CHAPTER 11

Heat Exchangers:
Design Considerations

Prepared by: Prof. Dr. Ş. Birgül Tantekin-Ersolmaz

Prof. Dr. Hüsnü Atakül
 Introduction
In this chapter our objectives are:

 to introduce performance parameters for assessing the

efficacy of a heat exchanger

 to develop methodologies for

• designing a heat exchanger or
• for predicting the performance of an existing
exchanger operating under prescribed conditions.

2
 Heat Exchangers
 Heat exchangers are devices that facilitate the exchange of heat
between two fluids that are at different temperatures while keeping
them from mixing with each other.
Heat transfer mechanisms: • convection in each fluid
• conduction through the wall
Examples: heating systems, air conditioning systems, chemical
processing, power production
 Heat exchangers are ubiquitous in energy conversion and utilization.
They encompass a wide range of flow configurations. Different heat
transfer applications require different types of hardware and different
configurations of heat transfer equipment.
 Heat exchangers are typically classified according to flow arrangement
and type of construction. The most common types are:
• Double-pipe heat exchangers • Cross-flow heat exchangers
• Shell-and-tube heat exchangers • Compact heat exchangers
• Plate and frame (or just plate) heat exchangers
3
 Heat Exchanger Types
 Concentric-Tube (Double-Pipe) Heat Exchangers
 Simplest configuration.
 Superior performance associated with counter flow.
hot and cold enter at the same end hot and cold enter at opposite ends

Parallel Flow Counterflow

4
 Cross-flow Heat Exchangers
 Heat exchanger performance is influenced by mixing.

Finned-Both Fluids Unmixed Unfinned-One Fluid Mixed the

Other Unmixed

 For cross-flow over the tubes, fluid motion, and hence mixing, in
the transverse direction (y) is prevented for the finned tubes,
but occurs for the unfinned condition.
5
  Shell-and-Tube Heat Exchangers
One Shell Pass & One Tube Pass
The most common type
of heat exchanger in
industrial applications

 Baffles are used to establish a cross-flow and to induce turbulent

mixing of the shell-side fluid, both of which enhance convection.
 The number of tube and shell passes may be varied, e.g.:

All the tubes make a U-turn

One Shell Pass, Two Tube Passes Two Shell Passes, Four Tube Passes 6
  Compact Heat Exchangers Also operate in cross-flow mode
 Widely used to achieve large heat rates per unit volume, particularly when
one or both fluids is a gas.
 Characterized by large heat transfer surface areas per unit volume, small
flow passages, and laminar flow.
(a) Fin-tube (flat tubes, (b) Fin-tube (circular tubes, (c) Fin-tube
Area Density continuous plate fins) continuous plate fins) (circular tubes,
circular fins)
heat transfer surface area

heat exchanger volume

If   400 m2/m3 for liquids and

 700 m2/m3 for gases
(d) Plate-fin (single pass) (e) Plate-fin (multipass)
 compact heat exchanger

Examples:
 car radiators (  1000 m2/m3)
 glass ceramic gas turbine h. e.
(  6000 m2/m3)
 human lung (  20 000 m2/m3) 7
 Overall Heat Transfer Coefficient
An essential, and often the most uncertain, part of any heat exchanger
analysis is determination of the overall heat transfer coefficient.
Recall: In Chp 3 it is defined in terms of the total thermal resistance to
heat transfer between two fluids.
A heat exchanger typically involves two flowing
fluids separated by a solid wall.
Heat transfer mechanisms:
• From hot fluid to wall:
by convection
• Through the wall:
by conduction
• From the wall to cold
fluid: by convection

Thermal resistance network involves two convection and one conduction

resistances. 8
 Overall Heat Transfer Coefficient
Subscripts c and h used to designate
the cold and hot fluids, respectively.
R  Rtotal  Rh  Rwall  Rc

