Design and Construction of Turboexpander based Nitrogen Liquefier Balaji Kumar Choudhury

Design and Construction of Turboexpander based Nitrogen Liquefier Balaji Kumar Choudhury
Design and Construction of
Turboexpander based Nitrogen Liquefier
A Thesis Submitted for Award of the Degree of
Doctor of Philosophy
Balaji Kumar Choudhury
Mechanical Engineering Department
National Institute of Technology
Rourkela 769008
Dedicated to my
PARENTS
NATIONAL INSTITUTE OF TECHNOLOGY
ROURKELA, INDIA
Ranjit Kr Sahoo
Professor
Sunil Kr Sarangi
Director
Mechanical Engg. Department
NIT Rourkela
NIT Rourkela
CERTIFICATE
Date: Dec 24, 2013
This is to certify that the thesis entitled ―Design and Construction of
Turboexpander based Nitrogen Liquefier‖, being submitted by Shri Balaji
Kumar Choudhury for the award of the degree of Doctor of Philosophy in
Mechanical Engineering, is a record of bonafide research carried out by him at
Mechanical Engineering Department, National Institute of Technology, Rourkela,
under our guidance and supervision. The work incorporated in this thesis has
not been, to the best of our knowledge, submitted to any other university or
institute for the award of any degree or diploma.
(Ranjit Kr Sahoo)
(Sunil Kr Sarangi)
Acknowledgement
I am extremely fortunate to be involved in an exciting and challenging research
project like ―Design and Construction of Turboexpander based Nitrogen
Liquefier‖. I have got an opportunity to look at the horizon of technology with a wide
view and to come in contact with people endowed with many superior qualities.
At First, I would like to express my deep sense of gratitude and respect to my
supervisors Prof. R. K. Sahoo and Prof. S. K. Sarangi for their excellent guidance,
suggestions and constructive criticism. I feel proud that I am one of their doctoral
students. I will always remember their helping hands and moral support in my good and
evil day during this period. The charming personality of Prof. Sarangi has been unified
perfectly with knowledge that creates a permanent impression in my mind. I am very
much inspired by the patience and confidence of Prof. R. K. Sahoo which will be fruitful
throughout my life. He also encouraged and stay with me during the period of testing
and commissioning of the plant. Without him I could not get confidence to do
experiment. I and my family members also remember the affectionate love and kind
support extended by Madam Sahoo and Madam Sarangi during our stay at Rourkela.
I take this opportunity to express my sincere gratitude to the members of my
doctoral scrutiny committee – Prof. K. P. Maity (HOD), Prof. A. K. Satatpathy, Prof. S.
Murugan of Mechanical Engineering Department and Prof. R. K. Singh of Chemical
Engineering Department for thoughtful advice and useful discussions. I am thankful to
prof. S. C. Mohanty, Prof. A. Satapathy and my other professors of the Mechanical
Engineering Department for constant encouragement and support in pursuing the Ph.D.
work.
I am indebted to Mr. Tilok Singh, Mr. Mukesh Goyal, Dr. Anindya Chhakravarty,
Mr. Rajendran S. Menon, Mr. Sandeep Nair of Cryogenic Division of Bhabha Atomic
Research Centre, Mumbai, for sharing their vast experience and provide kind support
regarding liquefaction plant and especially turboexpander. I am very much thankful to
Mr. N. Siva Rama Krishna and Mr. A. K. Pradhan of Central Tool Room and Training
Center, Bhubaneswar for understanding the requirements for fabricating the
turboexpander.
iv
I take this opportunity to express my heartfelt gratitude to Mr. Somnath Das
and Mr. Binaya kumar Kar for his cooperation and technical support to build the plant.
Beside them I am also grateful towards Mr. Jyana Ranjan Nayak, Mr. Naren Bisoi and
Mr. Pradeep kumar Mohanty for providing technical helping hand and Mr. H. Barkey for
assistance in official matters.
I record my appreciation and thanks for the helping hand extended by Dr.
Sidramappa Alur during my research work. I feel lucky to have Mr. Sachindra Kumar
Rout as my co-research fellow. I am also thankful to my friends Sanjay Kumar Swain,
Pankaj Kumar, Ajay kumar Gupta for their friendship during my stay at NIT Rourkela.
Thanks goes to my parents Sri Rajmohan Choudhury and Smt. Rajeswari
Choudhury, my brother Mr. Saroj kumar Choudhury, my sister in law Mrs. Mamata
Choudhury, my sisters and other relatives for their loving support and encouragement
for my PhD study. I am most grateful to my beloved wife Mrs. Prachetasi Choudhury for
her loving support and coperation.
I am happy with my daughter Padmini for not
disturbing during my PhD work.
(December 24, 2013)
Balaji Kumar Choudhury
v
Abstract
Cryogenic refrigerators are becoming increasingly popular particularly in the
areas of superconducting magnet applications, particle accelerators and medical
imaging systems, etc. It has also got wide applications in preservation of live biological
materials as well as in scientific equipment. In spite of nearly half a century of R & D
experience, our country is still dependent on imports for most of its needs in cryogenic
refrigerators and liquefiers. These components are enormously expensive to buy and to
maintain. The customers are often forced to buy equipment due to non-availability of
proprietary spares. It is imperative that our country develops an indigenous nitrogen
liquefier of capacity in the range 10 to 50 litre/hour. With the support from the
Department of Atomic Energy, our institute has initiated a programme on development
and study of a turboexpander based nitrogen liquefier of intermediate capacity (20 l/h).
The focus of this project is to build a turbine based liquid nitrogen generator of capacity
20 l/h using indigenous technology. This technology and expertise will be extended for
the liquefaction of helium in future.
The development of the turboexpander based nitrogen liquefier begins with the
process design of the cycle. The simulation of the cycles has been done using the
software Aspen HYSYS. All the state points are fixed and each equipment specifications
are determined. While designing the process, equipment availability, constraints and
cost is to be kept in mind. Process design also includes the setting the parameters up to
the optimum condition so that maximum amount of liquid will be obtained. After
process design the thermodynamics parameters of all the components are available.
As per process the nitrogen gas is compressed in the compressor upto 8 bar.
The compressed gas passes through the first heat exchanger. Some amount of the gas
is diverted through the turboexpander and remaining gas flow through the second heat
exchanger. A JT valve is used to expand the liquid which is collected in the phase
separator at a pressure just above ambient (1.2 bar). The vapour comes out of phase
separator mixes with the cold gas from the turboexpander and the resultant stream
meets at the suction side of the compressor, after passing through the second and first
heat exchanger as the reversed stream.
vi
The compressor unit is available in the laboratory which will discharge 336
nm3/hr of air and maximum working pressure is 10 bar. This is an oil injected twin
screw compressor.
Heat exchanger is a necessary component. Due to the requirement of high
effectiveness, two number of aluminum brazed plate fin heat exchangers are used. The
design of heat exchanger has been done using the software Aspen MUSE and also by
using the correlations by Maiti and Sarangi and Manglik and Bergles considering the size
and pressure drop as per the need of the process. The fabrication of the plate fin heat
exchangers has been done by APOLLO HEAT EXCHANGERS Pvt. Ltd.
The turboexpander is a vital component for the liquefier. Because it helps to
further lower down the temperature. A general design procedure is developed which will
able to design turboexpander with all pressure ratios. At first the turbine wheel is
designed followed by design of nozzles, diffuser, shaft, brake compressor, bearings and
other housing components. The fabrication of turbine wheel and brake compressor
wheel has been done by TURBOCAM Pvt. Ltd., Goa and rest of components has been
done by Central Tool and Training Center, Bhubaneswar.
The JT valve is also necessary for isenthalpic expansion. A suitable modification
has been done with a precession needle valve to operate as long stem JT valve. A
phase separator is designed and fabricated to separate the liquid nitrogen. All the
components are hanged inside a double walled vacuum insulated cold box.
The necessary pressure, temperature and flow measurement instruments are
mounted on the node points. Valves and safety devices are mounted on the liquefier.
Arrangement has been done to supply the gaseous nitrogen to the liquid nitrogen plant
from a liquid nitrogen tank through a LN2 vaporizer and a gas bag. After successful
running of the liquefier, for the flow rate 336 nm3/hr of gaseous nitrogen, it will delivers
17.44 lit/hr of liquid.
vii
Contents
Certificate
iii
Acknowledgement
iv
Abstract
vi
Contents
viii
List of Figures
xi
List of Tables
xv
Nomenclature
xvi
1.
2.
Introduction
1
1.1
Cryogenic refrigeration and liquefaction
1
1.2
Methods to produce low temperatures
2
1.3
Turboexpander
4
1.4
Requirement of liquid nitrogen
5
1.5
Objective of the work
7
1.6
Organization of the thesis
7
Literature Review
9
2.1
Introduction
9
2.2
Liquefaction of gases
9
2.3
Cryogenic liquefaction
10
2.4
Turboexpander for cryogenic liquefaction
11
2.5
Heat exchangers for cryogenic liquefaction
14
2.6
Design Methods of heat exchanger
18
2.7
Process design and simulation
19
2.8
Major industries supplying liquefaction plants
21
viii
3.
4.
5.
6.
7.
Process Design of Nitrogen Liquefaction Cycle
24
3.1
Introduction
24
3.2
Description of process cycles
25
3.3
Simulation of process cycles
26
3.4
Calculation of process parameters of selected cycle
29
3.5
Parametric analysis of selected cycle
32
3.6
Performance of nitrogen liquefaction plant
36
Design of Heat Exchanger
39
4.1
Introduction
39
4.2
Plate fin heat exchanger design procedure
39
4.3
Design of first heat exchanger
46
4.4
Design of second heat exchanger
48
Design of Turboexpander
50
5.1
Introduction
50
5.2
Design of turbine wheel
52
5.3
Design of nozzles
59
5.4
Design of diffuser
63
5.5
Design of shaft
65
5.6
Design of brake compressor
66
5.7
Selection of journal and thrust bearing
71
5.8
Supporting structures
75
5.9
Other turboexpander components
77
Assembly and Instrumentation
80
6.1
Available equipment
80
6.2
Fabricated components
82
6.3
Instrumentation
88
6.4
Assembly of components
90
Testing and Commissioning of the Liquefier
ix
94
8.
7.1
Introduction
94
7.2
Testing of turboexpander
94
7.3
Plant pipeline setup
98
7.4
Commissioning of the plant
103
7.5
Performance of the plant
103
Conclusions
105
References
107
Appendices
A.
Production Drawings of Turboexpander
118
B.
Production Drawings of Heat exchanger
149
Curriculum Vitae
155
x
List of Figures
Page No.
Figure 1-1 Joule Thomson inversion curve [1] ............................................................ 3
Figure 1-2 Shaft with brake compressor and turbine wheel ......................................... 5
Figure 2-1 Concentric tube heat exchanger [1] ......................................................... 15
Figure 2-2 Tube heat exchanger with wire spacer [1] ............................................... 15
Figure 2-3 Multi tube heat exchanger [1] ................................................................. 16
Figure 2-4 The Giauque Hampson heat exchanger [52] ............................................. 16
Figure 2-5 The Collins heat exchanger [52] .............................................................. 16
Figure 2-6 Perforated plate heat exchanger [52] ...................................................... 17
Figure 2-7 Plate fin heat exchanger [52] .................................................................. 18
Figure 3-1 Schematic diagram of Claude cycle (Case-1) ............................................ 25
Figure 3-2 Schematic diagram of modified Claude cycle (Case-2) ............................... 25
Figure 3-3 Schematic diagram of modified Claude cycle (Case-3) ............................... 26
Figure 3-4 Yield at different pressures in the three cases .......................................... 28
Figure 3-5 Heat load of heat exchangers at different pressures in case-1 ................... 28
Figure 3-6 Process flow diagram .............................................................................. 29
Figure 3-7 Pinch point of HX-2 ................................................................................. 30
Figure 3-8 Expansion in Turboexpander ................................................................... 31
Figure 3-9 Variation of yield with the change of mass fraction through turboexpander at
different operating pressures ................................................................................... 33
Figure 3-10 Variation of yield with effectiveness of HX-1 at different operating pressures
............................................................................................................................. 34
Figure 3-11 Variation of yield with pinch point of HX-2 at different operating pressures
............................................................................................................................. 34
Figure 3-12 Variation of yield with turboexpander efficiency at different operating
pressures ............................................................................................................... 35
Figure 3-13 Variation of compressor work per liquid mass produced with operating
pressure ................................................................................................................. 35
xi
Figure 3-14 Temperature entropy diagram of Nitrogen liquefier ................................. 38
Figure 4-1 Geometry of a typical offset strip fin surface ............................................ 40
Figure 4-2 Dimension of first plate fin heat exchanger .............................................. 47
Figure 4-3 Condensing h.t.c for nitrogen as a function of temperature difference [114]
............................................................................................................................. 48
Figure 4-4. Dimension of second plate fin heat exchanger ......................................... 49
Figure 5-1 Longitudinal section of Turboexpander ..................................................... 51
Figure 5-2 State points at nozzles, turbine wheel and diffuser ................................... 52
Figure 5-3 Velocity diagrams for turbine ................................................................... 56
Figure 5-4 Turbine wheel ........................................................................................ 59
Figure 5-5 Major Dimensions of Nozzle .................................................................... 60
Figure 5-6 Stagger angle deviation graph for different cascade angle [120] ............... 63
Figure 5-7 Nozzle Diffuser ....................................................................................... 64
Figure 5-8 Nozzle cover........................................................................................... 64
Figure 5-9 Shaft ..................................................................................................... 65
Figure 5-10 Brake compressor ................................................................................. 71
Figure 5-11 Pad ...................................................................................................... 72
Figure 5-12 Pivot less tilting pad journal bearing....................................................... 73
Figure 5-13 Pad and Rotor geometry ....................................................................... 73
Figure 5-14 Aerostatic thrust bearing ....................................................................... 75
Figure 5-15 Exhaust gas plate ................................................................................. 75
Figure 5-16 Cold end housing .................................................................................. 76
Figure 5-17 Bearing housing .................................................................................... 76
Figure 5-18 Warm end housing ............................................................................... 77
Figure 5-19 Principle of Labyrinth Sealing ................................................................. 78
Figure 5-20 Labyrinth Seal ...................................................................................... 78
Figure 5-21 Thermal Insulation ................................................................................ 79
Figure 5-22 Spacer ................................................................................................. 79
xii
Figure 5-23 Lock Nut turbine Side ............................................................................ 79
Figure 5-24 Lock Nut compressor Side ..................................................................... 79
Figure 6-1 Photograph of the compressor ................................................................ 81
Figure 6-2 Arrangement for regulating the pressure and flow rate ............................. 81
Figure 6-3. Photograph first heat exchanger ............................................................. 82
Figure 6-4. Photograph second heat exchanger ....................................................... 82
Figure 6-5 Long stem handle assembly for the J-T expansion valve ........................... 85
Figure 6-6 Photograph of expansion valve ................................................................ 85
Figure 6-7 Dimensions of phase separator ................................................................ 86
Figure 6-8 Cover plate of phase separator ................................................................ 86
Figure 6-9 Photograph cold box ............................................................................... 87
Figure 6-10 Holes on the cold box flange ................................................................. 87
Figure 6-11 Photograph of the RTD ......................................................................... 88
Figure 6-12 Orifice plate calibration ......................................................................... 89
Figure 6-13 Photograph of the accelerometer used for speed measurement ............... 89
Figure 6-14 Connection of turboexpander with the pipelines ...................................... 90
Figure 6-15 P & I diagram of Nitrogen liquefier......................................................... 91
Figure 6-16 3-D model assembly of nitrogen liquefier inside cold box ......................... 92
Figure 6-17 Assembly photograph of nitrogen liquefier.............................................. 93
Figure 6-18 Photograph of cold box flange ............................................................... 93
Figure 7-1 Turboexpander test set up ...................................................................... 95
Figure 7-2 Damaged surface of the thrust bearing .................................................... 96
Figure 7-3 Damaged surface of the shaft collar......................................................... 96
Figure 7-4 Damaged shaft surface by rubbing with tilting pad bearing ....................... 97
Figure 7-5 Filter used to remove micron dust particles .............................................. 97
Figure 7-6 FFT diagram for the speed of turbine wheel at 5 bar of inlet pressure ........ 97
Figure 7-7 FFT diagram for the speed of turbine wheel at 6 bar of inlet pressure ........ 98
xiii
Figure 7-8 LN2 Dewar to vaporizer .......................................................................... 99
Figure 7-9 Gas bag for gaseous nitrogen ................................................................ 100
Figure 7-10 Oil safety valve ................................................................................... 100
Figure 7-11 Coil heat exchanger for pre-cooling...................................................... 101
Figure 7-12 Arrangement for supply of process gas to cold box and turbine bearing . 101
Figure 7-13 P & I Diagram of the liquid nitrogen plant ............................................ 102
Figure 7-14 Temperature monitoring and recording using data acquisition system .... 103
Figure 7-15 Turboexpander exit temperature with time ........................................... 104
xiv
List of Tables
Page No.
Table 1-1 Maximum Inversion Temperature of Cryogenic Fluids .................................. 3
Table 3-1 Basic specifications of liquefier components .............................................. 37
Table 3-2 Thermodynamic state points of the process cycle ...................................... 37
Table 4-1 Thermal data for First Heat exchanger ...................................................... 46
Table 4-2 Fin specifications for first heat exchanger .................................................. 46
Table 4-3 Overall dimension of first heat exchanger .................................................. 47
Table 4-4 Thermal data of second heat exchanger .................................................... 48
Table 4-5 Fin specifications for second heat exchanger ............................................. 49
Table 5-1 Basic input values for turboexpander design .............................................. 50
Table 5-2 Basic input parameters for design of brake compressor .............................. 66
Table 5-3 Input parametrs to determine pad geometry ............................................. 73
Table 5-4 Pad geometry .......................................................................................... 74
Table 5-5 Aerostatic thrust bearing input parameters ................................................ 74
Table 5-6 Aerostatic thrust bearing clearance at load and no load ............................. 75
Table 6-1 Specification of the compressor ................................................................ 80
Table 6-2 Balancing report of shaft .......................................................................... 83
Table 6-3 Specification of the accelerometer ............................................................ 90
xv
Nomenclature
aff
=
Free flow area/fin (m2)
afr
=
Frontal area/fin (m2)
af
=
Fin surface area (m2)
as
=
Heat transfer area/fin (m2)
ap
=
Plate thickness (m)
aw
=
Total wall cross sectional area for longitudinal conduction (m2)
Afr
=
Frontal area available for heat exchanger (m2)
Aff
=
Free flow area available for heat exchanger (m2)
As
=
Heat transfer area of the heat exchanger (m2)
Aw
=
Total wall area for transverse heat conduction (m2)
A
=
cross sectional area normal to flow direction (m2)
b
=
height (nozzle, wheel blade) (m)
bp
=
distance between heat exchanger plates (m)
Cn
=
chord length of nozzle (m)
C
=
absolute velocity of fluid stream (m/s)
C0
=
spouting velocity (m/s)
Cd
=
Coefficient of discharge (dimensionless)
CP
=
specific heat at constant pressure (J/kg K)
Cs
=
velocity of sound (m/s)
Ch
=
heat capacity rate hot side of heat exchanger (W/K)
Cc
=
heat capacity rate cold side of heat exchanger (W/K)
C min =
Minimum of hot and cold capacity ratio (W/K)
Cr
=
Heat capacity rate ratio (dimensionless)
d
=
diameter (shaft) (m)
ds
=
specific diameter (dimensionless)
D
=
diameter (m)
De
=
Equivalent diameter of the flow passage (m)
E
=
Young‘s modulus (N/m2)
xvi
fq
=
vibration frequency (Hz)
f
=
Fanning friction factor (dimensionless)
ff
=
Fin frequency, Number of fins per meter length (fins/m)
G
=
Core mass velocity (kg/m2s)
h
=
enthalpy (J/kg)
hf
=
Height of fins (m)
hconv =
Convective heat transfer coefficient (W/m2 K)
H hx
=
No flow height (stack height) of the heat exchanger core (m)
j
=
The Colburn factor (dimensionless)
k1
=
Pressure recovery factor (dimensionless)
k2
=
Temperature and Density recovery factor (dimensionless)
Kf
=
Conductivity of the fin material (W/m- K)
Kw
=
Conductivity of the wall material (W/m- K)
M
=
Mach number (dimensionless)
m
=
mass of nitrogen delivered from compressor (kg/s)
mf
=
Rate of mass of liquid nitrogen produced (kg/s)
mtr
=
mass of nitrogen gas diverted through turboexpander (kg/s)
N
=
Rotational speed (rpm)
ns
=
specific speed (dimensionless)
nl
=
Total number of layers or total number of fluid passages (dimensionless)
N tu
=
Number of heat transfer units, UA / C min (dimensionless)
Lhx
=
Fluid flow (core) length on one side of the heat exchanger (m)
lf
=
Fin flow length on one side of a heat exchanger (m)
P
=
power output of the turbine (W)
p
=
pressure (N/m2)
p
=
Pressure drops (Pa)
pf
=
Fin pitch (m)
pr
=
Prandtl number of the fluid (dimensionless)
Q
=
volumetric flow rate (m3/s)
QH
=
Heat load (W)
R
=
Gas constant of the working fluid (J/kg K)
xvii
r
=
Radius (m)
Re
=
Reynolds number (dimensionless)
Re* =
Critical Reynolds number (dimensionless)
s
=
Specific entropy (J/kg K)
sf
=
Spacing between adjacent fins (m)
tf
=
Thickness of fin (m)
T
=
Temperature (K)
t
=
Blade thickness (m)
U
=
Blade velocity (in tangential direction) (m/s)
Uo
=
Overall heat transfer coefficient (W/m2 K)
x
=
dryness fraction (dimensionless)
W
=
velocity of fluid stream relative to blade surface (m/s)
wn
=
nozzle width (m)
Whx
=
Width of the core (m)
y
=
yield, mass of liquid produced per mass of gas compressed (dimensionless)
Z
=
number of blades (dimensionless)
Greek symbols

