Alur Final Thesis

Experimental Studies on Plate Fin
Heat Exchangers
A Thesis Submitted for Award of the Degree of
Doctor of Philosophy
Sidramappa Alur
Mechanical Engineering Department

National Institute of Technology
Rourkela
Dedicated to
The Memory of My Parents
NATIONAL INSTITUTE OF TECHNOLOGY

ROURKELA 769008
INDIA
Ranjit Kr Sahoo
Professor
Sunil Kr Sarangi
Director
Mechanical Engg. Department

NIT Rourkela
NIT Rourkela
CERTIFICATE
This is to certify that the thesis entitled Experimental Studies on Plate Fin Heat
Exchangers, being submitted by Shri Sidramappa Alur, 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)
Date: January 6, 2012

2006
(Sunil Kr Sarangi)
Acknowledgements
I would like to express my deep sense of gratitude and respect to my supervisors
Prof S.K.Sarangi and Prof R.K.Sahoo for their excellent guidance, suggestions and
constructive criticism. I feel proud that I am one of their doctoral students. The
charming personality of
Prof Sarangi has been unified perfectly with knowledge that
creates a permanent impression in my mind. It is my proud privilege to work under not

only a wonderful supervisor but also a sympathetic person, who always stood on my side
at tough times during my stay at N.I.T,Rourkela. I and my family members also
remember the affectionate love and kind support extended by Madam Sarangi during
our stay at Rourkela.
I also feel lucky to get Prof R.K.Sahoo as one of my supervisors. His invaluable
academic and family support and creative suggestions helped me a lot to complete the
task successfully. His liberal attitude encouraged me to have a friendly interaction, which
never made me feel the burden of work. I record my deepest gratitude to Madam Sahoo
for family support all the times during our stay at Rourkela.
I record my gratitude to cryogenic division of Bhabha Atomic research
centre,Mumbai for giving us the heat exchanger for testing in our laboratory. I am
grateful to Mr. Tilok Singh, Head, the Cryogenic Division of Bhabha Atomic Research
Centre (BARC), Mumbai ,Mr. Mukesh Goyal and Dr Anindya Chakravarty for sharing
their knowledge and experience on plate fin heat exchangers and for the encouragement
throughout this work. I take this opportunity to express my heartfelt thanks to the
members of my doctoral scrutiny committee for thoughtful advice and useful
suggestions.
I take this opportunity to express my heartfelt gratitude to all the staff members
of Mechanical Engineering Department, NIT, Rourkela for their valuable suggestions and
timely support. I am deeply indebted to Mr. Biswanath Mukherjee, technical assistant,
for the fabrication of the heat exchanger test rig. The hard work put in by Mr. Jitendra
Kumar and Mr. Akash Pandey, M.Tech students for making the test rig leak proof and
insulated deserves special appreciation. I am also thankful to Raman, Ajay Sutar and
Rohit Mukare for their help in organising the seminars. I am also thankful to Mr. Binay
Kar, technical assistant, Refrigeration laboratory, for his helping hand during the
calibration of the instruments.
i
I feel lucky to have Mr. Balaji Kumar Choudhury as my co-research fellow.

Working with him was really a wonderful and fascinating experience. He has been with
me for four years of my research at N.I.T, Rourkela. I am thankful to Mr. Sachindra
Kumar Rout, Mr. L.N.Patra, Mr. Prakash , research Scholars for their kind help and
valuable suggestions. I am also grateful to Mr. Tapas Sarangi and Mr. Harihar Barkey for
the kind help extended in academic matters.
I record my sincere apologies to those whose names I have inadvertently missed
despite their meaning contribution during the course of this work.
I am really grateful to my brother and my sisters, in laws and my relatives for
their perseverance, encouragement with support of all kinds and their unconditional
affection. With a smile on their faces but anxiety in their minds, they stood by my side in
times of need. Their presence itself came as a soothing solace. My beloved wife had to
manage all kinds of difficult situations all alone during the correction and submission of
my thesis. I am sorry for this but feel proud of her. I feel sorry for my kids, Vaibhav and
Vaishnavi for disturbing their education in the beginning of my research at N.I.T,
Rourkela. I feel proud of their understanding of my situation, their preservance and their
adaptability which is beyond their age. They filled my life with joy and laughter. I never
felt the burden of the work in their company.
I am grateful to Dr U.C.Kapale, our beloved Principal, for his encouragement and
cooperation throughout my research. I wholeheartedly thank my colleagues at H.I.T,
Nidasoshi for their well wishes and encouragement. I thank Mr. S.N.Toppannavar, for his
kind cooperation and assistance during the correction and submission of the thesis.
I am very grateful to the Boarders of S.S.B. Hall of residence for their love and
affection during our stay at N.I.T, Rourkela. Special thanks to Mr. Nimain, for attending
to all the essential requirements in the hall.
Sidramappa Alur
(January 6, 2012)
ii
Abstract
Heat exchangers used in cryogenic applications need to have very high
effectiveness to preserve the refrigerating effect produced. Normally the heat
exchangers used in cryogenic refrigerators and liquefiers have effectiveness of the order
of 0.95 or higher. If the effectiveness of the heat exchangers falls below the design
value, there may not be any liquid yield. Plate fin heat exchangers, because of their
compactness, low weight and high effectiveness, are widely used in aerospace and
cryogenic applications. Such heat exchangers have closely spaced fins and offer narrow
and intricate passages for the fluid flow which often leads to significant pressure drop.
The stringent requirement of high effectiveness in cryogenic refrigerators and liquefiers
and high pressure drop occurring in plate fin heat exchangers make it necessary to test
the heat exchanger before putting into operation in a liquefier.
Plate fin heat exchanger (PFHE) is a type of compact exchanger that consists of a
stack of alternate flat plates called parting sheets and corrugated fins, both being
brazed together as a block. Streams exchange heat by flowing along the passages made
by the fins between the parting sheets. Separating plates act as the primary heat
transfer surfaces and the appendages known as fins act as the secondary heat transfer
surfaces intimately bonded to the primary surface. Aluminum is the most commonly used
material and stainless steel is employed in high pressure and high temperature
applications.
Extensive research has been done on plate and fin heat exchangers over the last
eight decades to understand the heat transfer phenomena occurring therein and to
determine the dimensionless heat transfer coefficient, j and the friction factor, f. Though
experimental investigations predominate in the literature, analytical modeling and
numerical solutions have also been carried out. The theoretical solutions often suffer
from oversimplification of fin channel geometry and simplifying assumptions made.
Experiments on heat transfer over plate fin surfaces are expensive and difficult.
Experimental results generated by reputed international laboratories are limited and
have remained almost totally proprietary. With successful fabrication of plate and fin
heat exchangers, it became necessary for us to devolp the methodology for the design,
fabrication and testing of plate fin heat exchangers. An experimental set up has been
built in the laboratory to test the plate fin heat exchanger. The validity of the existing
correlations is checked by conducting performance test on a counter flow heat
exchanger.
iii
One of the earliest and most comprehensive works on compact heat exchangers
was carried out by Kays and London at Stanford University in late 1940s. Their report
published in 1948 is the most authoritative and reliable source for j and f factors and
they are still used today for the geometries tested by the authors. The set-up of the test
bench at Stanford was used for accurate measurement of the basic heat transfer and
flow friction characteristics of the plate fin surfaces. Several researchers used the
experimental technique of Kays and London to develop j and f data for many other
surfaces. Several empirical correlations were generated from the data base of Kays and
London and other experimental works.The description of the experimental set up used
and methodology adopted at that time is described in chapter-II
for comparison with
the experimental work carried out in this thesis.

The experimental set adopted in this investigation consists of a counter flow heat
exchanger. High pressure cold air from a compressor is made to flow through one
channel of the heat exchanger. On exit from channel-I, the stream of air is heated in a
heating unit and is made to flow through the other channel in the reverse direction. The
pressures at the inlet of both the fluids are noted from the pressure gauges. The
temperatures at inlet and outlet of both the fluid streams are measured by resistance
temperature detectors. Mass flow rate is measured by a rotameter placed at the exit of
the hot fluid. The effectiveness is calculated from the measured temperature values for
balanced flow. Predicted value of the effectiveness considering longitudinal heat
conduction loss is calculated by the rating procedure. Pressure drop across the core for
both the fluids is measured by a U- tube manometer. Minor losses such as loss due to
bends and pressure losses in the headers are subtracted from the measured pressure
drop to get the core loss. The effect of heat loss to the ambient on the effectiveness is
obtained by the difference of energy unbalance between the streams.
The temperature difference is a measure of effectiveness of heat exchanger
directly from the experiment. The effectiveness of plate fin heat exchanger is an
important parameter for the heat exchanger used for cryogenic applications and is used
for the calculation of other process parameters. Hence results are expressed as
effectiveness and pressure drop versus mass flow rate. On the contrary, representation
of effectiveness versus NTU requires estimation of two major parameters; heat transfer
coefficient and effective area of heat transfer. These parameters are to be estimated
from the correlations. Since there are four such correlations it will be difficult to use any
one of them.
iv
Values of heat exchanger effectiveness and pressure drop obtained from

experiments are compared with those evaluated by using correlations developed by
various investigators. The experimental effectiveness is also compared with the value
obtained by simulation software, Aspen MUSE. Heat loss to the ambient leads to two
values of effectiveness - h , the effectiveness based on the hot fluid and c , the
effectiveness based on the cold fluid. These effectiveness values are compared with the
respective effectiveness values obtained by simulation software, Aspen MUSE. The
comparison shows that the two effectiveness values agree within 2.75 %. The mean
effectiveness values are compared with the theoretical values obtained (without
considering the heat loss) by using the correlations. Comparison shows that correlations
developed by Maiti and Sarangi are in better agreement with the experimental data
compared to the other correlations where percentage deviation varies from 6.42 to
4.57%.
A large amount of deviation is observed between the measured pressures drop
and that computed from various correlations. The difference between the pressure drop
values obtained from experiments and by simulation using Aspen MUSE is also
significant. Comparison is made between the experimental arrangements used in the
present work and that used by Kays and London to explore the possible causes of this
deviation.
Uncertainty analysis of results is an essential component in experimental
procedure. The uncertainty in the effectiveness of heat exchangers arises from those in
mass flow and temperature measurements. Gas volume flow rates are measured with a
rotameter and temperature is measured with a platinum resistance thermometer. The
uncertainty in experiment is estimated to be 4.8% for an effectiveness of 89%.
Contents
Acknowledgements
Abstract
iii
Contents
vi
List of Figures
vii
List of Tables
Nomenclature
xii
1.
2.
3.
Introduction
1.1.
Plate fin heat exchanger
1.2.
Plate fin heat transfer surfaces
1.3.
Heat transfer and flow friction characteristics
1.4.
Measurement principles
1.5.
Objectives of the Present Investigation
1.6.
Organization of the Thesis
10
Literature Review
2.1.
Plate fin heat exchanger
13
2.2.
Experimental studies
16
2.3.
Analytical and Numerical studies
21
2.4.
Heat transfer and flow friction characteristics
24
2.5.
Secondary Irreversibilitys
29
Design of Plate Fin Heat Exchanger

3.1.
Summary of Design procedure
35
3.2.
Design inputs and specification of fin geometry
37
3.3.
Correlation-based Design of Heat exchanger
39
3.4.
Design of Heat exchanger using Simulation software
45
3.5.
Concluding dimensions of The Heat exchanger
46
vi
4.
5.
6.
7.
Rating of Plate fin heat exchanger

4.1.
Details of Given heat exchanger and input data
48
4.2.
Rating of given heat exchanger using different correlations.
49
4.3.
Rating of given heat exchanger using simulation software
56
4.4.
Effect of heat transfer to the ambient
57
The Experimental Apparatus

5.1.
Experimental set up and operation
58
5.2.
Calculation procedure
61
5.3.
Effect of heat transfer from the ambient
62
5.4.
Description of various equipment and instruments
63
5.5.
Error Analysis
71
Performance Analysis
6.1
Experimental results
75
6.2
Variation of effectiveness with mass flow rate.
77
6.3
Effect of heat transfer to the ambient
80
6.4
Comparison of effectiveness obtained with and without heat loss
81
6.5
Error estimation in experimental results
82
6.6
Variation of pressure drop of cold fluid with mass flow rate
83
6.7
Variation of pressure drop of hot fluid with mass flow rate
86
6.8
Results and Discussion
87
Conclusion
7.1
Concluding Remarks
90
7.2
Scope for Future Work
92
References
93
Curriculum Vitae
vii
List of Figures
Page No.
Chapter 1
1.1
Plate fin heat exchanger assembly and details
1.2
Cross flow arrangement
1.3
Counter flow arrangement
1.4
Cross counter flow arrangement
1.5
Some of the common fin geometries
1.6
Details of boundary layer and flow across offset strip and wavy fin
Chapter 5
5.1
Schematic P&I diagram of the Experimental Test Rig
59
5.2
Photograph of the experimental set up
60
5.3
Photograph of the experimental set up (with insulation)
61
5.4
Plate fin heat exchanger.
64
5.5
R.T.D Construction.
67
5.6
Photograph of the set up used for calibration
68
5.7
R.T.D calibration graph
69
5.8
Orifice plate.
70
Chapter 6
6.1
Variation of effectiveness with mass flow rate (hot inlet

temperature =369 K)
6.2
79

temperature =359 K)
6.3
79

temperature =349 K)
6.4
80

temperature =339 K)
80
viii
6.5
Comparison of effectiveness obtained by experiment and by simulation

With heat leak at different mass flow rates.
6.6
Variation of effectiveness with mass flow rate including heat leak

(hot inlet temperature =369 K)
6.7
82
Variation of pressure drop of cold fluid with mass flow rate for hot inlet
temperature =369 K
6.8
84
temperature =359 K
6.9
84
temperature =349 K
6.10
85
temperature =339 K
6.11
85
Variation of pressure drop of hot fluid with mass flow rate for hot inlet
temperature =369 K
6.12
86
temperature =359 K
6.13
86
temperature =349 K
6.14
81
87
temperature = 339 K.
87
ix
List of Tables
Page No.
Chapter 3
3.1
Fin geometry used in heat exchangers
38
3.2
Concluding dimensions of heat exchanger
46
Chapter 4
4.1
Dimensions of the Heat exchanger core.
48
4.2
Fin geometry of heat exchangers
48
4.3
Predicted value of effectiveness using different correlations for mass flow

rate of 5.77 g/sec operating between 315 K and 369 K.
56
Predicted value of pressure drop of cold fluid using different correlations

For mass flow rate of 5.77 g/sec operating between 315 K
and 369 K.
57
4.4
Chapter 5
5.1
Flow arrangement for the designed heat exchanger
63
5.2
Core dimensions of the test heat exchanger
63
5.3
Fin geometry in the given heat exchanger
65
5.4
Design data of the given heat exchanger
65
5.5
Compressor specifications
66
5.6
Calibration chart
69
Chapter 6
6.1
Experimentally observed data at different mass flow rates for hot fluid
inlet temperature of 369 K
75
6.2
6.3
6.4
6.5
76
76
77
Uncertainties obtained at different mass flow rates for hot fluid

83
xi
Nomenclature
A
Heat transfer area of the heat exchanger with subscripts h or c denoting hot and
cold fluid, m2
A ff Free flow area available for hot or cold fluid with subscripts h or c respectively, m2
A fr Frontal area available for hot or cold fluid with subscripts h or c respectively, m2
Aw
Total wall area for transverse heat conduction from the hot fluid to cold fluid, m2
Plate thickness, m
af
Fin surface area, m2
a ff
Free flow area/fin, m2
a fr
Frontal area/fin, m2
as
Heat transfer area/fin, m2
aw
Total wall cross sectional area for longitudinal conduction, m2
Flow stream heat capacity rate with subscript h or c for hot and cold fluids, W/K.
Cd
Coefficient of discharge, dimensionless
C min Minimum of C c or C h , W/K

Cp
Specific heat at constant pressure, J/kg-K
Cr
Heat capacity rate ratio, dimensionless
De
Equivalent diameter of the flow passage, m
Fin frequency, Number of fins per meter length, fins/m
Fanning friction factor, dimensionless
Core mass velocity, kg/m2s
No flow height (stack height) of the heat exchanger core, m
Height of fins, m
xii
Convective heat transfer coefficient, W/m2 K
The Colburn factor, non-dimensional heat transfer characteristic
Kc
Contraction coefficient, no units

Expansion coefficient ,no units
Ke
Conductivity of the fin material, W/m- K
Kf
K w Conductivity of the wall material, W/m- K
Fluid flow (core) length on one side of the heat exchanger, m
Fin flow length on one side of a heat exchanger, m
le
Effective fin length for efficiency determination with subscripts h and c denoting
hot and cold fluids, m
Mass flow rate, kg/sec.
Total number of layers or total number of fluid passages
N tu Number of heat transfer units, UA / Cmin , dimensionless

ntuc Number of heat transfer units based on cold fluid side, ( 0 hA) c / Cc
ntuh Number of heat transfer units based on the hot fluid side, ( 0 hA) h / Ch
Fin pitch, 1 / f , m
pf
Pr
Q
Prandtl number of the fluid
= Heat load, W
Re Reynolds number, dimensionless
Re*
Critical Reynolds number with subscripts j or f for heat transfer and pressure drop
considerations
Spacing between adjacent fins, p f t , m
Temperature of the fluid (with subscripts c, h or i, o)
Thickness of fin, m
xiii
Uo
Overall heat transfer coefficient. W/m2 K
Width of the core, m
Greek symbols
Ratio of free flow area to frontal area, a ff / a fr , dimensionless
Longitudinal conduction parameter, dimensionless
Fluid dynamic viscosity, pa-sec

Fluid density, kg/m3
0 hA Convection conductance, W/ K
Effectiveness of heat exchanger, dimensionless
Fin efficiency, dimensionless.
o Overall surface effectiveness of the extended fin surfaces (secondary surfaces) with
subscripts c or h denoting cold and hot fluids, dimensionless.
Pressure drops of hot and cold fluid with subscripts h or c, Pa.
m1 The uncertainty in mass of hot fluid, kg/sec
m2 The uncertainty in mass of cold fluid, kg/sec

T
The uncertainty in temperature, K
The uncertainty in effectiveness
Subscripts:
b
Bulk or mean fluid
Cold fluid side
Hot fluid side
Inlet
xiv
Max Maximum
Min Minimum
mean
Overall
Wall or properties at the wall temperature
xv
Chapter 1
Introduction
Chapter I
INTRODUCTION
Heat exchangers used in cryogenic applications need to have very high
effectiveness to preserve the refrigerating effect produced. Normally the heat exchangers
used in liquefiers have the effectiveness of the order of 0.95 or higher. If the
effectiveness of heat exchangers falls below the design value, there may not be any
liquid yield [1]. The minimum effectiveness of heat exchanger devices required in
regenerative refrigerators stands at 95-98%. In aircrafts where the demand on
performance is not high, the volume and weight of the heat exchanger should be kept at
minimum. These requirements have led to the development of a unique class of heat
exchangers known as compact heat exchangers. Compact heat exchangers present a
large surface area (area to volume ratio greater than 700 m2 /m3).
1.1 Plate fin heat exchanger

Plate fin exchanger is a type of compact heat exchanger where the heat transfer
surface area is enhanced by providing extended metal surface, interfaced between the
two fluids and is called the fins. Out of the various compact heat exchangers, plate fin
heat exchangers are unique due to their superior construction and performance. They are
characterized by high effectiveness, compactness, low weight and moderate cost. As the
name suggests, a plate fin heat exchanger (PFHE) is a type of compact exchanger that
consists of a stack of alternate flat plates called parting sheets and corrugated fins
brazed together as a block. Streams exchange heat by flowing along the passages made
by the fins between the parting sheets. Separating plates act as the primary heat transfer
surfaces and the appendages known as fins act as the secondary heat transfer surfaces
intimately bonded to the primary surfaces. Fins not only form the extended heat transfer
surfaces, but also work as structural supports against internal pressure difference. The
side bars prevent the fluid from spilling over and mixing with the second fluid or leaking
to outside. The fins and side bars are brazed with the parting sheets to ensure good
thermal link and to provide mechanical stability. Figure 1.1 shows an exploded view of
two layers of a plate fin heat exchanger. Such layers are arranged together in a
monolithic block to form a heat exchanger.
Figure 1.1: Plate fin heat exchanger assembly and details [45]
A. Advantages and disadvantages

