S" Then

ElseIf Unit1 = "MPA-S" Then
'Note: This is milliPa-s, not MPa-s

ElseIf Unit1 = "UPA-S" Then
ElseIf Unit1 = "G/CM-S" Or Unit1 = "POISE" Then
ElseIf Unit1 = "CENTIPOISE" Then
ElseIf Unit1 = "MILLIPOISE" Or Unit1 = "MPOISE" Then
ElseIf Unit1 = "MICROPOISE" Or Unit1 = "UPOISE" Then
ElseIf Unit1 = "LBM/FT-S" Or Unit1 = "LB/FT-S" Then
Value = Value * (LBMtoKG / FTtoM) ^ Drct
ElseIf Unit1 = "LBF-S/FT^2" Then
Value = Value * (LBFtoN / FTtoM ^ 2) ^ Drct
ElseIf Unit1 = "LBM/FT-H" Or Unit1 = "LB/FT-H" Then
Value = Value * (LBMtoKG / FTtoM / HtoS) ^ Drct
Else
End If
'----------------------------------------------------------------------' Thermal Conductivity Conversion

'----------------------------------------------------------------------ElseIf Tpe = "K" Then
If Unit1 = "MW/M-K" Then
ElseIf Unit1 = "W/M-K" Then
248

ElseIf Unit1 = "G-CM/S^3-K" Then
ElseIf Unit1 = "KG-M/S^3-K" Then
ElseIf Unit1 = "CAL/S-CM-K" Then
Value = Value * (CALtoJ * 100000) ^ Drct
ElseIf Unit1 = "KCAL/HR-M-K" Then
Value = Value * (CALtoJ * 100000 * 1000 / 100 / 3600) ^ Drct
ElseIf Unit1 = "LBM-FT/S^3-F" Or Unit1 = "LB-FT/S^3-F" Then
Value = Value * (1000 * LBMtoKG * FTtoM / RtoK) ^ Drct
ElseIf Unit1 = "LBF/S-F" Then
Value = Value * (1000 * LBFtoN / RtoK) ^ Drct
ElseIf Unit1 = "BTU/H-FT-F" Then
Value = Value * (1000 * BTUtoW / HtoS / FTtoM / RtoK) ^ Drct
Else
End If
'----------------------------------------------------------------------' Joule-Thomson Conversion

'----------------------------------------------------------------------ElseIf Tpe = "JT" Then
If Unit1 = "K/MPA" Or Unit1 = "C/MPA" Then
ElseIf Unit1 = "K/KPA" Or Unit1 = "C/KPA" Then
ElseIf Unit1 = "K/PA" Or Unit1 = "C/PA" Then
ElseIf Unit1 = "C/ATM" Then
249
Value = Value / ATMtoMPa ^ Drct

ElseIf Unit1 = "C/BAR" Then
Value = Value / BARtoMPA ^ Drct
ElseIf Unit1 = "K/PSI" Or Unit1 = "K/PSIA" Then
Value = Value / PSIAtoMPA ^ Drct
ElseIf Unit1 = "F/PSI" Or Unit1 = "F/PSIA" Or Unit1 = "R/PSIA" Then
Value = Value / (PSIAtoMPA / RtoK) ^ Drct
Else
End If
'----------------------------------------------------------------------' Length Conversion

'----------------------------------------------------------------------ElseIf Tpe = "L" Then
If Unit1 = "METER" Or Unit1 = "M" Then
ElseIf Unit1 = "DM" Then
ElseIf Unit1 = "CM" Then
ElseIf Unit1 = "MM" Then
ElseIf Unit1 = "KM" Then
ElseIf Unit1 = "INCH" Or Unit1 = "IN" Then
Value = Value * INtoM ^ Drct
ElseIf Unit1 = "FOOT" Or Unit1 = "FT" Then
Value = Value * FTtoM ^ Drct
ElseIf Unit1 = "YARD" Or Unit1 = "YD" Then
250
Value = Value * (INtoM * 36) ^ Drct

ElseIf Unit1 = "MILE" Or Unit1 = "MI" Then
Value = Value * (INtoM * 63360) ^ Drct
ElseIf Unit1 = "LIGHT YEAR" Then
Value = Value * 9.46055E+15 ^ Drct
ElseIf Unit1 = "ANGSTROM" Then
Value = Value / 10000000000# ^ Drct
ElseIf Unit1 = "FATHOM" Then
Value = Value * (FTtoM * 6) ^ Drct
ElseIf Unit1 = "MIL" Then
Value = Value * (INtoM / 1000) ^ Drct
ElseIf Unit1 = "ROD" Then
Value = Value * (INtoM * 16.5 * 12) ^ Drct
ElseIf Unit1 = "PARSEC" Then
Value = Value * (30837400000000# * 1000) ^ Drct
Else
End If
'----------------------------------------------------------------------' Area Conversion

'----------------------------------------------------------------------ElseIf Tpe = "A" Then
If Unit1 = "METER^2" Or Unit1 = "M^2" Then
ElseIf Unit1 = "CM^2" Then
ElseIf Unit1 = "MM^2" Then
ElseIf Unit1 = "KM^2" Then
251