=
absolute velocity angle (radian)
t
=
throat angle (radian)
0
=
inlet flow angle (radian)
tr
=
mass fraction of nitrogen diverted through turboexpander (dimensionless)

=
relative velocity angle (radian)

=
specific heat ratio (dimensionless)

=
Effectiveness of heat exchanger (dimensionless)

=
dynamic viscosity (Pa s)

=
density (kg/m3)

=
rotational speed (rad/s)

=
tangential coordinate (dimensionless)

=
Inlet turbine wheel diameter to exit tip diameter ratio (dimensionless)
tr
=
Hub diameter to tip diameter ratio (dimensionless)
xviii

=
Longitudinal conduction parameter, dimensionless

=
Ratio of free flow area to frontal area (dimensionless)

=
Isentropic efficiency (dimensionless)
T st =
total-to-static efficiency (dimensionless)
T T =
total-to-total efficiency (dimensionless)
f
=
Fin efficiency (dimensionless)
o
=
Overall surface effectiveness of the extended fin surfaces (dimensionless)
Subscripts
0
=
stagnation condition
ad
=
adiabatic
m
=
meridional direction
r
=
radial direction
s
=
isentropic
t
=
throat
tr
=
turbine

=
tangential direction
w
=
Wall or properties at the wall temperature
h
=
Hot fluid side
c
=
Cold fluid side
i
=
Inlet
o
=
Outlet
m
=
mean
max =
Maximum
min =
Minimum
hub =
hub of turbine wheel at exit
tip
tip of turbine wheel at exit
=
mean =
average of tip and hub
xix
1. Chapter I
Introduction
1.1 Cryogenic refrigeration and liquefaction
The literal meaning of ―cryogenics‖ is production of icy cold or low temperature.
A logical dividing line has chosen by the workers at National Bureau of Standards at
Boulder, Colorado for the field of cryogenics is below 123 K. The normal boiling points
of the permanent gases, such as oxygen, air, nitrogen, neon, hydrogen, helium lie
below 123 K.
In a thermodynamic process when the process fluid absorbs heat at
temperatures below that of the environment is called refrigeration. Liquefaction of
gases is always accomplished by refrigerating the gas to the temperature below its
critical temperature so that liquid can be formed at some suitable pressure below the
critical pressure. Thus gas liquefaction is a special case of gas refrigeration and cannot
be separated from it. In both cases, the gas is first compressed to an elevated pressure
in an isothermal compression process. This high-pressure gas is passed through a
countercurrent recuperative heat exchanger to a throttling valve or expansion engine.
Upon expanding to the lower pressure, cooling takes place, and leads to formation of
liquid. The cold, low-pressure gas returns to the compressor inlet to repeat the cycle.
The purpose of the countercurrent heat exchanger is to warm the low-pressure gas
prior to recompression and simultaneously it cools the high-pressure gas to the lowest
temperature possible prior to expansion. Both refrigerators and liquefiers operate on
this basic principle.
There is no accumulation of refrigerant in any part of the system in a continuous
refrigeration process. But in a gas liquefying system, the liquid accumulates and is
withdrawn. Thus, in a liquefying system, the total mass of gas that is warmed in the
1
countercurrent heat exchanger is less than that of the gas to be cooled by an amount
liquefied, creating an imbalance of mass flow in the heat exchanger. In a refrigerator
the warm and cool gas flows are equal, creating a balanced flow in the heat exchanger.
The thermodynamic principles of refrigeration and liquefaction are identical. However
the analysis and design of the two systems are quite different because of the condition
of balanced flow in the refrigerator and unbalanced flow in liquefier systems.
1.2 Methods to produce low temperatures
Based on the method of production of low temperature, cryogenic refrigeration
and liquefaction cycles can be grouped under three broad categories.
(i) Process with J-T valve.
(ii) Process with expansion engine or turbine.
(iii) Process with regenerative cycles.
(i) Process with J-T valve
The most usual process for production of low temperature is isenthalpic
process by using JT Valves. The operation of JT valve depends upon JouleThomson coefficient which is a gas property. It is the effect of change in
temperature with change in pressure under constant enthalpy. The JouleThomson coefficient can be expressed as,
 T 

 p h
JT  
(1.1)
It is a function of temperature and pressure. The isenthalpic curves are
shown in Figure 1-1. The slope of the curve gives the Joule-Thomson
coefficients and this may be positive, negative, or zero. When Joule-Thomson
coefficient is zero, then that point is called inversion point. The locus of such
points different enthalpy forms the inversion curve. The area inside the inversion
curve gives cooling effect by isenthalpic expansion while area outside the
inversion curve has reverse effect. The Table 1-1 gives the maximum inversion
temperature of some cryogenic fluids. Nitrogen has maximum inversion
temperature of 622 K which is well above atmospheric temperature. So for
nitrogen, JT valve can be used at atmospheric conditions to decrease its
temperature. But for neon, hydrogen and helium, the JT valve can only be used
if their temperature has cooled below to their inversion temperature by using
2
other precooling methods. Vapor compression cycle, cascade vapor compression
cycle, Mixed refrigerant cascade cycle, Linde cycle etc., are the examples which
uses only J-T Valve as expansion device.
Figure 1-1 Joule Thomson inversion curve [1]
Table 1-1 Maximum Inversion Temperature of Cryogenic Fluids
Fluid
Maximum inversion
temperature (K)
Oxygen
761
Argon
722
Nitrogen
622
Air
603
Neon
250
Hydrogen
202
Helium
40
(ii) Process with expansion engine or turbine
Another method of producing low temperatures is the adiabatic
expansion of the gas through a work-producing device such as an expansion
engine. In the ideal case, the expansion would be reversible and adiabatic and
therefore isentropic. In this case, one can define the isentropic expansion
3
coefficient which expresses the temperature change due to a pressure change at
constant entropy. The isentropic expansion process removes energy from the
gas in the form of external work, so this method of low-temperature production
is sometimes called the external work method.
In any liquefier, the expansion engine or turbine could not be used alone
without J-T valve. Because it is difficult to produce wet expansion turbine but a
J-T Valve could handle two phase easily. Therefore most cycles use combination
of both the expansion methods. Claude cycle, Brayton cycle, Kapitza cycle,
Heylandt cycle, Collins cycle are using both J-T valve and expansion turbine for
refrigeration and liquefaction.
(iii) Process with regenerative cycles
The Process uses regenerative type of heat exchanger. The regenerative
type of heat exchanger has single set of flow passages through which hot and
cold fluid passes alternately and continuously. Refrigerators and liquefiers with
regenerative heat exchangers are Stirling, Pulse tube, Gifford-McMahon etc. This
class of cycle uses working fluid such as helium and a condenser for
refrigeration/liquefaction of gases including helium.
1.3 Turboexpander
Generally, the word “Turboexpander” is used to define an expander and a
compressor as a single unit. It consists two primary components i.e., radial or mixed
flow expansion turbine wheel and a centrifugal compressor wheel. Both the wheels are
connected by a single shaft as shown in Figure 1-2. The high pressure process gas
flows through the turbine wheel to produce power and cause rotation of the shaft by
the expense of the kinetic energy. The centrifugal compressor acts as a loading device
and is used to extract work output of the turbine. Generally, the shaft is mounted in
vertical orientation to reduce the radial load on the bearings. Two number of radial
journal and two number of axial thrust bearings are used to keep the shaft in proper
alignment and to absorb the radial and axial load.
4
Figure 1-2 Shaft with brake compressor and turbine wheel
1.4 Requirement of liquid nitrogen
In 1772, David Rutherford discovered the chemical element nitrogen. It is a
colourless, odourless, tasteless and unreactive gas. It is the most abundant gas in the
air and constitutes 78% of air.
The liquid nitrogen have low production cost and relatively higher levels of
safety, it is the most common cooling medium in the cryogenic temperature range
above 77 K. The application covers such diverse areas as:
(i)
Pre coolant: The liquid nitrogen has low temperature up to 77 K. So it is
used for precooling the helium to bring the temperature of helium down.
After lowering the temperature it is used in any usual cycle to produce liquid
5
helium. It is not only used for precooling helium but also for other gases like
hydrogen, Neon etc.
(ii)
Coolant: Due to its very low temperature it is used as a coolant in many
industrial, medical and laboratory instruments. In NMR for Medical imaging
system it is used as coolant.
(iii)
Cryo-treatment: The process of treating the metals at cryogenic temperature
is known as cryo-treatment. Metallic components such as hubs, milling
cutters, knives, rollers, needles, dies and punches, bearings and precision
measuring
equipment
undergo
cryo-treatment
for
advancement
of
mechanical properties. The cryo-treatment renders improved mechanical
properties, such as longer life, less failure due to cracking, improved thermal
properties, better electrical properties with less electrical resistance, reduced
coefficient of friction, improved flatness, fine grain structure etc.
(iv)
Cryo-Preservation: Preservation of live biological material such as blood,
animal and human sperms, embryos, bacterial cultures etc. using liquid
nitrogen is known as cryo-preservation. It is the safest method for
preservation that can be freezed up to one decade.
(v)
Shrink fitting and Press freeing: It is a cost efficient method of assembling
and disassembling new and replacement of fine tolerance components. The
components with fine tolerance can be assembled by putting the parts inside
liquid nitrogen. Due to low temperature it will shrink a little and then it may
be fitted with another part.
(vi)
When a component is press-jammed, then liquid nitrogen can be used to
freeing the jam. It saves the production halt time, money and save the
component.
(vii) Cryotherapy: As liquid nitrogen has extreme cold temperature, any cells that
are touched by it will be instantly frozen. After freezing cells will die and fall
off. This allows liquid nitrogen to be an effective treatment for wart
removal or the removal of small skin cancers. Cryosurgery also done by
using the liquid nitrogen.
(viii) Food preparation and preservation: There are a number of dishes could be
made using liquid nitrogen. Quick ice could be made from liquid nitrogen.
The foods are stored using liquid nitrogen.
6
(ix)
Cold trap in vacuum systems and in adsorption pumps
(x)
It is used as low temperature dielectric and susceptibility measurement. Due
to its inertness property and low temperature, it is used in many chemical
applications. Apart from this it has miscellaneous laboratory and industrial
applications.
1.5 Objective of the work
Cryogenic refrigeration and liquefaction plants are enormously expensive to buy
and to maintain and owners are often forced to buy new plants due to non-availability
of proprietary spares. It is the need of our country to develop an indigenous nitrogen
liquefier of capacity in the range 10 to 50 litres/hour.
With the support from the Department of Atomic Energy, National Institute of
Technology, Rourkela has initiated a programme on development of turboexpander
based cryogenic refrigerator and liquefier of capacity in the range of 10 to 50
liters/hour. The objectives of this work are as follows:
(i) Process design and freezing the process parameters based on aspen software.
(ii) Design and fabrication of major components such as plate fin heat exchanger,
turboexpander, JT Valve etc.
(iii) Assembly of the nitrogen liquefier by connecting with pipelines and mounting
with safety and measuring instruments.
(iv) Commissioning and performance study of the liquefier.
(v) Proper documentation of the components design with fabrication and assembly
drawing.
1.6 Organization of the thesis
The thesis has been arranged in eight chapters and appendices. Chapter I deal
with a general introduction to cryogenic refrigeration and liquefaction processes. The
methods to obtain the cryogenic temperatures along with general cryogenic cycles are
described. It shows the properties of nitrogen and focuses the requirement of liquid
nitrogen and finally it defines the objective of the present work.
Chapter II is the literature review part of the thesis. It describes history of
cryogenic liquefaction and development of turboexpander. It points out some major
7
suppliers of liquid nitrogen plants. It also focuses on the process design techniques,
thermodynamic equations for plant performance.
Chapter III presents process design of the turboexpander based nitrogen
liquefaction cycles with optimum state points and component specification. These state
points and component specification are used for the design of other components.
Chapter IV includes the design of two plate fin heat exchangers as per the
requirement for the nitrogen liquefaction plant. The design of the heat exchangers is
made by following a general design procedure using different corelations available in
open literature. In addition Aspen MUSE software is used to validate the dimensions and
pressure drop.
Chapter V comprises with the design of the turboexpander. It consists of the
design of turbine wheel, nozzle, diffuser, shaft, brake compressor, bearings and
supporting components.
Chapter VI illustrates the fabrication of remaining components used in the
liquefier. It also covers assembly of the fabricated components and instrumentation.
Chapter VII shows testing performance of the turboexpander. It also includes operation
and performance study of the liquid nitrogen plant.
Chapter VIII presents the concluding remarks and recommendation for future
work. And finally references are presented which utilized to develop the turboexpander
based nitrogen liquefaction plant. It consists of appendices which contain fabrication
drawings and photographs of the turboexpander parts, heat exchanger and other
components of the plant.
8
2. Chapter II
Literature Review
2.1 Introduction
The chapter describes the innovation of gas liquefaction techniques and focuses
on the chronological development of cryogenic liquefaction plants. The commonly used
components in the liquefaction plant are turboexpander and heat exchanger. So the
development and use of turboexpander in the cryogenic liquefaction plants are
described. This chapter also explains the type of heat exchangers used for cryogenic
application and their design methods. This chapter also emphasizes on the process
cycle design methods and software for process simulation. It also highlights some major
cryogenic industries for supplying liquefaction plants.
2.2 Liquefaction of gases
Wolfgang [2] reported the history of liquefaction of common gases as well as
the permanent gases. For the first time, at around 1780 Louis Clouet and Gaspard
Monge was successfully liquefied a real gas (SO2) by compressing and cooling. After the
liquefaction of SO2, Ammonia gas was liquefied by Martinus van Marum and Adriaan
Paets van Troostwijk in 1787. But Fourcroy and Vauquelin was able to liquefy ammonia
at ambient pressure in 1799 and Guyton de Morveau in 1804. In 1823, Michael Faraday
published the liquefaction techniques of SO2, H2S, CO2, N2O , C2H2, NH3 and HCl. Again
in 1845, Faraday had published his second paper regarding gas liquefaction. At that
time Faraday had much better equipment to compress the gases up to 40 bar and using
refrigerating bath to liquefy them to -110 °C. By using this technique he was able to
liquefy a number of gases. But he was unable to liquefy CH4, O2, CO, N2, NO and H2.
Those gases were, therefore, called ‗‗permanent gases‘‘. The Viennese physician
9
Johannes Natterer tried to liquefy the permanent gases by compressing them up to 360
bar but unable to lower the temperature simultaneously below the critical temperature.
2.3 Cryogenic liquefaction
The general information about the cryogenic liquefaction are available in
standard text books [1, 3-5]. A brief history of cryogenics has been described by
Scurlock [6] in 1989.
In 1877, Louis Cailletet in Paris and at the same time Raoul Pictet in Geneva
attempted to liquefy ‗‗permanent‘‘ gases. Louis Cailletet had compressed oxygen gas
with a hand operated screw jack up to 200 bar and then cooled to -110° C by enclosing
the strong walled glass tube apparatus with liquid ethylene. Suddenly it was expanded
by releasing the pressure, he observed a momentary fog of oxygen droplets inside the
glass tube. Raoul Pictet had used the cascaded refrigeration system to liquefy oxygen.
He used sulphur dioxide and liquid carbon dioxide in the heat exchanger to cool the
oxygen gas. He used two numbers of compressors to drive sulphur dioxide and carbon
dioxide. The gas was expanded by opening the valve at the end of heat exchanger. By
doing so a transitory jet of liquid oxygen was formed. But both were unable to collect
the liquid oxygen.
In 1883, further improvement Cailletet's apparatus had been done by the Polish
scientists Olzewski and Wroblewski, at Cracow. They added an inverted U glass tube
and reduced the ethylene temperature to - 136°C by pumping it below atmospheric
pressure. These modifications enabled them to produce small quantities of liquid oxygen
in the U tube and to liquefy carbon monoxide and nitrogen for a few seconds. But this
production of oxygen was discontinuous and the quantities of liquid produced were still
very small.
In 1895 the air was liquefied by Carl von Linde in Munich and William Hampson
in London. Air was compressed up to 200 bar and cooled to ambient temperature using
water cooler. The precooled air was fed into counter flow coiled heat exchanger. Linde
has achieved isenthalpic expansion by utilizing the Joule-Thomson effect from a JTvalve. After expansion, liquid was collected but it took three days to cool down the
system and achieve steady-state. The yield of the air liquefier is approximately 3 liters
per hour. William Hampson had also achieved the success of liquefaction in the same
10
year in the similar process of Linde. He gave license to Brins Oxygen Company and
supplied liquefaction plants to several scientific institutions.
By 1897, Charles Tripler in New York also had built a similar but larger
liquefaction plant, which was capable to produced 25 liters per hour [6].He had used a
75 kW steam engine to drive the compressor. In 1898, Sir James Dewar liquefied
hydrogen by using the same technique of Linde. He had precooled hydrogen with liquid
air. Hydrogen was compressed to 180 bar and cooled with counter-current heat
exchanger and finally expanded using a JT-valve.
In 1902, George Claude a scientist of France had improved the Linde process by
adding two extra heat exchangers and an expansion engine. This was the first time an
expansion engine used in a liquefaction cycle successfully. The expansion engine was
reciprocating type.
In 1907, Linde installed the first air liquefaction plant in America. Kamerlingh
Onnes build up a cryogenic laboratory at the leiden on the Netherlands in 1895 but in
1908 he was successfully liquefied the Helium gas. There after a lot of liquefaction
plants for air, neon, helium were developed for commercial purpose and installed.
Another remarkable breakthrough was made when Kapitza developed a rotary
expansion engine for helium in 1934. And in 1939 Kapitza modified the basic Claude
system by eliminating the third or low temperature heat exchanger. He used the rotary
expansion engine instead of a reciprocating expander and a set of valved regenerators
instead of recuperators.
Around 1942 Samuel C. Collins developed an efficient liquid helium laboratory
facility. He developed Collins helium cryostat resulted to economical and safe production
of liquid helium. Furthermore a large liquefaction plants were developed to get large
amount of liquid. The capacity of large liquefaction plants was more than 100 ton/day.
The increase in the efficiency of the turboexpander and increase in effectiveness of the
heat exchanger had lead to the better efficiency of the plant. The basic process cycles
remaining same attention was focused to develop and increase the efficiency of the
components of the plant, i.e. expansion engines and heat exchangers.
2.4 Turboexpander for cryogenic liquefaction
The fundamental principles and governing equation can be found in several text
books and reports of fluid mechanics and Turbo-machinery [7-9]. A detail review on
11
turboexpander development was presented by Collins and Cannaday [10] and Sixsmith
[11]. In 1898, Lord Rayleigh [10] was the first person to introduce the concept of
turbine and it could be used to produce low temperature in liquefaction cycles.
Considering this suggestion some patents were made on expansion turbine. Among
them the patents of Edgar C. Thrupp, Joseph E. Johnson, Charles and Commett, Davis
are important. But the commercial development of expansion turbine for gas
liquefaction was done by Linde Works in Germany at around 1934 [11]. It was an axial
flow single stage impulse turbine. Guido Zerkowitz a scientist of Italy modified the
turbine and made it radial flow of impulse cantilever type. The rotational speed of the
turbine was 7000 rpm.
In the year 1939, a Russian physicist Peter Kapitza, had made a low pressure
cycle with expansion turbine. He made some revolutionary conclusions in paper
published in journal of Physics [12]. He proved by giving thermodynamic reason that a
low pressure liquefier using an expansion turbine is better than a high pressure liquefier
using a reciprocating expander. And the cost of low pressure plant is also very cheap.
He also concluded by doing both analytical and experimental studies that a radial inflow
turbine would be preferable to an axial impulse type machine. The turbine was rotating
at a speed of 40,000 rpm and measured efficiency was 80%.
Swearingen [13] described about a radial inflow, reaction type turbine that was
designed by Elliot Company and constructed by the Sharples company in USA. The
turbine was supported on ball bearings and design speed is 22,000 rpm. The selection
of turbine [14] depends on the parameters like specific speed (ns) and flow coefficient
() etc. From this it was concluded that for low flow rate and medium head, the radial
inflow configuration gives the maximum efficiency [15] and it became a standard
configuration of turbine. Linhardt [16] was also designed a large power output
turboexpander and studied the influence of design parameters.
Further much smaller and high speed turbines was made in about 1950 at the
University of Reading, England by Sixsmith [17]. The diameter of turbine wheel was
14.28 mm and designed speed was 240,000 rpm. The financial support for the
development of this turbine was given by British National Research and Development
Corporation. This turbine was employed as a source of refrigeration in a small air
liquefier. Based on this design, in 1959 the British Oxygen Company (BOC)
manufactured expansion turbines for application in air separation plants. BOC also built
12
the world‘s first commercial turbine-based helium liquefier for the Rutherford Laboratory
in Oxford. By 1958, the Lucas Company in England had developed a range of gas
lubricated radial inward flow turbines for Petrocarbon Development Corporation [18].
The Cryogenic Engineering Division of National Bureau of Standards had followed the
work of Reading University and developed a helium expansion turbine. This was used in
a helium refrigerator in 1964. The design speed was about 600,000 rpm while it was
rotating at about 720,000 rpm and maximum efficiency was 79.8 %.
In Winthertur of Switzerland, Sulzer Brothers [19-21] had constructed a small
turbine with gas lubricated bearings. The La Fleur Corporation [22], Lucas Corporation
[18] and Linde, Germany had also developed expansion turbine with gas lubricated
bearing. The General Electric Company, USA also worked on cryogenic refrigeration
systems based on miniature turbomachine [23, 24]. In 1962, L‘Air Liquide had
developed a radial inflow turbine with high reliability and had high performance having
expansion ratios up to 15.
Izumi et. al [25] at Hitachi Ltd., Japan had described the development of a
turboexpander for a small helium refrigerator. Yang et. al [26-29] had discussed about
the development of miniature turbines at the Cryogenic Engineering Laboratory of the
Chinese Academy of Sciences. Naka Fusion Research Centre of Japan Atomic Energy
Institute [30-32] had developed large turboexpanders for Helium liquefier. A high
expansion ratio (15.3 bar to 1.2 bar) helium turboexpander was developed by Ino et al.
[33, 34] for a 70 MW superconducting generator. The turbine impeller was backward
swept and was rotating at 230,000 rpm. Further development was continued by
Davydenkov [35-37] at around 1990s. Mikrokryogenmash company of Russia [38] had
developed small turboexpander for micro-cryogenic systems.
Beasley and Halford [18] had developed a nitrogen turbine supported on gas
bearings. Linde Division of Union Carbide Corporation had also developed turbines with
gas bearing for large air separation. Kun et. al. [39-41] had described a detailed design
methodology which has become a guideline for researchers. In collaboration with
NASA/GSFC , Sixsmith et.al. [42] had developed miniature turbines for Brayton Cycle
cryocoolers. The turbine had 31.8 mm in diameter and designed to rotate at a speed of
570,000 [42].
In 1985, Central Mechanical Engineering Research Institute (CMERI) Durgapur
of India has developed an expansion turbine [43]. It is a radial inflow radial turbine and
13
is used for the nitrogen plants. They also suggested advantage of rotary expansion
engine over reciprocating engine[44].The turboexpander developed was rotating at
30,000 rpm without any vibration. The reliability and machining process had been
discussed for the indigenous developed turbine [45].Then further improvements in
efficiency had been noticed for cryogenic turboexpander [46, 47].
In 2002, turboexpander was developed and tested at IIT Kharagpur [48], India.
They have developed high stabilized gas bearings. The turbine was designed for
140,000 rpm and tested to run at 80,000 rpm. Due to the increase in vibration
amplitude above 80,000 rpm they were unable to go beyond that speed.
In 2008, a radial turboexpander was developed at NIT Rourkela [49]. It was
designed to work with air or nitrogen to rotate at a speed of 220,000 rpm. Aerodynamic
bearings were used for the turboexpander. Recently BARC, Mumbai has developed high
speed miniature helium turboexpander [50, 51] which rotates at 2,70,000 rpm. The
turbine is tested and found to have efficiency of 65 %.
2.5 Heat exchangers for cryogenic liquefaction
Heat exchangers are the most important and critical components in any
cryogenic liquefaction systems. The main function of the heat exchanger is to conserve
cold. Using heat exchangers, the heat from the compressed process fluid extracted and
transferred to the low pressure stream.
The heat exchangers used for cryogenic liquefaction can be of two types, i.e,
recuperative and regenerative. Generally among recuperative type heat exchanger
tubular heat exchangers, Giauque-Hampson exchanger, Plate fin heat exchangers,
perforated plate heat exchangers are important. The details about the different heat
exchanger used for cryogenic application is given by Barron [1, 52].
(i) Concentric tube heat exchanger
For the first time simple tube in tube heat exchanger was used by Linde
for liquefaction of air in 1985 [6, 53]. The tube in tube heat exchanger consists
of a small inner tube and a co-axial larger tube. In the smaller tube usually high
pressure stream flows and in the annular space the low pressure stream flows.
The figure of the tube in tube heat exchanger is shown in Figure 2-1. The flow
can be parallel or counter type.
14
Later on the plastic or wire spacer is placed inside tube in tube heat
exchanger [Figure 2-2]. By this arrangement better heat transfer occurs due to
increase in the heat transfer coefficient. When more than two streams are
involved smaller tubes are placed inside a large tube. It is called multi-tube heat
exchanger [Figure 2-3].
(ii) Coil wound heat exchanger
Giauque-Hampson exchanger [Figure 2-4] is a coil wound type heat
exchanger, one of the most classical heat exchanger used in large liquefaction
systems. It consists of small diameter tubes that are wound in several layers
over a cylindrical mandrel. The successive layers are wound in opposite direction
and provided with spacing strips [54]. The tubes are joined at the ends with a
header. All the tubes are covered with an outer casing and the entire unit is
insulated. These types of heat exchangers are constructed up to large size which
can be transported easily [55]. The major problem with these type of heat
exchnagers are expensive and only produces by Linde group and APCI [56] .
Prof. Collins [5, 10] had made a special extended type surfaces inside the tubes
for his helium liquefier [Figure 2-5].
Figure 2-1 Concentric tube heat exchanger [1]
Figure 2-2 Tube heat exchanger with wire spacer [1]
15
Figure 2-3 Multi tube heat exchanger [1]
Figure 2-4 The Giauque Hampson heat
Figure 2-5 The Collins heat exchanger [52]
exchanger [52]
(iii) Perforated plate heat exchanger
The perforated plate heat exchangers are used in small scale
refrigerators and liquefiers. A Schematic diagram of the perforated plate heat
exchangers is shown in Figure 2-6 . It consists of a series of parallel perforated
plates separated by spacers or gaskets. The spacers are bonded to the plates
tightly to prevent leakages. Generally the plates are made of relatively high
thermal conductivity material like aluminium for better heat transfer and the
spacers with low thermal conductivity material like plastic material to avoid
16
longitudinal heat conduction. A comprehensive review of the history and
applications of perforated plate heat exchanger is given by Venkatarathnam and
Sarangi [57] and design and optimization of the plate heat exchanger is given by
Venkatarathnam [58, 59].
Figure 2-6 Perforated plate heat exchanger [52]
(iv) Plate fin heat exchanger
This type of heat exchanger is used widely because of its high heat
transfer area density, compactness, light weight and high effectiveness. It is also
much cheaper than the coil wound heat exchanger. It consists of a set of layers
of corrugated plates served as fins and these layers are separated by a
separating sheet [Figure 2-7]. It is manufactured by many industries around the
world grouped in the Brazed Aluminium Plate-Fin Heat Exchanger Manufacturers
Association [60]. This associations includes five major brazed aluminium plate fin
heat exchanger manufacturers i.e., Chart Energy & Chemicals Inc of USA [61],
Fives Cryo [62], Kobe Steel Ltd. of Japan [63] , Linde AG of Germany [64],
Sumitomo Precision Products Co. Ltd of Japan.[65].In addition several smaller
companies manufacture plate fin heat exchangers such as Appolo Heat
exchangers Pvt. Ltd. Several specialized laboratories like Heat Transfer and Fluid
Flow Services (HTFS)[33] in England and Heat Transfer Research Inc (HTRI)
[34] in USA
made significant contribution to the research on plate fin heat
exchangers. The growth of these organizations occurs due to the support of
many industries and institutions around the world.
17
Figure 2-7 Plate fin heat exchanger [52]
2.6 Design Methods of heat exchanger
There are a number of text books [66-71] which describes the heat transfer
phenomena and design of heat exchangers. The design and simulation methodology of
two streams plate fin heat exchangers are enriched by other books and literatures [7277]. Pacio and Dorao [78] has classified the design methods as below
(i) Lumped parameters
(ii) Distributed parameters
(iii) Stream evolution
(i) Lumped parameters
This type of design method [79-81] is adopted for two streams with
single phase. Sizing and rating can be done by considering the energy balance
equation. Lumped parameters model includes five different techniques i.e.,
Mean temperature difference, -NTU [82], P-NTU, -P [83] , P1-P2 [84]. But
first two techniques are widely used for designing heat exchangers for cryogenic
application. But the drawback with the lumped parameter method is that they
consider overall heat transfer coefficient and constant specific heat.
(ii) Distributed parameters
This method involves with dividing the heat exchanger into several zones
or elements [85] and applying the lumped parameter method to the individual
zones or elements. When phase change occurs inside a heat exchanger then it is
divided into zones i.e. single phase vapor, two phase and single phase liquid.
18
When the heat exchanger divided by considering some definite length of the
heat exchanger, each divided part is called element. By dividing the heat
exchanger into number of elements, lumped parameter method could be applied
and making energy balance at both the ends. In this technique one could
consider the variable specific heat which is very important in cryogenic
application. If the stream is in different phases then the heat exchanger can be
divided into zones and each zone is then divided into several elements [86].
(iii) Stream evolution
This method based on steady state one dimensional mass, momentum
and energy balance equations for each individual stream. This can be applied to
more complicated geometry and more than two streams. It correlates among
the fluid properties, heat transfer and pressure drop characteristics. Aspen MUSE
of AspenTech [87] and GENIUS by Linde AG are the software that uses this
stream evaluation method.
The plate fin heat exchanger is designed by using commercially available
software Aspen–MUSE [88], the simulation software. The software takes care of
the various losses occurring in the heat exchanger like flow maldistribution at
the headers, longitudinal heat conduction, heat losses to the ambient, pressure
losses in the headers etc. It has been accepted as dedicated tool for the design
of plate fin heat exchangers in industrial applications.
For the thermal and hydraulic design of coil wound heat exchanger
GENIUS proprietary software from Linde is used. This computer program plots
the temperature and pressure profiles of the individual streams and calculates
the distribution of the tubes to the various layers.
2.7 Process design and simulation
Process simulation is a very essential technique in the design, analysis, and
optimization of a cryogenic process plants [89, 90]. Simulations could be done using
computer programs or simulation software, that simulate the behavior of the process
plants using appropriate mathematical models. Process simulators are used to
determine the detailed specifications of all units of a process by considering material
balance and energy balance equations. This also helps to troubleshoot startup and shut-
19
down operations, determine performance under off-design conditions, design and
troubleshoot control strategies.
The cryogenic refrigeration and liquefaction processes are different from general
processes. This is due to the following reasons