Plate fin heat exchangers offer several advantages over the other types of heat
exchanger:
i)
Compactness: Large heat transfer surface area per unit volume (typically 1000
m2/m3), is usually provided by plate fin heat exchangers. Small passage size
produces a high overall heat transfer coefficient because of the heat transfer
associated with the narrow passages and corrugated surfaces.
ii) Effectiveness: Very high thermal effectiveness more than 95% can be obtained.
iii) Temperature control: The plate fin heat exchanger can operate with small
temperature differences. A close temperature approach (temperature approach
as low as 3K) is obtained for a heat exchanger exchanging heat with single phase
fluid streams. This is an advantage when high temperatures need be avoided.
Local overheating and possibility of stagnant zones can also be reduced by the
form of the flow passage.
iv) Flexibility: Changes can be made to heat exchanger performance by utilizing a
wide range of fluids and conditions that can be modified to adapt to various
design specifications. Multi stream operation is possible up to 10 streams.
v) Counter flow: True counter-flow operation (Unlike the shell and tube heat
exchanger, where the shell side flow is usually a mixture of cross and counter
flow) is possible in a plate fin heat exchanger.
The main disadvantages of a plate fin heat exchanger are:
i)
The rectangular geometry used puts a limit on operating range of pressure

and temperatures
ii) Difficulty in cleaning of passages, which limits its application to clean and
relatively non-corrosive fluids, and
iii) Difficulty of repair in case of failure or leakage between passages.
iv) Relatively high pressure drop due to narrow and constricted passages.
B. Manufacturing process
The basic principles of plate fin heat exchanger manufacturing process are the
same for all sizes and all materials. The heat exchanger is assembled from a series of flat
sheets and corrugated fins in a sandwich construction. Separating plates (i.e. parting
sheets) provide the primary heat transfer surface. Separating plates are positioned
alternatively with the layers of fins in the stack to form the containment between
individual layers. These elements i.e., corrugations, side-bars, parting sheets and cap
sheets are held together in a jig under a predefined load, and placed in a brazing furnace
to form the plate fin heat exchanger block. After this, the header tanks and nozzles are
welded to the block, taking care that the brazed joints remain intact during the welding
process. Differences arise in the manner in which the brazing process is carried out. The
methods in common use are salt bath brazing and vacuum brazing. In the salt bath
process, the stacked assembly is preheated in a furnace to about 5500 C, and then
dipped into a bath of fused salt composed mainly of fluorides or chlorides of alkali
metals. The molten salt works as both flux and heating agent, maintaining the furnace at
a uniform temperature. In case of heat exchangers made up of aluminum, the molten
salt removes grease and the tenacious layer of aluminum oxide, which would otherwise
weaken the joints. Brazing takes place in the bath when the temperature is raised above
the melting point of the brazing alloy. The brazed block is cleaned of the residual
solidified salt by dissolving in water, and is then thoroughly dried. In the vacuum brazing
process, no flux or separate pre-heating furnace is required. The assembled block is
heated to brazing temperature by radiation from electric heaters and by conduction from
the exposed surfaces into the interior of the block. The absence of oxygen in the brazing
environment is ensured by application of high vacuum (Pressure 10-6 millibar). The

composition of the residual gas is further improved (lower oxygen content) by alternate
evacuation and filling with an inert gas as many times as experience dictates. No washing
or drying of the brazed block is required. Many metals, such as aluminum, stainless steel,
copper and nickel alloys can be brazed satisfactorily in a vacuum furnace. In recent times
vacuum brazing process has been used almost exclusively for manufacture of plate fin
heat exchangers.
C. Applications
The plate-fin heat exchanger is suitable for use over a wide range of
temperatures and pressures for gas-gas, gas-liquid and multi-phase duties. They are
used in a variety of applications. They are mainly employed in the field of cryogenics for
separation and liquefaction of air, natural gas processing and liquefaction, production of
petrochemicals and large refrigeration systems. The exchangers that are used for
cryogenic air separation and LPG fractionation are the largest and most complex units of
the plate fin type and a single unit can be of several meters in length. Brazed aluminum
plate fin exchangers are widely used in the aerospace industry because of their low
weight to volume ratio and compactness. They are being used mainly in environment
control system of the aircraft, avionics cooling, hydraulic oil cooling and fuel heating.
Making heat exchangers as compact as possible has been an everlasting demand in
automobile and air conditioning industries as both are space conscious. In the automobile
sector they are used for making the radiators. The other miscellaneous applications are:
i)
Fuel cells
ii)
Process heat exchangers
iii) Heat recovery plants

iv) Pollution control systems
v)
Fuel processing and conditioning plants
vi) Ethylene and propylene production plants
D. Flow arrangement
A plate fin heat exchanger can have two or more streams, which may flow in
directions parallel or perpendicular to one another. When the flow directions are parallel,
the streams may flow in the same or in opposite sense. So there are three primary flow
arrangements for a plate fin heat exchanger (i) parallel flow, (ii) counter-flow and (iii)
cross flow. Thermodynamically, the counter-flow arrangement provides the highest heat
(or cold) recovery, while the parallel flow geometry gives the lowest. The cross flow
arrangement, gives an intermediate thermodynamic performance, by offering superior
heat transfer properties and easier mechanical layout. Under some circumstances, a
hybrid cross counter-flow geometry provides greater heat (or cold) recovery with
superior heat transfer performance. Thus in general engineering practice, there are three
main configurations for the plate fin heat exchangers: (a) cross flow, (b) counter-flow
and (c) cross-counter flow.
(a)
Cross flow:
Cross flow and counter flow arrangement of fluids in heat exchangers is as shown
in Figure (1.2). In cross flow heat exchangers, the fluids flow in directions normal to each
other. Thermodynamically the effectiveness for cross flow heat exchangers falls in
between that for the counter flow and parallel flow arrangements. The largest structural
temperature difference exists at the corner of the entering hot and cold fluids. Only two
streams are handled in a cross flow type of a heat exchanger which eliminates the need
for distributors. For this type of heat exchangers the header tanks are located on all four
sides of the heat exchanger core, making this arrangement simple and cheap. If high
effectiveness is not necessary, and if the two fluid streams have widely differing volume
flow rates, or if either one or both streams have constant temperature, the cross flow
arrangement should be preferred. Typical applications include automobile radiators and
some aircraft heat exchangers.
(b)
Counter flow:
In a counter flows heat exchanger the two fluids flow parallel to each other but in
opposite directions. The counter-flow heat exchanger provides the most thermally
effective arrangement for recovery of heat or cold from process streams. A counter flow
arrangement is thermodynamically superior to any other flow arrangement. It is the most
efficient flow arrangement, producing the highest temperature change in each fluid
compared to any other two-fluid arrangement for a given overall thermal conductance
(UA), fluid flow rates and fluid inlet temperatures. Cryogenic refrigeration and
liquefaction equipment use this geometry almost exclusively. But these type of heat
exchangers demand proper design because of the complex geometry of headers.
(b) Counter flow
(a) Cross Flow
(b) Cross -Counter Flow
Figure 1.2: Cross flow and counter flow arrangement[45]
(c)
Cross-counter flow:
The cross-counter flow geometry is a hybrid of counter-flow and cross flow
arrangements, delivering the thermal effectiveness of counter-flow heat exchanger with

the superior heat transfer characteristics of the cross flow configuration. In this
arrangement, one of the streams flows in a straight path, whereas the second stream
follows a zigzag path normal to that of the first stream. While moving along the zigzag
path, the second fluid stream covers the length of the heat exchanger in a direction
opposite to that of the direct stream. Thus the flow pattern can be assumed to be
globally counter flow while remaining locally cross flow. Cross-counter flow PFHEs are
used in applications similar to those of simple cross flow exchangers, but they allow more
flexibility in design and fabrication. They are particularly suited for the applications where
the two streams have considerably different volume flow rates, or permit significantly
different pressure drops. The fluid with the larger volume flow rate or that with the
smaller value of allowable pressure drop is made to flow through the straight channel,
while the other stream follows the zigzag path. For example, in a liquid-to-gas heat
exchanger, the gas stream with a large volume flow rate and low allowable pressure drop
is assigned the straight path, while the liquid stream with a high allowable pressure drop
flows normal to it over a zigzag path. This arrangement optimizes the overall geometry.
(Figure 1.2 shows a cross-counter flow arrangement for heat exchanger)
1.2 Plate fin heat transfer surfaces

The plate fin exchangers are mainly employed for liquid-to-gas and gas-to-gas
applications. Due to the low heat transfer coefficients in gas flows, extended surfaces are
commonly employed in plate-fin heat exchangers. By using specially configured extended
surfaces, heat transfer coefficients can also be enhanced. While such special surface
geometries provide much higher heat transfer coefficients than plain extended surfaces,
but at the same time, the pressure drop penalties are also high, though they may not be
severe enough to negate the thermal benefits. A variety of extended surfaces like the
plain trapezoidal, plain rectangular shown in Figure 1.3 can perform such function. The
offset strip fin geometry is included in the present study.
In order to improve the gas side heat transfer coefficients, surface features are
needed to be provided on the gas side. These features may be divided into two
categories: the first, in which the surface remains continuous (wavy and herring-bone
fins) and the second in which it is cut (offset, louvered). In a continuous type fin, the
corrugations cause the gas to make sudden direction changes so that locally the velocity
and temperature gradients are increased (Figure 1.4). This results in local enhancement
of heat transfer coefficient. But an undesirable consequence of such enhancement in
heat transfer coefficient is an increase in the friction factor and pressure drop whereas in
a discontinuous type of fin geometry boundary layers are interrupted, otherwise this
would have formed on a continuous plate. Adjacent to the leading edge of the fin, both
heat transfer coefficient and friction factor are high due to generation of fresh boundary
layers. But in addition to this friction drag, form drag is also formed due to the finite
thickness of the fin.
Although friction drag is associated with high heat transfer
coefficient, form drag has no counterpart and represents one form of wasted energy. The
form drag can be substantial depending on the quality of the cutting edge. However,
machined-formed fins are generally free from this problem.
Figure 1.3: Types of plate fin surfaces: (a) Plain rectangular (b) Plain trapezoidal (c)
Wavy (d) Serrated or offset strip fin (e) Louvered (f) Perforated [45]
Figure 1.4: Details of boundary layer for flow across offset strip and wavy fin [44]
1.3 Heat transfer and flow friction characteristics

The heat transfer and flow friction characteristics of a heat exchanger surfaces
are commonly expressed in non-dimensional form and are simply referred to as the basic
characteristic or basic data of the surface. Various correlations are available in literatures
which express the Colburn factor, j and friction factor, f as functions of Reynolds number
and other geometrical properties. The Colburn and friction factors are defined by the
relations:
h
(Pr) 2 / 3
GCP
4 flG 2
p
2 Dh
Where,
(1.1)
(1.2)
h =heat transfer coefficient (W/m2-K)
G Mass velocity (kg/s-m2) [on the basis of minimum free flow area]
L Length of flow passage (m)

Dh Hydraulic diameter (m)
Mean density of fluid (kg/m3)
1.4 The measurement principles

The experimental data given by Kays and London [2] are the most authoritative
and reliable sources of heat transfer and pressure drop characteristic of the plate fin heat
exchangers. The experimental set up used by Kays and London [2] consists of cross flow
heat exchanger. One channel of the cross flow heat exchanger is made of the surface to
be characterized. The fluid flowing over this test section should preferably be one which
is likely to be used in service. A condensing steam is made to flow through the other
channel giving a very high heat transfer coefficient so that thermal resistance on that
side is neglected. The longitudinal heat loss through the walls is negligible and is not
considered in determining the heat transfer characteristics.
The experimental set up used here consists of a counter flow heat exchanger.
Cold air from the compressor is made to flow through one channel where as the hot air
coming from a heating unit is made to flow through another channel in the counter flow
direction. The pressures at the inlet of both the fluids are noted from the pressure
gauges. The temperatures at inlet and outlet of both the fluids are measured by
resistance temperature detectors. From the measurement of effectiveness, overall heat
transfer coefficient and the number of transfer units are calculated. The effectiveness
considering the longitudinal heat loss is calculated using Kreogers [84] equation. The
effectiveness is calculated from measurement of the temperature and mass flow rates.
1.5 Objectives of the study

In open literature many correlations of heat transfer coefficient and friction factor
are available for design and rating of PFHE. In many cases, reports on validation of these
correlations with experiments are scarce. Hence the main objective of the present work is
to evaluate the performance parameters of a counter flow plate and fin heat exchanger
(PPHE) which includes the following steps.
i)
Design of a PFHE based on a chosen correlation.
ii) Industrial fabrication of the PFHE with the above supplied design data.
iii) Fabrication of a test rig for testing.

iv) Comparison of the values of effectiveness, overall heat transfer coefficient and
pressure drop obtained by experiment with the rating values of PFHE based on
various other correlations.
The hot test method is adopted for this study. The correlation used for design and
rating of PFHE are listed as
i)
Maiti- Sarangi correlation[45]
ii) Manglik- Bergles correlation[70]

iii) Joshi-Webb correlation[68]
iv) Simulation software of Aspen-MUSE.[113]
1.6 Organization of the Thesis

The thesis has been arranged into seven chapters. Chapter I deals with
general introduction to compact plate fin heat exchanger and enumerates the objective
of the present investigation. In Chapter II, a brief review of relevant literature covering
theoretical and experimental studies undertaken for determination of heat transfer and
flow friction characteristics of surfaces, and review of various correlations available in
literature for determination of heat transfer and flow friction characteristics of offset strip
fin surfaces have been presented. Chapter II also covers review of literature on
irreversibilities that affect heat exchanger performance.
In Chapter III, design procedure for plate fin heat exchanger is outlined. Design
of the given plate fin heat exchanger using correlations developed by Maiti and Sarangi
[45], Manglik and Bergles[70], Joshi and Webb [68] and simulation software, Aspen-MUSE
[113] have been presented. Rating i.e., calculation of performance parameters of plate
fin heat exchanger using the correlations developed by Maiti and Sarangi [45], Manglik
and Bergles [70], Joshi and Webb [68] and simulation software, Aspen-MUSE [113] has
been presented in Chapter IV.
A detailed description of the experimental set up and the operating procedure for
the hot test has been given in Chapter V. Description of the different components of the
experimental set up, instruments used and their calibration procedure are given.
10
Uncertainty analysis of the results has also been included in Chapter V. Experimental
results have been presented in Chapter VI. Chapter VI also contains the graphs showing
the comparison between the theoretical and experimental values of performance
parameters and discussion on the results.
Chapter VII, the last and final chapter, is devoted to concluding remarks and for
defining the scope of future work.
11
Chapter 2
Literature Review
Chapter II
LITERATURE SURVEY
Plate fin heat exchangers, because of their compactness, low weight and high
effectiveness are widely used in aerospace and cryogenic applications. Cryogenic
liquefiers need heat exchangers of very high effectiveness(of the order of 0.95 or more)
and the liquefiers cease to produce any liquid if the effectiveness of heat exchangers
falls below the design value[1].Correct design and quality construction of heat
exchangers is essential for proper functioning of such systems.
The heat transfer coefficient and the flow resistance are expressed in non
dimensional form as Colburn factor, j and the friction factor, f. Accurate prediction of the
heat transfer coefficient and friction factor is essential for proper design of heat
exchangers. The Colburn factor, j and the friction factor, f are expressed as functions of
Reynolds number and other geometrical parameters. j and f factors can be determined
by numerical modeling of the flow field through CFD.In spite of the progress in
computing power, it is not possible to predict j and f data by numerical solution. This is
because the models are usually based on certain simplifying assumptions. Numerical
solution along with flow visualization, however, helps in understanding the flow physics
associated with heat transfer enhancement. It is also possible to carry out a parametric
study on the effect of geometrical parameters on the performance of finned surfaces.
Fundamental relations describing various types of heat transfer phenomena and
heat exchanger design techniques have been discussed in well known text books [28].The book Compact Heat Exchangers by Kays and London [2] provides an excellent
introduction to the analysis of plate fin heat exchangers, and contains a valuable
database on the heat transfer and flow friction characteristics of several fin geometries.
The recent work of Shah [7-8] provides the most comprehensive information on the
subject, particularly on compact plate fin heat exchangers. Several specialized
monographs and conference proceedings, covering basic heat transfer, heat transfer
augmentation and design and simulation methodologies have further enriched the
literature [9-13].
Journals on thermal engineering and heat transfer devote a sizable portion of their
content to research findings on heat exchangers [14-17].Two major journals: Heat
Transfer Engineering [18] and International Journal of Heat exchangers [19] are almost
exclusively dedicated to the subject of heat exchangers.
One of the earliest and most comprehensive experimental works on compact
heat exchangers was carried out by Kays and London at Stanford University in late
1940.Their report [37] published in 1948 gives the complete methodology and details of
the experimental set up. Several researchers used the same experimental technique for
the experimental determination of non dimensional heat transfer coefficient and friction
factor. Later on several empirical correlations were generated from the experimental
data of Kays and London [2] and other experimental works.
Analytical determination of non dimensional heat transfer and pressure drop
characteristics is difficult. This is because the heat transfer and flow friction characteristic
of a surface are strong functions of geometrical parameters such as fin height, fin
spacing, fin thickness, etc; each fin type needs to be characterized separately.
This literature review focuses primarily on plate fin heat exchangers, details of
the experimental, numerical and analytical studies, predictions of j and f factors for
offset strip fin surfaces and a brief review on the secondary irreversibilities.
2.1 Plate Fin Heat Exchangers

The plate fin heat exchanger, a member of the compact heat exchanger family,
has found wide applications in various fields of engineering. They are widely used in
aircraft and automobile industry, chemical process plants and cryogenic engineering.
History of Development
Early developments of compact heat exchangers were stimulated by their
applications in the automobile and aircraft industries. During the 1930s,the secondary
surface plate and corrugation construction became established for aero engine radiators
using dip soldered copper as the material of construction [20].Development of compact
heat exchangers in the automobile and air conditioning industries has been reviewed by
Mori et al [21] and Cowell et al [22].In the early 1940s, the introduction of the aluminum
dip-brazing process made it possible to manufacture aircraft heat exchangers with
aluminum and resulted in substantial reduction in weight[23]. Subsequently, brazed
aluminum heat exchangers, fabricated from plate pairs, were employed as aircraft
engine intercoolers [24].Development of machines capable of producing very precise
13
corrugated fins with varying height and spacing lead to mass production. Continued
reduction in weight, increase of surface area density, enhanced reliability and flexibility
that it offers to the manufacturer have made the plate fin heat exchanger indispensable
in gas to gas heat exchange applications in cryogenic and chemical industries. More
recently their application has been extended to boiling and condensation duties [25,
26].Dip brazing was first employed commercially by the Trane Company in USA during
World War II and the first industrial size exchangers were manufactured in 1949 [24].
No loss of strength and ductility at low temperatures made aluminum extremely
suitable for cryogenic applications. Tubular heat exchangers used in cryogenic
applications were replaced by aluminum plate and fin heat exchangers. Features such as
compact shape, low weight, and design flexibility available with plate fin heat
exchangers led the way for their application on a much wider scale.Devolpment of large
aluminum plate fin heat exchangers and that of tonnage air separation plant supported
each other for further growth.
Today, brazed aluminum plate fin exchangers are being designed and
manufactured by several reputed companies around the globe. Information on these
companies and their products are available from the web sites of the Aluminium Plate
Fin Heat Exchanger Manufacturers Association [27].The five members of the
organization are Chart Heat Exchanger, USA[28], Kobe Steel Ltd, Japan [29], Linde AG,
Germany [30], Nordon Cryogenie,France [31], and Sumitomo Precision Products Co
Ltd,Japan [32]. In addition, several smaller but knowledge based companies
manufacture heat exchangers for specialized applications. The automobile industry is
another major manufacturer and user of aluminium plate fin exchangers.
Several specialized laboratories also made significant contribution to the research
on plate fin heat exchangers. Most notable among them are the Heat Transfer and Fluid
Flow Services (HTFS)[33] in England and Heat Transfer Research Inc (HTRI) [34] in
USA. These organizations, supported by industry and institutions from around the world
continue to produce most advanced and authentic information on the subject of plate fin
heat exchangers.
Fabrication of heat exchangers

The basic approach to fabricate plate fin heat exchangers is to assemble the
parting sheets, fins, side bars and top plates together in a fixture and to braze the
assembly to form the heat exchanger core. While the side bars and the parting sheets
are cut to size by milling and shearing, electro discharge machining(EDM) is employed
14
for cutting the preformed thin walled fins (thickness of 0.1 mm) to the required shape
and size. Typical materials include alloys of aluminum or stainless steel. Manganese
based 3000-series aluminium alloy (e.g Al 3003) can be easily brazed using 7 % Silicon
based aluminium filler metal (e.g.4004).For stainless steel the material of construction is
SS-304 and the filler material is a low melting nickel-iron alloy.
The brazing of fins to parting sheets is done either by dip brazing or, more
commonly, by vacuum brazing technique. Most metals, such as stainless steel, copper
and nickel alloys, can be brazed satisfactorily in a vacuum brazing furnace.Aluminium,
because of the tenacious oxide layer that forms quickly on the surface, requires either a
molten salt bath to dissolve the oxide or a very high vacuum [12].Detailed information
on different brazing techniques is available in literature[12,35-36].
In the dip brazing technique, the stacked assembly is heated in a furnace to a
temperature few tens of degrees below the melting point of the brazing alloy.The
preheated assembly is then dipped into a bath of fused salt mainly composed of
fluorides and chlorides of alkaline metals. The molten salt bath, with its temperature
carefully controlled, acts both as a flux and as the heating agent. On entry into the salt
bath, the brazing alloy melts and flows by surface tension along the joints between the
separating sheets, fins and sidebars. On completion of salt-bath brazing, the unit is
washed with clean water, followed by a wash with dilute nitric acid. Finally the unit is
washed thoroughly with dematerialized water [35] to remove traces of the acid.
In the vacuum brazing process, the stacked assembly is heated up to the brazing
temperature by radiation heating in a vacuum furnace. High vacuum (10-6 torr) ensures
a very low partial pressure of oxygen in the atmosphere, and facilitates chemical splitting
of the oxides).The size of the exchanger determines the temperature ramp and the
soaking time to be employed. In case of vacuum brazing of aluminium,the narrow
margin between the melting point of the brazing alloy and the parent metal (about 50 K)
makes it mandatory to employ close temperature control in the furnace [36].For small
heat exchangers, an independent six-zone control with 1 K variation is adopted, while
large units demand still more specialized control strategy. It is done to ensure complete
and uniform heating and chemical decomposition of the oxide layer. The system
temperature is raised quickly to the melting temperature of the brazing alloy, held for
one or two minutes, and brought down to a temperature 50 K below the brazing
temperature. This is done to avoid creeping deformation of the assembly under its own
weight. In stainless steel exchangers, because of the large difference in the melting
temperature of the braze alloy and the parent metal, such care is not strictly necessary.
15
The headers and the nozzle are welded to the heat exchanger core by TIG
welding. Care must be taken to ensure that remelting of the already brazed joints does
not take place. This is a serious problem in stainless steel exchangers where the
difference in melting temperatures of the parent metal and brazing alloy can exceed 50
K. The heat exchanger thus fabricated is then subjected to a series of tests viz. leak test,
pressure test etc before final acceptance for service.
2.2 Experimental studies

The prediction of non-dimensional heat transfer coefficient, j and friction factor, f
by numerical solution has not yielded accurate results because of the limitation of the
computing resources and the simplifying assumptions made. Empirical relations for the j
and f factors can be successfully used in less critical designs but for the more critical
design such as application in cryogenic systems, experimental determination of the non
dimensional heat transfer coefficient, j and the friction factor, f remains the only choice.
The experimental work can be conducted to check whether a given heat exchanger
meets the prescribed thermal performance and pressure drop requirement, and to
analyze the various causes of degradation and malfunctioning.
The experimental methods can be broadly classified as follows:
1. Steady state technique where heat is transferred from one fluid to another
through a separating wall (recuperative heat exchanger).
2. Transient technique where heat is exchanged with a solid matrix (regenerative
heat exchanger). This technique is further classified as
a. Single blow method, and
b. Periodic test methods.
A.
Steady state methods
a) Kays and London

One of the earliest and most comprehensive works on compact heat exchangers
was carried out by Kays and London at Stanford University in late 1940.Their report
published in 1948 is the most authoritative and reliable source for j and f factors till
today. Most of the correlations developed later are reworking of these correlations. Their
set up for the test bench at Stanford is used for the accurate measurement of the basic
16
heat transfer and flow friction characteristics. The methodology they adopted and
conditions in which these correlations are derived has been discussed here for
comparison with the experimental work carried out in this study.
In the steady state method, a cross flow type heat exchanger is usually employed
as the test exchanger. One channel of the cross flow heat exchanger is made of the
surface to be characterised.The fluid flowing over this test surface should preferably be
one which is likely to be used in service. Because a majority of plate fin heat exchangers
are used in gas to gas applications, and because most gases have comparable physical
properties, air is conventionally used as the testing medium. The fluid flowing over the
second channel must provide high heat transfer rate and low pressure drop to improve
accuracy. The list of fluids with high heat transfer coefficient includes condensing steam,
hot water and oil.
In a steady state experiment, measurement of temperatures and mass flow
rates in the two sides provides the required information to compute the heat exchanger
effectiveness. An Ntu relation, appropriate to the cross flow arrangement, is applied
to determine the Ntu and hence the overall heat transfer coefficient (U), the inverse of
which is related to the resistances of individual sides and that of the separating wall.
Assuming that the fouling resistances are negligible; the overall thermal resistance
(1/UA) is expressed by the following relation:
1
1
1
R wall
UA ( 0 hA) unknown ( 0 hA) known
(2.1)
where,
0 1 (a f / as ) (1 f )
(2.2)
where
a f / a s is the ratio of fins to the total surface area, and f is the fin efficiency.
The fin efficiency f is calculated by the formula
n f tanh(Ml ) /( Ml )
(2.3)
with
(2 h )
(K f t)
(2.4)
17
Once the surface area and the geometry are known for the extended surfaces, h
and 0 are computed iteratively from equations (2.1) to (2.4), the j factor is then
calculated from its definition
h
(Pr) 2 / 3
Gc p
(2.5)
The plate-fin heat exchangers are commonly used for gas-to gas heat exchange,
and the pressure drop for each stream is an important design factor. The overall
pressure drop through the plate fin heat exchanger involves four components: (1) the
pressure drop at the inlet, as the fluid leaves the inlet header and enters the finned
section (heat exchanger core), (2) the frictional pressure drop in the finned section or
core, (3) the pressure drop (or possibly a pressure rise) at the outlet, as the fluid leaves
the core and enters the outlet header, and (4) the momentum pressure drop ( or rise )
due to the velocity changes in the heat exchanger core resulting from changes in density
of the fluid. Fanning friction factor f is obtained from the following
m Dh 2P
f d Ld
1
1
2
(K c 1 2 )
)
K e 1 (1 4
2
4L G
inlet
outlet
Dd
(2.6)
Where K c and K e are the contraction and expansion coefficients respectively

Following points can be observed from the above experimental set up:
1. Condensing vapor (i.e., steam) is used as the heat transfer medium in the second
channel. Heat transfer coefficients are very high in condensation heat
transfer.Seperating walls are made thin. Thus the magnitude of the wall
resistance and the thermal resistance of the second channel are minimized. This
increases the accuracy in the measurement of hA over the test surface. Flow of
condensing vapor in the other channel gives a thermal boundary condition of a
uniform wall temperature with zero heat capacity rate (Cr). Longitudinal heat
conduction is normally negligible compared to the high rate of heat transfer in
the lateral direction.
The number of transfer units can be found out from the relation:
N tu ln(
T Tinlet
1
ln( s
)
(1 )
Ts T outlet
(2.7)
18
where
Ts is the saturation temperature of condensing fluid (steam) at its inlet condition.