ElseIf Unit1 = "INCH^2" Or Unit1 = "IN^2" Then
Value = Value * (INtoM ^ 2) ^ Drct
ElseIf Unit1 = "FOOT^2" Or Unit1 = "FT^2" Then
Value = Value * (FTtoM ^ 2) ^ Drct
ElseIf Unit1 = "YARD^2" Or Unit1 = "YD^2" Then
Value = Value * ((INtoM * 36) ^ 2) ^ Drct
ElseIf Unit1 = "MILE^2" Or Unit1 = "MI^2" Then
Value = Value * ((INtoM * 63360) ^ 2) ^ Drct
ElseIf Unit1 = "ACRE" Then
Value = Value * ((INtoM * 36) ^ 2 * 4840) ^ Drct
ElseIf Unit1 = "BARN" Then
Value = Value * 1E-28 ^ Drct
ElseIf Unit1 = "HECTARE" Then
Else
End If
'----------------------------------------------------------------------' Volume Conversion (Note: not specific volume)

'----------------------------------------------------------------------ElseIf Tpe = "V" Then
If Unit1 = "METER^3" Or Unit1 = "M^3" Then
ElseIf Unit1 = "CM^3" Then
ElseIf Unit1 = "LITER" Or Unit1 = "L" Or Unit1 = "DM^3" Then
ElseIf Unit1 = "INCH^3" Or Unit1 = "IN^3" Then
252
Value = Value * IN3toM3 ^ Drct

ElseIf Unit1 = "FOOT^3" Or Unit1 = "FT^3" Then
Value = Value * (IN3toM3 * 12 ^ 3) ^ Drct
ElseIf Unit1 = "YARD^3" Or Unit1 = "YD^3" Then
Value = Value * (IN3toM3 * 36 ^ 3) ^ Drct
ElseIf Unit1 = "GALLON" Or Unit1 = "GAL" Then
Value = Value * GALLONtoM3 ^ Drct
ElseIf Unit1 = "QUART" Or Unit1 = "QT" Then
Value = Value * (GALLONtoM3 / 4) ^ Drct
ElseIf Unit1 = "PINT" Or Unit1 = "PT" Then
ElseIf Unit1 = "CUP" Then
ElseIf Unit1 = "OUNCE" Then
ElseIf Unit1 = "TABLESPOON" Or Unit1 = "TBSP" Then
ElseIf Unit1 = "TEASPOON" Or Unit1 = "TSP" Then
ElseIf Unit1 = "CORD" Then
Value = Value * (FT3toM3 * 128) ^ Drct
ElseIf Unit1 = "BARREL" Then
Value = Value * (GALLONtoM3 * 42) ^ Drct
ElseIf Unit1 = "BOARD FOOT" Then
Value = Value * (IN3toM3 * 144) ^ Drct
ElseIf Unit1 = "BUSHEL" Then
Value = Value * 0.03523907016688 ^ Drct
Else
253
End If
'----------------------------------------------------------------------' Mass Conversion

'----------------------------------------------------------------------ElseIf Tpe = "M" Then
If Unit1 = "KG" Then
ElseIf Unit1 = "G" Then
ElseIf Unit1 = "MG" Then
'milligram

ElseIf Unit1 = "LBM" Or Unit1 = "LB" Then
Value = Value * LBMtoKG ^ Drct
ElseIf Unit1 = "GRAIN" Then
Value = Value * (LBMtoKG / 7000) ^ Drct
ElseIf Unit1 = "SLUG" Then
Value = Value * (KGFtoN * LBMtoKG / FTtoM) ^ Drct
ElseIf Unit1 = "TON" Then
Value = Value * (LBMtoKG * 2000) ^ Drct
ElseIf Unit1 = "TONNE" Then
Else
End If
'----------------------------------------------------------------------' Force Conversion

'----------------------------------------------------------------------ElseIf Tpe = "F" Then
254
If Unit1 = "NEWTON" Or Unit1 = "N" Then

ElseIf Unit1 = "MN" Then 'milliNewtons
ElseIf Unit1 = "KGF" Then
Value = Value * KGFtoN ^ Drct
ElseIf Unit1 = "DYNE" Then
ElseIf Unit1 = "LBF" Then
Value = Value * LBFtoN ^ Drct
ElseIf Unit1 = "POUNDAL" Then
Value = Value * (LBMtoKG * FTtoM) ^ Drct
ElseIf Unit1 = "OZF" Then
Value = Value * (LBFtoN / 16) ^ Drct
Else
End If
'----------------------------------------------------------------------' Energy Conversion

'----------------------------------------------------------------------ElseIf Tpe = "E" Then
If Unit1 = "JOULE" Or Unit1 = "J" Then
ElseIf Unit1 = "KJ" Then
ElseIf Unit1 = "MJ" Then
ElseIf Unit1 = "KW-H" Then
Value = Value * (HtoS * 1000) ^ Drct
ElseIf Unit1 = "CAL" Then
255

ElseIf Unit1 = "KCAL" Then
ElseIf Unit1 = "ERG" Then
ElseIf Unit1 = "BTU" Then
Value = Value * (BTUtoKJ * 1000) ^ Drct
ElseIf Unit1 = "FT-LBF" Then
Value = Value * FTLBFtoJ ^ Drct
Else
End If
'----------------------------------------------------------------------' Power Conversion

'----------------------------------------------------------------------ElseIf Tpe = "Q" Then
If Unit1 = "WATT" Or Unit1 = "W" Then
ElseIf Unit1 = "KWATT" Or Unit1 = "KW" Then
ElseIf Unit1 = "BTU/S" Then
Value = Value * BTUtoW ^ Drct
ElseIf Unit1 = "BTU/MIN" Then
Value = Value * (BTUtoW / 60) ^ Drct
ElseIf Unit1 = "BTU/H" Then
Value = Value * (BTUtoW / HtoS) ^ Drct
ElseIf Unit1 = "CAL/S" Then
Value = Value * CALtoJ ^ Drct
ElseIf Unit1 = "KCAL/S" Then
256