Large variation in thermo physical properties i.e., specific heat, thermal
conductivity etc.

Use of multistream LNG heat exchangers

Consideration of pinch point of heat exchanger

Double distillation columns

Phase separators
There are a number of software that are used for process simulation i.e., Aspen
HYSYS, Aspen Plus, CRYOSIM, ChemCAD etc. They have features that are required for
the simulation of cryogenic liquefaction and refrigeration processes.
Process simulation can be done by three methods,
(i) Sequential modular method,
(ii) Equation-oriented method,
(iii) Simultaneous modular method.
(i) Sequential modular method
Sequential modular method [91-94] is widely used to simulate cryogenic
refrigeration and liquefaction process. Every component of a process is
represented by a mathematical model and a program is written separately as
subroutines in these simulators. The mathematical models are developed to give
the thermodynamic parameters including pressure, temperature, enthalpy,
entropy, etc as output for the given input condition and component specification
such as pressure ratio, outlet pressure, efficiency of the equipment, etc. The
output of one component will be treated as input to the next component
attached with it. The simulation proceeds component by component from the
feed to the product streams. When there are recycle loops present in the
process, the recycle loops are torn at suitable points and estimated values are
assigned to these streams. Recycle loops are sequentially solved until the
assumed values of the tear streams match the computed stream information.
20
(ii) Equation-oriented method
The governing equations like mass balance, energy balance equations
and the governing equations of each process unit are solved one at a time,
sequentially, in the case of a sequential modular approach. But in equation
oriented method the governing equations of all the units are solved together
simultaneously in an equation-oriented approach. As all the equations are solved
simultaneously there is no need for tearing the streams. Also there is no need
for nested iteration loops.
The equation-oriented approach requires a suitable initial guess value of
variables for convergence. It is difficult to handle the error and also creates
problem at the time adding new unit or component. A robust nonlinear equation
solver is required to solve the equations simultaneously. Due to these problems,
the process simulation software has both the capabilities. Aspen Plus [95] ,
Aspen hysys uses both the method; sequential and equation oriented method.
(iii) Simultaneous modular method
When a process has a no. of tear streams with nested loops then this
technique is applied. By the sequential method, it takes more time with large no.
of iterations. Simultaneous modular method is a combination of sequential and
equation oriented method. It is also called two-tier method. Basically the process
is solved by sequential method but the nested loops are solved by equation
oriented method. A combination of simultaneous and sequential modular
approaches is sometimes preferred. The simultaneous modular approach is also
used in CRYOSIM [96] for optimization studies.
The dynamic simulation also can be done by the commercially available
software like Hysys dynamics [97-100], Cryogenic Process Real-time SimulaTor
(C-PREST) [101] etc.
2.8 Major industries supplying liquefaction plants
In some parts of our country, it is possible to buy liquid nitrogen from bulk
suppliers at low cost. The steel manufacturing companies are using liquid oxygen plant
and liquid nitrogen are waste product for them. So they supply liquid nitrogen at low
21
cost. But in most cases, including some major metropolitan areas, a laboratory needs to
operate its own liquid nitrogen generator.
There are three major international suppliers of nitrogen liquefiers to our
country:

Stirling Cryogenics of Netherlands,

Linde CryoPlants, UK and

Consolidated Pacific Industries, USA.

Air Liquide, France
The liquefier from Stirling Cryogenics [102] of Netherlands is based on the
integral Philips-Stirling Cycle, while the latter two use turbine for cold production.
Linde CryoPlant [103] of UK is leading supplier of liquid nitrogen plant. LINIT is
a turboexpander based nitrogen liquefier which has starting range of liquid nitrogen
production of 25 litres per hour. Also LINIT model with 50 and 100 litres per hour
capacity available and they are using gas bearings for the turboexpander.
Consolidated Pacific Industries [104] manufactures liquid nitrogen plant of 1.5
TPD. Liquid nitrogen produced from air by following the cryogenic distillation technology
due to different boiling points of oxygen and nitrogen.
Air Liquide of Canada is one of the industries who produce liquid oxygen,
nitrogen and hydrogen etc.
Cryomech, Inc. [105] of USA is manufacturers of fully automatic liquid nitrogen
plant ranging from 10 to 240 litres per day. It is based on Gifford-McMahon Cycle.
Kelvin International Corporation [106] produced two types of models. M is for
small capacity modules and NL for fully assembled industrial models. The ranges starts
from 15 to 120 litres per day.
Liquid-nitrogen-plant.com is among one of the leading supplier of Liquid
Nitrogen plants incorporating USA-technologies. The liquid Nitrogen is produced by
separating the nitrogen from air using membrane technology and then liquefying the
gaseous nitrogen using a USA technology built liquid helium refrigerator.
22
In association with a company named ING. L&A. Boschi of Italy, a liquid nitrogen
plant production factory was setup in 1985 in New Delhi. The plants are based on Linde
and Claude cycle.
Wuxi Victor Hengsheng Machinery Manufacturing Co., Ltd. is a company from
China established in 1999, which produces liquid nitrogen plant. The range of the
production is 220 to 2500 litres per hour. Hangzhou Union Industrial Gas-Equipment
Co., Ltd. and Hangzhou Kaihe Air Equipment Co.,Ltd. are also dealing with supplying
the liquid nitrogen plants.
23
3. Chapter III
Process Design of Nitrogen Liquefaction Cycle
3.1 Introduction
A turbo expander based cryogenic refrigerator or cryocooler consists of following
parts:

Compressor

Heat exchangers

Turboexpander

JT valve

Liquid nitrogen separator along with transfer line

Cold box

Piping

Instrumentation
A screw compressor is installed to provide the compressed nitrogen gas. Heat
exchangers are vital components of any cryogenic refrigerator or liquefier. To exchange
more amount of heat in small area plate fin heat exchangers are used. The expansion
turbine is the heart of the liquefier and it can use for lowering the temperature to
desired value adiabatically. J-T Valve is used for isenthalpic expansion. Phase separator
is used to separate liquid and gas phases. Piping and other instrumentations are
required to connect and to control the systems. The integrated system is kept inside the
cold box.
Three types of cycles have been selected for simulation i.e., Claude cycle [89],
modified Claude cycle eliminating the third heat exchanger and modified Claude cycle
eliminating the first heat exchanger. Out of these three process cycles one process cycle
is selected which gives optimum amount of liquid nitrogen. The selected process cycle is
further simulated to give optimum parameters like pressure of compression,
24
effectiveness of heat exchangers, mass fraction diverted through the turboexpander
and efficiency of turboexpander. These parameters are utilized in designing the
components like heat exchangers, turboexpander etc.
3.2 Description of process cycles
(i) Case-1:
This is a Claude cycle consists of a compressor, a turboexpander, a J-T
valve and three numbers of heat exchangers as shown in Figure 3-1. The
compressed nitrogen gas passes through HX-1 (first heat exchanger) and a
fraction of gas is diverted through the turboexpander, the remaining gas is
passes through the HX-2 (second heat exchanger) and HX-3 (third heat
exchanger). Then isenthalpic expansion is done by using J-T valve. The phase
separators separates liquid and remaining vapour passes through HX-3 as cold
stream. The cold stream exits from the HX-3 and mixes with the expanded
stream from the turboexpander. It is feed to the compressor through HX-2 and
HX-1.
Figure 3-1 Schematic diagram of Claude cycle (Case-1)
(ii) Case-2:
Figure 3-2 Schematic diagram of modified Claude cycle (Case-2)
25
It is a modified Claude cycles in which placement of all other components
remain same as in Claude cycle except the third heat exchanger. The HX-3 is
eliminated as shown in Figure 3-2.
(iii) Case-3:
It is also a modified Claude cycle in which all components are same as in
Claude cycle except the first heat exchanger. Here HX-1 is eliminated as shown
in Figure 3-3. The compressed gas divided into two streams before cooling by
any heat exchanger.
Figure 3-3 Schematic diagram of modified Claude cycle (Case-3)
3.3 Simulation of process cycles
Steady state simulation [89] is done on these three cycles using Aspen
®
Hysys [107]. Before simulating these cycles the following assumptions are made.
(i) Pressure drop in heat exchangers and pipes are assumed to be negligible.
(ii) No external heat transfer.
Mass flow rate, efficiency of compressor and turboexpander and pinch point of
Heat exchangers are kept constant for each process cycles and in every simulation.
Yield is estimated for different pressures of compression (10 bar to 100 bar).
The input conditions for the process cycles are
(i) Compressor

Mass flow rate

Inlet temperature = 300 K

Inlet pressure

Delivery pressure (from 10 bar to 100 bar)
=1 kg/s
= 1.1 bar
26
(ii) After Cooler

Outlet temperature = 300 K
(iii) HX-1 (only for case-1 and case-2)

Pinch point (minimum temperature difference between two streams of
heat exchanger) =5 K
(iv) HX-2

Pinch point =3 K
(v) HX-3 (only for case-1 and case-3)

Pinch point =1 K
(vi) Tee (From where diversion occurs to the turboexpander)

Mass fraction (ratio of mass flow diverted to the turboexpander with
compressed mass flow) through Turbo expander (change from 0.5 to 1
to get maximum yield)
(vii) Turbo expander