2. With high N tu heat exchangers, thermodynamic limitation restricts the change of
outlet fluid temperatures, making them less sensitive to changes in heat transfer
coefficient
and making the j factor measurement less accurate. Therefore N tu
of the test core in the steady state experiment is restricted between 1.0 and 3.0
to minimize the error in j and f measurement.
3. The measurement of friction characteristics is rather simple. Measurement of
fluid flow rate, inlet temperature, pressure and pressure drop across the core is
sufficient to determine the Fanning friction factor, f. The loss of pressure due to
flow through elbows and headers is to be deducted from the measured pressure
drop to get the pressure drop across the core.
b) Other experimental works

In extension of the work by Kays and London, London and Shah [38] have
reported measurement on eight high performance surfaces, all of the offset strip fin
geometry. The third and last edition of the book by Kays and London [2] presents all
available data in a systematic manner. More recently, Shah has suggested a modified
Wilson Plot technique [39] for determining heat transfer in both sides of a heat
exchanger simultaneously.
Davenport [40] conducted experiments on eight louvered fin surfaces to study
the effect of louver pitch with all other geometrical properties held constant. His test
core had a square crossection, 152 mm on each side, with a length of 40 mm along the
flow direction. Davenport used water at 850C temperature on the second channel. Heat
transfer coefficient on this side was found out by Dittus Boelter equation. The friction
factor was derived from the relation:
( inletV 2 )
P
Ke Kc f A
A
ff
2
The
experimental
error
involved
(2.8)
has
been
attributed
mainly
to
flow
measurement. Maximum errors on Stanton number (St) and friction factor (f) have been
estimated at 5% and 12% respectively.
Sunden and Svantesson [41] used the same experimental scheme as adopted by
Davenport [40] but they used a single channel of width 80 mm with fins of height 12.5
19
mm.The length of the core along the direction of air flow was 60 mm. Heat was provided
to the single channel heat exchanger from a constant oil bath maintained at 600C.They
used the Dittus Boelter equation to calculate the oil side heat transfer coefficient. The
estimated uncertainties in the measurement of St and f were 15 % and 6 %
respectively.
Wang et al [42] carried out heat transfer and flow friction experiments to study
the role of fin frequency on heat transfer and pressure drop.Lozza et al [43] have also
conducted steady state experiments on fin and tube heat exchangers having different fin
geometries using air and hot water at 600C as working fluids.
Recently Ghosh [44] at I.I.T., Kharagpur used the experimental technique of
Kays and London [2] and conducted experiments on three wavy and six offset strip fin
surfaces. The details of the geometry used are given in his thesis [44]. The results so
obtained were combined with the numerical results obtained in the same laboratory by
Maiti and Sarangi [45] to generate separate correlations for laminar and turbulent zones.
Some of the constants were obtained by multiple regressions over the numerically
computed results whereas the remaining constants were obtained from the experimental
data.
S.Freund and S. Kabelac[46] have used TOIRT method (Temperature oscillation
IR thermography) for determining the local heat transfer coefficients for plate heat
exchangers. In TOIRT method, temperature measurements are taken on outer surface
of a heat transferring wall with an IR camera and temperature oscillations are generated
by radiant heating. C.F.D. models for turbulent flow were correlated by using the
experimental values.
B.
Transient technique:
The single blow transient method is an alternative method of characterizing
heat transfer surfaces. This technique is used for calculating average heat transfer
coefficient of packed bed regenerator and matrix type high N tu heat exchanger
surfaces. In this method, a compact heat exchanger matrix, or a packed bed, is first
allowed to come in equilibrium with the process fluid temperature. Another cooler fluid is
then allowed to flow through the matrix. The fluid exchanges heat with the matrix. A
three way valve is used to switch from one fluid stream to another flowing through the
matrix. Another alternative is to employ a low thermal capacity electric heater upstream
of the matrix.
20
Heating is continued until the core reaches a uniform temperature manifested by

a negligible difference between the temperature of air stream at inlet and exit. The
electricity supply is switched off instantly to generate the step change. The fluid outlet
temperature is recorded during the cooling period up to the new equilibrated
temperature. This measured temperature response is matched with exit fluid
temperature history derived from a mathematical model of the system. From the
parameters of the mathematical model and the operating condition it is possible to
determine the heat transfer coefficient.
There are several computational methods to analyze the measured data for
determination of Ntu. The most prominent among them are:(1)The maximum slope
method,(2) the zero intercept method,(3) the direct curve matching method, and (4) the
first moment of area method. Detailed discussions on these techniques have been given
in references [47] and [48].Typical uncertainties in the final values of Colburn j factor
has been reported [49] to be 13% with overall Ntu of 3.5 for the test core.
Though transient tests are relatively easy to perform, the data reduction
procedures are significantly more complex compared to those in the steady state
technique. Transient tests are ideal for large Ntu heat exchangers, as the data reduction
procedures have substantial errors when used for low Ntu.The steady state technique
yields more accurate results for high performance surfaces.
2.3 Analytical and numerical studies

Unlike simpler geometries, the performance of a plate and fin heat exchanger is
not uniquely determined by the hydraulic diameter. Other geometrical parameters such
as fin spacing(s), fin height (h), fin thickness (t), offset strip length (l), wavelength ( ),
and wave amplitude (a) etc play significant roles. It will be very expensive and time
consuming to fabricate heat exchanger cores and conduct experiments over reasonable
ranges of all the geometric variables and Reynolds numbers. In contrast, it is relatively
easy and cost effective to carry out a parametric study through numerical simulation and
derive acceptable correlations for use by the heat exchanger industry. With the
development of more powerful computational tools, numerical prediction of j and f
factors are now feasible by solving the continuity, momentum and energy equations.
Patankar [50] provides a comprehensive summary of CFD equations relevant to compact
heat exchanger passages and techniques employed for their solution. Levent Bilir et
al[51] used CFD program FLUENT to analyze the effect of three different types of
21
vortex generators on the performance of fin tube heat exchangers. They found that the
three vortex generators when placed suitably will increase the heat transfer with
moderate increase in pressure drop. Numerical studies, supplemented by flow
visualization, can definitely be a means for the understanding of the heat transfer
enhancement mechanism. A detailed discussion of
the physics of the heat transfer
process has been given by Jacobi and Shah[52].Shah et al have also presented a
comprehensive review of numerical analysis of some of the important fin geometries
employed in compact heat exchangers[53]. Results of numerical studies on several plate
fin geometries have been summarized. This review also contains a discussion on the
physics of the flow process, as determined from experimentation and flow visualization.
This information will be useful in further refinement of the numerical techniques in
future.
Because of the extensive practical applications, louvered and offset strip fins
have attracted the attention of researchers more than other geometries. A brief review
of literature on analytical and computational studies on offset strip and louver fins is
presented in the following sections.
Offset strip fin surfaces:

Sparrow, Patankar and coworkers [54] were the first to use numerical (CFD)
techniques for prediction of j and f data in offset strip fin heat exchangers. Patankar and
Prakash [55] extended their work further and compared their numerical results for a two
dimensional heat transfer matrix having offset strip fins with the experimental results of
London and Shah [38].The results indicated reasonable agreement for the f factors. But
the predicted j factors were about twice as large as the experimental data.
Suzuki et al (56) took a different numerical approach by solving elliptic
differential equations of momentum and energy to study the thermal performance of a
staggered array of vertical flat plates at low Reynolds number. The validation of their
numerical model was done by carrying out experiments on a two dimensional system,
followed by those on a practical offset strip fin heat exchanger. The experimental results
were in good agreement with the computed values in the Reynolds number range
Re<800.Zhang et al (57) has attempted solving the unsteady Navier-Stokes and energy
equations on a massively parallel computer. Their study shows that the inclusion of flow
unsteadiness plays an important role in accurate prediction of j and f factors.
22
Louvered fins:
Louvered fins have found wide application in the aerospace industry. In the
1990s several workers devolped CFD codes based on non-orthogonal boundary fitted
meshes to compute the flow over louvered fins. Others used non-orthogonal meshes in
conjunction with commercial CFD codes [58-60].
Achaichia et al [58] investigated the variation in flow alignment with Reynolds
number using the mean flow angle defined by Achaichia and Cowell [61] as a
measure of the local degree of alignment. They found that the maximum value reached
by was less than the louver angle, but approached it at high Reynolds number.
Atkinson et al [60] analyzed two and three dimensional numerical models of louvered fin
arrays on a powerful work station using a commercial CFD package. They compared
their numerical results with experimental data of Achaichaia and Cowell [61] and
concluded that the heat transfer predictions of the 3D model were in agreement with
experimental observations. Ha et al [59] computed the overall Nusselt number and
friction factor for a limited number of louver angles, fin pitches and Reynolds numbers.
They found that the Nusselt number and the friction factor increase with that of louver
angle and decrease with reduction of fin pitch.
Springer and Thole [62] carried out a combined experimental and computational
study of flow through a louvered fin array at two different Reynolds numbers. The
experiments were conducted on a 20:1 scaled up model of a 19 row louvered fin array
with louver angle of 270 and fin pitch to louver pitch ratio of 0.76.Numerical simulation
was carried out for a single row of louvers assuming periodic boundary conditions and
two-dimensional, steady, laminar flow. Good agreement was found between the
computational predictions and the experimental measurements made with a two
component Laser Doppler Velocimeter.
All the studies discussed so far have assumed steady laminar flow and thus are
incapable of predicting time depentment phenomena such as flow separation and vortex
shedding. With the advent of high speed parallel computers, it has become possible to
solve the time dependent CFD equations. Tafti et al [63, 64] have used an efficient time
dependent calculation procedure for studying both fully developed and developing
unsteady flow and heat transfer in louvered fin heat exchangers. Their result shows that,
in the transitional regime, local heat transfer is strongly influenced by large-scale
vortices generated at the leading edge of the louvers.
23
2.4 Heat transfer and flow friction characteristics

One of the earliest and the most authoritative sources of experimental j and f
data on plate and fin surfaces is the monograph of compact heat exchangers by Kays
and London [2].Kays and London conducted experiments on different types of plate and
fin surfaces and observed from experiments that the heat transfer coefficient and friction
factor f of surfaces having the same effective diameter differed mutually according to
the fin geometrical properties like h/s, l/s and t/s etc. Therefore, it is imperative that the
j and f factors are obtained experimentally as functions of Reynolds number and other
geometrical properties. The expression for j and f data is obtained separately for each
surface type. J and f so presented are applicable to surfaces of any hydraulic diameter,
provided a complete geometric similarity is maintained. Different heat transfer
correlations for offset strip fins are given as below:
Offset strip fin surfaces:

Manson [65] appears to have made the first attempt at developing predictive
equations. However the data base he employed consisted of dissimilar geometries:
scaled up and actual offset strip fins, louvered fins and finned flat tubes. Kays [66] made
one of the first attempts at analytical modeling of heat transfer and friction loss in offset
strip fins and proposed a modified laminar boundary layer solution that includes form
drag contribution of blunt fin edges.
Wieting [67] developed an empirical correlation from experimental heat transfer
and flow friction data on 22 offset strip fin surfaces of Kays and London[2], London and
Shah[38], Walters[112] etc over two Reynolds number ranges:Re< 1000 and Re >2000.
For Re 1000
j 0.483(l / Dh ) 0.162 (s / h) 0.184 (Re) 0.536
(2.11)
f 7.661(l / Dh ) 0.384 (s / h) 0.092 (Re) 0.712
(2.12)
For
Re 2000
j 0.242(l / Dh ) 0.322 (t / Dh ) 0.08 (Re) 0.368
(2.13)
f 7.661(l / Dh ) 0.384 (s / h) 0.092 (Re) 0.712
(2.14)
24
For predicting j and f in the transition zone, extrapolating the equations upto
their respective transition zone boundaries was suggested. Although 85% of all available
data were correlated within 15 % for friction factor and 10 % for heat transfer,a few
points showed discrepancy as high as 40 %.Wietings correlation can be successfully
used for the design of practical heat exchangers, but care should be taken in
extrapolating the data to fins with geometrical parameters outside the recommended
range.
Joshi and Webb [68] conducted flow visualization experiments to identify the
transition from laminar flow. As the flow rate increases, oscillating velocities develop in
the wakes, leading to vortex shedding with further increase in Re. The onset of
oscillating flow and the consequent change in the wake structure were found to
correspond approximately to the departure from the laminar log linear behavoiur of j and
f. A wake width based equation was devolved to determine the critical Reynolds number.
They developed an analytical model in the laminar zone based on the numerical solution
done by Sparrow and Liu [69] and a semi empirical method has been used for the
turbulent region.
For laminar range ( Re Re )
j c 0.53(Re) 0.5 (l / De ) 0.15 (s / h) 0.14
(2.15)
f 8.12.(Re) 0.74 (l / De ) 0.41 (s / h) 0.02
(2.16)
For turbulent range ( Re Re 1000 )
j c 0.21(Re) 0.4 (l / De ) 0.24 (t / D h ) 0.02
(2.17)
f 1.12.(Re) 036 (l / De ) 0.65 (t / De ) 0.17
(2.18)
The critical Reynolds number for heat transfer and pressure drop considerations
is given by
Re
Re * 257(l / s)1.23 (t / l ) 0.58 De t 1.328
lD e
0.5
(2.19)
The empirical correlation for j and f factors proposed by the authors were verified
with experimental data on 21 heat exchanger geometries and their own observations on
scaled up geometries. They were able to correlate 82% of the f data and 91% of the j
data within 15 %.
25
Manglik and Bergles [70] examined the heat transfer and friction data for 18
offset strip fin surfaces given by Kays and London [2], Walters [112] & London and Shah
[38], and analyzed the effect of various geometrical attributes of offset strip fins. The
equations that describe the asymptotic behavior of the data in the deep laminar and fully
turbulent zones have been devolped.These asymptotes have been combined to give the
single predictive equation for j and f which are valid for laminar, turbulent and transition
zones.
j 0.6522 Re 0.5403( s / h) 0.1541(t / l ) 0.1499(t / s) 0.0678
1 5.269 10
Re1.340 ( s / h) 0.504 (t / l ) 0.546 (t / s) 1.055
0.1
f 9.6243 Re 0.7422( s / h) 0.1856(t / l ) 0.3053(t / s) 0.2659
1 7.669 10
Re 4.429 ( s / h) 0.920 (t / l ) 3.767 (t / s) 0.236
0.1
(2.20)
(2.21)
These equations predict all of the heat transfer data and approximately 90% of
the friction data within 20 %.
Maiti and Sarangi [45] used CFD as numerical tool for computing velocity,
pressure and temperature fields in plate and fin passages. They obtained correlations for
the non dimensional heat transfer coefficient,j
and pressure drop characteristic ,f in
terms of Reynolds number and other geometrical parameters using both computed and
experimental results . Some of the constants in the correlation are found by multiple
regression from the numerically computed results and the rest of the constants from
experimental data on the same geometry by another worker in the laboratory. They thus
combined both the experimental and computational methods. They also obtained the
expression for the transition Reynolds number.
For laminar range ( Re Re *)
j 0.36(Re) 0.51 (h / s) 0.275 (l / s) 0.27 (t / s) 0.063
(2.22)
f 4.67(Re) 0.70 (h / s)0.196(l / s)0.181(t / s)0.104
(2.23)
Turbulent range ( Re Re* )
j 0.18(Re) 0.42 (h / s) 0.288 (l / s) 0.184 (t / s) 0.05
(2.24)
f 0.32(Re) 0.286(h / s)0.221(l / s)0.185(t / s)0.023
(2.25)
26
The critical Reynolds number for heat transfer coefficient is given by
Re * 1568.58(h / s) 0.217 (l / s) 1.433(t / s) 0.217
(2.26)
The critical Reynolds number for pressure drop is given by
Re* 648.23(h / s) 0.06 (l / s) 0.1 (t / s) 0.196
(2.27)
Mochizuki et al [71] correlations are once again a reworking of the Wieting [67]
equations, with the coefficients and exponents modified to fit their own experimental
data for five scaled up rectangular offset strip fin surfaces. Only fully laminar flow and
fully turbulent flow are considered, with an abrupt change of flow regime at Re=2000.
Muzychka and Yovanovich [72] developed a new model to predict the heat
transfer and flow friction performance of offset strip fin geometries. They considered the
offset strip fins as an array of short channels or straight ducts. They developed simple
analytical models for the laminar or turbulent wake regions and suitably combined the
resulting asymptotic relations to create expressions for the turbulent zone. Their
correlation predicted the data in Ref [2] within 20 % for 96% of f data and 82% of j
data.
Hydraulic diameter
Hydraulic diameter is given by the following definition
Dh =
4 Ac
4 Ac
4 Freeflowvo lume
P
totalheatt ransferarea
A/l
(2.28)
The terms Ac, P and A have been evaluated differently by various investigators;
so there are different expressions for hydraulic diameter in the literature. At least three
different expressions can be identified in the literature, which are as given below
Manglik and Bergles [70]
Free flow area is taken Ac sh .In evaluating the heat transfer area A; the blunt
fin edges, both vertical and lateral, have been included in the channel surface area. Heat
transfer area is given by the expression:
A 2(sl hl ht ) ts
Therefore hydraulic diameter is given by the formula:
Dh
4shl
2( sl hl ht ) ts
(2.29)
27
Joshi and Webb [68], and Maiti and Sarangi [45]

Free flow area and heat transfer area are given as
Free flow area, Ac ( s t )h
Heat transfer area, A 2(sl hl ht )
Therefore hydraulic diameter is given by the formula:
Dh
2( s t )h
( sl hl ht )
(2.30)
Wieting [67] and Kays and London [2]

Considering a rectangular channel of cross section, sh , hydraulic diameter is
defined as :
Dh
4sh
2sh
2( s h) ( s h)
(2.31)
Critical Reynolds number and the transition zone

Joshi and Webb [68] conducted flow visualization experiments to identify the
flow structure at transition. They observed that a transition from steady laminar flow to
an oscillating or vortex shedding flow occurs at higher flow rates [72, 68, 73]; the flow is
generally characterized by a progression of laminar, second laminar (transitional, or
vortex shedding, or oscillating flow), and turbulent flow regimes [74].They gave a wake
width based definition for the critical Reynolds number and developed an expression of
the critical Reynolds number in terms of Reynolds number and other geometrical
parameters. Maiti and Sarangi [45] have developed the expression for the critical
Reynolds number as a function of geometrical parameters only.
In all the correlations mentioned above, only established laminar or turbulent
flow is considered and the transition zone is ignored. This extends over a fairly large
Reynolds number of 1000 in case of Wieting[67] and Joshi and Webb[68].Mochizuki et al
[71] and Dubrovsky and Vasilev[75] completely ignore the transition zone and consider
an abrupt change from laminar to turbulent flow; the evidence in the data for actual
cores[2,38] is contrary to this.
The change in the wake flow affects the j and f characteristics as seen in a
typical j and f versus Re plot. The departure from the log linear behavior of the j and f
28
curve is observed at higher mass flow rates. The intersection point of the two zones
gives the critical Reynolds number.
Mullisen and Loehrke [76] reported a direct correlation between the onset of
oscillating flows and the generation of audible tones. Other studies have found that the
development of oscillating flows and the consequent vortex shedding are influenced by
fin separation (or length) and fin offset[55,74,68].The heat exchanger core acts as a
flute for certain flow arrangements and flow rates.
2.5 Secondary irreversibilities

The performance of heat exchangers is adversely affected by the following
physical factors:
i)
Longitudinal heat conduction along the separating wall of the two fluids in the
heat exchanger.
ii) Heat transfer from or to the ambient,

iii) Flow maldistibution at the headers and due to manufacturing errors.
iv) Fluid property variation at low temperatures.
A. Effect of longitudinal heat conduction.

Most of the literature on the role of such secondary irreversibilities considers only
two stream heat exchangers. The effect of axial heat conduction in two stream heat
exchangers has been discussed by Barron [77] and by Shah [78]. Barron and Yeh [79]
computed temperature distribution and heat exchanger effectiveness considering the
effect of axial conduction along the separating walls. The temperature of the cold fluid
stream was assumed to be constant and the deterioration in performance was found to
depend on the axial conduction parameter.
Kays and London [2] expressed the ineffectiveness due to longitudinal heat
conduction in terms of a longitudinal heat conduction parameter, defined as:
kAc
LCmin
Chowdhury and Sarangi [80] have obtained the expression for the efficiency of
heat exchangers considering both axial conduction and lateral resistance due to the
separating wall in terms of relevant non-dimensional parameters. They gave an
expression for the optimum thermal conductivity of the separating wall for maximum
efficiency of heat exchangers.
29
Hausen [81] explained the deviation of temperature profiles from that of the heat
exchanger without longitudinal heat conduction assuming average temperature
properties. He thus presented an approximate method of predicting the performance
deterioration due to longitudinal heat conduction. The longitudinal heat conduction in a
single pass counter flow heat exchanger was studied by Hahnemann [82] and
complicated expressions were presented for evaluating the effectiveness of a heat
exchanger subject to longitudinal heat conduction. Bahnke and Howard [83] compared
their results with that of Hahnemann [82] and observed that deterioration of the heat
exchanger performance is maximum when the ratio of the flow stream capacity rates is
same.
The most comprehensive work on the performance degradation of heat
exchanger was carried out by Kroeger [84].He solved the governing equations of a two
stream counter flow heat exchanger, taking into account the effect of longitudinal heat
conduction. He presented a closed form solution for finding the ineffectiveness of a
balanced flow ( C r =1) heat exchanger as follows:
1
1 [ Ntu (1 B) /(1 Ntu )]
Where
1/ 2
Ntu
B
1 N tu
N tu
tanh
1/ 2
[Ntu /(1 Ntu )]
(3.7)
The parameter, ( 1 ) is also the dimensionless hot fluid exit temperature. Kroeger
presented the ineffectiveness of unbalanced flow ( Cr < 1 ) heat exchangers graphically as a
function of Ntu and longitudinal heat conduction parameter, .While Chiou [85-86] examined
its effect on the performance of cross flow heat exchangers.
In heat exchangers with large axial temperature gradient e.g. cryogenic heat
exchangers, the effect of axial heat conduction on heat exchanger performance is
significant, independent of number of streams. Venkatarathnam and Narayanan [87] have
studied the effect of longitudinal heat conduction from outer wall to the ambient on the
effectiveness of the heat exchanger. They have studied the performance of perforated plate
matrix heat exchangers and found that the performance degradation due to longitudinal
heat conduction through the walls separating the streams from the environment is therefore
nonnegligible in such heat exchangers. They also observed that the degradation of
30
performance of a high effectiveness heat exchanger due to longitudinal heat conduction

through the outer walls is non negligible even when the thermal resistance to heat transfer
between the stream and the outer wall is 10% of that between the fluids. The effect of
longitudinal heat conduction on performance of storage type heat exchangers have also
been investigated by several workers [88-89].
Recently Narayanam and Venkatarathnam [90] studied the performance of high
effectiveness counter flow heat exchangers subject to longitudinal heat conduction. They
derived the expression for the effectiveness of very high Ntu heat exchangers as a function
of the longitudinal heat conduction parameter, and heat capacity ratio, Cr
B. Effect of heat transfer to the ambient.