1 ln( Do / Di ) 1
Rh  Rwall  Rc 
hh Ah 2kL hc Ac

1 ln( Do / Di ) 1 Ah = DiL
R  
hh Ah 2kL hc Ac Ac = DoL

In the analysis of heat exchangers, it is

convenient to combine all thermal
Th resistances between the two fluids:
T
q  UAT  U h Ah T  U c Ac T
Tc R
U = the overall heat transfer coefficient (W/m2.oC)

1 1 1 1 1
  R  Rwall 
UA U h Ah U c Ac hh Ah hc Ac 9
 Overall Heat Transfer Coefficient
1 1 1 1 1  i=inside 
In general:   R  Rwall   
UA U i Ai U o Ao hi Ai ho Ao  o=outside 

• Ai and Ao are not equal to each other especially when one side of the
tube wall is finned and the other side is not.
• Note that: U i Ai  U o Ao , but U i  U o , unless Ai  Ao
Therefore, U must be specified based on A

Special case: If wall thickness is small and ktube is large

Rwall  0, 1 1 1
   U  Ui  Uo
Ai  Ao  As U hi ho
If one of the convection coefficients U is dominated by the smaller
is much smaller than the other  convection coefficient
Example: gas to liquid heat transfer
hi  ho  1/hi  1/ho  U  hi That is why fins are commonly used
on the gas side 10
 Fouling Factor
 During operation, surfaces are often subject to fouling by fluid
impurities, rust formation, or other reactions between the fluid
and the wall material.
Result: deposition of a film or scale on the surface

Accumulation of deposits on heat transfer surfaces


Additional resistance to heat transfer

Rate of heat transfer decrease

The net effect of these accumulations on heat
transfer is represented by a fouling factor

Fouling factor (Rf) = a measure of the thermal resistance from fouling

11
 Fouling Factor
Fouling factor (Rf) = a measure of the thermal resistance from fouling

For unclean (fouled) surfaces: Fouling resistances

1 1 1 1 R ''f ,h ln  Do / Di  R ''f ,c 1
  R    
UA U h Ah U c Ac hh Ah Ah 2 kL Ac hc Ac

Its value depends on:

• Temperature
• Fluid velocity
• Service time

Rf  Fouling factor for a unit surface area (m2  K/W)  Table 11.1
12
 Finned Surfaces
 When fins are added to the surfaces exposed to either or both
fluids to increase the heat transfer area, they reduce the overall
resistance to heat transfer.
With inclusion of surface fouling and fin (extended surface) effects, the
overall heat transfer coefficient is modified as follows:

1 1 R f ,h ln  Do / Di  R f ,c 1
R    
UA o hh Ah o Ah 2 kL c Ac o hc Ac
o = the overall surface efficiency or temperature effectiveness of a
finned surface (Section 3.6.5)
A = At = the total surface area
 Af  (fins + exposed base)
o , c or h  1  1  f   Af = the surface area of fins only
 A  c or h
f = the efficiency of a single fin

For an adiabatic tip,  f  

 tanh  mL  

the fin efficiency is:  mL  13
 Finned Surfaces
 tanh  mL   No fouling: mh or c   2h kL h or c
 f , i or o  
 mL i or o With fouling: mh or c   2U p kL h or c
 h 
U p , i or o   Partial overall heat transfer coefficient
 1  hR"
 f i or o

Hot fluid: U p , h  hh 1  hh R"f ,h 

Cold fluid U p , c  hc 1  hc R"f ,c 

The overall heat transfer coefficient equation is simpified, if the
convection coefficients in the first and fifth terms are replaced by Up,c
and Up,h, respectively:
1 1 ln  Do / Di  1
R  
UA oU p ,h Ah 2 kL oU p ,c Ac
14
 Overall Heat Transfer Coefficient
Summary

 An essential requirement for heat exchanger design or

performance calculations (often the most uncertain).
 Contributing factors include convection and conduction
associated with the two fluids and the intermediate solid, as
well as the potential use of fins on both sides and the effects
of time-dependent surface fouling.
 With subscripts i and o used to designate the inside and
outside fluids, respectively, the most general expression for
the overall coefficient is:

1 1 1 1 R ''f ,h R ''f ,c 1
     Rw 
UA (UA) h (UA)c 0 hA h 0 A h 0 A c 0 hAc
15
 Analysis of Heat Exchangers
Engineers are often required to:
 select a heat exchanger that will achieve a specified temperature
change in a fluid stream of known mass flow-rate or
 predict the outlet temperatures of the hot and cold streams in a
specified heat exchanger.
There are two methods to analyze heat exchangers:
1. Log mean temperature difference method (LMTD): Best suited for the
selection of heat exchangers
2. Effectiveness-NTU method: Best suited for the prediction of outlet
temperatures
Assumptions that apply to both methods:
• steady flow
• mass flow rate of each fluid is constant
• fluid properties (T, u) at inlet and outlet remain same
• Cp of each fluid is constant (independent of T)
• kinetic and potential energy changes are negligible
• axial heat conduction along the tube is negligible
• outer surface of the heat exchanger is perfectly insulated 16
A Methodology for Heat Exchanger Design Calculations
The Log Mean Temperature Difference (LMTD) Method
To design and to predict performance of a heat exchanger, it is essential to
relate the total heat transfer rate to inlet and outlet fluid temperatures,
overall heat transfer coefficient, and total surface area for heat transfer.
 Application of overall energy balance to the hot (h) and cold (c) fluids:

Since we assumed negligible heat transfer between the exchanger and its
surroundings and negligible potential and kinetic energy changes for each fluid:

q  m h  ih ,i  ih ,o  q  m c  ic ,o  ic ,i 
 
i  fluid enthalpy
17
 The LMTD Method  Overall Energy Balance
In case of constant specific heats and no phase change:

q  m h  ih ,i  ih ,o   q  m h c p ,h Th ,i  Th ,o   Ch Th,i  Th ,o 
 

 
q  m c  ic ,o  ic ,i   q  m c c p ,c Tc ,o  Tc ,i  Cc Tc ,o  Tc ,i 
 

Ch , Cc  Heat capacity rates Temperatures refer to the mean fluid

temperatures at the designated locations

Note that in a heat exchanger:

• The fluid with large C will experience a small temperature change
• The fluid with small C will experience a large temperature change
Another expression may be obtained by relating the total heat transfer
rate to the temperature difference between the hot and cold fluids
through Newton’s Law. Mean temperature difference

T  Th  Tc    q  UA Tm

T varies with position
18
 The LMTD Method
The Parallel-Flow Heat Exchanger
The form of Tm may be determined by applying an energy balance to
differential elements in the hot and cold fluids.

Notice that: Cp, C, Cp, h, and U are

average values for the heat exchanger

T1  Th ,1  Tc ,1
 Th ,i  Tc ,i

T2  Th ,2  Tc ,2
 Th ,o  Tc ,o

Note that Tc,o cannot exceed Th,o for a Parallel-Flow Heat Exchanger. 19
 q  UA Tm
The Tm may be determined by applying an
energy balance to differential elements in hot
and cold fluids.

heat lost by the hot fluid = heat gained by the

cold fluid

Hot fluid: dq  m
 hC phdTh  Ch dTh
Cold fluid: dq  m
 cC pc dTc  Cc dTc
Heat transfer across the surface area dA: dq  U (Th  Tc )dA
the local insert
temperature T  Th  Tc 
in differential form
 d  T   dTh  dTc  d Th  Tc 
difference
between the hot
 1 1 
and cold fluids
d  T   dq   
 Ch Cc  20
  1 1 
dq  U  T  dA
insert
d  T   U  T  dA   
 Ch Cc 
d  T   1 1   dA  2 Rdx 
   UdA   
T  Ch Cc   A  2 RL 
Integrate from the inlet to the outlet of the heat exchanger:

2
d  T   1 1 
L
 T2   1 1 
1 T    2πR U  Ch  Cc  0 dx  ln     UA  
 T1   Ch Cc 
q q
Cc  Ch  insert
Tc,o  Tc,i  Th,i  Th,o 
 T2   Th ,i  Th ,o Tc ,o  Tc ,i 
Th ,i  Th ,o   Tc ,o  Tc ,i  
UA
ln    UA     
 T1   q q  q  
21
 T2 T1
 T2   Th ,i  Th ,o Tc ,o  Tc ,i 
Th ,i  Tc ,i   Th ,o  Tc ,o  
UA
ln    UA   
 T1   q q  q  

q  UATlm

T1  T2 T2  T1 log mean

Tlm   temperature
ln  T1 T2  ln  T2 T1  difference

• Exact representation of the average T

• It makes no difference which end of the HE is designated as the inlet
and the outlet
• Tlm < Tam = (T1+T2)/2 and therefore, use of Tam will overestimate q
• If T1 differs from T2 less than 40%: using Tam  1% error
• If T1 differs from T2 more than 40%: using Tam  considerable error
22
 The LMTD Method
The Counter-Flow Heat Exchanger

T1  Th ,1  Tc ,1 T2  Th ,2  Tc ,2
 Th ,i  Tc ,o  Th ,o  Tc ,i

T1  T2
Tlm 
ln  T1 / T2 

q  UA Tlm
For equivalent values of UA
and inlet temperatures:

The outlet temp. of the cold fluid may exceed T m ,CF  T m ,PF
the outlet temp. of the hot fluid (but can never
exceed the inlet temp. of the hot fluid). 23
 Special Operating Conditions
C  mC p

• Case (a): Ch >> Cc or h is a condensing vapor  Ch    .

– Negligible or no change in Th Th ,o  Th ,i  .

• Case (b): Cc >> Ch or C is an evaporating liquid  Cc    .

– Negligible or no change in Tc Tc ,o  Tc ,i  .

• Case (c): Ch=Cc  T1  T2  T m

24
 Shell-and-Tube and Cross-Flow Heat Exchangers
Use of a Correction Factor
 The Tlm relation developed earlier is limited to double-pipe (PF & CF)
heat exchangers only
 Similar relations are developed for cross-flow and multipass shell-
and-tube heat exchangers but resulting expressions are too complex
 In such cases, it is convenient to use a correction factor (F)

Tlm  FTlm,CF Tlm,CF = log mean temperature for counter-flow

 F depends on:
 the geometry of the heat exchanger
 the inlet and outlet temperatures of the hot and cold streams
F 1 for cross-flow and shell-and-tube
F=1 means counter-flow

F  Figures 11S.1 - 11S.4

25
 F Charts for Common Heat Exchangers
t2  t1 T1  T2  mC p tube side
P R 
T1  t1 t2  t1  mC p 
shell side

T : shell side, t : tube side

26
F Charts for Common Heat Exchangers

27
 The LMTD Method
Summary

Suitable for determining the size of a heat exchanger when the

inlet and outlet temperatures of the hot and cold fluids are known
or can be determined from an energy balance.

Procedure of Problem Solution by LMTD Method

1. Select the type of heat exchanger suitable for the application

2. Determine any unknown inlet and outlet temperature
3. Calculate Tlm and F
4. Obtain (select or calculate) U
5. Calculate the heat transfer area A
28
Example 1: Condensing steam in a power plant
The condenser of a large steam power plant is a heat exchanger in which
steam is condensed to liquid water. Assume the condenser to be a shell-
and-tube heat exchanger consisting of a single shell and 30 000 tubes,
each executing two passes. The tubes are of thin wall construction with
D = 25 mm, and steam condenses on their outer surface with an
associated convection coefficient of ho = 11000 w/m2.K. The heat
transfer rate that must be effected by the exchanger is Q = 2x109 W, and
this is accomplished by passing cooling water through the tubes at a rate
of 3x104 kg/s (the flow rate per tube is therefore 1 kg/s). The tube side
convection coefficient of hi = 7552 w/m2.K. The water enters at 20oC,
while the steam condenses at 50oC.

• What is the temperature of

the cooling water emerging
from the condenser?
• What is the required tube
length L per pass ?
29
Known: Heat exchanger consisting of single shell and 30000 tubes with
two passes each

Find: Outlet temperature of the cooling water and tube length per pass
to achieve the required q

Schematic:

Assumptions: • Steady operation

• Negligible heat loss to surroundings, and negligible EK & Ep
changes
• Constant properties
• Tube internal flow and thermal conditions fully developed
• Negligible thermal resistance of tube material and fouling
effects
30
Properties: Water: Assume Tave = 27oC = 300 K, from related tables in the textbook:
C p  4197 J/kg.K   548 x10-6 N.s/m 2 k  0.613 W/m.K Pr  5.85

Analysis:
1. Outlet temperature of cooling water may obtained from overall energy balance:

q  mcC p ,c Tc ,o  Tc ,o   2 109  3 104  4197  Tc ,o  20 

Tc,o  36 o C
2. Tube length per pass:

The problem may be classified as one requiring a heat exchanger design

calculations. Accordingly, the LMTD method may be conveniently applied.

q  UFATlm,CF A  N  2L  D
1 1 1
Simplification: (1) Thin wall (Ah = Ac), (2) No fouling   
U hh hc
1 1 1
   U  4477 W/m 2  K
U 11000 7552 31
The correction factor F:

t2  t1 36  20
P   0.53 From Fig. 11S.1
T1  t1 50  20
 F 1
T1  T2 50  50
R  0
t2  t1 36  20
Th,o  Th,i  50 oC
T1  T2 14  30
Tlm    21 o C
ln  T1 T2  ln 14 / 30  Tc,o  20 oC

Tc,i  20 oC
q
L
U  N 2 D  F Tlm,CF
T1  50  36
L  4.51 m T2  50  20

32
Comments:
• Recognize that L= tube length per pass. Ltot= 9.02 m.
• Over time, the performance of the heat exchanger would be
degraded by fouling on both the inner and outer tube surfaces. When
Rfi = Rfo = 10-4 mK/W, a mantainance schedule should be applied.

1 1 1 1 R ''f ,h ln  Do / Di  R ''f ,c 1
To include fouling:   R    
UA U i Ai U o Ao hh Ah Ah 2 kL Ac hc Ac

• To maintain the requirement of Q = 2x109 W with max. allowable

fouling and restriction of mci = 1 kg/s, the tube length or the number
of tube would have to be increased.

33
Example 2: Overall Heat Transfer Coefficient
A heat recovery device involves
transferring energy from the hot flue
gases passing through an annular region
to pressurized water flowing through the
inner tube of annulus. The inner tube has
inner and outer diameters of 24 and 30
mm and its connected by eight struts to
an insulated outer tube of 60-mm
diameter. Each strut is 3 mm thick and is
integrally fabricated with the inner tube
from carbon steel (k = 50 W/m.K).

Consider conditions for which water at 300 K flows through the inner tube
at 0.161 kg/s providing a convection heat transfer coefficient of 1883
W/m2.K while flue gases at 800 K flow through the annulus, maintaining a
convection heat transfer coefficient of 100 W/m2.K on both the struts and
the outer surface of the inner tube. What is the rate of heat transfer per
unit length of tube from gas to water?
34
Known: Geometry of finned, annular heat exchanger. Gas-side
temperature and convection coefficient. Water-side convection
coefficient and temperature.
Find: Heat rate per unit length

Schematic:

Assumptions: • Steady operation

• One-dimensional conduction in strut
• Adiabatic outer surface conditions
• Negligible gas-side radiation
• Constant properties
• Fully-developed internal flow
• Negligible fouling effects
35
Properties: Water, at T = 300 K, from the related table in the textbook:
C p  4197 J/kg.K   655 x10-6 N.s/m 2 k  0.613 W/m.K Pr  5.83

Analysis: The heat transfer rate is

q  UA c Tm,h  Tm ,c  
 Tm,h  Tm ,c 
R
1 1 1 1 R ''f ,h R ''f ,c 1
     Rw 
UA (UA) h (UA)c 0 hA h 0 A h 0 A c 0 hAc
1 1 ln  Di ,2 Di ,1  1 No fin on the cold
 R  Rc  Rw  Rh    side; no fouling
UAc  hAc 2 kL 0 hAh
 1883 W/m .K    0.024 m 1 m   7.043 103 K/W
1 1
Rc  2

 hAc
ln  Di ,2 Di ,1  ln  30 / 24 
Rw    7.10 104 K/W
2 kL 2  50 W/m  K  36
  Af 
The overall fin efficiency is: o  1    1   f 
 A

Af  8  2  Lw   8  2  0.015 m 1 m   0.24 m 2

A  Af   Di ,2  8t  w  0.24 m 2    0.03 m  8  0.003 m   0.31 m 2

 tanh  mL  
 f ,h  
 mL h
m   2h kt    2 100 W/m 2  K 50 W/m.K  0.003 m  
1/ 2
 36.5 m -1
1/ 2

mL  36.5 m-1  0.015 m  0.55

tanh  0.55  0.499
 f ,h    0.911
0.55 0.55

o  1   0.24 / 0.31 (1  0.911)  0.931

37
   0.931100 W/m  K  0.31 m   0.0347 K/W
1 2 1
Rh  2

o hAh

  7.043 103  7.1104  0.0347  K/W  0.0424 K/W

1
UAc
UAc  23.6 W/K
Hence the rate of heat transfer per unit length of tube from gas to water
section is

q  23.6 W/K  800  300  K  11,800 W

Comments:
• The gas-side resistance is substantially decreased by using the fins
and q is increased  A'f   Di ,2  .
• Heat transfer enhancement by the fins could be increased further
by using a material of larger k, but material selection would be
limited by the large value Tm,h
38
Example 3: Ocean Thermal Energy Conversion
Design of a two-pass, shell-and-tube heat exchanger to supply vapor for
the turbine of an ocean thermal energy conversion system based on a
standard (Rankine) power cycle. The power cycle is to generate 2 MWe at
an efficiency of 3%. Ocean water enters the tubes of the exchanger at
300K, and its desired outlet temperature is 292K. The working fluid of the
power cycle is evaporated in the tubes of the exchanger at its phase
change temperature of 290K, and the overall heat transfer coefficient is
known.

Known: A two-pass, shell-and-tube heat exchanger with known inlet

and outlet temperature of cold fluid
Find: (a) Evaporator area, (b) Water flow rate
SCHEMATIC:

Schematic:

39
Assumptions: • Steady operation
• Negligible heat loss to surroundings, and negligible EK and Ep
changes
• Constant properties
• Tube internal flow and thermal conditions fully developed
• Negligible thermal resistance of tube material & fouling effects

Properties: Water, at T = 296 K, from the related table in the textbook:

C p  4181 J/kg  K
Th ,i  300K
Analysis: Th ,o  392K
W 2 MW
a) The efficiency is    0.03 Tc,i  Tc,o  290 K
q q
Hence the required heat transfer rate is
T  300  290  10 K
1
2 MW
q  66.7 MW T2  292  290  2 K
0.03

Tlm ,CF 
T1  T2

 300  290    292  290   5 o C
ln  T1 T 2   300  290 
ln  
 292  290  40
Hence the required heat transfer rate is t2  t1 290  290
P  0
P0 From Fig. 11S.1 T1  t1 300  290

R F=1
R
T1  T2 300  292
 
THe heat transfer area can be calculated as
t2  t1 290  290

q 6.67 107 W
A   11.1 m 2

UF Tlm,CF 1200 W/m 2  K  1 5 C

b) The water flow rate through the evaporator is

q 6.67 107 W
mh    1994 kg/s
C p ,h Th ,i  Th ,i   4181 J/kg  K    300  292  C

Comments:
• The required heat exchange size is enormous due to the small
temperature differences involved.
• The concept was considered during the energy crisis of the mid 1970s
but has not since been implemented. 41