Efficiency of turbo expander=50%

Outlet pressure =1.1 bar
(viii) JT Valve

Outlet pressure = 1.1 bar
Turboexpander efficiency is taken as 50%. Since the available turboexpander
efficiency is limited to this value. The Plate fin heat exchangers have found to give
effectiveness of about 0.95 – 0.98. So pinch point is taken as 5 K of HX-1. The
maximum yield occurs at a particular mass fraction diverted through the turboexpander.
A graph between yield and pressure at exit of compressor is plotted as shown in
Figure 3-4. It shows that under similar operating condition and component specification
the yield of case-1 and case-2 is approximately equal. Although yield of all cycles are
increasing with the increase in pressure of compression but the yield of case-3 is lowest
as compared to the other two process cycles. In case-3 the mass fraction diverted to
turboexpander is at higher temperature than the other two cases due to absence of HX1. The other two heat exchangers HX-2 and HX-3 along with the turboexpander are
unable to lower the temperature to produce appreciable amount of liquid.
27
Figure 3-4 Yield at different pressures in the three cases
Figure 3-5 Heat load of heat exchangers at different pressures in case-1
A graph has been plotted to compare the amount heat load of three heat
exchangers of case-1 with increasing pressure. The heat load of HX-3 in case-1 is found
to be very low (1 kW at 15 bar) as compared to the other two heat exchangers (175.9
kW of HX-1 and 18 kW of HX-2) as shown in Figure 3-5. So at pressures (10-50 bar) the
28
HX-3 can be neglected. If this is eliminated, it will became case-2 i.e., modified Claude
cycle with eliminating the last heat exchanger.
Among the three process cycles case-3 has very low yield and at low pressure
and case-2 can give yield of case-1 due to negligible load in HX-3 of case-1. So case-2
is selected as the liquefaction cycle [108].
3.4 Calculation of process parameters of selected cycle
The design of process parameters depends on the nodal temperature across the
heat exchangers and turbo-expanders. Figure 3-6 shows the process diagram and given
below are the set of equations used for the calculation of process parameters.
Figure 3-6 Process flow diagram
(i) Input parameters
a. Effectiveness of heat exchanger 1, 1  0.98
b. Pinch point for heat exchanger 2 =1 K
c. Efficiency of turboexpander, tx  0.5
d. Compressor discharge temperature, T 2  310 K
e. Mass fraction diverted through turboexpander, tx  0.94
f.
Corresponding property values at saturated liquid and saturated vapor
states are taken from property package, ALLPROPS [109].
(ii) Assumptions
a. Pressure drop in the heat exchanger= 0.05 bar.
b. No external heat transfer.
29
(iii) Component analysis
a.
Pinch Point Specification of Heat Exchanger2: Heat exchanger 2
divided into two parts: First part of heat exchanger is used to cool the
hot nitrogen gas in the high pressure side up to the saturation
temperature of 100.2 K and this stream is condensed in the second part
of heat exchanger. The minimum temperature difference occurs at the
point where the condensation begins and is called as pinch point. For the
specified pinch value of 1K for second heat exchanger gives,
Tp  T4g  1
b.
(3.1)
Heat Exchanger 1: For the specified value of effectiveness of first heat
exchanger and the pinch point specification for heat exchanger 2, the
inlet and outlet enthalpies of hot and cold fluids at node points 8, 3 and
9 are calculated. The enthalpies h8, h3 and h9 are calculated from
equations(3.4), (3.5) and (3.6) respectively. Since the enthalpy at node
point 8 ( h8 ) requires a prior estimation of y and h9 , initial guess values
are provided by equations (3.2) and (3.3) respectively.
Figure 3-7 Pinch point of HX-2
y  (1   )  for initial guess 
(3.2)
h9  1h2 '  for initial guess 
(3.3)
Considering the energy balance equation between HX-1 and HX-2a,
h8 
[h9 (1  y )(1   )  h2 (1   )  h4 g (1   )  hp (1  y )]
[( )(1  y )]
(3.4)
Now considering energy balance across HX-2a alone,
h3 
[h4 g (1   )  (1  y )(h8  hp )]
(1   )
(3.5)
30
Rearranging the effectiveness equation of HX-1,
h9  1h2 ' (1  1)h8
(3.6)
The iterations continued until the two subsequent iterative values of
h9 are within a specified limit (10-4 kJ/kg).
c. Turboexpander: In the turboexpander, the process 3-6s is the
isentropic expansion and the process 3-6 is the actual expansion. The
entropy at node point 3 ( s 3 ) for specified values of h3 and p 3 is taken
from property package [109]. The high pressure stream after the
expansion through the turbine is always to be dry which ensures that
s 6 is always greater than s 6 g . The enthalpy at the end of expansion is
estimated as,
Figure 3-8 Expansion in Turboexpander
s 6s  s 3
(3.7)
 (s 6 g  s 3 ) 



cp


T 6s  T 6 g exp 
d.
(3.8)
h6s  h6 g  c p (T 6 g T 6s )
(3.9)
h6  h3   (h3  h6s )
(3.10)
Mixer: Applying energy balance equation for the mixer, enthalpy at
outlet of mixer is,
1    y  h g  h
5
h7 
6
 1  y  h7
[ h6  (1    y )h5 g ]
(1  y )
(3.11)
31
e.
Heat Exchanger 2: Enthalpy at outlet of hot fluid is found out by
energy balance between hot and cold fluids as
h4 
[h3 (1   )  (1  y )(h8  h7 )]
(1   )
(3.12)
f. Throttle Valve: Throttling is an isenthalpic process which yields,
h5  h4
g.
(3.13)
Phase Separator: The dryness fraction, x 5 of the stream entering the
phase separator is estimated as
x5 
(h5  hf 5 )
(hg 5  hf 5 )
(3.14)
h. Yield: The liquid yield obtained per kg of gas passing through the
throttling valve, is (1  x 5 ) . Hence for (1   ) kg of gas passing through
the throttling valve, the yield is
y  (1   )(1  x 5 )
(3.15)
To obtain the process parameters the iteration process is repeated starting from
equations (3.4) to (3.15) till the tolerance in y is less than 10-5.
3.5 Parametric analysis of selected cycle
Parametric analysis was carried out to optimize the selected process cycle.
Following are the key parameters which influences the liquid yield.
(i)
Mass fraction diverted through the turboexpander
(ii)
Effectiveness of HX-1
(iii)
Pinch point of HX-2
(iv)
Efficiency of turboexpander
(v)
Pressure of compression
(i) Effect of mass fraction diverted through turboexpander
The effect of the variation of mass fraction through turboexpander on
the yield at different operating pressures is studied by keeping the values of
32
turbine efficiency, effectiveness of HX-1 and pinch point of HX-2 constant. From
the Figure 3-9, it is found that yield increases with the mass fraction diverted
through turboexpander, and after a certain value it starts decreasing [91]. This
is because of insufficient amount of fluid in the high pressure side of HX-2.
Figure 3-9 Variation of yield with the change of mass fraction through turboexpander at
different operating pressures
The yield also increases with increase in pressure. At 8 bar, the
maximum yield occurs when 94% of mass diverted through turboexpander. But
at 30 bar it occurs when 86 % passed through turboexpander. It means at high
pressures, maximum yield can be obtained with less mass fraction diverted
through turboexpander.
(ii) Effect of variation of effectiveness of heat exchanger 1
The liquid yield increases in direct proportion with the effectiveness of
HX-1, for the given efficiency of turbine, expander flow ratio and pinch point of
HX-2. With the increase in effectiveness of HX-1, the temperature at exit of high
pressure side decreases. This helps to produce more liquid.
It is also found that liquid yield is not possible, after certain effectiveness
of HX-1 as shown in Figure 3-10. The lower limit effectiveness of HX-1 decreases
with increasing the operating pressure. At 8 bar the lower limit of effectiveness
is 0.88. But at 30 bar it is 0.69. If it is operating at high pressures then a low
effectiveness HX-1 can be used.
33
Figure 3-10 Variation of yield with effectiveness of HX-1 at different operating pressures
(iii) Effect of variation of pinch point over the performance of the plant
The liquid yield decreases with increase in pinch point of HX-2, for the
given efficiency of turbine, expander flow and effectiveness of HX-1, as shown in
Figure 3-11. Thus pinch point of HX-2 has not effect much on the yield but it
should be lower as much as possible.
Figure 3-11 Variation of yield with pinch point of HX-2 at different operating pressures
34
(iv) Effect of efficiency of turboexpander
The effect of turbine efficiency on the liquid yield is shown in the Figure
3-12. For the given effectiveness of HX-1, pinch point of HX-2 the liquid yield
increases with the efficiency of turboexpander. With the increase in efficiency of
turboexpander the temperature drop in turboexpander increases. This causes to
improve the liquid yield. It is observed that the turboexpander with high
efficiency can provide a better yield with lesser mass flow rate through it. With
increasing operating pressure the yield also increases.
Figure 3-12 Variation of yield with turboexpander efficiency at different operating
pressures
Figure 3-13 Variation of compressor work per liquid mass produced with operating
pressure
35
(v) Effect of pressure of compression
The effects of pressure of compression on yield are shown in Figure 3-9,
Figure 3-10, Figure 3-11 and Figure 3-12. Yield increases with increase in
pressure of compression. But work required by the compressor to produce unit
mass of liquid also increases with increase in pressure as shown in Figure 3-13.
3.6 Performance of nitrogen liquefaction plant
At low pressure (10 bar) the compressor work done per unit mass of liquid is
less. Also availability of high effectiveness heat exchangers and difficulties in
manufacturing turboexpander at high pressures (above 15 bar), the case-2 cycle should
be selected to operate at 8-10 bar. Operating the case-2 cycle at 8 bar, it is found that,
the optimum mass fraction to be diverted through turboexpander for the maximum
yield is 0.93. For the better performance, the effectiveness of heat exchanger should be
as high as possible (0.99), pinch temperature of HX-2 should be as low as possible (1K)
and turboexpander efficiency must be as high as possible.
Applying the First Law for steady flow process to the system as shown in Figure
3-6, the liquid yield can be expressed as,
mh2  mh3  mh6  m  mf  h9  mf h5f  0
(3.16)
Rearranging the equation(3.16),
y 
h  h2
h  h6
mf
 9
 3
m h9  h5f
h9  h5f
(3.17)
An oil injected twin screw compressor is used to deliver the compressed nitrogen
at 8 bar pressure. Seshaiah [110] has done experimental studies on the above
compressor and has evaluated calculate flow rate of nitrogen at different pressure ratio.
For the required pressure of 8 bar, the compressor will deliver a mass flow rate of 296
kg/hr.
Finally the nitrogen liquefaction plant is fixed with component specifications
given in Table 3-1, and the thermodynamics state points given in Table 3-2. As per
design it will produce about 13.95 kg/s of liquid nitrogen which is equivalent to 17.44
liters/hr. The yield is calculated to be 4.71 %. The temperature entropy diagram of the
selected process cycle is shown in Figure 3-14.
36
Table 3-1 Basic specifications of liquefier components
Sl. No.
Name of the component
Specifications
1.
Heat exchanger (HX-1)
Effectiveness = 99 %
2.
Turboexpander (Tex)
Efficiency = 50 %
Flow through turboexpander
= 93 %
3.
Heat exchanger (HX-2)
Pinch temperature =1 K
Table 3-2 Thermodynamic state points of the process cycle
State
Pressure
Temperature
Points
(bar)
(K)
Quality
Mass Flow
Rate
(kg/hr)
1
Compressor inlet
1.1
300.00
296
2
Compressor exit
8
310.00
296
7.95
119.66
275.27
7.9
100.13
0.1125
20.73
1.2
78.78
0.3265
20.73
and HX-1 inlet
3
HX-1 exit and
Tex,HX-2 inlet
4
HX-2 exit and J-T
valve inlet
5
J-T exit and inside
phase separator
13.95
5f
Liquid produced
1.2
78.78
6s
Isentropic exit from
1.3
79.49
1.3
90.01
275.27
1.2
89.54
282.05
0.9451
Tex
6
Actual exit from
Tex
7
Mixing of phase
separator vapor
with exit of Tex and
LP inlet of HX-2
P
Pinch point in HX-2
1.15
99.13
8
LP exit of HX-2 and
1.15
100.93
282.05
1.1
307.89
282.05
LP inlet of HX-1
9
LP exit of HX-1
37
Figure 3-14 Temperature entropy diagram of Nitrogen liquefier
38
4. Chapter IV
Design of Heat Exchanger
4.1 Introduction
As per the process design described in chapter-II, the liquefier needs two
numbers of heat exchangers. The first heat exchanger should have high effectiveness.
So plate fin heat exchangers are to be used which will give high effectiveness up to
0.98. Design or sizing of the heat exchangers consists of determination of the heat
exchanger dimensions for the specified heat transfer and pressure drop. The amount of
heat transfer and pressure drop is taken from the process cycle state points. The first
heat exchanger used is having both streams in single phase but the second heat
exchanger under goes a phase change in high pressure stream.
The design of heat exchanger needs the determination of heat transfer
coefficients, flow friction which depends on accurate prediction of colburn, j
and
friction factor, f .Various correlations are available in literature for the determination of
colburn, j and friction factor, f . In the present design procedure for the plate and fin
heat exchanger is carried out by using the correlations developed by Maiti and Sarangi
[111], Manglik et.al [112] and Joshi et.al [113].The heat exchanger is also designed by
using commercially available heat exchanger design software, Aspen MUSE [88].
4.2 Plate fin heat exchanger design procedure
The design of heat exchanger needs temperature and pressure at inlet and exit
of both the streams along with the mass flow rate and allowable pressure drop in each
stream. All properties like density, enthalpy, specific heat, viscosity, prandtl number are
determined at mean temperature and pressure. Effectiveness, UA, heat load is also
39
calculated for the above inlet and exit conditions. As the heat capacity of cold stream is
mimimum, effectiveness is found out by the following formula.

hc ,o  hc ,i
hh' ,i  hc ,i
(4.1)
Where, hc ,i = Enthalpy at inlet of the cold stream
hc ,o = Enthalpy at outlet of the cold stream
hh' ,i = Enthalpy at hot stream temperature and cold stream pressure
Figure 4-1 Geometry of a typical offset strip fin surface
The dimensions for offset strip fin geometry are shown in Figure 4-1. The input
parameters of basic fin specifications for both the streams (i.e. for hot and cold layers)
are
(i) Fin frequency, f f
(ii) Fin thickness, t f
(iii) Fin length, l f
(iv) Fin height, hf
(v) Plate thickness, a p
40
The expressions for fin details for both the streams are,
Fin spacing,
sf 
(1  f f t f )
ff
(4.2)
Plate spacing,
bp  hf  t f
(4.3)
aff  (s f  t f )hf
(4.4)
afr  (s f  t f )(hf  t f )
(4.5)
as  2hf l f  2hf t f  2s f l f
(4.6)
af  2hf l f  2hf t f
(4.7)
Free flow area per fin,
Frontal area per fin,
Heat transfer area,
Fin area,
Equivalent diameter,
De 
2(s f  t f )hf l f
4  total free flow area  length

total heat transfer area
hf l f  hf t f  s f l f
(4.8)
Ratio of fin area with total surface area,

af
as
(4.9)
Frontal area ratio,
 
aff
afr
(4.10)

hf
sf
(4.11)
 
lf
sf
(4.12)
 
tf
sf
(4.13)
Dimensionless parameters for the fin,
Assume width of heat exchanger, W and number of layers in hot side and cold side, n h
and nc .
41
Total area between plates,
Afr  bp  nh Whx
(4.14)
Aff    Afr
(4.15)
Total free flow area,
Core mass velocity,
G 
m
Aff
(4.16)
Re 
GDe

(4.17)
Reynolds number,
Following correlations are used to determine the Colburn, j and friction factor, f .
Maiti and Sarangi correlations :
Critical Reynolds number
Re j *  1568.58( )0.217 ( )1.433 ( )0.217
(4.18)
Ref *  648.23( )0.06 ( )0.1 ( )0.196
(4.19)
j  0.36(Re)0.51 ( )0.275 ( )0.27 ( )0.063
(4.20)
f  4.67(Re)0.70 ( )0.196 ( )0.181 ( )0.104
(4.21)
j  0.18(Re)0.42 ( )0.288 ( )0.184 ( )0.005
(4.22)
f  0.32(Re)0.286 ( )0.221 ( )0.185 ( )0.023
(4.23)
For Re<Re*
For Re>Re*
Manglik and Bergles correlations :
j  0.6522Re0.5403  0.1541 0.1499 0.0678 1  5.269  105 Re1.340  0.504 0.546 1.055 
f  9.6243Re0.7422  0.1856 0.3053 0.2659 1  7.669  108 Re4.429  0.920 3.767 0.236 
0.1
0.1
(4.24)
(4.25)
Where

sf
hf
(4.26)
42
 
tf
lf
(4.27)
 
tf
sf
(4.28)
and
De 
2s f hf l f
2 hf l f  hf t f  s f l f   t f s f
(4.29)
After determining the j factor, convective heat transfer coefficient, can be
calculated from the formula
hconv 
(G  j  c )
(Pr)0.667
(4.30)
Fin parameter,
M 
(2  hconv ,c )
(4.31)
(K f  t f )
Fin effectiveness,
f 
tanh(Ml e )
(Ml e )
(4.32)
Where l e = effective length of fins
In cryogenic applications, heat loss to the cold fluid stream in the heat
exchanger is minimized by placing layers in between two hot fluid layers. The number
of layers through which the hot fluid passes will be one unit more than that of cold fluid
layer. This arrangement is done to expose the hot fluid layer at both the ends. To take
into account of heat losses to the ambient, the fin conduction lengths for the outer
layers on the hot side will be taken as b p whereas for the inner layers of the hot fluid,
the conduction length is taken as bp / 2 . However for the cold layers placed between the
hot layers the fin conduction length is taken as bp / 2 as it present only inner side and
adjacent layers share the same lengths.
The overall surface effectiveness of hot side,
 a f 
n h  2   a f 
2
     (1  f ) 
  (1  f ) 
nh   as 
nh
 as 
oh  1  



(4.33)
43
and for cold side
a
oc  1  ( f
as
)  (1  f )
(4.34)
The ratio of total heat transfer surface area to the separating surface area (wall area)
is,
Ao / Aw 
1  Nf f * t f
1  ( Af / Ao )
(4.35)
Overall heat transfer coefficient is,
1
Uo

1
ηoh hh

a p Ao (Aoc / Aoh )

K W AW
ηoc hc
(4.36)
Heat transfer area may be calculated as
A0 
U 0 A0
U0
(4.37)
The required length of the heat exchanger is calculated from the equivalent diameter
definition, as
L
De A0
4 Aff
(4.38)
The pressure drop is determined using the friction factor calculated from the
correlations (4.21) or (4.23) and (4.25).
p 
fLG 2
2De bp
(4.39)
If the pressure drop obtained is less than allowable pressure drop then
calculation is proceed to the estimation of longitudinal conduction of heat. Otherwise
the number of layers and fin parameters are modified to get the desired pressure drop.
Frontal area of fin,
Afrt  Whx  H hx
(4.40)
Aw  Afrt  Affh  Affc
(4.41)
Wall conduction area,
where, Affh and Affc are the free flow area of hot side and cold side respectively.
N.T.U required is determined as,
N tu 
1  C R  
1
ln 
1  C R  1   
(4.42)
44
Assuming a Factor of safety, F .S , NTU (considering longitudinal heat conduction)
required is,
(N tu )lc  (N tu )  F .S
(4.43)
(UA)lc  (N tu )lc * C min
(4.44)
and,
Area considering longitudinal conduction is,
(UA )lc
A
U
(4.45)
Length of the heat exchanger (considering longitudinal heat conduction) is,
De  Ao
4  Aff
L
(4.46)
The effect of longitudinal heat conduction is to reduce the effectiveness of heat
exchanger. The decrease in the effectiveness of heat exchanger is found out using
Kroeger‘s equation.
Wall conduction parameter is defined as,

K w Aw
LC min
(4.47)
The effectiveness of heat exchanger with longitudinal heat conduction is,
y   * N tu * Cr
 