The performance of most heat exchangers may be seriously affected by heat
exchange with the surroundings. This is particularly true for cryogenic heat exchangers
because of the large temperature differences between the ambient and the operating
temperatures. This problem can be reduced by using highly effective insulation. But
factors like cost, weight and volume of insulation, difficulties in fabrication and ageing of
insulation limit the extent to which heat transfer may be reduced.
Wood and Kern [91] have solved the governing differential equations and
obtained a closed form solution for studying longitudinal heat conduction through wall
and heat leak from surroundings independently. Their solution requires an iterative
procedure to determine heat exchanger effectiveness considering heat loss to the
surroundings and is of only limited value to the design engineer. Chowdhury and Sarangi
[92] have also obtained an expression for the effectiveness considering heat loss in
terms of a heat leak parameter. They give the effectiveness without an iterative
procedure unlike Wood and Kern [91]. Heat transfer from the surroundings to either one
or both of the fluid streams in a binary heat exchanger is expressed in dimensionless
form and a direct expression for the outlet temperature has been obtained by Barron
[93]. Gupta et al [94] conducted experiments on a coiled tube- in- tube heat exchanger
and found that heat in - leak from the atmosphere to the cold fluid, flowing in the
annular region of the tube-in-tube heat exchanger is significant. They developed a
numerical model and compared the numerical results with experiment.
31
C. Effect of flow maldistribution :

In the design of heat exchangers, it was assumed that the fluid is uniformly
distributed across the heat exchanger cores. In practice, however, it is impossible to
distribute the fluid uniformly because of improper inlet configuration, imperfect design
and complex heat transfer process [95]. The gross flow maldistribution and passage to
passage flow nonuniformity exist in plate fin heat exchangers. Mueller and Chiou [96]
summarized various types of flow maldistribution in heat exchangers and discussed the
reasons leading to flow maldistribution.The combined effects of wall longitudinal heat
conduction, inlet flow nonuniformity and temperature nonuniformity on compact plate fin
heat exchangers using finite element method have been investigated by Ranganayakalu
and Seetharamu[97].Jiao[98] investigated experimentally and analyzed theoretically the
combined effects of distributors inlet angle and mass flow rate on flow maldistribution.
Some distributors with different inlet angles were studied under similar conditions and
optimum performance was obtained at inlet angle of 450.Zhang [99] proposed a
structure of two stage distribution and the numerical investigation showed that the flow
distribution in plate fin heat exchangers is more uniform if the ratio of outlet to inlet
equivalent diameters is equal for both headers. Wen [100] employed C.F.D technique to
simulate and analyze fluid flow distribution and pressure drop in the header of plate fin
heat exchangers. A baffle with small sized holes is recommended to install in the header
to improve the performance of flow distribution. Most of the previous works mainly
focused on the effect of flow non uniformity on heat exchanger performance
deterioration based on their own flow maldistribution model. Flemming [101]
investigated flow maldistribution in paired channel heat exchangers and investigated the
effect of flow maldistribution on performance deterioration. Jian Wen et al [102]
investigated the flow characteristics of flow field in the entrance of plate fin heat
exchangers by means of particle image velocimetry (PIV). PIV is an instantaneous flow
field measurement technique, which uses a pulsed light sheet to illuminate a gas flow
seeded with tracer particles. They investigated the effect of configuration changes on
flow field uniformity.
D. The effect of variable fluid properties

The effect of variable fluid properties on the performance of cryogenic heat
exchangers has been examined by Chowdhury and Sarangi [103]. The local heat transfer
coefficient (h) usually varies both as a function of temperature and flow velocity. Change
32
in the fluid properties with temperature can alter the local heat transfer coefficient. The
heat transfer coefficient also varies along the length due to boundary layer development.
Shah [104] has made an extensive review of this subject and has outlined a method for
incorporating the influence of these effects on two stream heat exchanger performance.
Paffenbarger[105] has incorporated the effect of longitudinal heat conduction and
variation of fluid properties with temperature in his computationally intensive numeric
model. He has also illustrated the effect of longitudinal heat conduction on multistream
heat exchanger performance through an example.
33
Chapter 3
Design of Plate Fin
Heat Exchanger
Chapter III
DESIGN OF PLATE FIN HEAT EXCHANGER
Design or sizing of heat exchangers covers determination of the heat exchanger
dimensions for the specified heat transfer and pressure drop performance. We can
reduce this problem to the rating problem by tentatively specifying the dimensions and
then calculating the performance for comparison with the specified performance. The
heat transfer coefficient of the surface increases with increase of flow Reynolds number.
But the pressure drop which is a function of flow velocity also increases. The optimum
velocity is to be found out for the specified pressure drop by using the core mass
velocity equation [7].
Accurate prediction of non-dimensional heat transfer coefficient, j and friction
factor, f is necessary for the correct design of heat exchangers. Various correlations are
available in literature for the determination of non dimensional heat transfer coefficient,
j and friction factor, f. It is difficult for the designer to choose the best among them. The
heat exchanger is designed by using the correlations developed by Maiti and Sarangi
[45], Manglik and Bergles [70], Joshi and Webb [68], and by using Aspen MUSE [113],
the simulation software. The performance of design procedures are validated by
experiment. The experimental aspect is described separately in the subsequent
chapters. In the present chapter the design procedure for the plate fin heat exchanger is
given.
Heat exchanger performance is deteriorated by various factors such as
longitudinal heat conduction, heat loss to the surroundings, flow maldistribution at the
headers, manufacturing irregularities etc. Longitudinal heat conduction through the
separating wall is the major contributor to the ineffectiveness of a heat exchanger. The
decrease in the effectiveness of heat exchangers due to longitudinal heat conduction is
found out by using Kroeger equation [84].
The plate fin heat exchanger consists of restricted and narrow passages. The
pressure drop will be high for gases passing through the heat exchanger. The headers
provided in the heat exchanger helps in the uniform flow distribution of the fluid among
the passages or channels. Cross sectional area of the header passages should be more
than the diameter of the tubes. Pressures at the inlet of both the fluids should be high
enough to overcome the pressure losses that occur during the flow through the heat
exchanger. In this way we can ensure that the flow channels are completely flooded and
there is no starvation anywhere.
Factors, j and f, are strong functions of surface geometry. Increase in heat
transfer performance is associated with increase in flow friction and vice versa. The ratio
of j/f is often taken as a measure of the goodness of a finned surface. Though the ideal
fin geometry should have high value of j/f, the selection of particular fin geometry is
primarily governed by the process requirement.
3.1 Summary of the design procedure:

The main steps involved in the design procedure are as follows.
a) Determination of the optimum mass velocity.
The optimum mass velocity for the specified pressure drop, p is calculated by
the relation:
p
2
G
o ( j / f )
2/3
ntu
(1 / m ) Pr
1/ 2
(3.1)
Referring to figure E 9.2 in pp.625 of Fundamentals of Heat Exchanger Design

by R.K.Shah [7], value of goodness factor, j / f is assumed to be 0.25 for the Reynolds
number ranging from 300 to 5000 for the specified geometry to calculate the initial
approximate value of G. The overall surface effectiveness of fins, 0 is assumed to be
80 % for the first iteration unless a better value is known from the past experience.
From the initial value of G calculated, individual j and f are calculated .After a number
of iterations, the core mass velocity G is obtained. It is also found that change in
Reynolds number has a very little effect on the ratio of j/f for the fin of particular
geometry.
b)
Assumptions on free flow area and frontal area.

The frontal area or the free flow area is assumed so that core mass velocity is
below the optimum value calculated as above. The length of the heat exchanger is
assumed.
c)
Heat transfer areas
The surface areas, Ah and AC of both the sides are then computed from the geometry.
35
than the diameter of the tubes. Pressures at the inlet of both the fluids should be high
enough to overcome the pressure losses that occur during the flow through the heat
exchanger. In this way we can ensure that the flow channels are completely flooded and
there is no starvation anywhere.
Factors, j and f, are strong functions of surface geometry. Increase in heat
transfer performance is associated with increase in flow friction and vice versa. The ratio
of j/f is often taken as a measure of the goodness of a finned surface. Though the ideal
fin geometry should have high value of j/f, the selection of particular fin geometry is
primarily governed by the process requirement.
3.1 Summary of the design procedure:

The main steps involved in the design procedure are as follows.
a) Determination of the optimum mass velocity.
The optimum mass velocity for the specified pressure drop, p is calculated by
the relation:
p
2
G
o ( j / f )
2/3
ntu
(1 / m ) Pr
1/ 2
(3.1)
Referring to figure E 9.2 in pp.625 of Fundamentals of Heat exchanger design

by R.K.Shah [7], value of goodness factor, j / f is assumed to be 0.25 for the Reynolds
number ranging from 300 to 5000 for the specified geometry to calculate the initial
approximate value of G. The overall surface effectiveness of fins, 0 is assumed to be
80 % for the first iteration unless a better value is known from the past experience.
From the initial value of G calculated, individual j and f are calculated .After a number
of iterations, the core mass velocity G is obtained. It is also found that change in
Reynolds number has a very little effect on the ratio of j/f for the fin of particular
geometry.
b)
Assumptions on free flow area and frontal area.

The frontal area or the free flow area is assumed so that core mass velocity is
below the optimum value calculated as above. The length of the heat exchanger is
assumed.
c)
Heat transfer areas
The surface areas, Ah and AC of both the sides are then computed from the geometry.
35
d) Fluid mean temperatures and fluid thermo physical properties

on each fluid side are calculated.
e) Calculation of heat transfer coefficient, j and friction factor, f
The Reynolds number and j and f factors on each side are calculated by using
available correlations.
f)
Determination of overall heat transfer coefficient.

The heat transfer coefficients on both hot and cold sides are computed. The
overall heat transfer coefficient is calculated by the formula:
1
1
a
1
(U O AO ) h ( ohhh Ah ) K w A w ( oc hc Ac )
(3.2)
where
0c overall surface effectiveness of fins
1 (a f / as ) (1 f )
(3.3)
with
n f fin efficiency tanh(Mle ) /( Mle )

where M is a fin parameter and is defined as
( 2 h)
(K f t)
(3.4)
g) Number of transfer units are calculated by the relation:
N tu
UA
C min
(3.5)
h) Effectiveness considering longitudinal heat conduction loss.

Longitudinal heat conduction along the separating surfaces of hot and cold fluids
causes serious performance deterioration. The decrease in the effectiveness of
the heat exchanger is found out by using Kroegers equation [84].
For the case of balanced operation, Cr 1 , Kroeger [84] has presented the
solution for the ineffectiveness as :
1
1 [ N tu (1 B) /(1 N tu )]
(3.6)
where
36
N tu
B
1 N tu
1/ 2
N tu
tanh
1/ 2
( N tu ) /(1 N tu
(3.7)
For the case of N tu >3, the argument of the hyperbolic tangent is greater than 3, and
tanh (3) = 0.995 or almost unity. For this case, the ineffectiveness given by equation
(3.6) becomes
1 N tu /(1 N tu )1 / 2
1 N tu
1 N tu
(3.8)
Kroeger [84] has developed an approximation for the ineffectiveness for the case of
unbalanced operation, Cr 1 as :
(1 Cr )
exp( r1 ) Cr
(3.9)
(1 C r ) N tu
1 ( N tu C r )
(3.10)
(1 )
where
r1
For the case of Cr 0.50 , the value of the function reduces to:
(1 )
(1 )
(3.11)
(1 ) y
1 (1 ) y
( y /(1 y)1 / 2
(3.12)
(1 Cr )
(1 Cr )(1 y)
(3.13)
y N tu Cr
(3.14)
3.2 Design inputs and specification of fin geometry:

The heat exchanger has to be designed for the following design conditions
A. Heat exchanger input data.
Fluid used: Nitrogen gas
Temperature of hot fluid at inlet, =310 K
Temperature of cold fluid at inlet, =83.65 K
37
Pressure at inlet of hot gas = 7.35 bar.

Pressure at inlet of cold gas = 1.15 bar.
Mass flow rate of both the fluids (hot and cold) = 5 g /s.
Allowable pressure drop, ph pc 0.05 bar
B. Fin geometry.
The heat exchanger is to be constructed of Aluminum alloy Al-3003 with rectangular
offset strip fins with the following basic dimensions.
Table 3.1 Fin geometry used in heat exchanger
Fin geometry
High pressure Side
Low pressure Side
01
Fin frequency,f
714 fins per metre
588 fins per metre
02
Length of fin, l
3 mm
5 mm
03
Fin thickness,t
0.2 mm
0.2 mm
04
Fin height,h
9.3 mm
9.3 mm
05
Number of layers
05
04
Geometrical characteristics related to selected fin geometry.

Other geometrical characteristics related to the fin geometry are calculated as follows
i)
Fin spacing, s (excluding the fin thickness)
(1 p f t )
( pf )
0.0012 m
ii) Free flow area to frontal area ratio,
a ff / a fr
( s t )h
(0.0012 0.0002) 0.0093
0.6992
(h t )(s t ) (0.0093 0.0002)(0.0012 0.0002)
iii) Heat transfer area / fin, as 2hl 2ht 2sl 0.00006672 m2

iv) Ratio of fin area to heat transfer area of fin,
2h(l t )
2 0.0093(0.003 0.0002)
0.8920
2(hl sl ht ) 2(0.0093 0.003 0.0012 0.003 0.0093 0.0002)
v) Equivalent Diameter, De
(4 Freeflowar ea length )
heattransferarea
2(s t )hl
2(0.0012 0.0002) 0.0093 0.003
0.001672662 m
hl sl ht (0.0093 0.003 0.0012 0.003 0.0093 0.0002)
vi) Distance between plates, b h t 0.0095 m
38
3.3 Correlation based design of heat exchanger

A. Design using correlation developed by Joshi and Webb[68]:
The design calculations for the given heat exchanger are as given below:
a) Desired performance of heat exchangers.
The desired effectiveness of heat exchanger is given by the formula:
Ch (Thi Tho )
25.9875(310 92.85)
0.96
Cmin (Thi Tci ) 25.8835(310 83.65)
Heat load, Q Ch (Thi Tho ) 25.9875(310 92.85) 5.56 kW
b) Estimation of dimensions of heat exchanger.

i)
Length of the heat exchanger = 900 mm
ii) Width of the core, W = 73 mm

iii) Total Number of layers, N = 9.
c) Calculation of heat transfer area, A
The heat transfer area for hot side is calculated as
Total area between plates, Afrh b N h W 0.0095 5 0.073 0.0035 m2
Total free flow area, A ffh A frh 0.002424 m2
Wall conduction area on hot side, awh A frh A ffh 0.00104286 m2
Total heat transfer area, Ah
4 A ffh L
De
4 0.0024246 0.9
5.2184 m2
0.00167266
Alternatively,
Total heat transfer area,
Ah
as p f W L N
l
0.00006672 714 0.073 0.9 5

5.2184 m2
0.003
Similarly free flow area and the heat transfer are calculated for the cold side.
Total wall conduction (longitudinal conduction) area, a wh a wc = 0.001043+0.00069736
= 0.00174 m2
39
d) Properties at average temperature.

The fluid properties at the estimated mean temperatures of 202.95 K and 191.16
K for hot and cold fluid are obtained from property package, GASPAK
The properties of hot helium gas at the mean film temperature:
Specific heat, cp =5197.5 J/kg- K
Viscosity, 0.0000154 Pa-s
Prandtl number, Pr 0.6685
Density, 1.7404 kg/m3
The properties of cold helium gas at the mean film temperature.

e) Heat transfer coefficients and surface effectiveness of fins

The core dimensions are calculated for the side having more stringent pressure
drop specification. The dimensions on the other side are then chosen such that the
calculated pressure is within the specified limit. The calculations for the heat transfer
coefficients for the hot and cold gas are similar. The calculations for hot fluid are given
below.
i)
The core mass velocity, G
ii) The Reynolds number, Re
m
0.005
2.062 kg/s-m2
A ffc 0.002424
GDe
2.062 0.0016766
223.98
0.0000154
iii) The critical Reynolds number for pressure drop and heat transfer coefficient is
given by
Re
Re * 257(l / s)1.23 (t / l ) 0.58 De t 1.328
lD e
0.5
40
0.5
223.98

1.23
0.58
257(3 / 1.2) (0.2 / 3) 0.001672 0.2 1.328

0
.
003
0
.
001672
691.75
iv) The Colburn j factor (for Re Re *) is given by correlation proposed by Joshi

and Webb [68] as
j c 0.57(Re) 0.5 (l / De ) 0.15 ( ) 0.14

0.57(223.98) 0.5 (0.003 / 0.0016726) 0.15 (0.1290) 0.14 0.04321
v) The convective heat transfer coefficient, hc is given by
hc
( j c cc G c )
0.667
(Pr)
(0.04321 5197.5 2.062)

605.804 W/m2-K
0.667
(0.6685)
vi) The fin parameter is given by
(2 hc )
(K f t)
(2 605.804)
(150 0.0002)
200.964 m-1
vii) Heat loss to the ambient decreases the overall surface effectiveness of fins and
finally the effectiveness of heat exchanger. In heat exchangers used for
cryogenic service, the layers through which the cold fluid passes are placed in
between the two hot layers .This minimizes the heat loss from the cold fluid. The
number of layers through which the hot fluid passes will be more than that of
cold fluid by one and are exposed more to the ambient. To take into account the
heat losses to the ambient, the fin conduction lengths for the outer layers on the
hot side will be taken as b whereas for the inner layers of the hot fluid, the
conduction length is taken as b/2.However for the cold layers placed between
the hot layers the fin conduction length is taken as b/2 for both the inner and
outer layers.
l e = effective length of fins for inner layers of hot fluid =b/2= 4.75 mm and
l e = effective length of fins for outer layers of hot fluid = b = 9.5 mm
viii) The fin effectiveness for a straight fin is
tanh(Mle ) /( Mle ) =0.7771 for inner layers.
41
For the outer layers of hot side the fin effectiveness is 0.50127.
ix) The overall surface effectiveness of fins on hot side is
0 h 1 (a f / as ) (1 f )
0 h 1 (0.8920) (1 0.771)
( N p2)
2
)
(a f / as ) (1 fo) (
N p
N p
(5 2)
2
(0.8920) (1 0.5012) ( ) 0.7027
5
5
f) The overall heat transfer coefficient and number of transfer units.

The overall heat transfer coefficient is given by the formula:
1
1
a
1
where,
Aw lateral conduction area W L (2 N Pc 2) 0.073 0.9(2 4 2) 0.657 m2

1
0.0008
1
1
(U O AO ) h (0.7027 605.8 5.2184) 150 0.657 (0.8214 524.945 3.454)

= 0.00113 K/W
(U O AO ) h 885.23 W/K
The overall heat transfer coefficient, U Oh
Number of transfer units, N tu
(U O Ao) h 885.23
169.63 W/m2-K
Aoh
5.2184
U O AO 885.23
34.206
C min
25.883
g) Effectiveness of heat exchanger, neglecting longitudinal heat conduction.

The effectiveness of heat exchanger, neglecting longitudinal heat conduction is
given by the relation [77]:
1 e Ntu (1Cr )
1 e 34.20(10.996)
0.9734
1 Cr e Ntu (1C r ) 1 0.996e 34.20(10.996)
h) The effect of longitudinal heat conduction

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
Kroegers equation [84].
42
i)
Wall conduction area, a w 0.00174 m2
ii) Conductivity of fin, K w 150 W/m-K

iii) Wall conduction parameter,
K w a w 150 0.00174
0.0112
LCmin
0.9 25.883
iv) y N tu Cr 0.0112 34.20 0.996 0.3817

v)
(1 Cr )
(1 0.996)
0.00145
(1 Cr )(1 y) (1 0.996)(1 0.3817)
(1 ) y
0.00029
1 (1 ) y
vi) ( y /(1 y )1 / 2
vii)
(1 ) (1 0.001011)
1.0000
(1 ) (1 0.001011)
viii) r1
(1 C r ) N tu
(1 0.996) 34.20
0.09901
1 N tu Cr 1 0.0112 34.20 0.996
ix) (1 )
x)
(1 Cr )
(1 0.996)
0.0368
exp( r1 ) Cr 1 exp(0.09901) 0.996
[1 (1 )] 0.9632
The actual effectiveness of heat exchanger taking into account longitudinal heat
conduction is 0.96 and matches with the desired effectiveness. The required number of
transfer units for achieving the desired performance of the heat exchangers is 34.21 and
length of the heat exchanger is 900 mm.
i) Pressure drop
Since pressure drop of cold fluid is more critical, the pressure drop calculations
for the cold fluid are presented here.
i)
The Colburn f factor ( for Re Re *) proposed by Joshi correlation [68] is as
f 8.12.(Re) 0.74 (l / De ) 0.41 ( ) 0.02

8.12.(223.98) 0.74 (1.7935) 0.41 (0.129) 0.02
0.12138
ii) The pressure drop, p
4 fLG 2 4 0.12138 0.9 (2.398) 2
2675.99 Pa
2 De b
2 0.001672 0.2808
The pressure drop is less than the allowable pressure drop of 5 kPa. Hence the
design does meet the hydraulic requirement.
43
B. Design using correlation developed by Maiti and Sarangi{45]:

Maiti and Sarangi[45] used a C.F.D tool to compute the pressure, velocity and
temperature fields in the plate and fin heat exchanger passages. They expressed the
non dimensional heat transfer coefficient, j as a function of the Reynolds number and
other geometrical properties. Some constants in the correlation have been found out by
multiple regression from the numerically computed results where as the rest of the
constants in the correlation are found out by fitting experimental data of Kays and
London [2].
The expressions proposed by Maiti and Sarangi [45] for Colburn, j factor for
laminar and turbulent flows (for Re Re *) are given as,
j 0.36(Re) 0.51 (h / s) 0.275 (l / s) 0.27 (t / s) 0.063
(3.15)
j 0.18(Re) 0.42 (h / s) 0.288 (l / s) 0.184 (t / s) 0.05
(3.16)
Maiti and Sarangi [45] gave two separate expressions for finding the critical
Reynold number for transition from laminar to turbulent flow for heat transfer and
pressure drop considerations.
The critical Reynolds number for heat transfer coefficient is given by
Re * 1568.58(h / s) 0.217 (l / s) 1.433(t / s) 0.217
(3.17)
The critical Reynolds number for pressure drop is given by
Re* 648.23(h / s) 0.06 (l / s) 0.1 (t / s) 0.196
(3.18)
The authors used the following definition of hydraulic diameter to calculate the
Reynolds number
Dh
2( s t )h
( sl hl ht )
(3.19)
The dimensions of the heat exchanger are found out for the specified
performance of the heat exchanger. The length of the heat exchanger for achieving the
desired effectiveness is 1020 mm.The maximum pressure drops is 0.02 bar for the cold
fluid. This is well below the available pressure drop of 0.05 bar. The other dimensions of
heat exchanger core are same as given earlier.
44
C. Design using correlation developed by Manglik and Bergles[70]:

Manglik and Bergles [70] examined the experimental data for air flows with
heat transfer for 18 different cores given by Kays and London [2], Walters [112] and
London and Shah [38]. They analyzed the effect of various geometrical attributes of
offset strip fins and gave the following expression for j and f factor which is valid
continuously for laminar, turbulent and transition zones.
j 0.6522 Re 0.5403( s / h) 0.1541(t / l ) 0.1499(t / s) 0.0678
1 5.269 10
Re1.340 ( s / h) 0.504 (t / l ) 0.546 (t / s) 1.055
0.1
f 9.6243 Re 0.7422( s / h) 0.1856(t / l ) 0.3053(t / s) 0.2659
1 7.669 10
Re 4.429 ( s / h) 0.920 (t / l ) 3.767 (t / s) 0.236
0.1
(3.20)
(3.21)
The authors have chosen the free flow or channel flow area as Ac sh with the
hydraulic diameter given by the following expression
De
4 Ac
4shl
A / l 2( sl hl th ) ts
(3.22)
The length of the heat exchanger for achieving the desired performance of
0.96 is found out to be 1008 mm .The other dimensions of the heat exchanger are given
as given above. The maximum pressure drops is 0.014 bar for the cold fluid. The
pressure drop is reasonably low and is within the permissible limit.
3.4 Design of heat exchanger using simulation software

The heat exchanger has also been designed by using AspenMUSE [113], the
simulation software. The Aspen design takes into account the various losses occurring in
the heat exchanger like longitudinal heat conduction, heat losses to the ambient, flow
maldistribution at the headers, pressure losses in the headers etc. It has been accepted
as a versatile tool for the design of plate fin heat exchangers in industrial applications.
The core length of the heat exchanger calculated by using AspenMUSE [113] is
1050 mm. The pressure drop of the cold fluid and the hot fluid are 0.0325 bar and
45
0.00781 bar including the pressure drop in the headers. This is well below the
permissible pressure drop of 0.05 bar
3.5 Concluding dimensions of the heat exchanger

The selected dimension is given in tabular form as given below in Table 3.2
Table 3.2 Concluding dimensions of heat exchanger.
ITEM
DIMENSIONS
ITEM
DIMENSIONS
CORE LENGTH
900 mm
TOTAL LENGTH
1000 mm
CORE WIDTH
73 mm
TOTAL WIDTH
85 mm
CORE HEIGHT
93 mm
TOTAL HEIGHT
105 mm
PLATE THICKNESS
0.8 mm
END PLATE THICKNESS
6 mm
END BAR THICKNESS
6 mm
46
Chapter 4
Rating of Plate Fin
Heat Exchanger
Chapter IV
RATING OF PLATE FIN HEAT EXCHANGER
Rating of a heat exchanger consists of the steps leading to finding the thermo
hydraulic performance of a given heat exchanger for known dimensions of the
exchanger and given fin geometry. Since the outlet temperatures are not known in a
rating problem, the mean temperatures of the fluids have to be estimated first. The heat
transfer coefficient and the effectiveness of the plate fin heat exchanger are found based
on different correlations available in literature. The outlet temperatures and the average
fluid temperatures are calculated for an assumed effectiveness which is verified with the
calculated value. This is an iterative procedure and is repeated until the calculated values
of the exit fluid temperatures matches with the assumed values.
Longitudinal heat conduction along the separating surfaces of the two streams
causes serious performance deterioration in heat exchangers. It is due to short
conduction lengths and higher number of transfer units ( N tu ). The effectiveness
deterioration caused by longitudinal heat conduction is obtained using Kroeger equation
[84]. He obtained the expression for the ineffectiveness due to longitudinal conduction
as a function of longitudinal heat conduction parameter, , heat capacity ratio, C r and
number of transfer units, N tu . Heat loss to the ambient causes an energy unbalance.
Two values of effectiveness are obtained - h , the effectiveness based on the hot fluid
and c , the effectiveness based on the cold fluid. The effectiveness considering the heat
loss is obtained using simulation software, Aspen- Muse by substituting the
experimentally obtained heat loss as input along with other inputs parameters. Aspen
MUSE [113] after an iterative procedure gives two values of effectiveness, h , the
effectiveness based on the hot fluid and c ,the effectiveness based on the cold fluid.
Besides this, the effectiveness of the heat exchanger is also affected by flow
maldistribution at the headers. The manufacturing irregularities such as burred edges,
separating plate roughness and bonding imperfections influence the thermo hydraulic
performance of the heat exchanger.
A performance test is conducted on the given plate fin heat exchanger core. The
values of the effectiveness and pressure drops obtained at different mass flow rates and
at different hot inlet temperatures by experiment are compared with theoretical
predictions. The details of the experimental set up are given in chapter V and
performance analysis is given in chapter VI.
4.1 Details of given heat exchanger and input data

Tables 4.1 and 4.2 give the details of the given heat exchanger core along with
the fin geometry used in the performance test. The thermo hydraulic performance of the
given heat exchanger is to be found for the given mass flow rate. The pressures and
temperatures of hot and cold fluid at inlet are as given below.
Table 4.1 Dimensions of the heat exchanger core
ITEM
DIMENSIONS
ITEM
DIMENSIONS
CORE LENGTH
900 mm
TOTAL LENGTH
1000 mm
CORE WIDTH
73 mm
TOTAL WIDTH
85 mm
CORE HEIGHT
93 mm
TOTAL HEIGHT
105 mm
PLATE THICKNESS
0.8 mm
END PLATE THICKNESS
6 mm
END BAR THICKNESS
6 mm
The heat exchanger is constructed of Aluminum alloy Al-3003 with rectangular offset
strip fins having the following basic dimensions.
Table 4.2 Fin geometry of the heat exchanger core
Fin geometry
High pressure Side
Low pressure Side
01
Fin frequency,f
714 fins per meter
588 fins per meter
02
Length of fin, l
3 mm
5 mm
03
Fin thickness, t
0.2 mm
0.2 mm
04
Fin height, h
9.3 mm
9.3 mm
05
Number of layers
05
04
48
A. Geometrical characteristics related to selected fin geometry

Other geometrical characteristics related to the fin geometry are calculated from the
given fin geometry .The calculations have been given in chapter III.
B. Heat exchanger input data.

Temperature at hot nitrogen gas at inlet, =368.96 K
Temperature at cold nitrogen gas at inlet, =315.24 K
Pressure at inlet of hot nitrogen gas = 1.0721 bar
Pressure at inlet of cold nitrogen gas = 1.0917 bar
Mass flow rate of hot nitrogen gas = 5.77 g/s
Mass flow rate of cold nitrogen gas = 5.77g/s
4.2 Rating of the given heat exchanger using different

correlations
Effectiveness and pressure drop are obtained for given mass flow rate at the
desired inlet temperatures of hot and cold fluids by using the correlations developed by
Maiti and Sarangi[45], Manglik and Bergles [70] and Joshi and Webb [68]. Simulated
values of the effectiveness and pressure drop are also obtained by using simulation
software, Aspen- MUSE [113].They are compared with the performance parameters
obtained by experiments in chapter V. The procedure for calculating the effectiveness
and pressure drop using different correlations and simulation software is given below.
A. Rating of the given heat exchanger using correlations developed by Maiti

and Sarangi [45]
a) Heat transfer area, A
The heat transfer area for high pressure side is calculated as follows:
Total area between plates, A frh b N h W 0.0095 5 0.073 0.0035 m2
Total free flow area, A ffh A frh 0.002425 m2
Wall conduction area on hot side, awh A frh A ffh 0.001043 m2
Total heat transfer area, Ah
4 A ffh L
De
4 0.0024246 0.9
5.215 m2
0.00167266
49
Alternatively,
Total heat transfer area,
Ah
as p f W L N
0.00006672 714 0.073 0.9 5

5.215 m2
0.003
Similarly free flow area and the heat transfer areas are calculated for the cold side.
Total wall conduction (longitudinal conduction) area, a wh a wc = 0.001043+0.00069736
= 0.00174 m2
b) Estimation of average temperature.
Since the outlet temperatures are not known for the rating problem, the mean
temperatures of the fluids have to be estimated first. The fluid properties at the
estimated mean temperatures of 344.15 K and 340.05 K for hot and cold fluid are
obtained from property package, GASPAK.
The properties of hot nitrogen gas at the mean film temperature are:
The properties of cold nitrogen gas at the mean film temperature are,
Specific heat, C p 1040.7 J/kg- K
c) Heat transfer coefficients and surface effectiveness of fins

The experimental set up uses atmospheric air from the compressor as the cold
fluid and heated air from the heater as the hot fluid. For the hot fluid test conducted, the
hot layer is sandwiched between the two outer cold layers. Therefore the number of
layers through which the cold air from the compressor passes is taken as 5 where as for
the hot air the number of layers is taken as 4.
50
The calculations for the heat transfer coefficients for the hot and cold gases are
similar. The calculations are presented for the cold nitrogen gas.
i)
The core mass velocity, G
ii) The Reynolds number, Re
mc
0.00577
2.38 kg/s-m2
A ffh 0.002425
GDe
2.38 0.001674
202.3
0.0000197
iii) The critical Reynolds number for heat transfer coefficient is given as:
Re * 1568.58(h / s) 0.217 (l / s) 1.433(t / s) 0.217

1568.58(7.743) 0.217 (2.498) 1.433(0.1665) 0.217 399.77
iv) The Colburn factor, j (for Re Re *) is given by correlation proposed by Maiti
and Sarangi [45] as:
jc 0.18(Re) 0.42 (h / s) 0.288 (l / s) 0.184 (t / s) 0.05

0.18(202.298) 0.42 (7.743) 0.288 (2.498) 0.184 (0.1665) 0.05 0.03685
v) The convective heat transfer coefficient is given as:
hc
( jc cc Gc ) (0.03685 1040.71 2.3804)
113.95 W/m2K
0.667
0.667
(Pr)
(0.7169)
vi) The fin parameter is calculated as:

a. M
(2 hc )
(K f t)
(2 113.954)
(170 0.0002)
81.87 m-1
vii) The hot layer is sandwitched between the two outer cold layers. The cold layers
are exposed more to the atmosphere and for calculating the fin effectiveness, fin
conduction lengths for the outer layers on the cold side will be taken as b to take
into account the heat losses from the ambient whereas for the inner layers the
conduction length is taken as b/2.However for the hot layers sandwiched
between the two cold layers, The fin conduction length is taken as b/2 for both
inner and outer layers.
51
l c height of fins for the inner layers of the cold fluid =b/2= 4.75 mm.
l c height of fins for outer layers of the cold fluid
= b = 9.5 mm
l h height of fins for both inner and outer layers of the hot fluid = b/2 = 4.75
mm
viii) The fin effectiveness for a straight fin is
n f tanh(Ml c ) /( Ml c ) 0.9525 for inner layers.

For the outer layers of cold side the fin effectiveness is 0.8375
ix) The overall surface effectiveness of fins on cold side is:
0c 1 (af / as ) (1 fi )(
(N P 2)
NP
0c 1 (0.8920) (1 0.9524)(
(a f /as ) (1 fo )(
NP
(5 2)
2
(0.8920) (1 0.8375)( )
5
5
=0.9166.
The overall surface effectiveness of fins on hot side is:
0h 1 (af / as ) (1 f ) 1 (0.8656)(1 0.9626) 0.9676

d) The overall heat transfer coefficient and number of transfer units
The overall heat transfer coefficient is given as:
1
1
a
1
where,
Aw lateral conduction area W L (2 N P 2) 0.073 0.9(2 4 2) 0.657 m2

1
0.0008
1
1
(U O AO ) h (0.9676 88.464 3.452) 170 0.657 (0.9166 113.954 5.215)

= 0.00522 K/W
(U O AO ) h 191.32 W/K
Overall heat transfer coefficient, U Oc
Number of transfer units, N tu
(U O Ao) c 191.32
36.68 W/m2-K
Aoc
5.215
U O AO 191.32
31.84
C min
6.0084
52
e) Effectiveness of heat exchanger, neglecting longitudinal heat conduction

The effectiveness of heat exchanger, neglecting longitudinal heat conduction is
given by the relation [77]:
1 e Ntu (1Cr )
1 e 31.84(10.9998)
0.9696
1 Cr e Ntu (1C r ) 1 0.9998e 31.84(10.996)
f) The effect of longitudinal heat conduction
The effect of longitudinal heat conduction is to reduce the effectiveness of heat
exchanger. The decrease in the effectiveness of heat exchanger is determined using
Kroegers equation [84].
i)
Wall conduction area, a w 0.00174 m2
ii) Conductivity of fin, K w 170 W/m-K

iii) Wall conduction parameter,
K w a w 170 0.00174
0.05467
LCmin
0.9 6.0084
iv) y N tu Cr 0.05467 31.843 0.996 1.741

v)
(1 Cr )
(1 0.996)
1.928 10 5
(1 Cr )(1 y) (1 0.996)(1 1.7408)
(1 ) y
2.675 10 5
1 (1 ) y
vi) ( y /(1 y )1 / 2
vii)
viii) r1
(1 ) (1 2.675 10 5 )
1.0000
(1 ) (1 2.675 10 5 )
(1 C r ) N tu
(1 0.996) 31.84
0.001228
1 N tu Cr 1 0.05467 31.84 0.996
ix) (1 )
(1 Cr )
(1 0.996)
0.0761
exp( r1 ) Cr 1 exp(0.001228) 0.996
x) [1 (1 )] 0.9238
This is the value of the actual effectiveness of heat exchanger after considering
longitudinal heat conduction. Outlet temperatures of fluids based on this value of
effectiveness are calculated as follows:
53
The outlet temperature of hot fluid is computed as:
Tho Thi
C min (Thi T ci )
Ch
368.96
0.9238 6.0084(368.96 315.24)

319.34 K
6.0084
The outlet temperature of cold fluid is computed as:
Tco Tci
C min (Thi T ci )
Cc
0.9238 6.0084(368.96 315.24)

364.87 K
6.0084
315.24
Mean temperatures of hot and cold fluid are given by

Mean temperature of hot fluid, Thm
Thi T ho 368.96 319.34
344.15 K
2
2
Mean temperature of cold fluid, Tcm
Tci Tco 315.24 364.87
340.05 K
2
2
g) Calculation of pressure drop, p

Since pressure drop of cold fluid stream is more critical, the pressure drop calculations
for the cold fluid are presented here.
i)
The critical Reynolds number for pressure drop is given calculated as:
Re* * 648.23(h / s) 0.06 (l / s) 0.1 (t / s) 0.196
648.23(7.7435) 0.06 (2.5) 0.1 (0.1667) 0.196 892.77

ii) The friction factor
If Re>Re**,
f 0.32(Re) 0.286 (h / s) 0.221(l / s) 0.185 (t / s) 0.023

0.32(202.298) 0.286 (7.743) 0.221(2.5) 0.185 (0.1665) 0.023
0.1731
4 fLG 2 4 0.1732 0.9 (2.3804) 2
980.67 Pa
iii) The pressure drop, p
2 De b
2 0.001674 1.07551
Since the pressure drop is less than the allowable pressure drop of 5 kPa, the
design does meet the hydraulic requirement.
Thus the value of the effectiveness of the heat exchanger obtained using the
correlations developed by Maiti and Sarangi[45] after considering the longitudinal
conduction losses is 0.9238 and the pressure drops in the cold and hot fluids are 0.0098
54
bar and 0.00746 bars respectively. This is well below the allowable pressure drop of
0.05 bar.
B. Rating of the given heat exchanger using correlations developed

by Manglik and Bergles[70]
The rating of the given heat exchanger is also done by correlations developed by
Manglik and Bergles [70] using the same procedure as given above. For better
readability, the (Equations 3.20, 3.21 and 3.22) of Manglik and Bergles [70] are
rewritten as
The non-dimensional heat transfer coefficient,j and friction factor,f are given by
the following expressions
j 0.6522 Re 0.5403( s / h) 0.1541(t / l ) 0.1499(t / s) 0.0678
1 5.269 10
Re1.340 ( s / h) 0.504 (t / l ) 0.546 (t / s) 1.055
0.1
f 9.6243 Re 0.7422( s / h) 0.1856(t / l ) 0.3053(t / s) 0.2659
1 7.669 10
Re 4.429 ( s / h) 0.920 (t / l ) 3.767 (t / s) 0.236
0.1
Hydraulic diameter is given by the following expression
De
4 Ac
4shl
A / l 2( sl hl th ) ts
The effectiveness of the heat exchanger obtained using the correlation developed
by Manglik and Bergles [70] and after considering the longitudinal conduction losses is
0.9424 and the pressure drops in the cold and hot fluids are 0.0064 bar and 0.005 bar
respectively. This is well below the allowable pressure drop of 0.05 bar
C Rating of the given heat exchanger using correlations developed

by Joshi and Webb[68]
Joshi and Webb [68] used the same expression for the hydraulic diameter as
defined by Maiti and Sarangi [45]. The non dimensional heat transfer coefficient, j and
friction factor, f are given by the following expressions.
The Colburn, j factor and friction factor, f (for Re Re *) are given by
correlations developed by Joshi and Webb [68].
55
j c 0.57(Re) 0.5 (l / Dh ) 0.15 ( ) 0.14
(4.1)
f 8.12.(Re) 0.74 (l / Dh ) 0.41 ( ) 0.02
(4.2)
The effectiveness of the heat exchanger obtained using the Joshi and Webb [68]
correlation and after considering the longitudinal conduction losses is 0.9303 and the
pressure drops in the cold and hot fluids are 0.0074 bar and 0.0053 bar respectively.
This is well below the allowable pressure drop of 0.05 bar
4.3 Rating of the given heat exchanger using simulation software,

Aspen-MUSE {113].
Aspen MUSE [113] is the simulation software which is being increasingly used
for the industrial design of heat exchangers. It takes into consideration the various
losses that affect the thermo hydraulic performance of the heat exchanger while
simulating for the given inputs. The value of the effectiveness using simulation software
is 0.89. The pressure drops of the cold and hot fluid are 0.008 bar and 0.00525 bar
respectively.
Table 4.3 and Table 4.4 show the predicted value of the effectiveness and
pressure drop of the cold fluid calculated using different correlations and by simulation
software, Aspen MUSE [113].
Table 4.3.The predicted value of the effectiveness using different correlations for a mass
flow rate of 5.77 g/s operating between 315 K and 369 K.
SL. No
Mass flow
Effectiveness
rate,g/s
Maiti
5.77
and
Manglik and
Joshi and
Sarangi
Bergles
Webb
correlation
correlation
correlation
0.9238
0.9434
0.9303
Aspen-MUSE
0.89
Table 4.4.The predicted value of the pressure drop of the cold fluid using different
correlations for a mass flow rate of 5.77 g/s operating between 315 K and 369 K.
SL. No
Mass flow
Pressure drop
rate,g/s
Maiti
5.77
and
Manglik and
Joshi and
Sarangi
Bergles
Webb
correlation
correlation
correlation
0.0098
0.00637
0.0074
Aspen-MUSE
0.008
56
4.4 Effect of heat transfer to the ambient.

The performance of cryogenic heat exchangers may be seriously affected by heat
exchange with the surroundings. Because of the large temperature difference between
the ambient and operating temperature, the quantity of heat leak is quite high. This
reduces the effectiveness of the heat exchangers. This problem can be reduced by using
highly effective insulation, but the amount of insulation to be used may be too high.
Cryogenic heat exchangers are essentially placed in the vacuum chamber to minimise
the heat loss.
In this analysis the hot test method is used to study the performance of the
given plate fin heat exchanger. The flow direction is reversed so that the hot layers
(layers through which the hot fluid passes) are placed inside the two outer cold layers.
In this experiment several layers of glass wool and thermocole insulations are used on
the heat exchanger to eliminate the heat transfer to the surroundings. A resistance
temperature detector is placed on the outer surface of the insulation to indicate the
temperature difference for assessment of heat losses to the surroundings.
Energy balance between the hot and cold fluids is lost if there is a heat loss to
the surroundings. The temperature difference between the two fluids at the hot end and
that at the cold end are noted down. The difference in the values of temperatures at hot
end and cold end indicates the loss of energy. Hence two values of effectiveness - h ,
the effectiveness based on the hot fluid and c , the effectiveness based on the cold fluid
are obtained.
57
Chapter 5
Experimental Apparatus
Chapter V
THE EXPERIMENTAL APPARATUS
The experimental set up used in this experiment consists of a counter flow plate
fin heat exchanger. Cold air from the compressor is made to flow through one channel
where as hot air coming from a heating unit is made to flow through the second channel
in a counter flow direction. This chapter presents the measurement principle, layout of
the experimental set up, description of the different components of the set up,
calibration procedure of the instruments used, and an analysis of possible experimental
errors.
5.1 Experimental set up and the operation

The experimental rig comprises of the air supply system, the heating unit, heat
exchanger core and the instrumentation / measurement system as shown in Figures 5.1
and 5.2. In this apparatus a counter flow type of heat exchanger is used as the test
section.
Air is used as a working fluid in this experiment. The heat exchanger is
connected to a screw compressor which is capable of continuously supplying dry air. A
control valve is used to regulate the flow rate through the heat exchanger. The cold air
enters the heat exchanger from the bottom and gets heated as it passes through the
exchanger. The air coming out then passes through the heating unit and gets further
heated. The hot air coming out of the heater is fed into the heat exchanger from the top
end. The amount of hot air entering the heater is regulated by a valve located at the
inlet of the heater. The bypass valve will be closed for the balanced case, i.e, when the
mass flow rates of both the fluids are equal. The heat supplied to the heater is controlled
with the help of variacs.
Pressure gauges are provided to measure the pressures at inlet of both the hot
and cold fluids. The pressure taps are located on the upstream and downstream of heat
exchanger to measure the pressure drop across the heat exchanger. These pressure
taps are connected to a U tube manometer to give the value of pressure drop. The inlet
and outlet temperatures of both the fluids are measured by using resistance temperature
detector (RTD). For balanced flow, a rotameter at the exit of the heat exchanger is used
for measuring the flow rate of the fluids directly in the circuit. In case of unbalanced
flow, orifice plates are used for measuring the flow rate of both the fluids. Rotameter
also helps in calibrating the flow rate of the orifice meters as and when required.
Figure 5.1: Schematic P&I diagram of the experimental test rig

1: Compressor
2: Control Valve
3,7: Pressure Taps
4,8: U- Tube manometer
5: Heater
6:Test section
T1, T2, T3, T4 are RTDs
9: Bypass valve
10: Flowmeter
The test section is carefully insulated by using glass wool and thermocole
(polystyrene foam) sheets to eliminate heat losses from the system to the surroundings.
Nearly 80 mm of glass wool insulation is used as shown in Figure 5.3. A resistance
temperature detector is placed on the outer surface of the insulation to indicate the
temperature difference for assessment of heat losses to the surroundings.
The by pass valve is closed for the balanced flow rate operation. The flow rate
through the test section is set at the desired value. The volume flow rate through the
test section can be observed in a rotameter placed at the exit of the heat exchanger test
rig at balanced condition. The variac is kept at a low value initially and then increased
59
gradually according to the desired hot inlet temperature. Pressures at the inlet of both
the fluids are noted from pressure gauges. Temperature of hot air at inlet of the heat
exchanger is maintained at the desired temperature by adjusting the wattage of the
variac. The system is allowed to run until the steady state conditions are reached. The
inlet and outlet temperatures of both the fluids are noted down from the measurement
of resistance temperature detectors (RTD). Pressure drop of both the fluids are read
from the Utube mercury manometers.
Figure 5.2: Photograph of the experimental set up

The performance parameters of the heat exchangers such as the effectiveness
and pressure drop of the fluids are calculated. The values of the effectiveness, pressure
drops of the fluids are also found out using the various correlations available in literature
and also from well known simulation software, Aspen MUSE [113]. The performance
parameters obtained by experimentation and that obtained by simulation are compared
with the theoretical values for analyzing the various losses.
60
Figure 5.3: Photograph of the experimental set up (with insulation)
5.2 Calculation procedure