ElseIf Unit1 = "CAL/MIN" Then
Value = Value * (CALtoJ / 60) ^ Drct
ElseIf Unit1 = "KCAL/MIN" Then
Value = Value * (CALtoJ / 60 * 1000) ^ Drct
ElseIf Unit1 = "FT-LBF/S" Then
Value = Value * FTLBFtoJ ^ Drct
ElseIf Unit1 = "FT-LBF/MIN" Then
Value = Value * (FTLBFtoJ / 60) ^ Drct
ElseIf Unit1 = "FT-LBF/H" Then
Value = Value * (FTLBFtoJ / HtoS) ^ Drct
ElseIf Unit1 = "HP" Then
Value = Value * HPtoW ^ Drct
Else
End If
'----------------------------------------------------------------------' Surface Tension Conversion

'----------------------------------------------------------------------ElseIf Tpe = "N" Then
If Unit1 = "N/M" Then
ElseIf Unit1 = "MN/M" Then
ElseIf Unit1 = "DYNE/CM" Or Unit1 = "DYN/CM" Then
ElseIf Unit1 = "LBF/FT" Then
Value = Value * LBFTtoNM ^ Drct
Else
257

End If
End If
Unit1 = Unit2
Next Drct
UnitConvert = Value
End Function
Sub SetupUnits(i)
'Warning: If any of these are changed (to make them the default) after the program has run,
' you will need to exit Excel and restart it so that it reinitializes
'Refprop Units
tUnits2 = "K"
taUnits2 = "K"
pUnits2 = "kPa"
dUnits2 = "mol/dm^3"
vUnits2 = "dm^3/mol"
hUnits2 = "J/mol"
sUnits2 = "J/mol-K"
wUnits2 = "m/s"
visUnits2 = "uPa-s"
tcxUnits2 = "W/m-K"
stUnits2 = "N/m"
'Default units: (SI)
tUnits(0) = "K"
taUnits(0) = "K"
pUnits(0) = "MPa"
258
dUnits(0) = "kg/m^3"
vUnits(0) = "m^3/kg"
hUnits(0) = "kJ/kg"
sUnits(0) = "kJ/kg-K"
wUnits(0) = "m/s"
visUnits(0) = "uPa-s"
tcxUnits(0) = "mW/m-K"
stUnits(0) = "mN/m"
'Default units but with K switch to C (SI with C)
tUnits(5) = "C"
taUnits(5) = "K"
pUnits(5) = "MPa"
hUnits(5) = "kJ/kg"
wUnits(5) = "m/s"
stUnits(5) = "mN/m"
'Default units on a molar basis (Molar SI)
tUnits(6) = "K"
taUnits(6) = "K"
pUnits(6) = "MPa"
dUnits(6) = "mol/dm^3"
vUnits(6) = "dm^3/mol"
hUnits(6) = "J/mol"
sUnits(6) = "J/mol-K"
wUnits(6) = "m/s"
259
stUnits(6) = "mN/m"
'mks (mks)
tUnits(1) = "K"
taUnits(1) = "K"
pUnits(1) = "kPa"
hUnits(1) = "kJ/kg"
wUnits(1) = "m/s"
tcxUnits(1) = "W/m-K"
stUnits(1) = "mN/m"
'cgs (cgs)
tUnits(2) = "K"
taUnits(2) = "K"
pUnits(2) = "MPa"
dUnits(2) = "g/cm^3"
vUnits(2) = "cm^3/g"
hUnits(2) = "J/g"
sUnits(2) = "J/g-K"
wUnits(2) = "cm/s"
stUnits(2) = "dyn/cm"
'English (E)
tUnits(3) = "F"
'See comments above
260
taUnits(3) = "R"
pUnits(3) = "psia"
dUnits(3) = "lbm/ft^3"
vUnits(3) = "ft^3/lbm"
hUnits(3) = "Btu/lbm"
sUnits(3) = "Btu/lbm-R"
wUnits(3) = "ft/s"
visUnits(3) = "lbm/ft-s"
tcxUnits(3) = "Btu/h-ft-F"
stUnits(3) = "lbf/ft"
'Mixed (M)
tUnits(4) = "K"
taUnits(4) = "K"
pUnits(4) = "psia"
dUnits(4) = "g/cm^3"
vUnits(4) = "cm^3/g"
hUnits(4) = "J/g"
sUnits(4) = "J/g-K"
wUnits(4) = "m/s"
stUnits(4) = "mN/m"
End Sub
Function ConvertUnits(InpCode, Units, Prop1, Prop2)

Dim i As Integer, at As String, bt As String, tConv As Double, DefaultUnits As Integer

If IsMissing(Units) Then Units = ""
261
If IsMissing(Prop1) Then Prop1 = 0

If IsMissing(Prop2) Then Prop2 = 0
If ierr > 0 Then ConvertUnits = Trim2(herr): Exit Function
If tUnits2 = "" Then
Call SetupUnits(0) 'If Default units are changed, this needs to be called again. Normally it
is skipped after the first entry
End If
'Change the 0 in the following line to 3 for default English units, 1 for mks, or 2 for cgs, etc.
DefaultUnits = 0
i = DefaultUnits
'Do not change the order of the next 7 statements
If Left(UCase(Units), 2) = "SI" Then i = 0
'SI
If UCase(Units) = "SI WITH C" Or UCase(Units) = "C" Then i = 5 'SI with C

If Left(UCase(Units), 1) = "M" Then i = 4
'Mixed
If UCase(Units) = "MOLAR SI" Then i = 6
'Molar SI
If UCase(Units) = "MKS" Then i = 1
'mks
If UCase(Units) = "CGS" Then i = 2
'cgs
If Left(UCase(Units), 1) = "E" Then i = 3
'English
at = UCase(Left(InpCode, 1))
bt = UCase(Mid(InpCode, 2, 1))
If at = "-" Then
ConvertUnits = Prop1
If Prop1 >= -9999999 And Prop1 <= -9999900 Then
If Prop1 = CLng(Prop1) Then
ConvertUnits = Trim2("Undefined")
Exit Function
End If
End If
262
'If Len(Trim(Prop1)) > 0 Then

If bt = "T" Then ConvertUnits = UnitConvert(Prop1, "T", tUnits2, tUnits(i))
If bt = "A" Then ConvertUnits = UnitConvert(Prop1, "T", taUnits2, taUnits(i))
If bt = "P" Then ConvertUnits = UnitConvert(Prop1, "P", pUnits2, pUnits(i))
If bt = "D" Then ConvertUnits = UnitConvert(Prop1, "D", dUnits2, dUnits(i))
If bt = "V" Then ConvertUnits = UnitConvert(Prop1, "D", vUnits2, vUnits(i))
If bt = "H" Or bt = "E" Then ConvertUnits = UnitConvert(Prop1, "H", hUnits2, hUnits(i))
If bt = "S" Then ConvertUnits = UnitConvert(Prop1, "S", sUnits2, sUnits(i))
If bt = "W" Then ConvertUnits = UnitConvert(Prop1, "W", wUnits2, wUnits(i))
If bt = "U" Then ConvertUnits = UnitConvert(Prop1, "U", visUnits2, visUnits(i))
If bt = "K" Then ConvertUnits = UnitConvert(Prop1, "K", tcxUnits2, tcxUnits(i))
If bt = "N" Then ConvertUnits = UnitConvert(Prop1, "N", stUnits2, stUnits(i))
'End If
If bt = "J" Then
tConv = 1
If tUnits(i) = "R" Or tUnits(i) = "F" Then tConv = 1 / RtoK
ConvertUnits = Prop1 * tConv / UnitConvert(1, "P", "kPa", pUnits(i))
End If
Else
If Len(Trim(Prop1)) > 0 Then
If at = "T" Then Prop1 = UnitConvert(Prop1, "T", tUnits(i), tUnits2)
If at = "A" Then Prop1 = UnitConvert(Prop1, "T", taUnits(i), taUnits2)
If at = "P" Then Prop1 = UnitConvert(Prop1, "P", pUnits(i), pUnits2)
If at = "D" Then Prop1 = UnitConvert(Prop1, "D", dUnits(i), dUnits2)
If at = "V" Then Prop1 = UnitConvert(Prop1, "D", vUnits(i), vUnits2)
If at = "H" Or at = "E" Then Prop1 = UnitConvert(Prop1, "H", hUnits(i), hUnits2)
If at = "S" Then Prop1 = UnitConvert(Prop1, "S", sUnits(i), sUnits2)
If at = "W" Then Prop1 = UnitConvert(Prop1, "W", wUnits(i), wUnits2)
If at = "U" Then Prop1 = UnitConvert(Prop1, "U", visUnits(i), visUnits2)
263
If at = "K" Then Prop1 = UnitConvert(Prop1, "K", tcxUnits(i), tcxUnits2)

If at = "N" Then Prop1 = UnitConvert(Prop1, "N", stUnits(i), stUnits2)
End If
If Len(Trim(Prop2)) > 0 Then

If bt = "T" Then Prop2 = UnitConvert(Prop2, "T", tUnits(i), tUnits2)
If bt = "A" Then Prop2 = UnitConvert(Prop2, "T", taUnits(i), taUnits2)
If bt = "P" Then Prop2 = UnitConvert(Prop2, "P", pUnits(i), pUnits2)
If bt = "D" Then Prop2 = UnitConvert(Prop2, "D", dUnits(i), dUnits2)
If bt = "V" Then Prop2 = UnitConvert(Prop2, "D", vUnits(i), vUnits2)
If bt = "H" Or bt = "E" Then Prop2 = UnitConvert(Prop2, "H", hUnits(i), hUnits2)
If bt = "S" Then Prop2 = UnitConvert(Prop2, "S", sUnits(i), sUnits2)
If bt = "W" Then Prop2 = UnitConvert(Prop2, "W", wUnits(i), wUnits2)
If bt = "U" Then Prop2 = UnitConvert(Prop2, "U", visUnits(i), visUnits2)
If bt = "K" Then Prop2 = UnitConvert(Prop2, "K", tcxUnits(i), tcxUnits2)
If bt = "N" Then Prop2 = UnitConvert(Prop2, "N", stUnits(i), stUnits2)
End If
End If
End Function
264
References
Alessio, E., A. Carbone, G. Castelli, and V. Frappietro. 2002. Second-order Moving Average
and Scalling of Stochastic time series. THe European Physical Journal B 27: 197200.
American Petroleum Institute. 2007. API RP 521-Pressure-relieving and Depressuring
Systems, edited by A. P. Institute.
American Petroleum Institute. 2008. API RP 520 Part 1 Sizing, Selection and Installation of
the Pressure-relieving devices in the Refineries, edited by A. P. Institute.
Bansal, R. K. 2005. Fluid mechanics and hydraulic machines. 9 ed. New Delhi: Laxmi
Publications (P) Ltd.
Bellman, R. 1970. Method of Non-Linear Analysis. 2 vols. Vol. 1, Mathematics in Science
and Engineering. New York: Academic Press, Inc.
Bernuth, R. D. V. 1990. Simple and Accurate Friction Loss Equation for Plastic Pipe.
Journal of Irrigation and Drainage Engineering 116 (2): 294-298.
BOC
Gases.
2006.
BOC
Gases
Reference
Manual.
https://boc.com.au/boc_sp/au/downloads/reference_manuals/industrial/BOC_IPRM_
S03-IndGas.pdf (accessed
Botros, K. K., W. M. Jungowski, and M. H. Weiss. 1989. Models and Methods of
Simulating Gas Pipeline Blowdown. The Canadian Journal of Chemical
Engineering 67: 529-539.
Brkic`, D. 2011. Review of Explicit Approximations to the Colebrook Relation for Flow
Friction. Journal of Petroleum Science and Engineering 77: 34-48.
Case, J., L. Chilver, and C. T. F. Ross. 1999. Strenght of Materials and Structures, ed M.
Flynn. London: Arnold (accessed.
Chan, S. K., and W. A. Woods. 1992. On Rayleigh and Fanno Flows of Homogenous
Equilibrium Two-phase fluids. International Journal of Heat and Fluid Flow 13 (3):
273-281.
265
Chen, J. R., S. Richardson, and G. Saville. 1995a. Modelling Of Two-Phase Blowdown

From Pipelines-1. A Hyperbolic Model Based On Variational Principles. Chemical
Engineering Science 50 (4): 695-713.
Chen, J. R., S. Richardson, and G. Saville. 1995b. Modelling of Two Phase Blowdown of
Pipelines: II. A Simplified Numerical Method for Multi-component Mixtures.
Chemical Engineering Science 50 (13): 2173-2187.
Chen, N. H. 1979. An Explicit Equation for Friction Factor in Pipe. Industrial and
Engineering Chemistry Fundamentals 18 (3): 296-297.
Churchill, S. W. 1977. Friction-factor Equation Spans all fluid Flow Regimes. Chemical
Engineering Journal 84 (24): 91-92.
Cochran, T. W. 1996. Calculate Pipeline Flow of Compressible Fluids. Chemical
Engineering:
Colebrook, C. F. 1939. Turbulent Flow in Pipes, with Particular Reference to the Transition
Region between the Smooth and Rough Pipe Laws. Journal of the Institute of Civil
Engineers 11 (4): 133-156.
Cumber, P. S. 2001. Predicting Outflow From High Pressure Vessels. Transactions IChemE
79, part B: 13-22.
Eggers, R., and V. Green. 1990. Pressure Discharge from a Pressure Vessel Filled with C02.
Journal of Loss Prevention in the Process Industries 3 (1): 59-63.
Ellenberger, P. 2010. Piping and Pipeline Calculations Manual Construction, Design
Fabrication and Examination. Burlington : Elsevier Science and Technology
(accessed.
Evanger, T., T. Bjorge, B. Magnussen, and A. Bratseth. 1995. Full-scale depressurization of
a section of a rizer platform. Measurements and comparison with simulations.
Journal of Harzardous Materials 46: 117-129.
Fairuzov, Y. V. 1998. Blowdown of Pipelines Carrying Flashing Liquids. AIChE Journal 44
(2): 245-254.
Farina, I. H. 1997. Critical Length Helps Calculate Compressible Flow. Chemical
Engineering: 88-92.
266
Franz, D. D., and C. S. Melching. ND. Full Equations (FEQ) Model for the Solution of the
Full, Dynamic Equations of Motion for One-Dimensional Unsteady Flow in Open
Channels and through Control Structures. Chicago, Illinois: U.S. Geological
Survey.
Franzini, J. B., E. J. Finnemore, and R. L. Daugherty. 1997. Fluid Mechanics with
Engineering Applications. 9th ed: McGraw Hill.
Gebbeken, B., and R. Eggers. 1995. Blowdown of carbon dioxide from initially supercritical
conditions. Journal of Loss Prevention in the Process Industries 9 (4): 285-293.
Glushkov, B., E. Selyanskaya, and S. Kas'yanov. 2003. Processing Gas-Dynamic Test Data
On Compressors and Flow Sections for Air and Natural Gas. Chemical and
Petroleum Engineering 39: 276-280.
Goh, C. B. 1989. Estimation of Flowrate Through a Ruptured Natural Gas Pipe.
International Journal of Heat and Fluid Flow 10 (2): 173-178.
Govier, G. W., and K. Aziz. 1972. The Flow of Complex Mixtures in Pipes. New York: Van
Nostrand Reinhold Co.
Gradle, R. J. 1984. Design of Gas Pipeline Blowdowns. Energy Processing Canada:
Hall, A. R. W., G. R. Butcher, and C. E. The. 1993. Proceedings of European Two Phase
Flow Group Meeting, Transient Simulation of Two-phase Hydrocarbon Flows in
Pipelines. Hannover, Germany: (accessed
Haque, A., S. Richardson, and G. Saville. 1992. Blowdown of pressure vessels I. Computer
model. Transactions IChemE 70 (Part B):
Haque, A., S. Richardson, G. Saville, and G. Chamberlain. 1989. Rapid depressurization of
pressure vessels. Journal of Loss Prevention in the Process Industries 3:
Haque, A., S. Richardson, G. Saville, G. Chamberlain, and L. Shirvill. 1992. Blowdown of
pressure vessels II. Experimental validation of computer model and case studies.
Transactions IChemE 70 (Part B):
Hervieu, E. 1991. Behaviour of a flashing liquid within a vessel following loss of
containment. Application to propane.
267
Holditch, S. A., and R. R. Chianelli. 2008. Factors That Will Influence Oil and Gas Supply
and Demand in the 21st Century. Harnessing Materials For Energy 33:
Hong, Y.-J., S.-J. Park, H.-B. Kim, and Y.-D. Choi. 2004. The Cool-down Characteristics of
a Miniature Joule-Thomson Refrigerator. Cryogenics 46: 391-395.
Jain, A. K. 1976. Accurate Explicit Equation for Friction Factor. American Society Civil
Engineers Journal of Hydraulic Engineering 102 (HY5): 674-677.
Jones, J. B., and G. A. Hawkins. 1986. Engineering Thermodynamics An Introductory
Textbook. Second ed: John Wiley & Sons, Inc.
Kaye, J. 1953. Survey of Friction Coefficients, Recovery Factors, and Heat-Transfer
Coefficients for Supersonic Flow. Cambridge, Massachuetts: Massachusetts Institute
of Technology.
Keenan,
M.
2009.
Brittle
Fracture.
http://www.communityhotline.com/News_
Photos/news_photos_DetailCustomer.php?newsid=2532 (accessed 28th December).

Khazrai, F., H. B. Haghighi, and H. Kordabadi. 2001. Avoid Brittle Fracture in Pressure
Vessels. Hydrocarbon Processing. http://proquest.umi.com.dbgw.lis. curtin.edu.au
/pqdweb?did=2311079201&sid=1&Fmt=3&clientId=22212&RQT=309&VName=P
QD (accessed
King, R. E. 2006. Auto-Refrigeration / Brittle Fracture Analysis of Existing Olefins Plants Translation of Lessons Learned to Other Processes. Journal of Hazardous Materials
142 (2007): 608-617.
Kirk-Burnnand, D. 2009. Gas Properties and DBNGP Data. Research Project Meeting.
05/06/2009, Perth.
Kunz, O., R. Klimeck, W. Wagner, and M. Jaeschke. 2007. The GERG-2004 Wide-Range
Equation of State for Natural Gases and Other Mixtures. Bochum, Germany:
Groupe Europeen de Recherches Gazieres.
Lemmon, D. E. W., D. M. L. Huber, and D. M. O. McLinden. 2009. REFPROP Manual.
Boulder: U. S. Department of Commerce.
Lloyd, D. K., and M. Lipow. 1962. Reliability: Management, Methods and Mathematics:
Prentice-Hall.
268
Mackerle, J. 1999. Finite elements in the analysis of pressure vessels and piping, an
addendum. International Journal of Pressure Vessels and Piping 76: 461-485.
Mahgerefteh, H., P. Saha, and I. G. Economou. 1999. Fast Numberical Simulation for Full
Bore Rupture of Pressurized Pipelines. AIChE Journal 45 (6): 1191-1201.
Mahgerefteh, H., and S. M. A. Wong. 1999. A numerical blowdown simulation
incorporating cubic equations of state. Computers and Chemical Engineering 23:
Marian, D. R., D. H. B. Vuthaluru, and V. Ghantala. PVBLOW: A Mathematical Model for
Predicting Minimum Vessel Wall Temperatures for Pressure Vessels Undergoing
Blowdown. Curtin Univerisity of Technology.
Maric, I. 2005. The Joule-Thomson effect in natural gas flow-rate measurements. Flow
Measurement and Instrumentation 16: 387-395.
McAdams, W. H., L. A. Nicolai, and J. H. Keenan. 1946. Measurement of Recovery Factors
and Coefficient of Heat Transfer in a Tube for Subsonic Flow of Air. American
Institute of Chemical Engineers 42: 907-925.
McCabe, W. L., J. C. Smith, and P. Harriott. 2001. Unit Operations of Chemical
Engineering. Sixth ed. Singapore: McGraw Hill.
Moody, M. L. 1947. An Approximate Formula for Pipe Friction Factors. Transaction of
ASME 69: 1005-1011.
Nageshwar, G. D. 2003. Chemical Engineering Thermodynamics: Vipul Prakashan, India.
Nagy, I. 1992. Introduction to Chemical Process Instrumentation. Vol. 3. Amsterdam, The
Netherlands: Elsevier Science Publishers.
Narasimhan, S., and C. Jordache. 2000. Data Reconciliation and Gross Error Detection An Intelligent Use of Process Data. Houstan, Texas: Gulf Publishing Company.
National Instruments. 2006. Reducing the Effects of Noise in a Data Acquisition System by
Averaging. http://zone.ni.com/devzone/cda/tut/p/id/3947?submitted=yes.
Nijsing, R., and N. A. Brinkhof. 1996. Emergency Pressure RELIEF Calculations Using the
Computer Package RELIEF. Journal of Hazardous Materials 46: 131-143.
Nolan, D. P. 1996. Handbook of Fire and Explosion Protection Engineering Principles for
Oil, Gas, Chemical, and Related Facilities Noyes Publications (accessed.
269
Norris III, H., and Exxon Production Research Co. 1994. Hydrocarbon Blowdown From
Vessels and Pipelines. In 69th Annual Technical Conference and Exhibition of the
Society of Petroleum Engineers. New Orleans, LA, USA. SPE.
Norris III, H. L., Exxon Production Research Co, and U. o. T. R.C. Puls. 1993. Single-Phase
or Multiphase Blowdown of Vessels or Pipelines. In 68th Annual Technical
Conference and Exhibition of the Society of Petroleum Engineers. Houston. SPE.
Ortega, J. M., and W. C. Rheinboldt. 1970. Iterative Solution of Non-Linear Equations in
Several Variables. Edited by W. Rheinboldt, Computer Science and Applied
Mathematics. New York: Academic Press, Inc.
Ouyang, L.-b., and K. Aziz. 1995. Steady-state Gas Flow in Pipes. Journal of Petroleum
Science and Engineering 14: 137-158.
Pareek, V. 2008. Introduction to HYSYS. ChE 312 Process Synthesis and Design 1. Lecture
notes. Perth: Department of Chemical Engineering, Curtin University.
Parker, G. J. 1985. 'Pop' Safefy Valves: A Compressible Flow Analysis. International
Journal of Heat and Fluid Flow 6 (4): 279-283.
Parker, G. J. 1989. Adiabatic Compressible Flow in Parallel Ducts: An Approximate but
Rapid Method of Solution. International Journal of Heat and Fluid Flow 10 (2):
179-181.
Peng, D.-Y., and D. B. Robinson. 1976. A New Two-Constant Equation of State. Industrial
& Engineering Chemistry Fundamentals 15 (1): 59-64.
Prandtl, L. 2004. Prandtl's Essentials of Fluid Mechanics, ed H. Oertel. New York:
Springer-Verlag New York, Inc (accessed 22/09/2010).
Ramirez, W. F. 1998. Computational Methods for Process Simulation. Second Edition ed.
Boulder, Colorado: Butterworth-Heinemann.
Richardson, S., and G. Saville. 1991. Blowdown of Pipelines. In Offshore Europe
Conference. Aberdeen. Society of Petroleum Engineers.
Richardson, S., and G. Saville. 1996. Blowdown of LPG Pipelines. Transactions IChemE 74
(Part B): 235-244.
270
Riggs, J. B., and M. N. Karim. 2006. Chemical and Bio-Process Control. Third ed.
Lubbock, Texas: Ferret Publishing.
Roberts, T. A., I. Buckland, L. C. Shirvill, B. J. Lowesmith, and P. Salater. 2004. Design and
Protection of Pressure Systems to Withstand Severe Fires. Process Safety and
Environmental Protection 82 (B2): 89-96.
Romeo, E., C. Royo, and A. Monzon. 2002. Improved Explicit Equationsfor Estimation of
the Friction Factor in Rough and Smooth Pipes. Chemical Engineering Journal 86
(3): 369-374.
Saad, M. A. 1993. Compressible Fluid Flow. 2 ed. New Jersey: Prentice Hall.
Serghides, T. K. 1984. Estimate Friction Factor Accurately. Chemical Engineering Journal
91 (5): 63-64.
Setzmann, U., and W. Wagner. 1991. A New Equation of State and Tables of
Thermodynamic Properties for Methane Covering the Range from the Melting Line
to 625 K at Pressures up to 1000 MPa. Journal of Physical and Chemical Reference
Data 20 (6): 1061-1155.
Shackelford, J. F. 2005. Introduction to Materials Science for Engineers. In Mechanical
Behaviour. Upper Saddle River, NJ: Pearson Education (accessed.
Shapiro, A. H. 1954. The Dynamics and Thermodynamics of Compressible Flow
The Ronald Press Company, New York.
Shapiro, A. H., and W. R. Hawthorne. 1947. The Mechanics and Thermodynamics of Steady
One-Dimensional Gas Flow. Journal of Applied Mechanics: A-317 - A-336.
Shi, W., M. Miyamoto, Y. Katoh, and J. Kurima. 2001. Choked Flow of Low Density Gas in
a Narrow Parallel-plate Channel with Adiabatic Walls. International Journal of Heat
and Mass Transfer 44: 2555-2565.
Shoemaker, D. P., C. W. Garland, and J. W. Nibler. 1996. Experiments in Physical
Chemistry. 6th ed. New York: McGraw-Hill.
Skouloudis, A. N. 1992. Benchmark exercises on the emergency venting of vessels. Journal
of Loss Prevention in the Process Industries 5 (2): 89-103.
271
Soave, G. 1972. Equilibrium Constants from a Modified Redlich-Kwong Equation of State.

Chemical Engineering Science 27 (6): 1197-1203.
Soave, G. S. 1995. A Non-Cubic Equation of State for the Treatment of Hydrocarbon Fluid
at Reservoir Conditions. Industrial and Engineering Chemistry Research 34: 39813994.
Sonnad, J. R., and C. T. Goudar. 2006. Turbulent Flow Friction Factor Calculation Using a
Mathematically Exact Alternative to the Colebrook-White Equation. Journal of
Hydraulic Engineering ASCE 132 (8): 863-867.
Speranza, A., and A. Terenzi. 2005. Blowdown of hydrocarbons pressure vessel with partial
phase separation. (accessed
Spirax Sarco Limited. 2011. Air Receiver with Automatic Isolating Valves & Blowdown
Valve : Clipart Drawing. http://www.spiraxsarco.com/resources/cad/applications/
process/preview/tanks-and-vats.asp?drawing_id=341&preview=clipart
(accessed
05/06/2011).
Stoop, P. M., J. P. A. V. d. Bogaard, and H. Koning. 1985. Dynamic Loads on an RPV Vent
Line Piping System during Hydrogen Relief. The International Journal of Pressure
Vessel and Piping 18 (3): 183-208.
Stoop, P. M., J. P. A. V. d. Bogaard, and H. Koning. 1986. CHARME-01, A
Thermohydraulic Code for the Calculation of Fast Transients Inside Piping Systems.
The International Journal of Pressure Vessel and Piping 24 (2): 97-121.
Swamee, P. K., and A. K. Jain. 1976. Explicit Equations for Pipe-flow Problems. Journal of
Hydraulic Engineering ASCE 102 (5): 657-664.
Tilley, B. J., and D. A. Shaw. 1990. SAFIRE Computer Program for Emergency Relief
Sizing. In Emergency Relief System Design Using DIERS Technology - The Design
Institute for Emergency Relief Systems (DIERS) Project ManualNew York: AIChE.
Weiss, M. H., K. K. Botros, and W. M. Jungowski. 1988. Simple Method Predicts Gas
Blowdown Time. Oil and Gas Journal 86: 55-58.
West, A. H., D. Posarac, and N. Ellis. 2008. Assessment of Four Biodiesel Production
Processes using HYSYS.Plant. Bioresource Technology 99: 6587-6601.
272
Westman, M. 1997. Rapidly Design Safety Relief Valve Inlet Piping Systems. Chemical
Engineering Progress: 80-88.
Wisniak, J., and H. Avraham. 1996. On the Joule-Thomson effect inversion curve.
Thermochimica Acta 286: 33-40.
Won, K. W., A. R. Smith, and G. A. Zeininger. 2005. Thermodynamic Methods for Pressure
Relief Design Parameters. Fluid Phase Equilibria 241 (2006): 41-50.
Wood, D. J. 1966. An Explicit Friction Factor Relationship. Civil Engineering 36 (12): 6061.
Worth, B., H. Staedtke, and G. Franchello. 1993. RELAP5-EUR/MF, a System Code for
Thermal-Hydraulic Networks. Ispra: Joint Research Council (JRC), European
Commission.
Yuhu, D., G. Huilin, Z. Jing'en, and F. Yaorong. 2002. Mathematical Modelling of Gas
Release Through Holes in Pipelines. Chemical Engineering Journal 92: 237-241.
Zigrang, D. J., and N. D. Slyvester. 1982. Explicit Approximations to the Solution of
Colebrook's Friction Factor Equation. American Institute of Chemical Engineers
Journal 28 (3): 514-515.
Every reasonable effort has been made to acknowledge the owners of copyright material. I
would be pleased to hear from any copyright owner who has been omitted or incorrectly
acknowledged
273

Gas Pipe Blowdown Investigation

Uploaded by

Copyright:

Available Formats

Gas Pipe Blowdown Investigation

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Gas Pipe Blowdown Investigation

Uploaded by

Copyright:

Available Formats

Citation:

Department of Chemical Engineering

Investigation of Compressible Fluid Behaviour in a Vent Pipe during

Farhan Liyakat Husain Rajiwate

This thesis is presented for the Degree of

To Mum and Dad

Brief Biography of Author

Investigations into Compressible Gas Behavior in a Vent Pipe during Blowdown, a

Thesis Outline ................................................................................................................ 4

Blowdown Process ......................................................................................................... 6

Blowdown of Pressure Vessel................................................................................ 7

Blowdown of Pipeline ............................................................................................ 8

Blowdown or Emergency Depressurising Systems ....................................................... 9

Pressure Safety Valve .......................................................................................... 10

Thermophysical Property ............................................................................................. 11

Blowdown Effects ........................................................................................................ 12

Investigations into Developed Simulation Codes and Models..................................... 13

Summary of Literature Review .................................................................................... 24

Development of Mathematical Models ........................................................................ 27

Basic Conservation Equations ............................................................................. 27

Theoretical Aspects Related to Compressible Fluid Flow Behavior ................... 31

Model Assumptions ............................................................................................. 32

Development of Adiabatic Frictional Model ....................................................... 34

Modeling Approach ..................................................................................................... 46

Simulation Object ................................................................................................ 46

Method of Solution .............................................................................................. 46

Experimental Design .................................................................................................... 53

Instrumentation and Data Collection ........................................................................... 55

Experimental Data Noise Reduction ............................................................................ 56

Experimental Analysis ................................................................................................. 61

Experimental Validation .............................................................................................. 62

Experiment VPM-1 .............................................................................................. 63

Experiment VPM-4 .............................................................................................. 67

Experiment VPM-7 .............................................................................................. 70

Validation with Hysys.................................................................................................. 75

Comparison with Hysys Simulation for Air......................................................... 75

Comparison with Hysys Simulation for Carbon-dioxide ..................................... 87

Comparison with Hysys Simulation for Methane ................................................ 96

Comparison with Hysys Simulation for DBNGP Mixture: ............................... 104

Conclusions ................................................................................................................ 112

Recommendations ...................................................................................................... 114

Specific heat at constant pressure

Specific heat at constant volume

Hydraulic diameter or pipe diameter

Gravitational constant (m/s2)

Critical length (m)

Mach number (-)

Mass flow rate (kg/s)

Universal gas constant

Adiabatic wall temperature

Compressibility factor (-)

Specific heat ratio

1.2 Thesis Outline

INTRODUCTION AND OBJECTIVES

RESULTS AND DISCUSSIONS

CONCLUSION AND RECOMMENDATION

Blowdown. According to API (American Petroleum Institute 2007), the depressurization of a

A vessel or pipeline opens completely immediately releasing its

The end of a pipe is fully opened exposing the CS area of the

Smaller diameter piping or valves may be opened or fail which