(1  Cr )
(1  Cr )(1  y )
(1   )y 

1   (1   )y 

   (y / (1  y )1/2 
(4.49)
(4.50)
(1   )
(1   )
(4.51)
(1  C r ) * N tu
1    N tu  Cr
(4.52)

r1 
(4.48)
Ineffectiveness
(1   ) 
(1  Cr )
 exp(r1 )  Cr
(4.53)
Effectiveness
  [1  (1   )]
(4.54)
45
This calculated effectiveness is to be checked with the previously calculated
effectiveness using equation (4.1). If it is less than before then the factor of safety is
increased till this effectiveness is greater than or equal to the previous effectiveness.
The length obtained is the final length of heat exchanger. And again calculate the
pressure drop using the equation (4.39) is calculated for the new length.
4.3 Design of first heat exchanger
The first heat exchanger is designed using the correlations from Maiti and
Sarangi [111] and Manglik and Bergles [112]. This design is further rectified by Aspen
MUSE software. With the data given in Table 4-1 and Table 4-2, the design is given in
Table 4-3 as per design procedure outlined earlier.
Table 4-1 Thermal data for First Heat exchanger
High Pressure side Low Pressure Side
Inlet temperature
310 K
100.74 K
Outlet temperature
120.45 K
307.89 K
Mass flow rate
82.22 g/sec
78.34 g/sec
Pressure at inlet
8 bar
1.125 bar
Allowable pressure drop 0.05 bar
0.05 bar
Table 4-2 Fin specifications for first heat exchanger
High Pressure side Low Pressure Side
Fin frequency
714 fins/m
714 fins/m
Fin metal thickness
0.2 mm
0.2 mm
Fin length
3 mm
3 mm
Fin height
6.3 mm
9.3 mm
Separating plate thickness 0.8 mm
0.8 mm
Keeping the core height, core width and number of layers same the length is
found out using correlations and Aspen MUSE as shown in Table 4-3. The length of the
heat exchanger varies from 1976 mm to 2150 mm. Out of these the length given by
Aspen MUSE is selected excluding header length of 90 mm as per Aspen MUSE.
To accommodate a large length of the heat exchanger inside the cold box is
difficult. As the cold box has height of 1800mm, the first heat exchanger is divided into
two parts of equal length. The divided parts are joined by pipe as shown in.
46
Figure 4-2. Dimension of first plate fin heat exchanger
Table 4-3 Overall dimension of first heat exchanger
Maiti and
Manglik and
Sarangi
Bergles
Apen MUSE
Core length (in mm)
2081
1976
2150
Core width (in mm)
180
180
180
Core Height (in mm)
165
165
165
Number of layers in hot side
10
10
10
Number of layers in cold side
9
9
9
47
4.4 Design of second heat exchanger
As per process design done in Chapter-II the minimum temperature approach is
1 K. The thermal data for the second heat exchanger is given in Table 4-4. Nitrogen gas
is cooled to the saturated temperature of 100.45 K at 7.95 bar and then condensation
takes place. The heat transfer coefficient on condensation part is high and it is obtained
from graph [114] as shown in Figure 4-3. Due to high heat transfer coefficient, the fin
frequency in the condensation part of the second heat exchanger is low and this also
facilitates the removal of condensed droplets which would otherwise obstruct the flow.
The fin specification for the second heat exchanger is shown in Table 4-5. The heat
exchanger dimensions are shown in Figure 4-4.
Figure 4-3 Condensing h.t.c for nitrogen as a function of temperature difference [114]
Table 4-4 Thermal data of second heat exchanger
High pressure side Low pressure side
Inlet temperature
120 K
89.1 K
Outlet temperature
100.45 K
101.2 K
Mass flow rate
5.756 g/sec
78.34 g/sec
Pressure at inlet
7.95 bar
1.2 bar
Allowable pressure drop 0.05 bar
0.05 bar
48
Table 4-5 Fin specifications for second heat exchanger
High pressure side
Low pressure side
Fin frequency
500 fins/m
714 fins/m
Fin metal thickness
0.2 mm
0.2 mm
Fin length
3 mm
3 mm
Fin height
6.3 mm
6.3 mm
Separating plate thickness 0.8 mm
0.8 mm
Figure 4-4. Dimension of second plate fin heat exchanger
49
5. Chapter V
Design of Turboexpander
5.1 Introduction
The performance of all cryogenic refrigeration and liquefaction plants mainly
depends on the efficiency of turboexpander. The turboexpander is an expansion device
where the process gas is expanded with reduction of temperature. It is a dynamic
system which responds to the variation in process stream parameters. To design the
turboexpander a fixed state of process stream parameters or design point is required.
The design point is fixed as per the process design done in chapter III. The basic input
parameters or design points are given in Table 5-1.
Table 5-1 Basic input values for turboexpander design
Working fluid
Nitrogen
Turbine inlet temperature, T in
124 K
Turbine inlet pressure, p in
7.97 bar
Discharge pressure, pex
1.2 bar
Mass flow rate, m
76.46 g/sec
The literatures of design methods for turboexpander [18, 26, 48, 49, 115, 116]
are available for low expansion ratio with purely radial turbines. But the above data
design point have high expansion ratio (i.e. 6.64). The high expansion ratio is achieved
by using backward curved vanes which increases the tangential force on the blades
thereby increasing the turbine speed for higher efficiency.
The present design procedure has following characteristics,
(i) Design for any expansion ratio
(ii) Design of radial, backward or forward turbine
50
The turboexpander consists of a turbine wheel and a brake wheel at the two
ends of a vertically placed shaft. The sectional view of the turboexpander is shown in
Figure 5-1. The shaft is supported by two number of tilting pad journal bearing and two
number of aerostatic thrust bearing. The whole components are place inside a bearing
housing. At the lower end of bearing housing, cold end housing is attached. The cold
end housing consists of nozzles and diffuser. At the top of bearing housing, warm end
housing is present. It consists of a nozzle to the brake compressor.
Figure 5-1 Longitudinal section of Turboexpander
51
The design of turboexpander mainly consists
(i)
Design of turbine wheel
(ii)
Design of Nozzles
(iii)
Design of Diffuser
(iv)
Design of Shaft
(v)
Design of Brake compressor
(vi)
Selection of journal and thrust bearing
(vii)
supporting and housing components
(viii)
Other turboexpander components
5.2 Design of turbine wheel
Figure 5-2 State points at nozzles, turbine wheel and diffuser
The two dimensionless parameters, specific speed and specific diameter
uniquely determine the major dimensions of the wheel and its inlet and exit velocity
triangles.
Specific speed, n s 
  Q3
 h 
3/4
(5.1)
3,s
Specific diameter, d s 
D 2   h3,s  1/4
Q3
(5.2)
Where, Q 3 is the volumetric flow rate at the exit of the turbine wheel and h3,s is the
isentropic enthalpy drop from inlet to the turbine exit. The values of specific speed and
specific diameter are chosen so that Mach number of the fluid at the nozzle exit is
maintained at or close to unity. If the velocity of fluid exceeds the speed of sound, the
flow gets choked leading to the creation of shock waves and flow at the exit of the
52
nozzle will be non-isentropic. The specific speed and specific diameter are chosen so as
to achieve the maximum possible efficiency within the subsonic zone.
Assume sd s  1.9
And specific speed s  0.5471
Then Specific diameter d s 
sd s
 3.4728
s
At known p in and T in , hin =118.92 kJ/kg and s in =5.2603 kJ/kg K
As s ex ,s  s in and pex ,s  pex , so s ex ,s  5.2603 kJ/kg K and pex ,s  1.2 bar
At known pex ,s and s ex ,s , hex ,s =69.18 kJ/kg
Efficiency of the turboexpander,

hin  hex
hin  hex ,s
(5.3)
Reasonable turbine efficiency th  75 % is assumed [49]
Substituting the values in equation(5.3), we get enthalpy at exit of turboexpander as,
hex  hin   (hin  hex ,s )  81.615 kJ/kg
Power produced,
P  m (hin  hex )  2.8523 kW
(5.4)
Isentropic enthalpy drop in turboexpander  hin  hex ,s  49.74 kJ/kg
Volume flow rates at diffuser exit, Qex 
m
 0.01480 m 3 /s
ex
(5.5)
The states at the exit of the turboexpander are known. But states at turbine exit
are not known. There is the difference between the states ‗3‘ and ‗ex‘ caused by
pressure recovery and consequent rise in temperature and density in the diffuser. The
two factors k 1 and k 2 are taken in account.
Where
k1 
Q3

 3
Qex ex
(5.6)
k2 
hin  h3s
hin  hex ,s
(5.7)
and
53
The value of k 1 is assumed to be 1.11 as taken by Ghosh [49] and k 2 value is
taken as 1.03 from the suggestion of Kun and Sentz [22, 117]
Substituting the values of k 1 and k 2 in equation(5.6) and (5.7) respectively we
get, isentropic enthalpy drop from turbine
h3,s  (hin  h3,s )  k 2 (h1  hex ,s )  51.23 kJ/kg
(5.8)
and volumetric flow rate at exit of turbine wheel
Q3  k1Qex  0.01643 m 3 /s
(5.9)
Again substituting the values of h3,s and Q 3 in the equations (5.1) and (5.2)
gives the speed and diameter of the turbine wheel.
Rotational speed of turbine wheel,

ns  h3s 
Q3
Inlet diameter of turbine, D 2 
3
4
 14534.67 rad/s  138778 rpm
d s Q3
 h3,s 
1
 29.6 mm
(5.11)
4
Blade velocity at inlet to the turbine, U 2 
D2
2
 215.11 m/s
Spouting velocity, C 0  2(h1  hex ,s )  315.4 m/s
Velocity ratio,V .R . 
(5.10)
U2
 0.682
C0
(5.12)
(5.13)
(5.14)
As all radial turbines are found to produce maximum efficiencies over the range
of velocity ratio 0.65 to 0.70 [118]. The current velocity ratio from equation (5.14) is
within limit.
Rohlik [118] prescribes that,  the ratio of eye tip diameter to inlet diameter
should be limited to a maximum value of 0.7 to avoid excessive shroud curvature. The
value of  is taken as 0.6, corresponding to the maximum efficiency within the subsonic
zone and for obtaining longer blade passages.

Dtip
 0.60
D2
(5.15)
Diameter at the eye tip, Dtip  0.6  D2  0.6  0.0296  0.0177 m  17.8 mm
54
Again from reference [41], the exit hub to tip diameter ratio should be limited to
a minimum value of 0.4 to avoid excessive hub blade blockage and energy loss. Kun
and Sentz [41] have taken a hub ratio of 0.35 citing mechanical considerations. Ino et.
al [33] have chosen a value of 0.588. For current design the value of  is taken as
0.425.

Dhub
 0.50
Dtip
(5.16)
Diameter at the hub, Dhub  0.425  Dtip  0.00888 m  8.9 mm
Mean outlet diameter of the turbine wheel,
D3,mean 
U 3,mean 
U 3,tip 
D3,hub  D3,tip
2
  D3,mean
2
  D3,tip
2
U 3,hub 
  D3,hub
2

0.0089  0.0177
 0.01332 m  13.32 mm
2
 96.801 m/s
(5.17)
(5.18)
=129.359 m/s
(5.19)
 64.679 m/s
(5.20)
Assume absolute meridian velocity C m 3 is 109.9 m/s. For small turbines, the hub
circumference at exit and diameter of milling cutters available determine the number of
blades. So number of blades, Z  10 [49], thickness of blades, t  0.6 mm [33] and
exit angle  3 is taken as 950 [22].
Absolute velocity at turbine exit,
C3 
C m3
 110.3198 m/s
sin  3
(5.21)
Absolute tangential component,
C  3  C 3Cos (3 )  9.615 m/s
(5.22)
Relative velocity at turbine exit,
W3  C 32  U 32  2C 3U 3 cos(3 )  152.978 m/s
3,mean  tan1
C 3 sin( 3 )
 45.92o
U 3,mean  C 3 cos(3 )
(5.23)
(5.24)
55
Flow through turbine

D
4
Q3  C 3 sin( 3 ) 

2
3,tip

 D3,hub 2 
Z tr t tr D 3,tip  D 3,hub  
2 sin 3,mean
  0.01643 m 3 /s (5.25)


Check the value of Q 3 in equation (5.25) with the previously calculated value
from equation (5.9) of Q 3 . If they are not equal then change the value of C m 3 .
Assume incidence angle, 2 as 260 . The amount of work can be extracted from a
turbine is calculated from change in momentum of the fluid in its passage through the
turbine wheel. From the energy conservation equation for turbo machine [22], we get
(h02  h03 )  C 2U 2 cos(2 )  U 3C 3 cos(3 )
(5.26)
From equation (5.26), we get
C2 
1000(h02  h03 )  U 3C 3 cos(3 )
 187.386 m/s
U 2 cos(2 )
(5.27)
C  2  C 2 cos(2 )  187.38cos(26)  168.42 m/s
(5.28)
C m 2  C 2 sin(2 )  187.38sin(26)  82.145 m/s
(5.29)
W2  C m 22  (U 2  C  2 )2  82.1432  (215.113  168.42)2  94.487 m/s
2  tan1
Cm2
82.145

 60.38o
(U 2  C  2 ) (215.113  168.42)
(5.30)
(5.31)
Figure 5-3 Velocity diagrams for turbine
56
Thermodynamic state at wheel discharge (state 3)
Exit stagnation enthalpy:
h0ex  hex 
C ex 2
2
 81.615 
172
 81.76 kJ/kg
2000
(5.32)
Exit stagnation pressure:
p0ex  pex 
1
ex C ex 2  1.2075 bar
2
(5.33)
Neglecting the losses in the diffuser, stagnation enthalpy at turbine outlet is
h03  h04 and static enthalpy at the outlet of turbine is given by
h3  h03 
C 32
3 
2
 81.76 
110.32
 75.674 kJ/kg
2  1000
m 0.07646

 4.6537 kg/m3
Q3 0.01643
(5.34)
(5.35)
Corresponding to h3 and 3 all the other properties at the turbine outlet are
calculated from property calculation software Allprops [109]. Static properties at the
turbine outlet are as follows,
p3  1.020 bar, T 3  77.373 K, C s 3  172.274 m/s
Relative velocity at eye tip,
W3,tip  C 32  U 3,2tip  2C 3U 3,tip cos(3 )  177.17 m/s
3,tip  tan1
Highest Mach number at eye tip =
C 3 sin(3 )
 38.33o
U 3,tip  C 3 cos(3 )
W3tip
Cs3

(5.36)
(5.37)
177.17
 1.0284
172.274
Where Cs 3 is the velocity of sound for the corresponding state point 3.
Similarly Relative velocity at hub,
W3,hub  C 32  U 3,2 hub  2C 3U 3,hub cos(3 )  132.65 m/s
3,hub  tan1
Mach number at hub =
C 3 sin( 3 )
 55.94 o
U 3,hub  C 3 cos(3 )
(5.38)
(5.39)
W3hub
132.65

 0.77
Cs3
172.274
57
Thermodynamic state at wheel inlet (state 2)
For computing the thermodynamic properties at wheel inlet (state 2, Figure 5-2),
the efficiency of the expansion process till state 2 is assumed. The nozzle efficiency s
needs to be between 0.9 and 0.95. So nozzle efficiency s  0.93 is assumed. Assuming
isentropic expansion in the vaneless space, the efficiency of the nozzle along with the
vane less space is defined as
n 
hin  h2
hin  h2s
(5.40)
Since h02  h01  h0in  hin  118.92 kJ/kg
h2  hin 
C 22
2
h2s 
 118.92 
187.3862
 101.363 kJ kg
2  1000
h01n  h01  h2
 100.04 kJ/kg
n
(5.41)
(5.42)
Also s in  s 1  s 2s  5.2603 kJ/kg
Knowing h2s and s2s the other properties at the point 2 are calculated as,
p2s  p2  4.319 bar
T 2  104.57 K, 2  15.0194 kg/m3 , s 2  5.273 kJ/kg K, Cs 2  199.6 m/s
From continuity equation, the blade height at entrance to the wheel is computed as:
.
mtr
b2 
( D 2  Z tr t tr ) 2C m 2

0.07646
  0.0296  10  0.6   15.0914  82.144
(5.43)
 0.709 mm
Absolute Mach number at the exit of nozzle or turbine wheel
C 2 187.38

 0.9388  1
C s 2 199.6
The absolute Mach number at exit of the nozzle should be less than 1.
Blade Profile of Turbine wheel:
The blade profile of the turbine wheel is determined by following the
computational procedure described by Hasselgruber
[119] by assuming pressure
balanced flow path. This technique gives three dimensional contours of the blades and
simultaneously determines the velocity, pressure and temperature profile in the turbine
wheel.
58
Figure 5-4 Turbine wheel
The turbine wheel was made with Aluminum alloy (i.e. Al-6160-T6). The
photograph of the turbine wheel is shown in Figure 5-4.
5.3 Design of nozzles
Thermodynamic state at the throat and vane less space, a convergent type of
nozzle, which gives subsonic flow at nozzle exit, is desired. The Figure 5-5 shows the
schematic diagram of nozzle ring with nozzle dimensions. If Dt is nozzle throat circle
diameter and C mt the meridian component of the nozzle throat velocity, then
considering the mass balance equation,
C mt 
m
 Dt bt t
(5.44)
Where, bt is the height of the nozzle. It is usually somewhat smaller than the turbine
inlet blade height. This allows some margin for expansion in annular space and also to
accommodate the axial misalignment. So it is taken as 0.8 times of the turbine inlet
blade height.
bt  0.8  b2  0.567 mm
Following Reference [41] Dt is 1.08 times of D 2 , but in the current design it is
taken as 1.068 times of D 2 to decrease the vane less space.
Dt  1.068  D2  0.0316 m  31.6 mm
59
Figure 5-5 Major Dimensions of Nozzle
The velocity at exit of the throat consists of two components, C mt and C t . The
meridian component is perpendicular to the nozzle throat circle diameter, which
determines the mass flow rate whereas the other component C t is tangential to the
throat. From conservation of angular momentum in free vortex flow over the vaneless
space, the tangential component of the throat velocity,
C t 
C  2D 2
Dt
(5.45)
Assuming isentropic expansion in the vane less space, s t  s 2
Corresponding
to
the
assumed
value
of
pressure
at
the
throat, pt  4.525 bar and s t  s 2  5.273 kJ/kg K the properties at the throat are
found from the properties package are as follows
Tt  106.01 K, t  15.61 kg/m 3 , ht  102.71 kJ/kg, Cst  200.97 m/s
Substituting the values of the properties, the velocities at the throat are found as
follows
C mt  86.95 m/s; C  t  157.698 m/s and C t  180.08 m/s
60
The enthalpy at the throat is given by
 h01 
Ct 2
 h01 
2
C mt 2
2

C t 2
2
86.952 157.6982
 118.92 

 102.71 kJ/kg
2  1000 2  1000
The value of enthalpy at the throat obtained from equation
matches with that
obtained from the assumed value of the pressure, hence pressure at the throat,
Pt  4.525 bar .
Mach number at the throat, M t 
C t 181.767

 0.906
C st
200.66
This leads to subsonic operation with no loss of energy which may be due to
aerodynamic shocks.
Sizing of the nozzle vanes
The correct throat angle for finite trailing edge thickness of nozzle is estimated
using the conservation of momentum & continuity of flows [40] . Aerodynamically, it is
desirable to make the trailing edge as thin as mechanical design consideration will
allow. Using the continuity equation and the density at the throat, the throat width
Wt and the throat angle t are calculated as follows.
Assuming the number of nozzles, Z n  23
Width at the throat
wt 
m tr
0.07646

 0.00208 m  2.08 mm (5.46)
Z n bt t C t 23  0.5672  15.61  180.081
and the throat angle,
 C mt 
o
  28.87
 C t 
t  tan1 
(5.47)
It may be noted that the throat outlet angle is different form the turbine blade inlet
angle and the discrepancy is due to the drifting of fluid in the vane less space.
The blade pitch length, p n is estimated as,
pn   Dt Z n  3.142  31.6 / 23  4.318 mm
(5.48)
61
t is the angle between the perpendicular to the throat width w t and the tangent to
the throat circle diameter. From Figure 5-5, the inner diameter of the nozzle ring is
calculated as,
Dn  Dt 2  w t 2  2w t Dt Cos t  29.8 mm
(5.49)
Where t is angle between Dt and w t .
In cascade theory, blade loading and cascade solidity are defined as:
u 
Vu
C mn
 cot t  cot  0
n 
From cascade notation, cot t  cot  
(5.50)
Chn
S
u
(5.51)
and cot  0  cot  
2
u
2
The separation limit in an approximate way is expressed by a minimum required
solidity. Its value is found from the aerodynamic load coefficient  z defined as the ratio
of actual tangential force to ideal tangential force, also known as Zweifel number. The
optimum value for the aerodynamic load coefficient is about 0.9. Thus the chord length
of nozzle can be found from the equation of solidity and expressed as
Chn 
2s  cot t  cot  0  sin2 t
 z sin s

2 u  S
2
 
u  
 z 1   cot  
 sin s
2  
 
(5.52)
Where
S = tangential vane spacing =
 Dn
 4.07 mm
Zn
 = cascade angle or mean vector angle
 cot t  cot  0
2

  cot 1 

0
  44.627

(5.53)
s = stagger angle =   m and
Nozzle inlet angle,  0 is taken as 78 [33]
Stagger deviation angle curve from the Figure 5-6 [120] gives  m as a function of
  0  t  for various values of  , yields  m = -2.4°, leading to:
u  1.6011,   44.6274, s  42.2274 and Chn  5.02 mm
62
Figure 5-6 Stagger angle deviation graph for different cascade angle [120]
5.4 Design of diffuser
Kinetic energy at the rotor outlet should be recovered using a diffuser. The best

exit diameter  throat diameter 
suited diffusing angle   tan1

2 * length of the diffuser


which minimizes the
loss in pressure recovery is 5o-6o and the aspect ratio is maintained, 1.4 to 3.3.
Diameter of the diffuser at inlet is equal to the turbine wheel inlet diameter with
recommended radial clearance of 2% of the turbine exit radius, which is approximately
0.1 mm for wheel. Diffuser outlet diameter is equal to the outlet piping diameter which
gives the length of the diverging section as 87.4 mm. The volume flow rate at the exit
from the diffuser is calculated from the equation(5.5).
Aex 
Qex 0.0148

 0.00087 m 2
C ex
17
(5.54)
Assuming radial clearance 0.1 mm
Diameter at inlet of diffuser, Dd ,in  D2  2  radial clearance=29.8 mm
Area at the inlet of diffuser, Ad ,in 
Diameter at exit of diffuser, Dex 

4
 Dd ,in 2  0.00069746 m 2
4  Aex

 0.0333m=33.3 mm
63
Diameter at throat of diffuser, Dd ,th  D3,tip  2  radial clearance = 18 mm
Taper angle of the diverging section, d  50
The length of the diverging section of the diffuser is given by
Ld 
(Dex  Dd ,th )
2 tan( d )