In the steady state experiment, measurement of temperature and mass flow
rates in the two sides provides the required information to compute the heat exchanger
effectiveness. For the flow rate of 300 liter/min operating between 315 K and 369 K, the
calculations of the performance parameters are given as below
i)
For a balanced mass flow rate of hot and cold fluids, the effectiveness is given by
(T2 T1 ) (T3 T4 )
(T3 T1 ) (T3 T1 )
(5.1)
where
T1 =Temperature at inlet of cold fluid
= 315.2 K
T2 =Temperature at outlet of cold fluid = 360.2 K
T3 =Temperature at inlet of hot fluid
= 368.96 K
T4 =Temperature at outlet of hot fluid = 321.1 K

we get
h = Effectiveness of hot fluid = 0.89

c = Effectiveness of cold fluid = 0.84
61
ii) Mass flow rate of the fluids flowing through the heat exchanger is measured by a
rotameter placed at the exit of the heat exchanger test rig.
Mass flow rate, m 4 Q
(5.2)
where
4 =Density of the hot fluid at exit =1.1542 kg/m3

Q = Volume flow rate measured by the rotameter = 0.005 m3/sec
Mass flow rate, m 1.1542 0.005 5.77 g/s.
iii) The pressure taps are located on the upstream and downstream of heat exchanger
to measure the pressure drop across the heat exchanger. The minor pressure loss
due to flow through elbows and pressure loss in the headers have to be subtracted
from the measured pressure drop to get the pressure drop in the heat exchanger
core. The pressure loss in the headers is found out by simulation with and without
headers.
Pressure drop in the core = measured pressure drop minor lossespressure loss in the headers
= 0.012-0.001-0.0056
= 0.0054 bar
The non-dimensional heat transfer coefficient, j and friction factor, f
are
calculated using the correlations developed by Maiti and Sarangi [45], Manglik and
Bergles [70], and Joshi and Webb [68]. The theoretical or the predicted values of the
effectiveness are calculated by using the rating procedure outlined in chapter IV. The
theoretical value of the effectiveness calculated using the above correlations and that
calculated using simulation software, Aspen-MUSE [113] are compared with the
effectiveness value obtained from experiment.
5.3 Effect of heat transfer from the ambient

The performance of heat exchangers may be seriously affected by heat leak
from the surroundings. Because of the large temperature difference between the
ambient and operating temperatures, the quantity of heat leak is quite high. This
problem can be reduced by using highly effective insulation. But factors like cost, weight
and volume of insulation, difficulties in fabrication and ageing of insulation limit the
extent to which heat transfer may be reduced. Heat transfer from the surroundings is
manifested as a reduction in the effectiveness of heat exchangers. In this experiment
62
several layers of glass wool and thermocole insulation are used on the heat exchanger to
eliminate the heat transfer from the surroundings. This can be ascertained from the
reading on the temperature detector placed on the outer surface of insulation.
5.4 Description of various equipment and instruments

A. Plate fin heat exchanger.
The test section consists of a counter flow plate fin heat exchanger with offset
strip fin geometry. The design procedure of the given plate fin heat exchanger has been
given in chapter III. Figure 5.4 shows the plate fin heat exchanger with all its
dimensions and arrangements of inlet and outlet ports. The cold layers (layers through
which the cold fluid passes) are sandwiched in between the two outer hot layers of the
plate and fin heat exchangers used in cryogenic applications. Table 5.1 shows the flow
arrangement for the hot and cold fluids intended for cryogenic applications. However for
the performance test conducted (Hot test), the flow direction has been reversed. The
cold fluid from the compressor is made to flow through the high pressure side (having 5
layers) whereas the hot fluid passes through the low Pressure side (having 4
layers).Thus a hot layer is sandwiched between the two outer cold layers. The Tables
5.2 and 5.3 provide the details of core dimensions and fin geometry used in the given
heat exchanger. The Table 5.4 gives the design data of the given heat exchanger.
Table 5.1: Flow arrangement for the designed heat exchanger
HIGH PRESSURE SIDE
LOW PRESSURE SIDE
(HOT SIDE)
(COLD SIDE)
OSF
OSF
NO. OF PASSAGE
NO. OF PASS
FLOW RATE
COUNTER FLOW
COUNTER FLOW
FIN
Table 5.2: Core dimensions of the test heat exchanger

CORE LENGTH
900 mm
TOTAL LENGTH
1000 mm
CORE WIDTH
73 mm
TOTAL WIDTH
85 mm
CORE HEIGHT
93 mm
TOTAL HEIGHT
105 mm
63
Figure 5.4: Plate fin heat exchanger
64
Table 5.3: Fin geometry in the given heat exchanger

Fin geometry
High pressure Side
Low pressure Side
01
Fin frequency, f
714 fins per metre
588 fins per metre
02
Length of fin, l
3 mm
5 mm
03
Fin thickness ,t
0.2 mm
0.2 mm
04
Fin height, h
9.3 mm
9.3 mm
05
Number of layers
05
04
Table 5.4: Design data of the given heat exchanger

HEAT LOAD
5.5 KW
HOT SIDE
COLD SIDE
HELIUM (HP)
HELIUM (LP)
FLOW RATE
5 g/s
4.8 g/s
INLET TEMP.
310 K
83.65 K
OUTLET TEMP.
92.85 K
301.67 K
ALLOWABLE PRESSURE DROP
0.05 bar
0.05 bar
PRESSURE AT INLET
7.35 bar
1.15 bar
FLUID
B. Twin screw compressor

The air supply system consists of a twin screw rotary compressor which is a
positive displacement type. In this oil flooded rotary compressor lubricating oil bridges
the space between the rotors, providing a hydraulic seal and transferring mechanical
energy between the driving and driven rotors. Gas enters at the suction side and
meshing rotors force the gas through the threads as the screws rotate. Screw
compressors have relatively high rotational speed compared to other types of positive
displacement machines which make them compact. Sufficient amount of oil provided
gives the cooling effect to maintain the temperature nearly constant. They have the
ability to maintain high volumetric efficiencies over a wide range of operating pressure
and flow rates. It has long service life and high reliability.
The Compressor specification is given below:
65
Table 5.5 Compressor Specifications:

Make:
Kaeser (Germany)
Model:
BSD 72
Profile of screw:
Sigma
Free air delivery:
336 m3 /hr
Suction pressure:
Atmospheric
Maximum Pressure:
11 bar
Operating temperature:
75- 1000C
Motor:
37kW, 74amps, 3, 50Hz,

415V10%, 3000rpm
Oil capacity:
24 L
Cooling:
Air
C. Heating device
This heating element was fully designed and developed in our laboratory. It
consists of a shell containing a number of heating elements. The cold fluid enters the
shell from one side and moves over the heater to leave at the other end of the shell. The
heating elements are arranged in a particular pattern which also acts as baffles for
better heat transfer. There are seventeen number of heating tubes each having a
capacity of 1500 W (220/230V, 50 Hz).
D. Resistance Temperature Detectors (RTDs)

Resistance thermometers also called as resistance temperature detectors (RTD)
or resistive thermal devices are temperature sensors that exploit the predictable change
in electrical resistance of some materials with changing temperature. It is a positive
coefficient device, which means that the resistance increases with temperature. So the
material whose resistance increases with temperature is used for making the RTD.
Typical elements used for RTD include nickel (Ni), copper (Cu), but platinum (Pt) is by
far the most common, because it has the best accuracy and stability in comparison to
other RTD materials. For platinum RTD the resistance versus temperature curve is fairly
linear and the temperature range is widest and has a very high resistivity. It means that
only a small amount of platinum is required to fabricate a sensor and making platinum
costs competitive with other RTD materials. The RTDs are slowly replacing the use of
66
thermocouples in many industrial applications below 600 0C, due to higher accuracy and
reliability.
Figure 5.5: RTDs construction

Figure 5.5 shows the construction of a RTD. RTDs are constructed by one of
two different manufacturing configurations. First one is the wire wound RTD which are
constructed by winding a thin wire into a coil. Second type, thin fin element is a more
common configuration, which consists of a very thin layer of metal laid out on a plastic
or ceramic substrate. Thin-film elements are cheaper and more widely available because
they can achieve higher nominal resistances with less platinum. In order to protect the
RTD, a metal sheath encloses the RTD element and lead wires are connected to it. RTDs
are available with three different lead wire configurations. The selection of a particular
configuration depends on the desired accuracy and instruments to be used for the
measurement.
(a) Two wire configuration
(b) Three wire configuration and
(c) Four wire configuration.
RTDs are popular because of their excellent stability and exhibit the most linear
signal with respect to temperature when compared to any other electronic temperature
sensor. They are generally more expensive than alternatives, however, because of the
careful construction and use of platinum, RTDs are also characterized by a slow
response time and low sensitivity; and because they require current excitation, they can
be prone to self heating. And there main limitation is that they cannot be used for
measurement of temperature above 660 0C and below -270 0C. Also they are less
sensitive to very small temperature changes.
RTDs are commonly categorized by their nominal resistance at 00C. By far the
most common RTD used in the industry have a nominal resistance of 100 Ohms at 00C
67
are called as the PT-100 sensors. The relationship between resistance and temperature
is nearly linear and follows the equation,
For < 0 0C
RT = R0 [1+ aT+bT2+cT3 (T-100)]
(5.4)
For > 0 0C
RT = R0 [1+ aT+bT2]
(5.5)
Where,
RT = resistance at temperature T
R0 = resistance at nominal temperature
a, b, and c are the constants used to scale the RTD.
Four numbers of RTDs are used for the measurement of inlet and exit
temperature of both the fluid streams. For accurate measurement of temperature, these
RTDS are to be calibrated with a single known temperature. Water is heated in a beaker
with the help of induction type heater. The resistance temperature detectors to be
calibrated are immersed in the water .The water is heated slowly and is stirred with the
help of stirrer for uniform distribution of heat. The temperatures indicated in a 16
channel temperature indicator are noted down.
The variation of temperature detector readings T2, T3 and T4 are plotted with
respect to temperature indicator reading T1The thermometer is also inserted to observe
the temperatures and it acts as a reference thermometer. The calibration graph and the
measured values are shown in fig 5.7 and Table 5.6.The set up used for calibration of
resistance temperature detectors is shown in fig 5.6.
Figure 5.6: Photograph of the set up used for calibration
68
Temperatures (T2,T3 and T4)
120
100
T2
80
T3
T4
60
40
20
20
40
60
80
100
120
Temperature,T1
Figure.5.7: RTD Calibration graph
Table 5.6 Calibration Chart
THERMOMETER(0C)
RTD1(0C)
RTD2(0C)
RTD3(0C)
RTD4(0C)
30.5
32.8
31.85
33.67
31.8
33.64
31.98
33.82
37.04
32.84
34
34.87
34.84
35.01
34.03
39
40.31
40.31
40.46
39.48
43.5
44.98
44.98
45.13
44.11
47
48.9
48.89
45.13
44.11
50
51.72
51.69
51.84
50.76
54
55.76
55.74
55.92
54.85
57
58.66
58.7
58.82
59.71
60
61.55
61.6
61.69
60.51
62.8
63.93
63.98
64.07
62.87
65
66.67
66.71
66.78
65.54
68
70.04
70.06
70.24
69.1
71
72.45
72.43
72.63
71.44
74.5
76.1
76.12
76.34
75.12
85
86.69
86.81
86.87
85.57
90
91.64
91.79
91.89
90.64
94.5
96.21
96.34
96.45
95.04
100
101.8
101.91
101.99
100.51
105.5
107.55
107.73
107.77
106.38
107.5
107.9
108.01
108.08
106.55
110.5
113.5
113.36
113.42
111.95
112.5
114.71
114.85
114.89
113.33
115
117.04
117.18
117.22
115.65
117
120
120.07
120.12
118.54
69
E. Orifice mass flow meter

A flow meter or flow sensor is an instrument used in process instrumentation to
measure the flow rate of liquid or gas. Its working is based on the Bernoullis principle
which says that there is a relationship between the pressure and the velocity of a fluid
stream. When the velocity increases, the pressure decreases and vice versa.
An orifice plate is basically a thin plate with a hole in the middle. It is usually
placed in a pipe in which fluid flows. When the fluid reaches the orifice plate, with the
hole in the middle, the fluid is forced to go through the small hole of varying cross
section causing the change in both the velocity and pressure of the fluid. The point of
maximum convergence actually occurs shortly at downstream of the physical orifice and
is called as the point of vena contracta. Beyond the vena contracta, the fluid expands
and the velocity and pressure change to normal. The volumetric and mass flow rates are
obtained from the Bernoullis equation by measuring the difference in fluid pressure
between the normal pipe section and at the vena contracta, as shown in figure 5.7.The
pressure recovery is limited for an orifice plate and the permanent pressure loss depends
primarily on the area ratio. For an area ratio of 0.5, the head loss is about 70 - 75% of
the orifice differential.
Fig.5.7 Orifice plate [114]

The volume flow rate, Q can be calculated from the formula given below:
Q CC d 2 gha
(5.6)
Where, C = area constant =
Cd = coefficient of discharge and ha = head of air = hw (
w
1)
a
Mass flow rate, m = a Q
70
F. Variac or Autotransformer
A variac, also called as an autotransformer is an electrical transformer with only
one winding. The auto prefix refers to the single coil acting on itself rather than any
automatic mechanism. In an autotransformer portions of the same winding act as both
the primary and secondary. The winding has at least three taps where electrical
connections are made. In India, autotransformers are used to step up or step down
between voltages in the 220-230-240-volt range.
The autotransformers are used to
regulate the voltage of the heating element to get the desired temperature of the heater
unit.
5.5 Error analysis:

The total uncertainty in an observable or measurable quantity is decomposed
into two parts: random errors and systematic errors. A random error is defined as the
uncertainty detected by repeating the measurement procedure under the same
conditions, while a systematic error is that which cannot be detected through this
method and is usually associated with bias in experimental data. Random errors are
generally caused by the imprecision of the measuring instruments and fluctuations in
environmental conditions. They can, in general, be bound within desired limits by using
precision instruments and by controlling environmental factors. But systematic errors are
inherent in the experimental process and a high degree of subjective judgment is
necessary to estimate these errors. For example, faulty design of the test system or lack
of homogeneity in the heat exchanger core leads to errors in velocity distribution, which
translate to errors in Re and f. Even if the replicated measurements give identical
results, this intrinsic defect will yield results that may not replicate in future. Calibration
of the experimental system can eliminate some of these errors.
The following major sources of systematic errors are identified in this
experimental system:
1.
Errors inherent in the design and execution of the test system: Faulty design of the
test system leads to error in velocity distribution, which leads to errors in j and f
values. This can contribute up to 1% in j and f data, as observed by Kays and
London [2].This in turn effect the measurement of the effectiveness.
71
2.
Errors in calibration of pressure transducers and temperature sensors. The

uncertainty in calibration of the sensors is estimated to be of the order of 1% for
pressure transducers and 0.5% for temperature sensors.
3.
Lack of homogeneity in test core construction such as uneven accumulation of

brazing alloy at the roots of the fins and compression of flow passages results in
errors in both friction factor and heat transfer measurements. The level of
uncertainty is of the order of 1% as reported by Kays and London [2].
The result R of an experiment is calculated from a set of measurements, for
example, x1, x2, x3 xn. Thus,

R=R(x1, x2, x3 xn)
Let R be the overall uncertainty in the result, representing the range (on both
the sides of R) in which the true value may lie, and x1,x2,x3.....xn the uncertainties in
the independent variables x1, x2, x3 xn respectively. In a simplistic way, the overall
uncertainty may be calculated, by adding the uncertainties caused by all independent
variables.
R
R
R
x1
x2 .......
x n
x1
x2
xn
(5.7)
In this expression, errors in x1, x2, x3 xn, with the same sign and the
maximum magnitude for each term, are combined in the worst possible way, resulting in
an overestimation of the experimental inaccuracy. This can happen only when the
variables are not really independent. A more realistic expression for the overall
uncertainty can be predicted by the root mean square error: [106-111]
2
2
R 2 R
R

R x1
x 2 ........
xn
x1 x2
xn

0.5
(5.8)
The effectiveness of the heat exchanger is given by
m1 (T2 T1 )
m2 (T3 T1 )
(T2 T1 )
m1 m2 (T3 T1 )
(5.9)
72
m1 (T2 T1 ) 1(T2 T1 )
(T T2 )
1
2
m2 m2 (T3 T1 ) m(T3 T1 ) m(T3 T1 )
T 1
m2 (T3 T1 )
(5.10)
m1 (T2 T1 )
m2 (T3 T1 )
m1 (T2 T1 )
T1
T1
2
m2 (T3 T1 ) 2
m2 (T3 T1 )(m1 ) m1 (T2 T1 )(m2 )
2
T 1
m2 (T3 T1 )2
=
(m2T3 m2T1 )(m1 ) (m1T2 m1T1 )(m2 )

2
m2 (T 3T1 )2
m1 m2T3 m1m2T1 m1m2T2 m1m2T1

2
m2 (T3 T1 ) 2
m1 (T2 T3 )
m2 (T2 T1 ) 2
(5.11)
m1 (T2 T1 )
T2 T2 m2 (T3 T1 )
=
m1T2
m1T1
T2 m2 (T3 T 1) m2 (T 3T1 )
T2 (T3 T1 )
(5.12)
m (T T ) x
1 2 1
T 3 x
x
T3
m1 (T2 T1 )
T 3 m2 (T3 T1 ) 2
(5.13)
(
m1 ) 2 (
.m2 ) 2
.T1
.T2
T3
m1
m2
T1
T2
T3
(5.14)
Where
73
m1 and m2 are the errors in mass flow rates
T1 , T2 And T 3 are the errors in temperatures which are equal to 0.1 K for RTD.
The mass flow rate is measured by a rotameter placed at the exit of the hot
fluid. The accuracy of the rotameter has been specified as 12 liter/min by the
instrument supplier. However the readings of the rotameter were checked by a thermal
mass flow meter. The least count of RTD is 0.05 K as specified by the supplier. In this
experiment the temperature difference is measured. Thus the total least count is
addition of individual least counts which comes to 0.1 K.
The error in mass flow rate = the error in volume flow rate density of hot fluid
at outlet
= 0.0002 1.15
0.000231 kg /sec
For the mass flow rate of 5.77 g/s operating between 315.4 K and 369.13 K, The
error terms in equation (5.14) are obtained by substituting the temperature and mass
values in equation (5.9) to (5.13).
The error terms in the equation are obtained as 145.09, 145.09, 0.00303, 0.0186
and 0.01558 and the errors in mass and temperatures to get the error in effectiveness
as
2
(
m1 ) 2 (
.m2 ) 2
.T1
.T2
T3
m1
m2
T1
T2
T3
(145 .09 0.000234 )2 (145 .09 0.000234 )2 0.00303 0.12 0.01861 0.12
0.015578 0.12
= 0.0475
Hence for an effectiveness of 89%, the percentage of uncertainty is 4.75%.
74
Chapter 6
Performance Analysis
Chapter VI
PERFORMANCE ANALYSIS
Hot fluid test is conducted to determine the performance parameters such as
effectiveness and pressure drop, p across the core for both the fluids and compare
them with the theoretical or predicted values. The experiment is conducted at different
mass flow rates (5.7 g/s to 14.2 g/s) and at different hot fluid inlet temperature to study
the variation of the performance parameters. The amount of air entering the heat
exchanger is controlled by a control valve placed at the inlet of the heat exchanger. The
temperature of the hot air at inlet is maintained at the desired value by using the auto
transformer. The values of the experimentally observed data have been tabulated in
Tables 6.1 to 6.4.
Table 6.1.Experimentally observed data at different mass flow rates for the hot fluid inlet
368.96
321.1
400
0.14
0.12
15
12
311.35
359.94
367.91
316.95
500
0.2
0.17
25
22
311.93
361.38
368.88
317
550
0.24
0.20
30
26
312.82
361.71
369.45
317.35
588
0.28
0.24
31
27
313.41
361.33
368.96
317.86
650
0.32
0.26
40
35
314.16
360.74
368.72
318.08
Hold fluid inlet
temperature, T2
fluid, mm of Hg
Pressure drop, hot
fluid, mm of Hg
temperature, T4(K)
360.22
Hold fluid outlet
315.24
temperature, T3 (K)
Cold fluid outlet
(K)
Cold fluid inlet
(mm of Hg)
temperature,T1(K)
Pressure drop, cold
0.06
inlet,P2(kg/cm2)
0.08
Pressure at hot
Pressure at cold
300
inlet,P1(kg/cm2)
Flow rate, Q (litr/min)
temperature of 369 K
358.83
319.3
400
0.14
0.1
15
13
313.6
352.88
358.86
318.43
500
0.2
0.16
24
20
312.7
353.05
358.69
317.35
550
0.24
0.19
30
26
315.08
353.06
358.86
318.99
588
0.28
0.23
34
31
316.55
353.16
358.83
320.3
650
0.34
0.28
38
35
315.75
352.39
359.32
319.06
Hold fluid inlet
temperature, T2
fluid, mm of Hg
Pressure drop, hot
fluid, mm of Hg
temperature, T4(K)
352.08
Hold fluid outlet
313.94
temperature, T3 (K)
Cold fluid outlet
10
(K)
Cold fluid inlet
(mm of Hg)
temperature,T1(K)
12
Pressure drop, cold
0.06
(kg/cm2)
0.09
Pressure at hot inlet,P2
Pressure at cold
300
inlet,P1(kg/cm2)
Flow rate ,Q (litr /min)
348.86
317.78
400
0.13
0.11
15
13
314.13
344.11
348.98
317.85
500
0.2
0.16
23
21
316.18
344.66
348.88
319.5
550
0.24
0.19
30
26
316.1
344.52
348.71
319.44
588
0.28
0.24
33
31
316.62
344.63
348.88
319.59
650
0.34
0.28
39
34
316.6
344.16
348.8
319.18
Hold fluid inlet
temperature, T2
fluid, mm of Hg
Pressure drop, hot
fluid, mm of Hg
temperature, T4(K)
343.27
Hold fluid outlet
313.32
temperature, T3 (K)
Cold fluid outlet
(K)
Cold fluid inlet
(mm of Hg)
temperature,T1(K)
2
(kg/cm
Pressure) drop, cold
0.06
inlet,P2
0.08
(kg/cm2)
Pressure at hot
Pressure at cold
300
inlet,P1
Flow rate,Q (litr /min)
76
316.94
400
0.14
0.11
16
14
315.77
335.86
339.26
318.45
500
0.2
0.16
24
22
312.51
335.42
338.9
315.55
550
0.24
0.19
30
26
316.46
336.01
338.83
318.86
588
0.28
0.23
33
31
312.99
335.34
338.8
315.57
650
0.34
0.28
37
34
315.72
335.67
339.16
317.93
Hold fluid inlet
temperature, T2
mm of Hg
temperature, T4(K)
339.31
Hold fluid outlet
335.01
temperature, T3 (K)
Cold fluid outlet
313.92
(K)
Cold fluid inlet
(mm of Hg)
temperature,T1(K)
Pressure drop, hot fluid,
mm of Hg
0.06
(kg/cm2)
0.08
Pressure at hot inlet,P2
Pressure at cold inlet,P1
300
(kg/cm2)
Flow rate ,Q (litr /min)
Pressure drop, cold fluid,
6.2 Variation of effectiveness with the mass flow rate:

The figures 6.1 to 6.4 show the comparison of the effectiveness values obtained
by various correlations and the simulation software with the experimental values at
different mass flow rates and at different hot inlet temperatures. The difference in the
values of temperatures at hot end and cold end indicates a loss of energy. In an ideal
situation without heat leak from the surroundings, the temperature drop in hot stream
must be equal to the temperature gain in cold stream. There is always some amount of
energy imbalance even with sufficient insulation. Hence two values of effectiveness - h ,
the effectiveness based on the hot fluid and c , the effectiveness based on the cold fluid
are obtained.The energy imbalance is minimum when the heat exchanger is operated at
500-550 litre/min (10.44 to 11.72 g/s). It is observed from experiment that hot
effectiveness, h (effectiveness measured on the basis of hot fluid) increases with the
mass flow rate where as the cold effectiveness, c ( the effectiveness based on the cold
fluid) increases up to certain mass flow rate, remains constant and then decreases with
77
further increase in mass flow rate. The mean of these effectiveness values is also found
out and variation of the mean effectiveness with mass is shown.
The effectiveness values obtained by using the correlations developed by Maiti
and Sarangi [45], Joshi and Webb [68] have found to increase up to the mass flow rate
of 11.72g/s and then remain constant. The values of effectiveness obtained by
simulation using Aspen MUSE [113] and the experimental mean effectiveness have also
shown the same trend. The effectiveness is directly related to NTU (hA / mc p ) .With
increase in mass flow rate, the Reynold number increases and thus increases the heat
transfer coefficient as per the correlation at a faster rate than the mass flow rate. This
increases NTU which gives higher effectiveness .Further increase in mass flow rate,
gives saturated condition where the heat transfer coefficient increases slowly compared
to the mass flow rate. The effectiveness thus increases very slowly or remains nearly
constant at higher mass flow rate.
The values of the effectiveness obtained by different correlations are without the
heat loss considerations and are hence compared with the mean experimental value.
The actual experimental values that may have been obtained without the heat loss may
be slightly higher (nearer to the hot effectiveness line) than these mean experimental
values. The percentage deviation between the mean experimental values and the
predicted values given by Maiti and Sarangi[45] varies from 6.42 % to 4.57%, while for
Manglik and Bergles[70],the variation is from 8.27% to 4.97 % and for Joshi and
Webb[68], the variation is from 7.07 % to 5.54 % when the inlet temperature of the hot
fluid is 369 K. The experimental values agree with the values obtained by simulation.
The percentage deviation between the mean experimental values and the simulated
values is from 3.83% to 2.6%. It can be seen that the percentage deviation between the
effectiveness values obtained by various correlations and the experimental values
decreases with increase in mass flow rate and with the increase in hot inlet temperature.
The effectiveness values with heat losses are also obtained using simulation
software, Aspen MUSE [113]. The amount of heat loss obtained from the experiments is
given as an input in the simulation software along with other inputs. Simulation gave two
values of the effectiveness, the hot effectiveness and the cold effectiveness. These are
compared with the respective hot and cold effectiveness value obtained from the
experiments as shown in figure 6.5. The percentage deviation between the effectiveness
value obtained by simulation software, Aspen (with heat leak considered) and the
78
experimental value varies from 2.26 % to 2.75 % at the hot inlet temperature 369 K.
This is well within the error band of the measurement error. More detailed description is
given in the next section.
Effectiveness v/s mass flow rate (369 K)

Experimental(Hot)
0.96
Experimental(Cold)
0.94
Dipak Maiti
Manglik
Effectiveness
0.92
Joshi
0.9
Muse
Expt mean
0.88
0.86
0.84
0.82
4
12
16
Mass flow rate(gm/sec)
Figure 6.1: Variation of effectiveness with mass flow rate (hot inlet temperature=369 K)
Effectiveness v/s mass flow rate(359 K)

0.96
0.94
Effectiveness
Experimental(Hot)
0.92
Experimental (cold)
Dipak
0.9
Manglik
0.88
Joshi
0.86
aspen MUSE
Experimental mean
0.84
0.82
4
8
12
Mass flow rate,gm/sec
16
79
Effectiveness v/s mass flow rate(339 K)

0.96
Experimental(Hot)
0.94
Experimental(Cold)
Effectiveness
0.92
0.9
Dipak Maiti
0.88
Manglik
0.86
Joshi
0.84
Muse
0.82
4
12
16
Experimental mean
6.3 Effect of heat transfer to the ambient

Energy balance between the hot and cold streams is lost if there is a heat loss to
the surroundings. The temperature difference between the two fluids at the hot end and
that at the cold end are noted down. The difference in the values of these temperatures
at hot end and cold end indicates a loss of energy. Hence there are two values of
80
effectiveness - h , the effectiveness based on the hot fluid and c , the effectiveness
based on the cold fluid. For balanced flow condition, the heat unbalance of the two
streams is given as,
mc p (T3 T2 ) (T4 T1 )
5.77 1.04 (8.75 5.82) 17.582 W.

Substituting the input heat value as 17.58 W and for the mass flow rate of 5.77
gm/sec operating between 369 K and 315.3 K, simulation by Muse gave the two values
of effectiveness, the hot effectiveness and the cold effectiveness. These effectiveness
followed the same trend as the respective experimental values as shown in the figure
6.5.
Effectiveness with heat leak at hot inlet temp of

369 K
0.96
Effectiveness
hot(muse)
Effectiveness
0.94
0.92
Effectiveness
cold (muse)
0.9
Experimental
(Hot)
0.88
0.86
Experimental
(cold)
0.84
0.82
4
12
16
Figure 6.5: Comparison of effectiveness obtained by experiment and by simulation

with heat leak at different mass flow rates.
The percentage deviation between the effectiveness values obtained by
simulation software, Aspen MUSE [113] (with heat leak considered) and the
experimental value varies from 2.26 % to 2.75 % at the hot inlet temperature 369 K.
6.4 Comparison of effectiveness obtained with and without heat

loss
Figure 6.6 shows the comparison of the effectiveness values obtained by
simulation software aspen-MUSE [113] with and without heat loss for the mass flow rate
of 5.77 g/s operating between 369 K and 315.3 K. As explained earlier, heat loss to the
ambient causes an energy imbalance between the hot and cold streams. Energy lost by
the hot fluid is not equal to energy gained by the cold fluid. By substituting the heat leak
81
in the simulation software, Aspen- MUSE[113], two values of the effectiveness are
obtained effectiveness - h , the effectiveness based on the hot fluid and c ,the
effectiveness based on the cold fluid. The mean of these effectiveness is calculated and
variation of this mean effectiveness with mass flow rate is shown in figure 6.6.
Variation of effectiveness with mass flow by Muse

(with and without heat leak)
0.96
Effectiveness
hot(muse)
Effectiveness
0.94
0.92
Effectiveness
cold (muse)
0.9
0.88
Muse(Without
heat leak)
0.86
0.84
4
8
12
16
Muse
mean(without
heat loss)
Figure 6.6: Variation of effectiveness with mass flow rate including heat leak (hot inlet
temperature=369 K)
The mean effectiveness is compared with the effectiveness obtained without
heat loss for the same mass flow rate. The comparison shows that the effectiveness
obtained without heat loss is slightly more than the mean effectiveness. Hot end of the
heat exchanger gets more affected by the heat loss to the surroundings. This means
that the effect of heat loss on effectiveness of the cold fluid is more compared to that on
hot fluid.
The effectiveness obtained by experiments is with heat loss while the predicted
values of the effectiveness obtained by different correlations are without heat loss. For
analysis purpose the mean of the experimental hot and cold effectivenesses can be
approximated as the effectiveness without heat loss. The mean experimental value is
compared with the predicted value of the effectiveness to find the percentage deviation
between the predicted and the experimental value of the effectiveness. Actual deviation
may be slightly lower than this deviation when there is no heat leak.
6.5 Error estimation in experimental results

The errors in the measurement of effectiveness have been estimated from the
measurement of individual variables for one set of data. Equations 5.9 to 5.14 of
chapter-V are used to calculate the uncertainty in the measurement of effectiveness.
82
Table 6.5 shows the uncertainties in the value of effectiveness obtained at different
mass flow rates when the inlet temperature of hot fluid is 369 K. It is found that the
uncertainties decrease with increase in mass flow rate.
Table 6.5. Uncertainties obtained at different mass flow rates for the hot fluid inlet
T1(K)
T2(K)
T3(K)
T4(K)
m1 m2
5.77
315.24
360.22
368.96
321.1
145.1
0.0030
0.0186
0.01557
0.04807
8.02
311.35
359.94
367.91
316.95
107.07
0.0025
0.0177
0.01518
0.03661
10.44
311.93
361.38
368.88
317
83.125
0.0023
0.0175
0.01523
0.0297
11.72
312.82
361.71
369.45
317.35
73.67
0.0024
0.0177
0.01525
0.02663
12.81
313.41
361.33
368.96
317.86
67.32
0.0025
0.018
0.01552
0.0253
14.48
314.16
360.74
368.72
318.08
58.96
0.0027
0.0183
0.01564
0.02284
T1
T2
T3
uncertainty
Mass flow rate,

g/s
6.6 Variations of pressure drop of cold fluid with the mass flow
rate:
Figures 6.7 to 6.10 show the comparison of the pressure drop values obtained by
various correlations and the simulation software with the experimental values at different
mass flow rates and at different hot inlet temperatures. It is seen that the pressure drop
increases continuously with mass flow rate. The pressure drop of cold fluid is below the
allowable pressure drop of 0.05 bar up to the Reynolds number of 550. The pressure
drop exceeds 0.05 bar when the Reynolds number of the flow exceeds 600.
A large amount of deviation is observed between the experimental and
theoretical pressure drops obtained by various correlations. The variation between the
pressure drops obtained by experiments and by simulation is also large. However the
pressure drop is not a serious concern since it is within the allowable limit.
83
Pressure drop,bar
Pressure drop of cold fluid v/s mass flow

rate(369 K)
0.045
0.04
0.035
0.03
0.025
0.02
0.015
0.01
0.005
0
Experimental(cold)
Dipak
Manglik
Joshi
Muse
12
16
Figure 6.7: Variation of pressure drop of cold fluid with mass flow rate (hot inlet
temperature = 369 K)
Pressure drop,bar
Pressure drop of cold fluid v/s mass flow rate

(359 K)
0.045
0.04
0.035
0.03
0.025
0.02
0.015
0.01
0.005
0
Experimental(cold)
Dipak
Manglik
Joshi
Muse
12
16
temperature =359 K)
84

rate(349 K)
0.045
Pressure drop,bar
0.04
Experimental(cold)
0.035
Dipak
0.03
0.025
Manglik
0.02
0.015
Joshi
0.01
Muse
0.005
0
4
12
16
temperature=349 K)

rate(339 K)
0.045
Pressure drop,bar
0.04
Experimental(cold)
0.035
Dipak
0.03
0.025
Manglik
0.02
0.015
Joshi
0.01
0.005
Muse
0
4
12
16
Mass flow rate,gm/sec.
temperature=339 K)
85
6.7 Variations of pressure drop of hot fluid with the mass flow
rate:
Variation of pressure drop of hot fluid with mass flow rate is also shown in
Figures 6.11 to 6.14. It is seen that the pressure drop increases continuously with the
mass flow rate. The pressure drop is below the allowable pressure drop of 0.05 bar over
the entire range of mass flow rate.
Pressure drop of hot fluid v/s mass flow

rate(369 K)
0.03
Pressure drop,bar
0.025
experimental(hot)
Dipak
Manglik
Joshi
Muse
0.02
0.015
0.01
0.005
0
4
12
16
Figure 6.11: Variation of pressure drop of hot fluid with mass flow rate (hot inlet
temperature=369 K)
rate(359 K)
0.04
Pressure drop,bar
0.035
Experimental(hot)
0.03
Dipak
0.025
0.02
Manglik
0.015
Joshi
0.01
Muse
0.005
0
12
16
Figure 6.12: Variation of pressure drop of hot fluid with mass flow rate
(hot inlet
temperature=359 K)
86

rate(349)
0.04
0.035
Experimental
(Hot)
Dipak
Pressure drop,bar
0.03
0.025
Manglik
0.02
0.015
Joshi
0.01
Muse
0.005
0
0
10
20
Figure 6.13: Variation of pressure drop of hot fluid with mass flow rate (hot inlet
temperature = 349 K)
Pressure drop of hot fluid v/s mass

flow rate(339)
0.04
Pressure drop,bar
0.035
0.03
Experimental(hot
)
Dipak
0.025
0.02
Manglik
0.015
0.01
Joshi
0.005
Muse
0
4
8
12
16
Figure. 6.14 Variation of pressure drop of hot fluid with mass flow rate (hot inlet
temperature=339 K)
6.8 Results and discussion

The values of
effectiveness obtained from experiments agree with the values
obtained by simulation software, Aspen-MUSE[113].The percentage deviation between

the effectiveness values obtained by simulation software, Aspen-MUSE (with heat leak
considered) and the experimental values varies from 2.26 % to 2.75 % at the hot inlet
87
temperature 369 K. This is well within the band of the measurement error. The
percentage deviation between the mean experimental values (estimated as obtained
without heat loss) and the predicted values given by various correlations shows that
correlations developed by Maiti and Sarangi[45] are in better agreement with the
experimental data compared to the other correlations, the percentage deviation between
the experimental values and the predicted values given by Maiti and Sarangi [45] varying
from 6.42 % to 4.57%. The value of uncertainties varies from 4.75 % to 2.28 % for the
variation of mass flow rate from 5.77 g/s to 14.48 g/s.
The heat loss to the ambient causes an energy imbalance between the hot and
cold fluids. Two values of effectiveness are measured, the hot effectiveness, h (
effectiveness based on the hot fluid) and the cold effectiveness, c (Effectiveness based
on the cold fluid).The mean value of the hot and cold effectiveness obtained from the
experiments is estimated as the effectiveness value obtained without heat loss and is
compared with the predicted values of effectiveness obtained by different correlations
as they are without heat loss consideration. The heat loss to the ambient causes the
decrease in effectiveness based on both the fluids, the cold fluid flowing in the outer
layers suffering the most. Normally heat exchangers are placed in a vacuum insulated
cold box to completely eliminate the heat loss when operated at cryogenic temperatures.
The pressure drop increases continuously with the mass flow rate. The pressure
drop of cold fluid is below 0.05 bar up to the Reynolds number of 550 and thereafter the
pressure drop increases rapidly. A large amount of deviation is obtained between the
pressure drop obtained by experiment and the pressure drop obtained by various
correlations or the simulation software, Aspen-MUSE [113].
Manufacturing irregularities such as burred edges, separating plate roughness
and
bonding imperfections influence the heat transfer and fluid flow in heat exchanger
cores. Burred fin ends causes an effective increase in fin thickness and therefore in form
drag. Top and bottom surface roughness may cause an increase in both heat transfer
and flow friction. Irregularities may not be uniform over the entire length of the heat
exchanger. Heat transfer and pressure drop is also affected by the vortices leaving the
trailing edges of the fin segments and their interaction with the fins downstream.
Analytical modeling consists of solving the energy and momentum equations on the unit
cell of the geometry. The unit cell is an idealization of the actual geometry considered
because it neglects the possible burrs on the fin ends and also the roughness on the top
88
and bottom of the channel. Numerical solution also depends upon certain simplifying
assumptions made.
The correlations for j and f factors developed experimentally are based on the
experimental technique of Kays and London [2]. The experimental set up used by Kays
and London [2] consists of a small Ntu heat exchanger with cross flow arrangement of
fluids of steam and air (with no header losses) or a channel of heat exchanger. The
arrangement may be an ideal arrangement for the measuring the j factors but not for
friction factors. It does not give a real estimate of pressure drop although pressure drop
per unit length is used for determining the friction factor. Any small deviation in friction
factor between the predicted and the experimental value will get reflected as a large
deviation in the pressure drop as the pressure drop is a function of mass velocity,
equivalent diameter and length of heat exchanger in addition to the friction factor.
Pressure drop also depends on density which varies with temperature.
89
Chapter 7
Conlusion
Chapter VII
CONCLUSION
7.1 Concluding remarks
An experimental set up has been built in the laboratory to test the plate fin heat
exchanger. A hot fluid test is conducted to determine the thermo hydraulic performance
of the given heat exchanger at different mass flow rates (5.8 g/s to 14.5 g/s) and at
different hot inlet temperatures. The values of the effectiveness and pressure drops
obtained are compared with the values obtained by using the correlations developed by
Maiti and Sarangi [45], Manglik and Bergles [70], Joshi and Webb [68].The effectiveness
values are also compared with the values obtained by simulation software, Aspen-MUSE
[113]. The following points are noted from comparison of the experimental with the
predicted and the simulated values.
i)
The value of the effectiveness obtained by the experiments agree with the values
obtained by simulation software, Aspen-MUSE[113].The percentage deviation
between the effectiveness value obtained from Aspen-MUSE [113] (with heat
leak considered) and the experimental value varies from 2.26 % to 2.75 % at the
hot inlet temperature of 369 K. This is well within the measurement error band.
ii) The percentage deviation between the mean experimental values and the
predicted values of effectiveness given by Maiti and Sarangi[45] varies from 6.42
% to 4.57%, while for Manglik and Bergles[70],the variation is from 8.27% to
4.97 % and for Joshi and Webb[68], the variation is from 7.07 % to 5.54 %
when the inlet temperature of the hot fluid is 369 K. The experimental values
agree with the values obtained by simulation. The percentage deviation between
the mean experimental values and the simulated values is from 3.83% to 2.6%.
iii) Correlations developed by Maiti and Sarangi [45] are in better agreement with
the experimental data compared to the other correlations.
iv) The pressure drop of the fluids is below the allowable pressure drop of 0.05 bar
up to the Reynolds number of 500 and thereafter the pressure drop increases
rapidly. A large amount of deviation is observed between the experimental and
theoretical pressure drops obtained by various correlations. The variation
between the pressure drops obtained by experiments and by simulation is also
large.
The experiment suggests that the plate fin heat exchanger is ideal for low
Reynold number applications (up to the Reynold number of 500-550).Correlations
developed by Dipak Maiti [45] are in better agreement with experimental data although
with correlation developed by Joshi and Webb [68], the deviation is slightly higher and
both the correlations can be used for the design of heat exchangers. Manglik and
Bergles [70] has neglected the thickness of fins while calculating the free flow area and
correlation developed by them can also be used if the correction is made for the same.
Now as regarding the pressure drop considerations, the plate fin heat exchanger should
be used for low Reynold number applications.
All the correlations including the simulation software, Aspen MUSE [113] have
under- predicted the pressure drop. A large amount of deviation observed between the
experimental and the theoretical pressure drops calls for alternative experimental set up
for determining the friction factor. The correlations developed by using the experimental
technique of Kays and London have predicted the j factors reasonably well but have
under- estimated the pressure drop. The experimental set up used by Kays and London
consisting of a cross flow heat exchanger (with no header losses) and small Ntu seems to
be ideal for predicting the j factors but not the friction factors. In addition to the
pressure drop taking place in the narrow and intricate passages of an exchanger,
pressure drop also takes place in the headers, while flowing through elbows and
connecting piping. Manufacturing irregularities such as burred edges, separating plate
roughness and bonding imperfections influence heat transfer and flow friction in heat
exchanger cores. The pressure drop occurring through the heat exchanger should be
estimated correctly and friction factor should be determined accurately. This is because
any small deviation in friction factor between the predicted and the experimental value
will bring about a large deviation in the pressure drop as the pressure drop is a function
of mass velocity, equivalent diameter and length of heat exchanger in addition to the
friction factor.
91
Maiti and Sarangi [45] have suggested a new approach to develop heat transfer
and flow friction correlations by combining computational and experimental data. While
numerical results are used to find the effect of fin geometry on j and f, experimental
data compensate the error due to simplifying assumptions taken in numerical simulation.
This method reduces the volume of experiments to be done and can use the available
computing resources in the laboratory. The approach provides a
workable solution and
can be used for the generation of j and f factors. For finding friction factors,
experimental set up used in this thesis can be used to generate the correlation for f
factors. The experimental arrangement used in the thesis gives a real estimate of the
pressure drop occurring in a practical heat exchanger set up. The hot test method is
simple compared to testing of plate fin heat exchanger at cryogenic temperature. The
heat exchanger used in this test rig will be applied to cryogenic temperature. Since the
heat exchanger performance is satisfactory at hot test, it can be presumed that
satisfactory result is expected at cryogenic temperature. It is because the dimensionless
parameters are same in both hot and cold tests.
7.2 Scope for future work

Various correlations are available in literature for predicting the non-dimensional
heat transfer and pressure drop characteristics of plate fin heat exchanger. The thesis
presents an experimental set up for finding the thermo hydraulic performance of a
specific plate fin heat exchanger and to check the validity of these correlations. Following
are some of the activities that are proposed to be taken in our laboratory in the near
feature:
All the correlations developed so far have predicted the j factors reasonably well
but have under predicted the friction factors considerably. The experimental set
up used in this thesis can be used for developing the correlations for friction
factors. The experimental set up used is a practical heat exchanger set up with
header and piping connections and gives a real estimate of the pressure drop.
Most of the available correlations for j and f factors have been developed by
using the experimental technique of Kays and London [2] for fluids at or above
room temperatures. Cold fluid test has to be conducted using fluids at cryogenic
temperatures to check the validity of these correlations at
cryogenic
temperatures.
It is found from the hot fluid tests that even with sufficient insulation, heat loss
could not be eliminated completely. Plate fin heat exchanger presents a large
92
surface area through which the heat gets dissipated to the surroundings. The
heat loss from the plate fin heat exchanger has to be obtained only
experimentally. Experiments have to be conducted on plate fin heat exchangers
of different geometries and heat loss has to be determined at different
temperature levels and at different mass flow rates. Heat loss to the
surroundings has to be taken into account while calculating the effectiveness of
heat exchanger using different correlations.
93
References
References
1.
Barron, R.F. Cryogenic systems, Oxford University Press (1985)
2.
Kays, W.M. and London, A.L. Compact Heat exchangers, McGraw-Hill, New
York (1984)
3.
Kern, D.Q., Kraus, A.D. Extended Surface Heat Transfer, McGraw-Hill, New York
(1972)
4.
Frass,A.P., Heat Exchanger Design, 2nd ed., Wiley-Inter science (1989)
5.
Ozisik, M.N., Heat Transfer-A Basic Approach, McGraw-Hill (1985)
6.
Incropera, F.P., Dewitt, D.P. Fundamentals of Heat Transfer 2nd ed. John
Wiley, New York, 1985
7.
Shah, R.K., Dusan,P., Sekulic,D.P. Fundamentals of Heat Exchanger Design,

John Wiley & Sons, (2003)
8.
Shah, R.K. Compact Heat Exchangers & Enhancement Technology, Begell House
Publication, (1999)
9.
Kakac,S., Bergles, A.E., Mayinger, F. Heat Exchangers, Thermal-Hydraulic

Fundamentals and Design, by Hemisphere Publication (1980)
10.
Shah, R.K., Subbarao, E.C. Mashelkar, R.A. Heat Transfer Equipment Design
edited by R.K.Shah, Publishing Corporation, Washington, DC, (1988)
11.
Taborek, J., Hewitt, G.F. and Afgtan,N Heat Exchangers-Theory and

Practice(Eds.) Hemisphere Publishing, New York (1983)
12.
Taylor, M.A. Plate Fin Heat Exchangers: Guide to their Specification and Use
HTFS, Oxon, UK (1987)
13.
Sunden, B., Faghiri, M. Computer Simulations in Compact Heat Exchangers 1st

ed. by Computational Mechanics Publications (1998)
14.
Applied Thermal Engineering, Editor-In-Chief D.A.

Reay(http://www.elsvier.com)
15.
Experimental Thermal and Fluid Science, Editor-In-Chief I. Kennedy and L.

Kennedy (http://www.elsevier.com)
16.
Journal of Heat Transfer, Editor-In-Chief V.K. Dhir (http://www.asme.org)
17.
International Journal of Heat and Mass transfer, Editor-In-Chief J. Harnet

and W. Minkowycz (http://www.elsevier.com)
18.
Heat Transfer Engineering. Editor-In-Chief Afshin J. Ghajar

(http://www.tandf.co.uk./journals/titles/01457632.asp)
19.
International Journal of Heat Exchangers, Editor-In-Chief Bengt Sunden

(http://www.edwardspub.com/journals/IJHEX)
20.
London, A.L. A Brief History of Compact Heat Exchanger Technology, in Compact

Heat Exchangers, History, Technological Advancement and Mechanical Design
Problems edited by R.K. Shah et al (1980) HTD-Vol.10 1-4
21.
Mori, Y. Nakayama, W. Recent Advances in Compact Heat Exchangers in

Japan, History, Technological Advancement and Mechanical Design Problems
edited by R.K. Shah et al (1980) HTD-Vol.10 5-16
22.
Cowel, T., Achaichia,N. Compact Heat Exchangers in the Automobile Industry

Proceedings of the International Conference on Compact Heat Exchangers in the
Process Industries (1997) 11-28
23.
Lenfestey, A.G. Low Temperature Heat Exchangers, Progress in Cryogenics

Heywood & Co. (1961) 25-47
95
24.
Butt, A.G. Mechanical Design of Cryogenic heat exchangers Compact Heat

Exchangers-history, Technical Advancement and Mechanical Design Problems(Eds.
Shah,R.K. et al) (1980)HTD-Vol10 161-190
25.
Panitsidis, H., Gresham, R.D. and Westwater, J.W. Boiling of liquids in a

compact plate fin heat exchangers, International
Journal of
Heat and Mass
Transfer, 18, 37-42, (1975)

26.
Robertson, J.M., Boiling heat transfer with liquid nitrogen in brazed-aluminum

plate-fin heart exchangers, American Institute of Chemical Engineers Symposium
Series, san diego, 75, 151-164 (1979)
27.
ALPEMA, The Standards for the Brazed Aluminium Plate Fin Heat Exchanger
Manufactures Association (http://www.alpema.org/)
28.
www.chart-ind.com
29.
www.kobelco.co.jp/eneka
30.
www.linde.com
31.
www.nordon-cryogenie.com
32.
www.spp.co.jp
33.
www.htfs.comj
34.
www.htri.com
35.
Kohlweiler,W.
http://www.brazetec.de/brazetec/content_en/articles/brazing_aluminium.pdf
36.
Giannettoni,R., Tirloni,A., Tonini,G. A technique for vacuum aluminum

brazing: development of a high vacuum furnace for aluminum brazing
http://www.tav-alto-vuoto.it/brazin.htm (2000)
96
37.
Kays, W.M. and London, A.L. Description of Test equipment and method of
Analysis for basic Heat transfer and flow friction test of high rating heat
exchanger surfaces Technical Report No. 2, Department of
Mechanical
Engineering, Stanford University (1948)

38.
London, A.L., and Shah,R.K. Offset Rectangular Plate fin surfaces-Heat

Transfer and Flow Friction Characteristics, Transactions of the ASME,Journal of
Engineering for Power (1968) 90 218-228.
39.
Shah, R.K. Assessment of Modified Wilson Plot Techniques for obtaining Heat
Exchanger Design Data Proceedings of 9th International Conference (1990) 5 5156
40.
Davenport, C.J. Heat Transfer and Flow Friction Characteristics of Louvered

Heat Exchanger, Heat Exchangers-Theory and Practice edited by J. Taborek,
G.F.Hewitt, N.Afgan, McGraw Hill, New York (1983) 387-412.
41.
Sunden, B., and Svantesson, J. Heat Transfer and Pressure Drop from
Louvered
Surfaces in Automotive Heat Exchangers, Experimental Heat Transfer
(1991) 4 111- 125.

42.
Wang, J., Hirs, G.G., Rollmann, P. The Performance of a New Gas to gas Heat
Exchanger with Strip Fin, Energy Conservation and Management (1999) 40 17431751.
43.
Lozza, G., Merlo, U. An Experimental Investigation of Heat Transfer and Friction

Losses of Interrupted and Wavy Fins for Fin and Tube Heat exchangers,
International Journal of Refrigeration (2001) 24 409-416.
44.
Indranil Ghosh, Experimental and Computational Studies on Plate Fin Heat

Exchanger. PhD Dissertation, Indian Institute of Technology, Kharagpur (2004)
45.
Maiti, D.K.,and Sarangi.S.K. Heat Transfer and Flow Friction Characteristics of

Plate Fin Heat Exchanger Surfaces- A Numerical Study PhD Dissertation, Indian
Institute of Technology, Kharagpur (2002)
97
46.
Freund,S., Kabelac,S. Investigation of local heat transfer coefficients in plate

heat exchangers with temperature oscillation IR thermography and CFD,
International Journal of Heat and Mass Transfer, 2010; 53: 3764-3781.
47.
Pucci, P.F., Howard, C. P.
And Piersall, C.H. The single blow Transient
Testing Technique for Compact Heat Exchanger Surfaces, Transactions
ASME
Journal of Engineering for Power 89(A) 29-40 (1967)

48.
Mondt, J.R. and Siegla, D.C. Performance of Perforated Heat Exchanger

Surfaces, Transactions
ASME Journal of
Engineering for Power 96(A) 81-86
(1974)
49.
Shah, R. K Compact Heat Exchangers in S. Kakak, A.E. Bergles and F.Maylingar

(Eds)
Heat
Exchangers
Thermal
Hydraulic
Fundamentals
and
Design,
Hemisphere Publishing Corp Washington DC, 111-151 (1981)

50.
Patankar, S.V. Numerical prediction of flow and heat transfer in compact heat
exchanger passages, in Compact Heat Exchangers edited by R.K.Shah, A.D.Kraus
and D.Metzger, Hemisphere Publishing Corp New York (1990) 191-204.
51.
Bilir, L., B. Ozerdem, A. Erek and Z. Ilken. Heat Transfer and Pressure Drop
Characteristics of Fin-Tube Heat Exchangers with Different Types of Vortex
Generator Configurations, Journal of Enhanced Heat Transfer, 17(3), 243-256
(2010)
52.
Jacobi, A.M., and Shah, R.K. Air Side Flow and Heat transfer in Compact Heat
Exchangers:
Discussion
of
Enhancement
Mechanisms,
Heat
Transfer
Engineering (1998) 19 (4), 29-41.

53.
Shah, R.K., Heikal, M.R., Thonon,B.,Touchon,P.Progress in numerical

analysis of compact heat exchanger surfaces, Advances in Heat Transfer (2001)
34 363-442.
98
54.
Patankar, S.V., Liu, C.H., and Sparrow, E.M. Fully Developed Flow and Heat
Transfer in Ducts Having Stream wise-Periodic Variations of Cross-Sectional Area,
Transactions of the ASME Journal of Heat Transfer (1977) 99 180-186.
55.
Patankar, S.V., and Prakash, C. An Analysis of the Effect of Plate Thickness on

Laminar Flow and Heat Transfer in Interruptedplate passages, International
Journal of Heat and Mass Transfer, (1981) 24(11)1801-1810.
56.
Suzuki, K., Hirai, E., Miyake, T. Numerical and Experimental Studies on a Twodimensional Model of an Offset-strip-fin Type Compact Heat Exchanger Used at
Low Reynolds Number, International Journal of Heat and Mass Transfer, (1985)
28(4)823-836
57.
Zhang, L.W., Tafti, D.K., Najjar, F.M., and Balachandar, S. Computations of

Flow and
Heat Transfer in Parallel-Plate Fin Heat Exchangers on the CM-5:
Effects of Flow Unsteadiness and Three-Dimensionality, International Journal of

Heat and Mass Transfer (1997) 40 (6) 1325-1341
58.
Achaichia, A. and Heikal, M. R., Sulaiman, Y. and Cowell, T.A. Numerical

Investigation of Flow and Friction in Louvered Fin Arrays, in Proc. 10th
International Heat Transfer Conference, Brighton (1994)
59.
Ha, M.Y., Kim, K.C., Koak,S.H., Kim,K.H., Kim,K.I., Kang,J.K.and Park,T.Y.

Fluid Flow and Heat Transfer Characteristics in Multi-louvered Fin Heat
Exchangers, SAE Paper No 950115 (1995)
60.
Atkinsona, K.N., Drakulic, R., Heikal, M.R. and Cowell, T.A. Two and Three
Dimensional Numerical Models of Flow and Heat Transfer over Louvered Fin
Arrays in Compact Heat Exchangers, International Journal of Heat and Mass
Transfer, 41 4063-4080 (1998).
61.
Achaichia, A. and Cowell, T.A. Heat transfer and pressure drop characteristics
of Flat tube and louvered plate fin surfaces, Experimental Thermal and Fluid
Science 1 147-157 (1988).
99
62.
Springer, M.E. and Thole, K.A. Experimental design for flow field studies of
louvered fins, Experimental Thermal and Fluid Science 18 258-269 (1998)
63.
Tafti, D.K., Zhang, L.W. and Wang,G. Time-Dependent Calculation Procedure

for Fully developed and developing Flow and Heat Transfer in Louvered Fin
Geometries, Numerical Heat Transfer Part A 35 225-249 (1999)
64.
Tafti, D.K., Wang,G. and Lin, W. Flow Trasition in a Multilouvered Fin Array,
International Journal of Heat and Mass Transfer, 43 901-919 (2000)
65.
Manson, S. V., Correlation of Heat Transfer Data and of friction Data for
Interrupted Plane fins Staggered in Successive rows, NACA Tech. Note 2237,
National Advisory Committee for Aeronautics, Washington, DC, December 1950.
66.
Kays, W.M., Compact Heat Exchangers, AGARD lecture Serr. No 57 on Heat

exchangers, AGARD-LS-57-72, NATO, Paris, 1972.
67.
Wieting, A. R. Empirical Correlations for Heat Transfer and Flow Friction

Characteristics of Rectangular Offset-fin Plate-fin Heat Exchangers, ASME Journal
of Heat Transfer (1975) 97 488-490.
68.
Joshi, H.M. and Webb, R.L. Heat Transfer and Friction in the Offset Strip-fin
Heat Exchanger, International Journal of Heat and Mass Transfer, (1987) 30 (1)
69-84
69.
Sparrow, L.U. and Liu, C.H. Heat Transfer, Pressure Drop and Performance
relations for
in-line, Staggered
and Continuous Plate Heat Exchangers,
International Journal of Heat and Mass Transfer, (1979) 22 1613-1625

70.
Manglik, R. M., and Bergles, A.E. Heat Transfer and Pressure Drop
Correlations For the Rectangular Offset strip Fin Compact Heat Exchanger,
Experimental Thermal and Fluid Science (1995) 10 171-180
100
71.
Mochizuki, S., Yagi, Y., and Yang, W.J., Transport phenomena in stacks of
Interrupted parallel plate surfaces, Experimental Heat Transfer 1, 127-140, 1987
72.
Muzychka, Y.S. and Yovanovich, M.M. Modeling the f and j Characteristics for
transverse flow through an offset strip fin at low Reynolds number, Journal of
Enhanced Heat Transfer (2001) 8 261-277
73.
Mochizuki, S., Yagi, Y., and Yang, W.J., Flow pattern and turbulence intensity
in stacks of interrupted parallel plate surfaces, Experimental Thermal and Fluid
Science -1, 51-57, 1988
74.
Mochizuki, S., Yagi, Y., Characteristics of vortex shedding in plate arrays , in

Flow visualization II, Merzkirch, Ed, pp- 99-103, Hemisphere, Washington, D.C,
1982
75.
Dubrovsky, E.V., and Vasileiev,V.Y, Enhancement of convective heat transfer

in reactangular ducts of interrupted surfaces, International Journal of Heat and
Mass Transfer, 31 807-818 (1988)
76.
Mullisen, R.S., and Loehrke, R.I., Study of the Flow Mechanisms Responsible
for Heat Transfer Enhancement in Interrupted Plate Heat Exchangers,
Transactions of the ASME Journal of Heat Transfer 108,377-385, 1986
77.
Barron, R.F. Cryogenic Heat Transfer, Taylor &Francis (1999) 311-318
78.
Shah, R.K. A Review of Longitudinal Wall Heat Conduction in Regenerators,

Journal of Energy, Heat and Mass Transfer (1994) 16 15-25
79.
Barron, R.L. and Yeh, S.L. Longitudinal Conduction in a Three-Fluid Heat

Exchanger, (1976) ASME Paper, 76-WA, HT-9 2-7
80.
Chowdhury, K., Sarangi, S. Effect of finite thermal conductivity of the

separating wall on the performance of counter flow heat exchangers, Cryogenics
(1983) 23 212-216
101
81.
Hausen .H, Heat transfer in counter flow, parallel flow and cross flow, New York:
Mc Graw Hill, 1983.
82.
Hahnemann. H.W, Approximate calculation of thermal ratios in heat exchangers

including heat conduction in direction of flow; National Gas Turbine Establishment
Memorandum, 1948:36.
83.
Bahnke. G.D., Howard. C.P., The effect of longitudinal heat conduction on

periodic flow heat exchanger performance, ASME Journal of Engineering for
Power 1964: 86: 105-20.
84.
Kroeger,P.G. Performance deterioration in high effectiveness heat
exchanger
due to Axial Conduction Effect, Advances in Cryogenic Engineering, Vol.12, pp.

363-372, 8 (1966)
85.
Chiou, J. P. The Effect of Longitudinal Heat Conduction on Cross flow Heat

Exchanger, Transactions ASME, Journal of Heat Transfer (1978) 100 436-441
86.
Chiou, J. P. The Advancement of Compact Heat Exchanger Theory considering

the Effects of Longitudinal Heat Conduction and Flow Non-uniformity in Compact
Heat Exchangers, History, Technological Advancement and Mechanical Design
Problems edited by R.K. Shah et al) (1980) HTD-Vol.10 101-121
87.
Venkatarathnam, G., Narayanan, S.P. Performance of a counter flow heat

exchanger with longitudinal heat conduction through the wall separating the fluid
streams from the environment, Cryogenics (1999) 39 811-819.
88.
Mondt, J.R. Correlating the Effects of Longitudinal Heat Conduction on Heat

Exchanger performance in Compact Heat Exchangers, History, Technological)
Advancement and Mechanical Design Problems edited by R.K. Shah et al) (1980)
HTD-Vol.10 123-134
89.
Shah, R.K. A Correlation for Longitudinal Heat Conduction Effects in Periodic flow
Heat Exchangers, Transactions ASME Journal of Engineering for Power (1975)
97 453- 454
102
90.
Narayanan. S. P, Venkatrathnam. G
Performance degradation due to
longitudinal heat conduction in very high Ntu counter flow heat exchangers,
Cryogenics 1998:
91.
38; 927-30
B.M.Wood and J.Kern, Design of Heat exchangers with heat loss to the
Surroundings,paper No 17, Second National Meeting of the South African
Institute of Chemical Engineers(S-123), Johnnesburg (1976).
92.
Chowdhury, K., Sarangi, S. Performance of Cryogenic Heat Exchangers with

Heat leak from the Surroundings in Advances in Cryogenics Engineering (1984)
29 273-280
93.
Barron, R.F. Effects of Heat Transfer from Ambient on Cryogenic Heat

Exchangers Performance in Advances in Cryogenics Engineering (1984) 29 265272
94.
Gupta, P., Atrey, M. D. Performance evaluation of counter flow heat exchangers

considering the effect of heat in leak and longitudinal conduction for lowtemperature applications, Cryogenics (2000) 40 (7) 469-474
95.
Kitto, J.B, Robertson, J.M. Effects of Maldistribution of Flow on Heat Transfer

Equipment Performance, Heat Transfer Engineering 1989; 10(1): 18-25.
96.
Mueller A. C, Chiou J. P Review of various Types of Flow Maldistribution in Heat

Exchangers, Heat Transfer Engineering 1988; 9(2); 36-50.
97.
Ranganayakulu Ch, Seetharamu. K.N. The combined effects of wall

longitudinal heat conduction, inlet fluid flow nonuniformity and temperature
nonuniformity in compact tube fin heat exchanger: A finite element method,
International Journal of Heat and Mass Transfer, 1999; 42: 263.73.
98.
Jiao AJ, Li YZ, Chen CZ, Zhang R. Experimental investigation on fluid flow
maldistribution in plate
fin heat exchangers, Heat Transfer Engineering 2003;
24(4): 25-31.
103
99.
Zhang Z, Li YZ. CFD simulation on inlet configuration of plate-fin heat

exchanger, Cryogenics (2003) 43: (12) 673-678.
100. Wen J, Li Y. Study of flow distribution and its improvement on the header of
plate fin heat exchanger, Cryogenics (2004) 44: 8 23-31.
101. Fleming, R. B. The effects of flow distribution in parallel channels of counter
flow heat exchangers, Advanced Cryogenic Engineering 1967 (December): 352.
102. Jian Wen, Yanzhong Li, Aimin Zhou, Ke Zhang.Jiang Wang. PIV
experimental investigation of entrance configuration on flow maldistribution in
plate fin heat exchanger, Cryogenics (2006) 46 37-48.
103. Chowdhury, K., Sarangi, S. The Effect of Variable specific Heat of the working
Fluid on the Performance of counter flow Heat Exchangers, Cryogenics (1984) 24
679-680
104. Shah, R.K. Non Uniform Heat Transfer Coefficients for Heat Exchanger Thermal
Design in Aerospace Heat Exchanger Technology edited by R. K.Shah
and A.
Hashem (1993) 417-445.
105. Paffenbarger, J. General Computer Analysis of Multistream Plate Fin Heat

Exchangers Compact Heat Exchangers A Festschrift for A.L. London, (Eds. Shah,
R.K., Kraus, A.D, and Metzger, D) Hemisphere Publishing New York (1990) 727746
106. Taylor, R.P., Hodge, B.K., James, C.A. Estimating uncertainty in thermal
systems analysis and design, Applied Thermal Engineering, (1999) 19 51-73
107. Moffat, R.J. Describe the Uncertainties in Experimental Results, Experimental
Thermal and Fluid Science (1988) 1 (1)3-17
108. Evans, D.M. The Role of Probable Uncertainty in the Design and Testing of
Compact Heat Exchanger in Compact Heat Exchangers, History, Technological
104
Advancement and Mechanical Design Problems edited by R.K. Shah et al) (1980)
HTD-Vol.10 135-143
109. Lestina, T. Scott, B. Assessing the Uncertainty of Thermal Performance
Measurements of Industrial Heat Exchangers Proceedings of the International
Conference on Compact Heat Exchangers in the Process Industries (1997) 401416.
110. Clarke, D.D., Vasquez, V.R., Whiting, W.B., Greiner, M. Sensitivity and
uncertainty analysis of heat exchanger designs to physical properties estimation,
Applied Thermal Engineering 2001 (21) 993-1017
111. Brown, K.K., Coleman, H.W., Steele W.G., A Methodology for Determining
Experimental Uncertainties in Regressions, Proceedings of the ASME Fluids
Engineering Division, ASME (1996) FED-Vol. 242 263-273
112. Walters, F.M., Hypersonic Research Engine Project phase II A, category I Test
report on fin Heat transfer and pressure drop testing, Data item No. 63.02, Ai
Research Manufacturing co., Doc, AP-69-5348, 1969
113. Aspen
Muse
Reference
guide,
Cambridge,
Massachusetts:
Aspen
Technology, Inc. V2006

114. Fribance, Austine. Industrial Instrumentation Fundamentals, McGraw-Hill
(1962)
105
Curriculum Vitae
SIDRAMAPPA ALUR
E-mail:
[email protected]
Permenant Address:
At Post Mannur
Tq Afzalpur.
Dist Gulbarga.
Pin Code 585246
Karnataka, India
Education:
2012
Dissertation submitted
1995
M.E, Gulbarga University, Gulbarga.
1989
B.E, Gulbarga University, Gulbarga.
Personal Information:
Date of Birth
29-09-1966
Nationality
Indian
Employment:
1990 1992
Lecturer, M.C.E, Nilanga, Latur, Maharashtra.
1995 1997
Lecturer, R.E.C, Bhalki, Bidar, Karnataka.
1997 Till Date
Lecturer, Senor lecturer and Asst Professor, H.I.T, Nidasoshi,

Belgaum, Karnataka.
Publications:
1. S.A.Alur., Balaji Kr Choudhury., R.K.Sahoo., Sunil Kr. Sarangi., 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.
2. S.A.Alur., Balaji Kr Choudhury., R.K.Sahoo., Sunil Kr. Sarangi., 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.
3. Balaji Kr Choudhury., S.A.Alur., R.K.Sahoo., Sunil Kr. Sarangi. Optimization
Analysis of Liquefaction Cycles for Nitrogen., Indian Journal of Cryogenics, 37 (1-4), pp
34, 2012.
4. S.A.Alur., Balaji Kr Choudhury., R.K.Sahoo., Sunil Kr. Sarangi., Performance
Analysis of plate fin heat exchanger at low Reynolds number. 23rd National Symposium on
cryogenics (NSC-23). October 26-30, 2010, National Institute of technology, Rourkela769008, Orissa, India.
5. S.A.Alur., Balaji Kr Choudhury., R.K.Sahoo., Sunil Kr. Sarangi., Performance

Analysis of plate fin heat exchanger. Asian conference on Applied Super conductivity and
cryogenics, November 16-18, 2011, I.U.A.C, .New Delhi.
Papers Communicated:
1. S.A.Alur., Balaji Kr Choudhury., R.K.Sahoo., Sunil Kr. Sarangi.,Thermo hydraulic
Performance of plate fin heat exchanger using hot fluid test at and above room
temperature, Experimental thermal and Fluid Sciences, International journal of
experimental heat transfer, Thermodynamics and Fluid Mechanics.

Alur Final Thesis

Uploaded by

Copyright:

Available Formats

Alur Final Thesis

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Alur Final Thesis

Uploaded by

Copyright:

Available Formats

Experimental Studies on Plate Fin

Mechanical Engineering Department

The Memory of My Parents

NATIONAL INSTITUTE OF TECHNOLOGY

Mechanical Engg. Department

Date: January 6, 2012

Prof Sarangi has been unified perfectly with knowledge that

creates a permanent impression in my mind. It is my proud privilege to work under not

I feel lucky to have Mr. Balaji Kumar Choudhury as my co-research fellow.

for comparison with

the experimental work carried out in this thesis.

Values of heat exchanger effectiveness and pressure drop obtained from

Plate fin heat exchanger

Plate fin heat transfer surfaces

Heat transfer and flow friction characteristics

Objectives of the Present Investigation

Organization of the Thesis

Plate fin heat exchanger

Analytical and Numerical studies

Heat transfer and flow friction characteristics

Design of Plate Fin Heat Exchanger

Summary of Design procedure

Design inputs and specification of fin geometry

Correlation-based Design of Heat exchanger

Design of Heat exchanger using Simulation software

Concluding dimensions of The Heat exchanger

Rating of Plate fin heat exchanger

Details of Given heat exchanger and input data

Rating of given heat exchanger using different correlations.

Rating of given heat exchanger using simulation software

Effect of heat transfer to the ambient

The Experimental Apparatus

Experimental set up and operation

Effect of heat transfer from the ambient

Description of various equipment and instruments

Variation of effectiveness with mass flow rate.

Effect of heat transfer to the ambient

Comparison of effectiveness obtained with and without heat loss

Error estimation in experimental results

Variation of pressure drop of cold fluid with mass flow rate

Variation of pressure drop of hot fluid with mass flow rate

Results and Discussion

Scope for Future Work

Plate fin heat exchanger assembly and details

Cross flow arrangement

Counter flow arrangement

Cross counter flow arrangement

Some of the common fin geometries

Schematic P&I diagram of the Experimental Test Rig

Photograph of the experimental set up

Photograph of the experimental set up (with insulation)

Plate fin heat exchanger.

Photograph of the set up used for calibration

R.T.D calibration graph

Variation of effectiveness with mass flow rate (hot inlet

Variation of effectiveness with mass flow rate (hot inlet

Variation of effectiveness with mass flow rate (hot inlet

Variation of effectiveness with mass flow rate (hot inlet

Comparison of effectiveness obtained by experiment and by simulation