0.0333  0.018
 87.4 mm
2 tan(5)
(5.55)
The length of the converging section depends on the height of the turbine wheel
and the turbine wheel height depends on blade profile program.
The nozzle and the diffuser are combined to make one component as shown in
Figure 5-7. In the Figure 5-8, the nozzle is covered with the nozzle cover and its bottom
part is the difusser.
Figure 5-7 Nozzle Diffuser
Figure 5-8 Nozzle cover
64
5.5 Design of shaft
The dimensions of the shaft based on data from comparable installations by
other workers are verified for maximum stress, critical speed and heat conduction. Ino
et. al. [33] have chosen a shaft diameter of 16 mm for their helium turbine rotating at
2,30,000 r.p.m, while Yang et al [26] have chosen 18 mm for their air turbine rotating
at 180,000 r/min. The shaft diameter 16 mm, length 108 mm with collar diameter 44
mm have been taken as design input for this work. The Figure 5-9 shows the fabricated
shaft.
Figure 5-9 Shaft
The peripheral speed on the shaft surface is computed to be
V surf 
d
2

14534.67  16
 116.27 m/s
2000
(5.56)
and that on the tip of the collar is 319.76 m/s.
Taking density of Monel K-500 material as 8440 kg/m3, a preliminary calculation
considering the collar as a solid disk gives [121]

1
1
ssV 2surf   8440  290.6932  287.65 MPa
3
3
(5.57)
This value is less than recommended design stress of 790 MPa for Monel K-500.
Shaft speed is generally limited by the first critical speed in bending. This
limitation for the given diameter determines the shaft length. The overhang distance
into the cold end, strongly affects the conductive heat leak penalty.
The first bending critical speed for a uniform shaft is given by the formula [120]
f  0.9 d l 2 
E
Hz

(5.58)
65
Where d is the diameter of the shaft, l is the length, E is the Young‘s modulus and  is
the density of the material. Considering the shaft to be monel K-500 of diameter 16 mm
and length 108 mm, the bending critical speed is
f  0.9 d l 2 
E
1.8  1011
 0.9  0.016 / 0.12 
 6650.09 Hz  399,005.4 rpm

8440
This is well above the operating speed of 1, 38,777 rpm.
5.6 Design of brake compressor
The shaft power generated by the turbine is transferred by using brake
compressor which is mounted on the shaft. Hence the brake compressor is designed to
transfer 2.852 kW of power.
The design inputs for design of brake compressor are given in Table 5-2.
Table 5-2 Basic input parameters for design of brake compressor
Process gas
Air/nitrogen
Power to be dissipated, P
2852 W
Angular speed, 
14534.67 rad/s (1, 38,777 rpm)
Inlet total pressure, p 04
4.1 bar
Inlet total temperature,T 04
300 K (ambient temperature)
Expected efficiency, b
60%
To determine the compressor discharge pressure and flow rate, an estimate of
the static thermodynamic properties at the inlet is needed.
Stagnation density at 4.1 bar and 300 K is 4.61 kg/m3
Let static density at inlet be,
4  0.96 04  0.96  4.61  4.4256 kg/m 3
(5.59)
From the s  d s diagram for a single stage centrifugal compressor, the
operating point is chosen in order to achieve proper velocity triangles within the
constraints of available power and rotational speed. The operating conditions are
s  1.95, d s  2.9 .
66
Where specific speed,
s 
 Q4
hs 3/4
(5.60)
and specific diameter,
ds 
D5 hs 1/4
(5.61)
Q4
hs is the isentropic static head drop across the compressor and D 5 is the diameter of
the impeller. Balje [120] has pointed out that mixed flow geometry is necessary to
obtain the highest efficiency at these values of s and d s
Assume, hs  7504.4 J/kg ;
From equation(5.60), volume flow rate at the inlet of compressor
2
2
 (hs )3/4 
1.95  (7504.4)3/4 
3
Q4   s



  0.0117 m /s

14534.67




(5.62)
From equation(5.61),
D5 
d s Q4
2.9 0.0117

 0.0337 m=33.7 mm
1/4
(hs )
(7504.4)1/4
(5.63)
Mass flow rate through the compressor,
mb  4Q4  4.4256  0.0117  0.05179 kg/s
(5.64)
Assuming zero swirls at inlet, Power input,
P   sf mbU 52
Where,
(5.65)
  power input factor  1.02 [122]
 sf  Slip factor  0.9
[122]
U 5  Peripheral speed at exit 
D5
2
mb  Mass flow rate through compressor  1Q1
Substituting the values in equation(5.65), we get
P 
1.02  0.9  0.05179  14534.672  0.03372
 2852.4 W
4
This agrees with the power to be dissipated.
Hence static enthalpy drop across the compressor is hs  7504.4 J/kg
67
Power dissipated,
P 
mb h0s 4Q 4 h0s

 2852 W
b
b
(5.66)
here h0 s is the total isentropic head drop across the compressor.
Substituting the value of Q1 in equation(5.66)
Total isentropic head drop, h0s 
2852.4  0.60
 33047.72 J/kg
4.4256  0.0117
Peripheral speed at exit, U 5  D5 / 2 
Assuming exit to inlet diameter ratio,
14534.67  0.0337
 244.94 m/s
2
D5
 2.25
D4
and blade height to diameter ratio at inlet as
b4
 0.2
D4
Inlet diameter, D5  D4 / 2.25  0.0337 / 2.25  0.015 m = 15 mm
Inlet blade height, b4  0.2  D4  0.2  0.015  3 mm
Number of blades of brake compressor
There are several empirical relations for determining the optimum number of
blades. Well known among them are
Eck
:z b  8.5 sin 5 (1 
D 4 1
)
D5
 D  D 5    4  5 
Pfleirderer : z b  6.5  4
 sin 

 D5  D 4   2 
1
Stepanoff : z b  5 with 5 given in degrees.
3
(5.67)
The equations(5.67) give 17, 16 and 18 blades respectively for the impeller. A
choice of 12 numbers of blades and thickness of 0.75 mm are assumed from the
present design point of view.
Inlet velocities
Assuming number of blades, z b  12 and a uniform thickness t b  0.075 mm , the
radial absolute velocity C m 4 (which is also equal to the absolute velocity C 4 in the
absence of inlet swirl) is given as:
Cm4  C 4 
Q4
 102.309 m/s
 D 4  Z b t b   b 4 )
68
The peripheral velocity at inlet is computed to be:
U4 
D 4
2
 109.01 m/s
The inlet blade angle  4 and the inlet relative velocity W 4 are computed from the inlet
velocity triangle shown as,
 4  tan1
Cm4
 43.18o ,
U4
W4  U 42  C 42  149.5 m/s
The relative Mach number at inlet,
MW 4  W4
 RT 4  0.422 .
This value indicates that the flow is subsonic in nature.
Thermodynamic variables at inlet and exit
Static temperature at inlet:
T 4  T 04 
C 42
 294.99 K
2C p
(5.68)
Inlet static pressure:

 T   1
p 4  p 04  4   3.868 bar
 T 04 
(5.69)
 being the specific heat ratio 1.41. The density at inlet is calculated as
4 
p4
 4.4235 kg/m 3
RT 4
(5.70)
This is close to the assumed value equation (5.59) of 4.4256 kg/m3.
The rise in stagnation temperature through the compressor can be obtained
from the power expended and the mass flow rate through the compressor. Thus
T 05  T 04 
P
2852.3
 300 
 352.688 K
mb C p
0.05179 *1045.29
(5.71)
The exit stagnation temperature for an isentropic compressor with isentropic
efficiency, b  0.6 is estimated as,
T 05s  T 04 
b P
 331.73 K
mb C p
(5.72)
69
The corresponding stagnation pressure is found to be:

p 05
 T   1
 p 04  05s   5.80 bar
 T 04 
(5.73)
The absolute exit velocity:
C 5  2 h05s  h5s   279.08 m/s
(5.74)
Absolute Mach number at the outlet of the compressor,
M5 
C 5 339.62

 0.921
C st 368.78
(5.75)
Using the value of 0.9 for the slip factor, the tangential velocity:
C  5  0.9 U 5  220.44 m/s
C m5 
C 52  C 25  171.14 m/s
(5.76)
(5.77)
The exit blade angle:
1

Cr5 
o
  81.85
U5 C5 
(5.78)
 C m5 
o
  41.104
C
 5 
(5.79)
5  tan 
and the absolute exit angle :
5  tan1 
The relative velocity at exit:
W5  C m 5 cos ec (5 )  172.88 m/s
(5.80)
Relative mach number at outlet,
Mw 5 
W5
172.88

 0.4767
C s 5 368.779
Exit temperature
T 5  T 04 
hadst
Cp
 315.22 K
(5.81)
and exit pressure:

 T   1
p5  p 04  5   4.45 bar
 T 04 
(5.82)
Density at exit:
5 
p5
 4.757 kg/m3
RT 5
(5.83)
70
The required blade height at exit,
b5 
mb
( D5  Z b t b ) 5C m 5
(5.84)
After obtaining all the dimensions the brake compressor has been fabricated.
Figure 5-10 shows the photograph of the brake compressor.
Figure 5-10 Brake compressor
5.7 Selection of journal and thrust bearing
Successful working of a turboexpander strongly depends on the performance of
the bearings and their protection systems. The main functions of the bearings are,
i)
To support the rotor in the correct position relative to the static parts of
the machine.
ii)
To permit the rotor to run stably up to the design speed.
iii)
To withstand the axial and radial force imposed on the rotor by the
working fluid. Thrust bearing supports the axial thrust load comprises of
the rotor weight and the difference of force due to pressure between the
turbine and the compressor ends. The radial load arises primarily due to
rotor imbalance and is taken up by a pair of aerodynamic journal
bearing.
For the turboexpander gas lubricated bearings are used, i.e., aerostatic thrust
bearings and the aerodynamic tilting pad journal bearings. The main advantages of
these bearings are
i)
Complete absence of oil contamination because the process gas
(Nitrogen) can be used for lubrication.
ii)
It can accept a wide range of operating temperature.
71
iii)
Extremely high speed can be achieved.
iv)
Heat in flow can be minimized.
v)
High stability to self-excitation and external dynamic load.
(i) Pivot-less tilting pad journal bearing
Two number of pivot less tilting pad journal bearings are used at the two
ends of the shaft as shown in Figure 5-1. A pivot-less tilting pad bearing consists
of three pads floating around the journal, within the pad housing, surrounded by
gas films on all sides. Each pad basically consists of a front face that forms the
bearing surface, and its back face, consists of the network of three holes. High
pressure from the bearing surface is communicated to the back face of the pad
through the holes. This generates a pressure profile at the back face. The forces
coming into picture are the aerodynamic load on the pad, the frictional force on
the bearing surface and the force due to pressure distribution at the pad back
face. The normal forces developed in the bearing clearance and at the back
face, along with the frictional force due to rotor motion, determine the
equilibrium of pad tilt. This type of tilting pad bearings is especially suited for
supporting small rotors.
Figure 5-11 Pad
The dimensions of the tilting pad journal bearings are determined by the
following the procedure of Chakraborty [123], the basic input parameters are
i)
Bearing gas (Air /Nitrogen)
ii)
Ambient conditions (i.e. Pressure and Temperature)
iii)
Shaft diameter
iv)
Rotational speed
72
Apart from these basic input parameters, some other additional input
parameters are required for the pad geometry as shown in Table 5-3. The
calculation of pad geometry is shown in Table 5-4.
Figure 5-12 Pivot less tilting pad journal bearing
Figure 5-13 Pad and Rotor geometry
Table 5-3 Input parametrs to determine pad geometry
Length to diameter ratio
1
Pad clearence with rotor,C
0.015 mm
Pad angle ratio,
1.25
Clearence with pad housing,Ho
0.005 mm
Wedge depth, 
1 mm
Total pad angle,'
111
Connecting hole diameter, dcon
1.75 mm
Bleed hole diameter,dbleed
1.3 mm
Included angle between connecting holes, holes
60
73
Table 5-4 Pad geometry
Pad radius,r
7.4638 mm
Pad housing radius, ro
7.505 mm
Corrected Total pad angle,'c
111.59
Angular extent od pad trailing edge wedge,
22.318
Wedge width, Dw
3.1023 mm
Angular extent od pad trailing edge wedge,c
28.9
Pad angle ratio,c
1.3495
Effective pad angle, 
82.69
(ii) Aerostatic thrust bearing
Two number of aerostatic thrust bearing are used at the two sides of the
collar as shown in Figure 5-1. As the thrust load of the aerostatic thrust bearings is
unidirectional, that‘s why a double thrust bearing is always provided as a stop in
case of accidental thrust reversal. The shaft is vertically oriented and runs at high
speed.
The aerostatic thrust bearing dimensions are checked by following the
procedure of Chakraborty [123] , for the 50 N of load with the conditions and
dimensions shown in Table 5-5. The aerostatic thrust bearing has three supply gas
lines. At the backside of the bearing there is a gas plate which collects the exhaust
gas comes out of the bearing.
Table 5-5 Aerostatic thrust bearing input parameters
Supply Gas Pressure, p0
8 bar
Exhaust Pressure, pa
1.2 bar
Ambient Temperature , T
300 K
Number of feed holes, n
18
Outer diameter of thrust bearing,d1
44 mm
Inner diameter of thrust bearing,d2
18 mm
Feed hole pitch circle diameter,d0
30 mm
Total bearing clearance, ht
0.03 mm
Feed hole coefficient of discharge, Cd
0.8
Hole diameter upper side, dhu
0.8 mm
Hole diameter lower side, dhl
0.4 mm
74
Table 5-6 Aerostatic thrust bearing clearance at load and no load
Bearing clearance at no load
Bearing clearance at on load
Upper side
17.57 m
16.47 m
Lower side
12.43 m
13.53 m
Total
30.00 m
30.00 m
Figure 5-14 Aerostatic thrust bearing
Figure 5-15 Exhaust gas plate
5.8 Supporting structures
The supporting structure of the turboexpander can be divided into 3 parts.
(i) Cold end housing
(ii) Bearing housing
(iii) Warm end housing
75
(i) Cold end housing
The cold end housing is the lower most part which is capable to hold the
Teflon insulation rings so that the heat could not enter into it. It contains the nozzle
diffuser centrally. It takes the process gas inside and cooled gas comes out centrally
from the diffuser.
Figure 5-16 Cold end housing
(ii) Bearing housing
The bearing housing is the central component providing support to the two
journal bearings and the two thrust bearings. It contains total six inlet tubes for
supply of bearing gas to the aerostatic thrust bearings. And an exit hole for exhaust
of the bearing gas. Two lock nuts are provided, one in turbine side and another on
compressor side to set all the bearings and insulator inside the housing. Flanges are
provided at the both ends of the housing to attach the warm end casing and cold
end casing.
Figure 5-17 Bearing housing
76
(iii) Warm end housing
The warm end housing has a nozzle to the brake compressor which is fitted
above brake compressor by shrink fit operation. There is an inlet and exit tube
through which air is sucked in and compressed air goes out. At the top of the warm
end housing there is a controller to control the gas inlet. The warm end housing is
made up of Aluminum alloy material.
Figure 5-18 Warm end housing
5.9 Other turboexpander components
(i) Labyrinth seals
For each succeeding stage the pressure increases and therefore it is
necessary to seal the interface between the dynamic and the stationary
components. The simplest of these inter-stage seals is the labyrinth seal that work
by creating turbulence in the cavities and thereby restricting flow from the high
pressure side to the low pressure side. Seals of this type are favorable because
there is no contact between the stationary and the moving parts; hence there is no
mechanical friction or wear. The principle is shown in Figure 5-19. It is made up of
Teflon.
77
Figure 5-19 Principle of Labyrinth Sealing
Figure 5-20 Labyrinth Seal
(ii) Thermal Insulations
The thermal insulators are required to separate the inlet process gas from the
expanded cold gas. Two number of nylon insulation rings are provided.
(iii) Spacer
There is a spacer required to maintain the distance between the two thrust
bearings. The width of the space plays an important role. It depends on the total
thrust bearing clearance between the thrust bearing and the collar width of the
shaft. The holes are provided along the circumference of the space. The function of
the holes is to outlet the bearing gases from the thrust bearing. Another function is
78
to insert the probe through any of the holes so that it will reach near to the collar of
the shaft to measure the rotation of the shaft.
Figure 5-21 Thermal Insulation
Figure 5-22 Spacer
(iv) Locknuts
Two number of locknuts are provided to keep the all the bearings along with
the shaft inside the bearing housing. One is kept on turbine side, while other one is
on the brake compressor side. Threads are provided on their circumference for
fastening in the bearing housing. Two numbers of holes are provided for tightening.
Figure 5-23 Lock Nut turbine Side
Figure 5-24 Lock Nut compressor Side
79
6. Chapter VI
Assembly and Instrumentation
6.1 Available equipment
(i) Compressor
The compressor is available in our laboratory. The compressor is a Kaerser
make oil injected twin screw compressor. The specifications of the available
compressor are given in the Table 6-1. The compressor has the capacity to deliver
336 m3/h of air at maximum working pressure. The flow rate requirement for the
liquefaction plant is fulfilled by the compressor. The photograph of the compressor
is shown in Figure 6-1. The compressor runs automatically and takes care of the
pressure range set for the operation. Beside the compressor there is an
arrangement to control the flow and regulate the delivery air pressure. The pipe line
and instrumentation diagram for compressor is shown in the Figure 6-2.
Table 6-1 Specification of the compressor
Make
Kaeser (Germany)
Model
BSD 72
Profile of screw
Sigma
Free air delivery
336 m3/hr (at 10 bar of working pressure)
Suction pressure
Atmospheric
Maximum Pressure
11 bar
Motor
37 kW, 74 amps, 3 Φ, 50 Hz, 415V±10%, 3000 rpm
(direct coupled)
Oil capacity
24 L
Cooling
Air
80
Figure 6-1 Photograph of the compressor
Figure 6-2 Arrangement for regulating the pressure and flow rate
81
6.2 Fabricated components
(i) Heat exchangers
As per design discussed in Chapter-IV, two numbers of brazed aluminum
plate fin heat exchangers are fabricated by APOLLO HEAT EXCHANGERS Pvt. Ltd,
Thane, Maharastra, India. At the inlet and exit end of the heat exchanger transition
joints are welded. The transition joints are procured from THEVENET + CLERJOUNIE
of France. Photographs of the first and the second heat exchanger are shown in
Figure 6-3 and Figure 6-4 respectively.
Figure 6-3. Photograph first heat
exchanger
Figure 6-4. Photograph second heat
exchanger
82
(ii) Turboexpander
Two set of turboexpander fabricated according to the dimensions determined
in chapter V. The turbine wheel and brake compressor have been manufactured at
TURBOCAM India Pvt., Goa, India Ltd. by using 5-axis CNC machine. The
turboexpander housing and other components of the turboexpander have been
manufactured at Central Tool and Training Centre, Bhubaneswar. The production
drawings of the turboexpander parts are given in appendix A.
The shaft of the turboexpander rotates at a high speed about 1, 40,000 rpm.
The clearance between the bearing and the shaft is also very small. So a precise
dynamic balancing of the shaft is required [124]. The dynamic balancing was done
by Schenck Ro Tec Gmbh make hard bearing type precision balancing machine at
BARC, Mumbai. The balancing machine is meant for small rotors and able to balance
less than 10 mg per plane. The residual unbalance after balancing of by removing
weight from the shaft is mentioned in Table 6-2.
Table 6-2 Balancing report of shaft
Shaft
Before balancing
After balancing
Turbine Side
1.16 mg 135°
2.30 mg 23°
Brake wheel side
37.9 mg 6°
3.43 mg 73°
Turbine Side
11.3 mg 50°
2.29 mg 25°
Brake wheel side
126 mg 236°
4.45 mg 162°
No.
1
2
Speed (rpm)
2111
2310
The assembly of the turboexpander components need to be done in a very
clean and dustless environment. Before proceed for the assembly of turboexpander
the shaft, the thrust bearing surface and the tilting pad bearing have been polished
to get mirror surface finish. At first the labyrinth seal is inserted inside the bearing
housing. Then lower tilting pad bearing is inserted. The lower thrust bearing is
inserted by taking care of the bearing gas feed holes of the bearing and housing are
matching. Then the shaft is inserted followed by the spacer, upper thrust bearing
and upper tilting pad bearing. Now both the ends of the housing are tightened by
using the locknuts. The turbine wheel is fitted in the lower side while brake wheel is
fitted on the upper side of bearing housing. The bearing housing is ready to
assemble with cold end housing. The nozzle diffuser is fitted inside the cold end
83
housing. Then it is covered with nozzle cover. The thermal insulator rings are
inserted inside the annular space in the cold end housing. The turbine end is taken
inside the cold end housing and bolted with placing an O-ring. The upper end flange
of the bearing housing bolted with the hot end housing. A nozzle for the brake
compressor is shrink-fitted inside the hot end housing.
(iii) J-T expansion valve
The Swagelok make needle valve is suitably modified to be used as expansion
valve. To operate it from the top of the cold box, a long stem is attached with the
valve along with a metering handle to know the percentage opening of the valve.
The long stem is made rigid and leak proof so that the valve can be operated from
the top without any leakages. The assembly diagram of the long stem handle is
shown in Figure 6-5 and the photograph of the expansion valve assembly is shown
in Figure 6-6.
(iv) Phase separator
The phase separator is a cylindrical container with rounded lower end and a
circular flange with eight numbers of holes at the upper end as shown in Figure 6-7.
On the both sides of the phase separator two no. of brackets are welded which will
help to hang the container. The container is designed with 25 litres of capacity by
volume. So it could accommodated easily inside the cold box and able to contain
liquid nitrogen for an hour. The phase separator is made up of stainless steel
(SS304).
There is a cover plate over the phase separator. The cover plate has a vapour
nitrogen outlet hole, a liquid-vapour nitrogen inlet hole, a liquid nitrogen outlet hole
and a feed through hole for temperature measurement. The cover plate has holes
to match with phase separator flange an it fitted with the nuts and bolts by placing
a Teflon sheet in between as gasket to prevent leakage. The cover plate diagram is
shown in Figure 6-8.
84
Figure 6-5 Long stem handle assembly for the J-T expansion valve
Figure 6-6 Photograph of expansion valve
85
Figure 6-7 Dimensions of phase separator
Figure 6-8 Cover plate of phase separator
86
(v) Cold box
All the equipment need to be housed in a vacuum vessel to reduce the cold
loss. The vessel is called cold box. A double walled cold box is fabricated with a
height of 1800 mm and diameter of 850 mm. A lower vacuum maintained inside the
main chamber while high vacuum is maintained at the annular space provided at the
double wall. On the flange of the cold box holes are drilled for the outlet of process
gas pipes and also to accommodate the turboexpander.
Figure 6-9 Photograph cold box
Figure 6-10 Holes on the cold box flange
87
6.3 Instrumentation
(i) Temperature measurement
To study the performance of the liquefier and to monitor the cool down
behavior, it is necessary to measure the temperature. So temperature is measured
by using Platinum Resistance Temperature Detector (RTD). It is the most linear and
stable temperature sensor. The Pt-100 type RTDs is fitted at different locations to
measure the temperatures. Prior to the using of the RTD all of them are calibrated
by dipping inside liquid nitrogen. ADAM-4015 data acquisition modules are used and
it provides data output as RS-485. This is again converted to RS-232 to view in PC
by using ADAM-4520 converter.
Figure 6-11 Photograph of the RTD
(ii) Pressure measurement
The pressure is measured by using WIKA make dial pressure gauges. For high
pressure stream, 0-16 bar (g) and for low pressure stream 0-4 bar (g) range
pressure gauges are used.
(iii) Measurement of flow rate
An orifice type flow meter is used to measure the flow rate of liquid or gas.
Its working is based on the Bernoulli‘s principle which relates the pressure and the
88
velocity of a fluid stream. When the velocity increases, the pressure drop across the
orifice plate increases and vice versa.
The volumetric and mass flow rates are obtained from the Bernoulli‘s
equation by measuring the difference in fluid pressure between the normal pipe
section and at the vena contracta of the orifice plate. The pressure difference is
measured by mercury u-tube manometer. The orifice plate is calibrated with a
rotameter as shown in photograph before use.
Figure 6-12 Orifice plate calibration
(iv) Turbine speed measurement
Turbine rpm is measured by a Brüel & Kjær make piezoelectric Miniature
DeltaTron™ Accelerometers. It has high sensitivity, low mass and small physical
dimensions and easy to mount on the turboexpander body using supplied glue. The
photograph and specification is shown in Figure 6-13 and Table 6-3 respectively.
Figure 6-13 Photograph of the accelerometer used for speed measurement
89
Table 6-3 Specification of the accelerometer
Model No.
4507
Frequency
0.3 - 6000 Hz
Sensitivity
100 mV/g
Temperature
-54 - 121 ºC
Residual Noise Level in Spec Freq Range (rms) ±
0.35 mg
Maximum Operational Level (peak)
70 g
Maximum Shock Level (± peak)
5000 g
Resonance Frequency
18 kHz
6.4 Assembly of components
The 3-D model of the cold box assembly is shown in Figure 6-16 . The two heat
exchanger and phase separator are hanged from the cold box flange using tie rod. The
one end of tie rod is inserted inside a welded socket on the cold box flange and bolted.
The other end is fixed with nut and bolt arrangement with the component. Slight
movement of the component could be done with adjusting the nuts. After hanging the
components, pipes are connected by TIG welding as per the process diagram shown in
Figure 6-15.
Figure 6-14 Connection of turboexpander with the pipelines
90
The critical connection is the turboexpander with the process pipelines. As the
turboexpander is to be taken out for maintenance, it could not be connected
permanently. So it is connected by brass connector with proper sealing as shown in
Figure 6-14.
Figure 6-15 P & I diagram of Nitrogen liquefier
91
Figure 6-16 3-D model assembly of nitrogen liquefier inside cold box
92
Figure 6-17 Assembly photograph of nitrogen liquefier
Figure 6-18 Photograph of cold box flange
93
7. Chapter VII
Testing and Commissioning of the Liquefier
7.1 Introduction
The assembled cold box is required to be connected with other components with
proper pipeline. To study the experimental performance of the plant, it is required run
the plant. Before running the plant to study the performance, the turboexpander is to
be tested by the trial run of the plant. After completing the trial run of the plant
successfully, vacuum is to be created inside the annular space of the cold box and all
the components inside the cold box have to be covered by super insulation. Then the
plant will be ready for the experimental performance evaluation.
7.2 Testing of turboexpander
The turboexpander is the heart of the liquefaction plant. So the turboexpander
should run at rated speed and able to lower the temperature having required efficiency.
This is evaluated by doing trial run of the turboexpander before using it on liquefaction
plant. By doing trial run of the turboexpander, the performance is evaluated as well as
the knowledge on operation is achieved.
(i) Turboexpander trial run set up
The testing of the turboexpander has been done with air as the process fluid.
The trial run set up consists of a compressor to supply the compressed air to the
turboexpander. The photograph of the test setup is shown in Figure 7-1. Some
fraction of the high pressure gas is supplied to the turbine inlet and some to the
turboexpander bearing. To make the flow smooth and to avoid flow fluctuation the
turboexpander bearing gas supply is made through a H.P. pressure vessel. The
bearing supply gas from the H.P. pressure vessel is divided into two streams. One is
94
supplied to the upper thrust bearing and other is feed to the lower thrust bearing
through two valves to control the flow rate. The exit air from the turboexpander is
discharged to atmosphere. The exit bearing gas also vent to the atmosphere. The
brake compressor of the turboexpander sucks the atmospheric air and also its
compressed hot air leaves to the atmosphere.
Figure 7-1 Turboexpander test set up
(ii) Problems occurred during turboexpander trial run
Trial run of the turboexpander has done using the compressed air from the
compressor. After assembly, at the fresh start of the turboexpander, it rotates
smoothly and steadily. But after some start and stop the rotation is not smooth as
compared to the first run. After disassembly of the turboexpander, it is found that
the thrust bearing surface has been damaged along with surface of the shaft and
shaft collar. The photo graph of the damaged surface is shown in Figure 7-2, Figure
7-3 and Figure 7-4. The damage occurs due to the small dust particles that come
from the compressor. The dust particles are rubbed with graphite pads of the
journal bearing and also in between the collar and the thrust bearing. This is
resulting wear in the surfaces. To avoid this, filters are provided with micron size of
SS mesh as shown in Figure 7-5.
95
(iii) Performance of the turboexpander
The bearings surfaces are buffed and made mirror finished. Due to this the
shaft rotates smoothly without any friction for a long time after several starts and
stops. The turbine speed is measured by placing the accelerometer on the flange of
the bearing housing. The rotating speed of turbine is about 1,04,760 rpm (1.746
kHz). It is observed that at an inlet pressure of 5 bar, it is able to decreases
temperature of 18 °C by expanding the gas to the atmosphere. The FFT graph is
shown in Figure 7-7.
Figure 7-2 Damaged surface of the thrust bearing
Figure 7-3 Damaged surface of the shaft collar
96
Figure 7-4 Damaged shaft surface by rubbing with tilting pad bearing
Figure 7-5 Filter used to remove micron dust particles
Figure 7-6 FFT diagram for the speed of turbine wheel at 5 bar of inlet pressure
97
Figure 7-7 FFT diagram for the speed of turbine wheel at 6 bar of inlet pressure
7.3 Plant pipeline setup
Pure nitrogen gas is required for the nitrogen liquefier. The air contained
nitrogen but it also contained objectionable amount of moisture, carbon dioxide and
oxygen. At low temperatures the small quantity of moisture will form ice and block the
flow passage. Hence for the experiment, the pure nitrogen is supplied from a liquid
nitrogen Dewar. From the Dewar the liquid goes to the nitrogen vaporizer by just
opening and pressurising the Dewar (Figure 7-8). Vapour nitrogen comes out from the
nitrogen vaporizer is feed to the gas bag. For the safety of the gas bag, it is equipped
with a level indicator and oil safety valve. The maximum and minimum level of the gas
bag is marked and controlled manually to maintain the level. The secondary safety is
the oil safety valve and it is shown in Figure 7-10. It is filled with compressor oil up to
the required level. The oil remains inside till it sustains the pressure of the gas bag.
When the pressure become access the oil comes out, this indicates that the bag is over
filled. The gas bag is connected with the suction of the compressor.
The discharge gas from the compressor is connected to coil type small heat
exchanger for pre-cooling and moisture removal from the process gas. The precooling is
done by using liquid nitrogen from a LN2 cylinder. The photograph of the pre-cooler is
shown in Figure 7-11.
The pre-cooled gas divided into two streams. One is supplied to the high
pressure end of the cold box through a orifice type flow meter and a dust filter. Another
stream is supplied to bearings of the turboexpander through a buffer. The buffer is
meant for the smooth supply of the gas to the bearing without any flow fluctuation.
98
The cold box flange has one high pressure inlet to HX1 and one low pressure
exit end from the HX1. The exit from the HX1 is connected to the suction side of the
compressor. In this way the process is made a closed cycle one. The process gas is
coming in to the cold box, again return to the compressor. If some liquefaction occurs
then the makeup gas is supplied by the gas bas connected through a non-return valve.
If there is some decrease in pressure in suction side, the non-return valve gets opened
and gas flows from the gas bag to the compressor.
The top flange of the cold box has liquid nitrogen transfer valve along with burst
disk and safety relief valve. The flange has also four number of feed through
connections and these are connected to temperature sensor mounted on components
inside the cold box. The other end of the feed through connection is connected to the
data acquisition system. The data is sent to the PC through a converter. The detail
pipeline assembly drawing is shown in Figure 7-13.
Figure 7-8 LN2 Dewar to vaporizer
99
Figure 7-9 Gas bag for gaseous nitrogen
Figure 7-10 Oil safety valve
100
Figure 7-11 Coil heat exchanger for pre-cooling
Figure 7-12 Arrangement for supply of process gas to cold box and turbine bearing
101
Figure 7-13 P & I Diagram of the liquid nitrogen plant
102
7.4 Commissioning of the plant
The plant starts with filling the gas bag with nitrogen gas up to a marked level.
This is done by opening the valve of liquid nitrogen Dewar. After filling the gas bag, the
compressor is started and the valve of the bearing supply is opened first. Then cold box
inlet valve is opened slowly. At this time the exit valve from the cold box is also opened.
The whole system is filled with air. By supplying nitrogen to remove the air from the
system is called purging. It took some hours to purge completely. The coil heat
exchanger present for pre-cooling removes the condensed moisture and also filters
present at the supply end of the cold box and bearing supply removes the moisture.
After ensuring the all gas are purged the cold box exit valve is closed so that the
process becomes closed cycle. The nitrogen gas is circulated in closed cycle.
7.5 Performance of the plant
The plant performance is observed from the temperature sensor mounted at the
node locations of the components. A graphical user interface is developed in PC to
monitor and also to record the temperature at different nodes of the liquefier. The PC
along with the data acquisition system is shown in Figure 7-14.
Figure 7-14 Temperature monitoring and recording using data acquisition system
103
Figure 7-15 Turboexpander exit temperature with time
The plant is required to run continuously for several hours for cool down the
entire mass of the system and produce liquid nitrogen. Figure 7-15 shows the graph of
the turboexpander exit temperatures with time. It took around 30 minutes to decrease
the temperature from 305 K to 289 K.
104
8. Chapter VIII
Conclusions
The work reported in this thesis is an attempt to design and construct a
turboexpander based nitrogen liquefier. The development includes turboexpander, heat
exchanger, J-T valve etc. During this period a lot of important technological ideas are
implemented. A liquefier is built which will produce liquid nitrogen. The following are the
significant contributions of the thesis.
1. Process design of the liquefaction cycle has been done by equation oriented
method and optimization of the cycle has been done by using the software
Aspen Hysys. Finally process design parameters and component specifications
are fixed. This will produce 17.44 liter of liquid nitrogen per hour by supplying
296 kg/hr of gaseous nitrogen.
2. The First heat exchanger has been designed for maximum effectiveness (0.99)
and second heat exchanger has been designed to give minimum pinch
temperature (1 K). Due to the large length of the first heat exchanger, it is
divided into two halves and joined together. Again the analysis is carried out to
split heat exchanger using the software Aspen MUSE. After successful design of
the heat exchangers, these are fabricated.
3. The turboexpander has been designed for an efficiency of 75%. All the parts of
the turboexpander have required dimensional accuracy. So these parts are
designed and manufactured carefully considering thermal, design and production
points of view. The turboexpander is tested in the test rig before using in the
liquefier. It is found that the turboexpander rotates at around 1,02,375 rpm and
results a temperature drop of 18 K at room temperature in an open cycle.
4. Modification in the JT valve has been done for ease of its operation from the top
of the cold box flange.
105
5. Other components like cold box and phase separator have been designed and
fabricated. Apart from these oil safety valve is fabricated for the safety of the
gas bag. Two dust filters are made to avoid the dust as well as moisture. A small
coil type heat exchanger is developed for pre cooling of the process gas.
6. Assembly of the components carried out inside cold box by TIG welding. Piping
and instrumentation have been done for complete plant.
7. The commissioning of the plant has been done and 16 K temperature drop has
been obtained in just 30 minutes. The plant is required to run for several hours
with proper insulation of the system and vacuum inside the cold box, so that it
will produce liquid nitrogen.
The present development of turboexpander based liquid nitrogen plant is not
sufficient. Further a lot work can be carried out. Further to improve the plant, the
following are major works that could be considered,
1. The performance study of the liquid nitrogen plant could be done with
optimization of the plant parameters.
2. The dynamic analysis of the plant could be done using the process
simulation software Aspen Hysys. This can be compared with the real
existing system.
3. The experimental study of the heat exchanger and comparing the
effectiveness with the rating programs or software Aspen MUSE.
4. Similarly the second heat exchanger is important because it is two phase
heat exchanger. Attention can be focused on experimental validation.
5. The main focus could be given to the heart of the system, the
turboexpander. As it rotates smoothly, but further modification is necessary
to decrease the bearing gas consumption.
A lot of efforts are given for developing liquid nitrogen plant. A complete design
methodology is provided along with the production drawing. This will be fruitful to
develop another plant indigenously with in-house development of turboexpander and
heat exchanger.
106
References
1.
Barron, R. F., Cryogenic Systems. Oxford University Press (1985),
2.
Wolfgang, F., History of cryogenics: the epoch of the pioneers from the
begining to the year 1911. Int. J. refrigeration (2002), 25, 283-292.
3.
Flynn, T. M., Cryogenics Engineering. Marcel-Dekker Inc (1996),
4.
Timmerhaus, K. D., and Flynn, T. M., Cryogenic Process Engineering. Plenum
Press (1989),
5.
Scott, R. B., Cryogenic Engineering. (1959),
6.
Scurlock, R. G., History and Origins of Cryogenics. Oxford, Clarendon Press.
(1993),
7.
White, F. M., Fluid mechanics. 2nd ed.; McGraw-Hill BookCompany,
8.
Simms, J., Fundamentals of the turboexpander " Basic Theory and Design".
SIMMS Machinery international,Inc (2009),
9.
Sulaiman, M. Y., Turbomachinery. (2005),
10.
Collins, S. C., and Cannaday, R. L., Expansion Machines for Low
Temperature Processes. Oxford University Press (1958),
11.
Sixsmith, H., Miniature cryogenic expansion turbines - a review. Adv Cryo Eng
(1984), 29, 511-523.
12.
Kapitza, P. J. Phys. Ac. Sci. USSR.1, No.7,; (1939); p 29.
13.
Swearingen, J. S., Turbo-expanders. Trans AIChE (1947), 43 (2), 85-90.
14.
Land, M. L., Expansion turbines and engines for low temperature processes.
Advances in Cryogenic Engineering (1957), V2, 251-260.
15.
Linhardt, H. D., Cryogenic turboexpanders in Vance, R. W. (Ed). Vol. 4, 89117,
107
16.
Linhardt, H. D., Process application and design of large power output
turboexpanders. J Engg. for Industry,Trans ASME (1973), 227-232.
17.
Clarke, M. E., A decade of involvement with small gas lubricated turbine. Adv
Cryo Eng (1974), 19, 200-208.
18.
Beasley, S. A., and Halford, P. , Development of a High Purity Nitrogen Plant
using Expansion Turbine with Gas Bearing. Advances in Cryogenic Engineering
(1965), 10B, 27-39.
19.
Trepp,
C.,
Refrigeration
system
for
temperatures
below
25
K
with
turboexpanders. Adv Cryo Eng (1962), 7, 251-261.
20.
Schmid, C. In Gas bearing turboexpanders for cryogenic plant, 6th
International Gas Bearing Symposium University of Southampton England,
University of Southampton England, (Year), 131 B1:1-8.
21.
Eber, N., Quack, H. and Schmid, C., Gas bearing turbines with dynamic gas
bearings. Cryogenics (1978), 18, 585-588.
22.
Blackford, J. E., Halford P. and Tantam, D. H., Expander and Pumps in G G
Haselden (Ed). In Cryogenic Fundamentals, Academic Press London: (1971); pp
403-449.
23.
Colyer, D. B., Miniature cryogenic refrigerator alternators. Adv Cryo Eng
(1968), 13, 405-415.
24.
Colyer, D. B., and Gessner, R. L., Miniature cryogenic refrigerator
Turbomachinary. Adv Cryo Eng (1968), 13, 484-493.
25.
Izumi, H., Harada, S. and Matsubara, K., Development of small size Claude
cyclehelium refrigerator with micro turbo-expander. Adv Cryo Eng (1986), 31,
811-818.
26.
Yang, K. J., He, H. B., Ke, G., and Li, G. Y. , Application and test of miniature
gas bearing turbines. Advances in Cryogenic Engineering (1990), V35, 997-1003.
108
27.
Yang, K. J., and He, H. B., Miniature expansion turbine with grooved selfacting gas bearings. In Proc Tenth Inti Cryo Eng Conf, Butterworth, Guildford,
UK, (1984); pp 660-663.
28.
Yang, K. J., and He, H. B., Design and test of a novel miniature cryogenic
expansion turbine. Adv Cryo Eng (1986), 31, 829-833.
29.
Yang, K. J., and Munday, A. J., A grooved self-acting gas bearing for use in
cryogenic expansion turbines. Cryogenic Process and Engineering 1982, AIChE
Symposium Series (1983), 79, 90-101.
30.
Kato, T., Yamaura, H., Kawno, K., Hiyama, T., Tada, E., Kakayama,
Y.,Kawashina, Sato, M., Yoshida, J., Ito, N., Sato, S. and Shimamato, S.
A., Large scale turboexpander d velopment and its performance test result. Adv
Cryo Eng (1990), 35, 1005-1012.
31.
Kato, T., Kamiyauchi, Y., Tada, E., Hiyama, T.,· Kawano, K., Sugimoto,
M., Kawageo, E., Ishida, H., Yoshida, J., Tsuji, H., Sato, S., Xakayama,
Y., Kawashima I, Development of a large helium turbo-expander with variable
capacity. Adv Cryo Eng (1992), 37B, 827.
32.
Kato, T., Miyake, A., Kawno, K., Hamada, K., Hiyama, T., Iwamoto, S.,
Ebisu, H., Tsuji, H., Saji, N., Kaneko, Y., Asakura, H., Kuboto, M. and
Nagai, S. , Design and test of wet type turbo-expander with an alternator as a
brake. Advances in Cryogenic Engineering (1994), 39, 917.
33.
Ino, N., Machida, A.,Ttsugawa, K. and Arai, Y., Development of high
expansion ratio Helium turbo expander. Advances in Cryogenic Engineering
(1992), V37B, 835-844.
34.
Ino, N., Machida, A. and Ttsugawa, K., Development of externally
pressurized thrust bearing for high gas expansion ratio expander. Adv Cryo Eng
(1992), 37B, 817-825.
35.
Davydenkov, I. A., Davydov, A. B. and Perestoronin, G. A., Hydrogen and
nitrogen turboexpanders with high gas expansion ratios. Cryogenics & 32 (1992)
ICEC supplement (1992), 84-86.
109
36.
Davydenkov, I. A., Ravikorichy, Yu A., Davydov, A. B., Ermilov, Yu.,
Zakharova, N. E., Adler, Yu. R and Schedukhin, Development of cryogenic
turboexpanders with gas dynamic foil bearings. Cryogenics (1992), 32, 80.
37.
Baranov, A., Duzev, V., Kashirskikh, G., Mikhailov, A., Ugrovatov, A. and
Zhulkin, V., An experience in the maintenance of a liquefier from the T-15
cryogenic system, results of its reliability and capacity enhancement. Adv Cryo
Eng (1996), 41 a, 737-743.
38.
Polishchuk, E. L., Shariklankin V. I. and Lyapin. V. I., Self-contained
microcryogenic system with a turboexpander Khimicheskoe I Neftyanoe
Mashinostroenie Trans: Chemical & Petroleum Engg (1991), 27 (3/4), 217.
39.
Kun, L. C., Expansion turbines and refrigeration for gas separation and
liquefaction. Adv Cryo Eng (1988), 33, 963-973.
40.
Kun, L. C., and Hanson, T. C., High efficiency turboexpander in a N2 liquefier.
In AIChE Spring meeting, Houston, Texas, (1985).
41.
Kun, L. C., and Sentz, R. N., High efficiency expansion turbines in air
separation and liquefaction plants. In International Conference on Production
and Purification of Coal Gas & Separation of Air, Beijing, China (1985); pp 1-21.
42.
Sixsmith, H., Valenjuela, J. and Swift, W. L. , Small Brayton cryocoolers
Adv Cryo Eng (1988), 34, 827-836.
43.
Jadeja, H. T., Mitter, A. and Chakrabarty, H. D., Turboexpander application
for cryoprocessing of nitrogen and related gases. In
Proceedings of
INCONCRYO85 Indian Cryogenic Council, Tata McGraw: (1985); pp 85-101.
44.
Jadeja, H. T., Kundu, K. M.,Chakrabarty, H. D. and Mitter, A. , IFR
expansion turbine for cryogenic plants. Institution of Engineers (India) J. Ill
Mech Eng (1987), 67, 106 -112.
45.
Jadeja, H. T., Chakrabarty, H. D. Nidhi, S. C. and Mitter, A., Mechanical
reliability
and
manufacturing
process
for
indigenous
development
of
turboexpander In Proceedings of INCONCRY0-88 Indian Cryogenic Council,
(1988); pp 331-337.
110
46.
Aghai, R. R., Lin, M.C. and Ershaghi, B., High Performance cryogenic
turboexpanders. Adv Cryo Eng (1996), 41, 941-947.
47.
Aghai, R. R., Lin, M.C. and Ershaghi, B., Improvements of the efficiency of
the turboexpanders in cryogenic applications Adv Cryo Eng ((1996) ), 41, 933940.
48.
Ghosh, P. Analytical and Experimental Studies on Cryogenic Turboexpanders.
(2002).
49.
Ghosh, S. K.
Experimental
and
Computational
Studies
on
Cryogenic
miniature
cryogenic
Turboexpander. (2008).
50.
Chakravarty,
A.,
and
Singh,
T.,
High
speed
turboexpander impellers at BARC. Indian Journal of Cryogenics (2011), 36, 1-9.
51.
Menon, R., Chakravarty, A.,Goyal, M., Jadav, M., Arun, S.,Bharti, S. K.,
and Singh, T., High speed cryogenic turboexpander rotor for stable operation
upto 4.5 kHz rotational speed. Indian Journal of Cryogenics (2012), 37 (1-4), 4045.
52.
Barron, R. F., Cryogenic Heat Transfer. (1985),
53.
Robertson, J., Heat exchnager equipment for cryogenic process industry. Heat
exchnager theory and practice (1983), 469-561.
54.
Abadzic, E. E., and Scholz, H. W., Coiled Tubular heat exchangers. Adv.
Cryogenics Engg. (1973), 18, 42-51.
55.
McKeever, J., Pillarella, M., and Bower, R., An ever evolving technology.
LNG Ind Spring (2008), 44–52.
56.
Tariq, S., LNG technology selection. Hydrocarb Eng (2004), 9 (2), 71-76.
57.
Venkatarathnam, G., and Sarangi, S., Matrix heat exchangers and their
application in cryogenic systems. Cryogenics (1990), 30 (11), 907-918.
58.
Venkatarathnam, G., Effectiveness-Ntu relationship in perforated plate matrix
heat exchangers. Cryogenics (1996), 36 (4), 235-241.
111
59.
Pavan, K., and G. Venkatarathnam, Optimization of Matrix Heat Exchanger
Geometry. Journal of Heat Transfer, ASME (200), 122, 579-586.
60.
ALPEMA. The Standards of the Brazed Aluminium Plate-Fin Heat Exchanger
Manufacturers‘ Association. (2000).
61.
Chart Energy & Chemicals Inc. www.chartindustries.com.
62.
Fives Cryo. www.fivesgroup.com/fivescryogenie.
63.
Kobe Steel, Ltd. www.kobelco.co.jp.
64.
Linde AG, Germany. http://www.linde-engineering.com.
65.
Sumitomo Precision Products Co., Ltd. www.spp.co.jp.
66.
Kays, W. M., and London, A. L., Compact Heat exchangers. McGraw-Hill New
York, (1984),
67.
Kern, D. Q., and Kraus, A. D., Extended Surface Heat Transfer. McGraw-Hill
New York, (1972),
68.
Ozisik, M. N., Heat Transfer-A Basic Approach. McGraw-Hill (1985),
69.
Frank, P. I., David P. DeWitt., Theodore L. Bergman., and Adrienne S.
Lavine, Fundamentals of Heat and Mass Transfer 6th ed. John Wiley New York,
(2006),
70.
Frass, A. P., Heat Exchanger Design, 2nd ed. Wiley-Interscience (1989),
71.
Shah, R. K., Cmpact Heat Exchangers & Enhancement Technology. Begell
House Publication (1999),
72.
Smith, E. M., Plate Fin Surface OptimisationUsing Direct-Sizing. Advances in
Enhanced Heat Transfer, ASME, (2000), 4, 105–115.
73.
Shah, R. K., Subbarao, E. C., and Mashelkar, R. A., Heat Transfer
Equipment Design edited by R.K.Shah. Publishing Corporation Washington DC,
(1988),
112
74.
Taborek, J., Hewitt, G. F., and Afgtan, N, Heat Exchangers-Theory and
Practice. Hemisphere Publishing New York, (1983),
75.
Taylor, M. A., Plate Fin Heat Exchangers: Guide to their Specification and Use
HTFS. Oxon UK, (1987),
76.
Aslam, B., Hayat, M. M., Nasir, Bashir., Khan, M. H.,Ahmad, A. R., and
Khan, A. N., CFD applications in various heat exchangers design: A review.
Applied Thermal Engineering (2012), 32 (0), 1-12.
77.
Das, P. K., and Ghosh, I., Thermal Design of Multistream Plate Fin Heat
Exchangers—A State-of-the-Art Review. Heat Transfer Engineering (2011), 33
(4-5), 284-300.
78.
Pacio, J. C., and Dorao, C. A., A review on heat exchanger thermal hydraulic
models for cryogenic applications. Cryogenics (2011), 51 (7), 366-379.
79.
Kakac, S., Bergles, A. E., and Mayinger, F., Heat Exchangers, Thermal-
Hydraulic Fundamentals and Design. Hemisphere Publication (1988),
80.
Kuppan, T., Heat exchanger design handbook. Mechanical engineering. A series
of textbooks and reference books. Marcel Dekker Inc. New York, (2000); Vol.
126,
81.
Shah, R. K., Dusan,P., and Sekulic,D. P., Fundamentals of Heat Exchanger
Design. John Wiley & Sons (2003),
82.
Taborek J., Evolution of heat exchanger design techniques. Heat Transfer Eng
(1979), 1 (1), 19-25.
83.
Mueller, A. In New charts for true mean temperature difference in heat
exchangers., AIChE paper 10, 9th National Heat Transfer Conference, (Year).
84.
Roetzel, W., and Spang, B., Verbessertes diagramm zur berechnung von
wärmenbertragern. Heat Mass Transfer (1990), 25, 259–64.
85.
Iu, I. Development of air-to-air heat pump simulation program with advanced
heat exchanger circuitry algorithm. Ph.D., Oklahoma State University, (2007).
113
86.
Orth, L., Zietlow, D. C., and Pedersen, C. O. Predicting refrigerant inventory
of HFC 134a in air cooled condensers. Technical Report ACRCTR-34.; Air
Conditioning and Refrigeration Center, University of Illinois: (1993).
87.
ASPENTech, MUSE. http://www.aspentech.com/products/aspen-plate-fin.cfm.
88.
Aspen
Muse
Reference
guide.
Aspen
Technology,
Inc.
Cambridge,
Massachusetts, Vol. V2006,
89.
Venkatarathnam, G., Cryogenic Mixed Refrigerant Processes. Springer-Verlag
New York Inc (2008),
90.
Sekwang, Y., Habin, C., Dong-H, L., and Jin-Kuk, K, Process Design and
Optimization of Natural Gas Liquefaction Processes. Chemiical Engiineeriing
Transactiions (2012), 29, 1585-1590.
91.
Atrey, M. D., Thermodynamic analysis of Collin‘s helium liquefaction cycle.
Cryogenics (1998), 38, 1199-1206.
92.
Narinsky, G. B., Temirova, V. V., and Chernokov, L. V. , Influence of
thermodynamic parameters on main characteristics of helium liquefaction and
refrigeration plant. Cryogenics (1995), 35, 483-487.
93.
Jianlin, Y., Improving the performance of small Joule–Thomson cryocooler.
Cryogenics (2008), 48, 426-431.
94.
Nandi, T. K., and Sarangi, S. , Performance And Optimization Of Hydrogen
Liquefaction Cycles. Int. J. Hydrogen Energ (1993), 18 (2), 131-139.
95.
Harvey, S., and Kane, N'Diaye, Analysis of a reheat gas turbine cycle with
chemical recuperation using Aspen. Energy Conversion and Management (1997),
38 (15–17), 1671-1679.
96.
Melaaen, E., Owren, G.,Wadahl, A., and Wagner, U., Simulation Program
for Cryogenic Plants at CERN. In Proceedings of the Sixteenth International
Cryogenic Engineering Conference/International Cryogenic Materials Conference,
Haruyama, T.; Mitsui, T.; K. YamafujiA2 - T. Haruyama, T. M.; Yamafuji, K., Eds.
Elsevier Science Oxford, (1997); pp 99-102.
114
97.
Maekawa, R., Ooba, K., Nobutoki,M., and Mito,T, Dynamic Simulation of a
Helium Liquefier. Advances in Cryogenic Engineering (2004), 49, 192-199.
98.
Deschildre, C., Barraud, A., Bonnay, P., Briend, P., Girard, A., Poncet, J.
M., Roussel,P., and Sequeira, S. E., Dynamic Simulation of an Helium
Refrigerator. Advances in Cryogenic Engineering (2008), 53, 475-482.
99.
Rogez, E., Bradu B.,Moraux, A.,Pezzetti, M.,Gayet, Ph., and Coppier, H.
In A simulation study for the virtual commissioning of the CERN central helium
liquefier, ICEC 22 - ICMC 2008 Conference, Seoul, Korea, Seoul, Korea, (Year),
1-5.
100.
Bradu, B., Gayet, Philippe., and Niculescu, Silviu-Iulian., A process and
control simulator for large scale cryogenic plants. Control Engineering Practice
(2009), 17 (12), 1388-1397.
101.
Maekawa, R., Ooba, Kouki.,Nobutoki, Minoru., and Mito, Toshiyuki.,
Dynamic simulation of the helium refrigerator/liquefier for LHD. Cryogenics
(2005), 45 (3), 199-211.
102.
Stirling Cryogenics. http://www.stirlingcryogenics.com/.
103.
Linde Cryoplants. http://www.lindecryoplants.com/en.php.
104.
Pacific Consolidated Industries. http://www.pci-intl.com/.
105.
Cryomech Inc. www.cryomech.com/.
106.
Kelvin International Corporation http://www.kelvinic.com.
107.
Aspen Hysys User's Guide. Aspen Technology, Inc. Cambridge, Massachusetts,
108.
Moon, J. W., Lee, Y. P., Jin, Y. W., Hong, E. S., and Chang, H. M.,
Cryogenic Refrigeration Cycle for Re-Liquefaction of LNG Boil-Off Gas. In
International Cryocooler Conference, Inc., Boulder, (2007); pp 629-635.
109.
Lemmon, E. W., Jacobsen, R. T., Penoncello, S. G., and Beyerlein, S. W.,
ALLPROPS 4.2 – Computer Programs for Calculating Thermodynamic Properties
115
of Fluids of Engineering Interest, Centre for Applied Thermodynamic Studies:
University of Idaho (1995).
110.
Seshaiah, N. Experimental and computational studies on oil injected twin screw
compressor. NIT Rourkela, (2006).
111.
Maiti, D. K. Heat Transfer and Flow Friction Characteristics of Plate Fin Heat
Exchanger Surfaces – A Numerical Study. (2002).
112.
Manglik, R. M., and Bergles, A. E, Heat Transfer and Pressure Drop
Correlations for the Rectangular Offset Strip Fin Compact Heat Exchanger
Experimental Thermal Fluid Science (1995), 10, 171-180.
113.
Joshi, H. M., and Webb, R. L., Heat Transfer and Friction of the Offset Stripfin Heat Exchanger. Int. J. Heat Mass Transfer (1987), 30 (1), 69-84.
114.
Frost, W., Heat Transfer At Low Temperatures. Plenum Press New York,
(1975); 208,
115.
Denton, J. D., The turboexpander - a design, make and test student project.
ASME-96-GT- 191 (1996).
116.
Jekat, W. K., An Impulse Type Expansion Turbine. Advances in Cryogenic
Engineering (1957), V2, 250-260.
117.
Whitfield, A., and Baines, N. C. , Design of Radial Turbomachines. Longman
Scientific & Technical (1990),
118.
RohliK, H. E., Analytical determination of radial inflow turbine geometry for
maximum efficiency. NASA TN D-4384 (1968).
119.
Hasselgruber,
H.,
Stromungsgerechte
gestaltung
der
laufrader
von
radialkompressoren mit axialelaufradeintrict. Konstruction (1958), 10 (1) 22 (in
German).
120.
Balje, O. E., Turbomachines. John Wiley and Sons (1981),
121.
Sixsmith, H., and Swift, W. L., Cryogenic turbines and pumps in Hands, B. A:
(Ed.). Academic Press (1986),
116
122.
Dixion., S. L., Fluid Mechanics and Thermodynamics of Turbomachinery 3rd
ed.; Pergamon Press (1978),
123.
Chakravarty, A. Analytical and Experimental Studies on Gas Bearings for
Cryogenic Turboexpanders (2000).
124.
Muhammad, S., Malik M. Nazeer, and Abdul, Qayyum, Balancing of a
cryogenic turbo-expander assembly. In 15th International Congress on Sound
and Vibration, Daejeon, Korea, (2008); pp 2381-2386.
117
Appendix A
Production Drawings of
Turboexpander
Turboexpander Part List :
SL. NO PART NO
PART NAME
1
1
Shaft
2
2
3
MATERIAL
QUANTITY
Monel K-500
1
Aerostatic thrust plate
SS304
2
3
Aerostatic thrust plate
SS304
2
4
4
Exhaust gas plate (lower)
SS304
2
5
5
Exhaust gas plate (upper)
SS304
2
6
6
Tilting pad housing
Monel K-500
2
7
7
Pad
High density metal
impregnated graphite
6
8
8
End pad plate
SS304
4
9
9
Spacer
SS304
1
10
10
Bearing block
SS304
1
11
11
Lock nut (Turbine side)
SS304
1
12
12
Lock nut (Ccompressor
SS304
1
13
13
Nozzle diffuser
Brass
1
14
14
Cold end housing
SS304
1
15
15
Cover T nozzle
SS304
1
16
16
Nozzle brake compressor
Al Alloy
1
17
17
Heat exchanger
Al Alloy
1
18
18
Stem
Brass
1
19
19
Stem tip
Brass
1
20
20
Thermal Insulator-1
Nylon-6
1
21
21
Thermal Insulator-2
Nylon-6
1
22
22
Thermal Insulator-3
Teflon
1
Figure A-1 Turbine wheel
120
121
122
Figure A-2 Brake compressor
123
124
125
Figure A-3 Shaft
126
Figure A-4 Aerostatic thrust plate (lower)
127
Figure A-5 Aerostatic thrust plate (upper)
128
Figure A-6 Exhaust gas plate (lower)
129
Figure A-7 Exhaust gas plate (upper)
130
Figure A-8 Tilting pad housing
131
Figure A-9 Pad
132
Figure A-10 End pad plate
133
Figure A-11 Spacer
134
Figure A-12 Bearing block
135
Figure A-13 Wilson coupling
136
Figure A-14 Lock nut (Turbine side)
137
Figure A-15 Lock nut (compressor side)
138
Figure A-16 Nozzle diffuser
139
Profile of Nozzle Diffuser
(DWG. No. TEX-04-10-13)
Radius r Height z
8.8534
8.8559
8.8904
8.9504
9.0317
9.1322
9.2514
9.3896
9.5484
9.7297
9.9362
10.1712
10.4384
10.7422
11.087
11.4776
11.9185
12.4133
12.9643
13.5711
14.2309
14.9
0
0.8446
1.6794
2.5037
3.3175
4.121
4.9143
5.6974
6.47
7.2314
7.9805
8.7157
9.4349
10.1351
10.8128
11.4629
12.0796
12.6556
13.183
13.6534
14.0596
14.397
140
Figure A-17 Cold end housing
141
Figure A-18 Nozzle cover
142
Figure A-19 Nozzle brake compressor
143
Profile of Nozzle Brake Compressor
(DWG. No. TEX-04-10-16)
Radius r Height z
7.9149
7.9287
8.0588
8.2581
8.5047
8.7871
9.0993
9.4385
9.8039
10.1958
10.6148
11.0616
11.5368
12.0405
12.5722
13.1306
13.7143
14.3211
14.9489
15.5953
16.258
16.86
0
0.8616
1.6485
2.3879
3.0935
3.7732
4.4311
5.0696
5.6891
6.2894
6.8694
7.4274
7.9617
8.4704
8.9517
9.404
9.8264
10.218
10.5787
10.9086
11.2084
11.479
144
Figure A-20 Warm end Housing
145
Figure A-21 Stem
146
Figure A-22 Stem tip
147
Figure A-23 Thermal insulators
148
Appendix B
Fabrication Drawings of
Heat exchanger
Figure B-1 HX1
150
Figure B-2 HX1-A Passage Details
151
Figure B-3 HX1-B Passage Details
152
Figure B-4 HX2
153
Figure B-5 HX2 Passage Details
154
Curriculum Vitae
NAME
: BALAJI KUMAR CHOUDHURY
E-mail
: [email protected]
Permenant Address : Salia bandha Street
Gosaninua Gaon
Berhampur-760003
Odisha,India
Education:
2013
Ph.D. Dissertation submitted, NIT Rourkela
2009
M.Tech., NIT Rourkela
2003
B.E., Degree Engg. College S.M.I.T., Ankushpur
Personal Information:
Date of Birth
01-06-1979
Nationality
Indian
Employment:
2006 – 2007
Lecturer, S.I.E.T. Dhenkanal
Publications:
1. Choudhury, B. K., Sahoo, R. K., and Sarangi, S. K., Design of Backward Swept Turbine
Wheel for Cryogenic Turboexpander,
Journal of Engineering Science and Technology,
V9(4),2014.
2. Choudhury, B. K., Sahoo, R. K., and Sarangi, S. K., Dynamic Simulation of Modified
Claude Cycle, Proceedings of the 24th National Symposium on Cryogenics, January 22-24,
2013., IPR, Ahmedabad, India.
3. Choudhury, B. K., Alur, S. A., Sahoo, R. K., and Sarangi, S. K., Optimization Analysis
of Liquefaction Cycles for Nitrogen, Indian Journal of Cryogenics, volume 37No(1-4), pp 34,
2012.
4. Alur, S. A., Choudhury, B. K., Sahoo, R. K., and Sarangi, S. K., Performance Analysis
of Plate Fin Heat Exchangers, Asian conference on Applied Super conductivity and Cryogenics
November 16-18, 2011, I.U.A.C, .New Delhi.
5. Alur, S. A., Choudhury, B. K., Sahoo, R. K., and
Sarangi, S. K., Simulation of
Turboexpander Based Nitrogen Liquefier. Proceedings of the 20th National and 9th
International ISHMT-ASME Heat and Mass Transfer Conference, January 4-6, 2010,
Mumbai, India.
155
6. Choudhury, B. K., Alur, S. A., Sahoo, R. K., and Sarangi, S. K.,
Design of High
Expansion Ratio Turboexpander for Nitrogen Liquefier. Proceedings of the 3rd International
Conference on Advances in Mechanical Engineering, January 4-6, 2010., S.V. National
Institute of Technology, Surat, India.
7. Choudhury, B. K., Rout, S. K., Sahoo, R. K., and Sarangi, S. K., Optimization analysis
of turbine wheel for cryogenic turboexpander, Experimental heat Transfer (Communicated)
156
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement