716 Chapter 11 䊏 Heat Exchangers

EXAMPLE 11.1

A counterflow, concentric tube heat exchanger is used to cool the lubricating oil for a large
industrial gas turbine engine. The flow rate of cooling water through the inner tube
('L ⫽ 25 mm) is 0.2 kg/s, while the flow rate of oil through the outer annulus ('R ⫽ 45 mm)
is 0.1 kg/s. The oil and water enter at temperatures of 100 and 30⬚C, respectively. How long
must the tube be made if the outlet temperature of the oil is to be 60⬚C?

SOLUTION

Known: Fluid flow rates and inlet temperatures for a counterflow, concentric tube heat
exchanger of prescribed inner and outer diameter.

Find: Tube length to achieve a desired hot fluid outlet temperature.

Schematic:

7K,L= 100°C

P• K= 0.1 kg/s

Oil

7 ([)
'L= 'R= 7K,R= 60°C
25 mm 45 mm P• F= 0.2 kg/s
Water
7F,R
7F,L= 30°C
[
[

Assumptions:
 Negligible heat loss to the surroundings.
 Negligible kinetic and potential energy changes.
 Constant properties.
 Negligible tube wall thermal resistance and fouling factors.
 Fully developed conditions for the water and oil (8 independent of [).

Properties: Table A.5, unused engine oil (7苶K ⫽ 80⬚C ⫽ 353 K): FS ⫽ 2131 J/kg 䡠 K, ␮ ⫽
3.25 ⫻ 10⫺2 N 䡠 s/m2, N ⫽ 0.138 W/m 䡠 K. Table A.6, water (7苶F ⬇ 35⬚C): FS ⫽ 4178 J/kg 䡠 K,
␮ ⫽ 725 ⫻ 10⫺6 N 䡠 s/m2, N ⫽ 0.625 W/m 䡠 K, 3U ⫽ 4.85.

Analysis: The required heat transfer rate may be obtained from the overall energy bal-
ance for the hot fluid, Equation 11.6b.

T ⫽ PÚ KFS,K(7K,L ⫺ 7K,R)
T ⫽ 0.1 kg/s ⫻ 2131 J/kg 䡠 K (100 ⫺ 60)⬚C ⫽ 8524 W
11.3 䊏 Heat Exchanger Analysis: Use of the Log Mean Temperature Difference 717

Applying Equation 11.7b, the water outlet temperature is

T
7F,R ⫽ ⫹ 7F,L
PÚ FFS,F

7F,R ⫽ 8524 W ⫹ 30⬚C ⫽ 40.2⬚C

0.2 kg/s ⫻ 4178 J/kg 䡠 K
Accordingly, use of 7F ⫽ 35⬚C to evaluate the water properties was a good choice. The
required heat exchanger length may now be obtained from Equation 11.14,
T ⫽ 8$ ⌬7lm
where $ ⫽ ␲'L/ and from Equations 11.15 and 11.17
(7K,L ⫺ 7F,R) ⫺ (7K,R ⫺ 7F,L) 59.8 ⫺ 30
⌬7lm ⫽ ⫽ ⫽ 43.2⬚C
ln [(7K,L ⫺ 7F,R)/(7K,R ⫺ 7F,L)] ln (59.8/30)
From Equation 11.5 the overall heat transfer coefficient is

8⫽ 1
(1/KL) ⫹ (1/KR)
For water flow through the tube,
4PÚ F 4 ⫻ 0.2 kg/s
5H' ⫽ ⫽ ⫽ 14,050
␲'L␮ ␲(0.025 m)725 ⫻ 10⫺6 N 䡠 s/m2
Accordingly, the flow is turbulent and the convection coefficient may be computed from
Equation 8.60
1X' ⫽ 0.023 5H4/5
' 3U
0.4

1X' ⫽ 0.023 (14,050)4/5 (4.85)0.4 ⫽ 90

Hence
90 ⫻ 0.625 W/m 䡠 K
KL ⫽ 1X' N ⫽ ⫽ 2250 W/m2 䡠 K
'L 0.025 m
For the flow of oil through the annulus, the hydraulic diameter is, from Equation 8.71,
'K ⫽ 'R ⫺ 'L ⫽ 0.02 m, and the Reynolds number is
␳XP 'K ␳('R ⫺ 'L) PÚ K
5H' ⫽ ␮ ⫽ ␮ ⫻
␳␲(' ⫺ '2)/4
2
R L

4PÚ K 4 ⫻ 0.1 kg/s

5H' ⫽ ⫽ ⫽ 56.0
␲ ('R ⫹ 'L) ␮ ␲(0.045 ⫹ 0.025) m ⫻ 3.25 ⫻ 10⫺2 kg/s 䡠 m
The annular flow is therefore laminar. Assuming uniform temperature along the inner surface
of the annulus and a perfectly insulated outer surface, the convection coefficient at the inner
surface may be obtained from Table 8.2. With ('L /'R) ⫽ 0.56, linear interpolation provides
KR 'K
1XL ⫽ ⫽ 5.63
N
718 Chapter 11 䊏 Heat Exchangers

and

KR ⫽ 5.63 0.138 W/m 䡠 K ⫽ 38.8 W/m2 䡠 K

0.020 m
The overall convection coefficient is then

8⫽ 1 ⫽ 38.1 W/m2 䡠 K
(1/2250 W/m2 䡠 K) ⫹ (1/38.8 W/m2 䡠 K)
and from the rate equation it follows that
T 8524 W
/⫽ ⫽ ⫽ 65.9 m 䉰
8␲'L ⌬7lm 38.1 W/m2 䡠 K ␲ (0.025 m) (43.2⬚C)

Comments:
 The hot side convection coefficient controls the rate of heat transfer between the two
fluids, and the low value of KR is responsible for the large value of /. Incorporation of
heat transfer enhancement methods, such as described in Section 8.7, could be used to
decrease the size of the heat exchanger.
 Because KL Ⰷ KR, the tube wall temperature will follow closely that of the coolant
water. Accordingly, the assumption of uniform wall temperature, which is inherent in
the use of Table 8.2 to obtain KR, is reasonable.

EXAMPLE 11.2

The counterflow, concentric tube heat exchanger of Example 11.1 is replaced with a compact,
plate-type heat exchanger that consists of a stack of thin metal sheets, separated by 1 gaps of
width D. The oil and water flows are subdivided into 1/2 individual flow streams, with the oil
and water moving in opposite directions within alternating gaps. It is desirable for the stack to
be of a cubical geometry, with a characteristic exterior dimension /. Determine the exterior
dimensions of the heat exchanger as a function of the number of gaps if the flow rates, inlet
temperatures, and desired oil outlet temperature are the same as in Example 11.1. Compare
the pressure drops of the water and oil streams within the plate-type heat exchanger to the
pressure drops of the flow streams in Example 11.1, if 60 gaps are specified.

SOLUTION

Known: Configuration of a plate-type heat exchanger. Fluid flow rates, inlet tempera-
tures, and desired oil outlet temperature.

Find:
 Exterior dimensions of the heat exchanger.
 Pressure drops within the plate-type heat exchanger with 1 ⫽ 60 gaps, and the concen-
tric tube heat exchanger of Example 11.1.
11.3 䊏 Heat Exchanger Analysis: Use of the Log Mean Temperature Difference 719

Schematic:
/

D 7K,L = 100°C
/
P• K= 0.1 kg/s

7 ([)
7K,R = 60°C
PF= 0.2 kg/s
•

/ 7F,R
[ 7F,L = 30°C
Oil Water [

Assumptions:
 Negligible heat loss to the surroundings.
 Negligible kinetic and potential energy changes.
 Constant properties.
 Negligible plate thermal resistance and fouling factors.
 Fully developed conditions for the water and oil.
 Identical gap-to-gap heat transfer coefficients.
 Heat exchanger exterior dimension is large compared to the gap width.

Properties: See Example 11.1. In addition, Table A.5, unused engine oil (7K ⫽ 353 K):
␳ ⫽ 852.1 kg/m3. Table A.6, water (7F 艐 35⬚C): ␳ ⫽ v⫺1
I ⫽ 994 kg/m .
3

Analysis:
 The gap width may be related to the overall dimension of the heat exchanger by the
expression D ⫽ //1, and the total heat transfer area is $ ⫽ /2 (1 ⫺ 1). Assuming D Ⰶ /
and the existence of laminar flow, the Nusselt number for each interior gap is provided
in Table 8.1 and is
K'
1X' ⫽ K ⫽ 7.54
N
From Equation 8.66, the hydraulic diameter is 'K ⫽ 2D. Combining the preceding
expressions yields for the water:
KF ⫽ 7.54N1/2/ ⫽ 7.54 ⫻ 0.625 W/m 䡠 K ⫻ 1/2/ ⫽ (2.36 W/m 䡠 K)1//
Likewise, for the oil:
KK ⫽ 7.54N1/2/ ⫽ 7.54 ⫻ 0.138 W/m 䡠 K ⫻ 1/2/ ⫽ (0.520 W/m 䡠 K)1//
and the overall convection coefficient is

8⫽ 1
1/KF ⫹ 1/KK
From Example 11.1, the required log mean temperature difference and heat transfer rate
are ⌬7lm ⫽ 43.2⬚C and T ⫽ 8524 W, respectively. From Equation 11.14 it follows that
/2(1 ⫺ 1) T
8$ ⫽ ⫽
1/KF ⫹ 1/KK ⌬7lm
720 Chapter 11 䊏 Heat Exchangers

which may be rearranged to yield

/⫽
T
冤1 ⫹ 1
⌬7lm(1 ⫺ 1) KF / KK / 冥
⫽ 8524 W
冤 1 ⫹ 1
43.2⬚C (1 ⫺ 1)1 2.36 W/m䡠 K 0.520 W/m䡠 K 冥
⫽ 463 m
1(1 ⫺ 1)
䉰

The size of the compact heat exchanger decreases as the number of gaps is increased,
as shown in the figure below.
1

0.8

0.6
/(m)

0.4

0.2

0
20 40 60 80
Number of Gaps, 1

 For 1 ⫽ 60 gaps, the stack dimension is / ⫽ 0.131 m from the results of part 1, and
the gap width is D ⫽ //1 ⫽ 0.131 m/60 ⫽ 0.00218 m.
The hydraulic diameter is 'K ⫽ 0.00436 m, and the mean velocity in each water-
filled gap is
Ú 2 ⫻ 0.2 kg/s
XP ⫽ P2 ⫽ ⫽ 0.0235 m/s
␳/ /2 994 kg/m3 ⫻ 0.1312 m2
providing a Reynolds number of
␳XP 'K 994 kg/m3 ⫻ 0.0235 m/s ⫻ 0.00436 m
5H' ⫽ ␮ ⫽ ⫽ 141
725 ⫻ 10⫺6 N 䡠 s/m2
For the oil-filled gaps
Ú 2 ⫻ 0.1 kg/s
XP ⫽ P2 ⫽ ⫽ 0.0137 m/s
␳/ /2 852.1 kg/m3 ⫻ 0.1312 m2
yielding a Reynolds number of
␳XP 'K 852.1 kg/m3 ⫻ 0.0137 m/s ⫻ 0.00436 m
5H' ⫽ ␮ ⫽ ⫽ 1.57
3.25 ⫻ 10⫺2 N 䡠 s/m2
Therefore, the flow is laminar for both fluids, as assumed in part 1. Equations 8.19 and
8.22a may be used to calculate the pressure drop for the water:
␳X2 994 kg/m3 ⫻ 0.02352 m2/s2
⌬S ⫽ 64 䡠 P 䡠 / ⫽ 64 ⫻ ⫻ 0.131 m
5H' 2'K 141 2 ⫻ 0.00436 m

⫽ 3.76 N/m2 䉰
11.3 䊏 Heat Exchanger Analysis: Use of the Log Mean Temperature Difference 721

Similarly, for the oil

␳X2 852.1 kg/m3 ⫻ 0.01372 m2/s2

⌬S ⫽ 64 䡠 P 䡠 / ⫽ 64 ⫻ ⫻ 0.131 m
5H' 2'K 1.57 2 ⫻ 0.00436 m
⫽ 98.2 N/m2 䉰

For Example 11.1, the friction factor associated with the water flow may be calcu-
lated using Equation 8.21, and for a smooth surface condition is I ⫽ (0.790 ln(14,050) ⫺

1.64)⫺2 ⫽ 0.0287. The mean velocity is XP ⫽ 4P/(␳␲ '2L ) ⫽ 4 ⫻ 0.2 kg/s/(994 kg/m3 ⫻
␲ ⫻ 0.025 m ) ⫽ 0.410 m/s, and the pressure drop is
2 2

␳X2P 994 kg/m3 ⫻ 0.4102 m2/s2

⌬S ⫽ I 䡠 䡠 / ⫽ 0.0287 ⫻ ⫻ 65.9 m
2'K 2 ⫻ 0.025 m
⫽ 6310 N/m2 䉰

For the oil flowing in the annular region, the mean velocity is XP ⫽ 4P/[ ␳␲
('2R ⫺ '2L )] ⫽ 4 ⫻ 0.1 kg/s/[852.1 kg/m3 ⫻ ␲ ⫻ (0.0452 ⫺ 0.0252)m2] ⫽ 0.107 m/s, and
the pressure drop is

␳X 2 852.1 kg/m3 ⫻ 0.1072 m2/s2

⌬S ⫽ 64 䡠 P 䡠 / ⫽ 64 ⫻ ⫻ 65.9 m
5H' 2'K 56 2 ⫻ 0.020 m
⫽ 18,300 N/m2 䉰

Comments:
 Increasing the number of gaps increases the 8$ product by simultaneously providing
more surface area and increasing the heat transfer coefficients associated with the flow
of the fluids through smaller passages.
 The area-to-volume ratio of the 1 ⫽ 60 heat exchanger is /2(1 ⫺ 1)//3 ⫽ (1 ⫺ 1)/
/ ⫽ (60 ⫺ 1)/0.131 m ⫽ 451 m2/m3.
 The volume occupied by the concentric tube heat exchanger is 9 ⫽ ␲ '2R //4 ⫽ ␲ ⫻
0.0452 m2 ⫻ 65.9 m/4 ⫽ 0.10 m3, while the volume of the compact plate-type
exchanger is 9 ⫽ /3 ⫽ 0.1313 m3 ⫽ 0.0022 m3. Use of the plate-type heat exchanger
results in a 97.8% reduction in volume relative to the conventional, concentric tube
heat exchanger.
 The pressure drops associated with use of the compact heat exchanger are significantly
less than for a conventional concentric tube configuration. Pressure drops are reduced
by 99.9% and 99.5% for the water and oil flows, respectively.
 Fouling of the heat transfer surfaces may result in a decrease in the gap width, as well
as an associated reduction in heat transfer rate and increase in pressure drop.
 Because KF ⬎ KK, the temperatures of the thin metal sheets will follow closely that of
the water, and, as in Example 11.1, the assumption of uniform temperature conditions
to obtain KF and KK is reasonable.
 One method to fabricate such a heat exchanger is presented in C. F. McDonald, $SSO
7KHUPDO(QJLQ., , 471, 2000.
728 Chapter 11 䊏 Heat Exchangers

EXAMPLE 11.3

Hot exhaust gases, which enter a finned-tube, cross-flow heat exchanger at 300⬚C and leave
at 100⬚C, are used to heat pressurized water at a flow rate of 1 kg/s from 35 to 125⬚C. The
overall heat transfer coefficient based on the gas-side surface area is 8K ⫽ 100 W/m2 䡠 K.
Determine the required gas-side surface area $K using the NTU method.

SOLUTION

Known: Inlet and outlet temperatures of hot gases and water used in a finned-tube, cross-
flow heat exchanger. Water flow rate and gas-side overall heat transfer coefficient.

Find: Required gas-side surface area.

Schematic:
Finned-tube, cross-flow
7K,L heat exchanger,
8K= 100 W/m2•K
Both fluids unmixed 7K,L= 300°C

P• K
7

7F,L 7F,R

P• F= 1 kg/s
7K,R= 100°C
7F,R= 125°C 7F,L= 35°C

7K,R

Assumptions:
 Negligible heat loss to the surroundings and kinetic and potential energy changes.
 Constant properties.

Properties: Table A.6, water (7F ⫽ 80⬚C): FS,F ⫽ 4197 J/kg 䡠 K.

Analysis: The required surface area may be obtained from knowledge of the number of
transfer units, which, in turn, may be obtained from knowledge of the ratio of heat capacity
rates and the effectiveness. To determine the minimum heat capacity rate, we begin by
computing
&F ⫽ PÚ FFS,F ⫽ 1 kg/s ⫻ 4197 J/kg 䡠 K ⫽ 4197 W/K
Since PÚ K is not specified, &K is obtained by combining the overall energy balances, Equa-
tions 11.6b and 11.7b:
7F,R ⫺ 7F,L
&K ⫽ PÚ KFS,K ⫽ &F ⫽ 4197 125 ⫺ 35 ⫽ 1889 W/K ⫽ &min
7K,L ⫺ 7K,R 300 ⫺ 100
From Equation 11.18

Tmax ⫽ &min(7K,L ⫺ 7F,L) ⫽ 1889 W/K (300 ⫺ 35)⬚C ⫽ 5.00 ⫻ 105 W

11.4 䊏 Heat Exchanger Analysis: The Effectiveness–NTU Method 729

From Equation 11.7b the actual heat transfer rate is

T ⫽ &F(7F,R ⫺ 7F,L) ⫽ 4197 W/K (125 ⫺ 35)⬚C

T ⫽ 3.78 ⫻ 105 W

Hence from Equation 11.19 the effectiveness is

T
⫽ 3.78 ⫻ 105 W ⫽ 0.755
5
␧⫽T
max 5.00 ⫻ 10 W
With
&min 1889
⫽ ⫽ 0.45
&max 4197
it follows from Figure 11.14 that
8K $ K
NTU ⫽ 艐 2.0
&min
or
2.0(1889 W/K)
$K ⫽ ⫽ 37.8 m2 䉰
100 W/m2 䡠 K

Comments:
 Equation 11.32 may be solved iteratively or by trial and error to yield NTU ⫽ 2.0,
which is in excellent agreement with the estimate obtained from the charts.
 With the heat exchanger sized ($K ⫽ 37.8 m2) and placed into operation, its actual per-
formance is subject to uncontrolled variations in the exhaust gas inlet temperature
(200 ⱕ 7K,L ⱕ 400⬚C) and to gradual degradation of the heat exchanger surfaces due to
fouling (8K decreasing from 100 to 60 W/m2 䡠 K). For a fixed value of &min ⫽ &K ⫽
1889 W/K, the reduction in 8K corresponds to a reduction in the NTU (to NTU ⬇ 1.20)
and hence to a reduction in the heat exchanger effectiveness, which can be computed
from Equation 11.32. The effect of the variations on the water outlet temperature has
been computed and is plotted as follows:

180

160

8K (W/m2•K)
140
7F,R (°C)

100
80
60
120
P• K = 1.89 kg/s
P• F = 1 kg/s
100
7F,L = 35°C

80
200 250 300 350 400
7K,L(°C)
730 Chapter 11 䊏 Heat Exchangers

If the intent is to maintain a fixed water outlet temperature of 7F,R ⫽ 125⬚C, adjust-
ments in the flow rates, PÚ F and PÚ K, could be made to compensate for the variations.
The model equations could be used to determine the adjustments and hence as a basis
for designing the requisite FRQWUROOHU

11.5 Heat Exchanger Design and Performance Calculations

Two general types of heat exchanger problems are commonly encountered by the practic-
ing engineer.
In the KHDW H[FKDQJHU GHVLJQ SUREOHP, the fluid inlet temperatures and flow rates, as
well as a desired hot or cold fluid outlet temperature, are prescribed. The design problem is
then one of specifying a specific heat exchanger type and determining its size—that is, the
heat transfer surface area $—required to achieve the desired outlet temperature. The design
problem is commonly encountered when a heat exchanger is to be custom-built for a specific
application. Alternatively, in a KHDW H[FKDQJHU SHUIRUPDQFH FDOFXODWLRQ, an existing heat
exchanger is analyzed to determine the heat transfer rate and the fluid outlet temperatures for
prescribed flow rates and inlet temperatures. The performance calculation is commonly
associated with the use of off-the-shelf heat exchanger types and sizes available from a
vendor.
For heat exchanger design problems, the NTU method may be used by first calculating
␧ and (&min /&max). The appropriate equation (or chart) may then be used to obtain the NTU
value, which in turn may be used to determine $. For a performance calculation, the NTU and
(&min/&max) values may be computed and ␧ may then be determined from the appropriate
equation (or chart) for a particular exchanger type. Since Tmax may also be computed from
Equation 11.18, it is a simple matter to determine the actual heat transfer rate from the
requirement that T ⫽ ␧Tmax. Both fluid outlet temperatures may then be determined from
Equations 11.6b and 11.7b.

EXAMPLE 11.4

Consider the heat exchanger design of Example 11.3, that is, a finned-tube, cross-flow
heat exchanger with a gas-side overall heat transfer coefficient and area of 100 W/m2 䡠 K
and 40 m2, respectively. The water flow rate and inlet temperature remain at 1 kg/s and
35⬚C. However, a change in operating conditions for the hot gas generator causes the
gases to now enter the heat exchanger with a flow rate of 1.5 kg/s and a temperature of
250⬚C. What is the rate of heat transfer by the exchanger, and what are the gas and water
outlet temperatures?

SOLUTION

Known: Hot and cold fluid inlet conditions for a finned-tube, cross-flow heat exchanger
of known surface area and overall heat transfer coefficient.

Find: Heat transfer rate and fluid outlet temperatures.

11.5 䊏 Heat Exchanger Design and Performance Calculations 731

Schematic:

7K,L= 250°C
$K = 40 m2
8K = 100 W/m2•K

P• K = 1.5 kg/s

7
7F,R
P• F = 1 kg/s
7K,R
7F,L= 35°C

Assumptions:
 Negligible heat loss to surroundings and kinetic and potential energy changes.
 Constant properties (unchanged from Example 11.3).

Analysis: The problem may be classified as one requiring a heat exchanger SHUIRUPDQFH
FDOFXODWLRQ The heat capacity rates are

&F ⫽ PÚ FFS,F ⫽ 1 kg/s ⫻ 4197 J/kg 䡠 K ⫽ 4197 W/K

&K ⫽ PÚ KFS,K ⫽ 1.5 kg/s ⫻ 1000 J/kg 䡠 K ⫽ 1500 W/K ⫽ &min

in which case

&min 1500
⫽ ⫽ 0.357
&max 4197

The number of transfer units is

8K $K 100 W/m2 䡠 K ⫻ 40 m2
NTU ⫽ ⫽ ⫽ 2.67
&min 1500 W/K

From Figure 11.14 the heat exchanger effectiveness is then ␧ ⬇ 0.82, and from Equation
11.18 the maximum possible heat transfer rate is

Tmax ⫽ &min(7K,L ⫺ 7F,L) ⫽ 1500 W/K (250 ⫺ 35)⬚C ⫽ 3.23 ⫻ 105 W

Accordingly, from the definition of ␧, Equation 11.19, the actual heat transfer rate is

T ⫽ ␧Tmax ⫽ 0.82 ⫻ 3.23 ⫻ 105 W ⫽ 2.65 ⫻ 105 W 䉰

It is now a simple matter to determine the outlet temperatures from the overall energy bal-
ances. From Equation 11.6b

T
⫽ 250⬚C ⫺ 2.65 ⫻ 10 W ⫽ 73.3⬚C
5
7K,R ⫽ 7K,L ⫺ 䉰
PÚ KFS,K 1500 W/K
732 Chapter 11 䊏 Heat Exchangers

and from Equation 11.7b

T
⫽ 35⬚C ⫹ 2.65 ⫻ 10 W ⫽ 98.1⬚C
5
7F,R ⫽ 7F,L ⫹ 䉰
PÚ FFS,F 4197 W/K

Comments:
 From Equation 11.32, ␧ ⫽ 0.845, which is in good agreement with the estimate
obtained from the charts.
 The overall heat transfer coefficient has tacitly been assumed to be unaffected by the
change in PÚ K. In fact, with an approximately 20% reduction in PÚ K, there would be a
significant, albeit smaller percentage, reduction in 8K.
 As discussed in the Comment of Example 11.3, flow rate adjustments could be made
to maintain a fixed water outlet temperature. If, for example, the outlet temperature
must be maintained at 7F,R ⫽ 125⬚C, the water flow rate could be reduced to an amount
prescribed by Equation 11.7b. That is,

T 2.65 ⫻ 105 W
PÚ F ⫽ ⫽ ⫽ 0.702 kg/s
FS,F(7F,R ⫺ 7F,L) 4197 J/kg 䡠 K (125 ⫺ 35)⬚C

The change in flow rate has again been presumed to have a negligible effect on 8K. In
this case the assumption is good, since the dominant contribution to 8K is made by the
gas-side, and not the water-side, convection coefficient.

EXAMPLE 11.5

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 VKHOODQGWXEH heat exchanger
consisting of a single shell and 30,000 tubes, each executing two passes. The tubes are of
thin wall construction with ' ⫽ 25 mm, and steam condenses on their outer surface with an
associated convection coefficient of KR ⫽ 11,000 W/m2 䡠 K. The heat transfer rate that must
be effected by the exchanger is T ⫽ 2 ⫻ 109 W, and this is accomplished by passing
cooling water through the tubes at a rate of 3 ⫻ 104 kg/s (the flow rate per tube is therefore
1 kg/s). The water enters at 20⬚C, while the steam condenses at 50⬚C. What is the tempera-
ture of the cooling water emerging from the condenser? What is the required tube length
/ per pass?

SOLUTION

Known: Heat exchanger consisting of single shell and 30,000 tubes with two passes each.

Find:
 Outlet temperature of the cooling water.
 Tube length per pass to achieve required heat transfer.
11.5 䊏 Heat Exchanger Design and Performance Calculations 733

Schematic:
Steam
7K,L= 50°C
T= 2 x 10 W
9
P• K
Tube data 7K,L
7K,R= 50°C
1 = 30,000 7F,L

7
' = 0.025 m P• F= 3 x 104 kg/s
KR = 11,000 W/m2•K / P• F
Water
7F,R
/ = length/pass 7F,R
P• = 1 kg/s 7F,L= 20°C
7K,R

P• K

Assumptions:
 Negligible heat transfer between exchanger and surroundings and negligible kinetic
and potential energy changes.
 Tube internal flow and thermal conditions fully developed.
 Negligible thermal resistance of tube material and fouling effects.
 Constant properties.
Properties: Table A.6, water (assume 7苶F ⬇ 27⬚C ⫽ 300 K): ␳ ⫽ 997 kg/m3, FS ⫽ 4179
J/kg 䡠 K, ␮ ⫽ 855 ⫻ 10⫺6 N 䡠 s/m2, N ⫽ 0.613 W/m 䡠 K, 3U ⫽ 5.83.
Analysis:
 The cooling water outlet temperature may be obtained from the overall energy bal-
ance, Equation 11.7b. Accordingly,
T 2 ⫻ 109 W
7F,R ⫽ 7F,L ⫹ ⫽ 20⬚C ⫹
PÚ FFS,F 3 ⫻ 10 kg/s ⫻ 4179 J/kg 䡠 K
4

7F,R ⫽ 36.0⬚C 䉰
 The problem may be classified as one requiring a KHDWH[FKDQJHUGHVLJQFDOFXODWLRQ.
First, we determine the overall heat transfer coefficient for use in the NTU method.
From Equation 11.5
8⫽ 1
(1/KL) ⫹ (1/KR)
where KL may be estimated from an internal flow correlation. With
Ú 4 ⫻ 1 kg/s
5H' ⫽ 4P ⫽ ⫽ 59,567
␲'␮ ␲(0.025 m)855 ⫻ 10⫺6 N 䡠 s/m2
the flow is turbulent and from Equation 8.60
1X' ⫽ 0.023 5H4/5
' 3U
0.4
⫽ 0.023(59,567)0.8(5.83)0.4 ⫽ 308
Hence

KL ⫽ 1X' N ⫽ 308 0.613 W/m 䡠 K ⫽ 7543 W/m2 䡠 K

' 0.025 m

8⫽ 1 ⫽ 4474 W/m2 䡠 K
[(1/7543) ⫹ (1/11,000)] m2 䡠 K/W
734 Chapter 11 䊏 Heat Exchangers

Using the design calculation methodology, we note that

&K ⫽ &max ⫽ 앝
and

&min ⫽ PÚ FFS,F ⫽ 3 ⫻ 104 kg/s ⫻ 4179 J/kg 䡠 K ⫽ 1.25 ⫻ 108 W/K

from which
&min
⫽ &U ⫽ 0
&max
The maximum possible heat transfer rate is
Tmax ⫽ &min(7K,L ⫺ 7F,L) ⫽ 1.25 ⫻ 108 W/K ⫻ (50 ⫺ 20) K ⫽ 3.76 ⫻ 109 W

from which
T
⫽ 2 ⫻ 10 W
9
␧⫽T ⫽ 0.532
max 3.76 ⫻ 109 W
From Equation 11.35b or Figure 11.12, we find NTU ⫽ 0.759. From Equation 11.24,
it follows that the tube length per pass is
NTU 䡠 &min 0.759 ⫻ 1.25 ⫻ 108 W/K
/⫽ ⫽ ⫽ 4.51 m 䉰
8(12␲') 4474 W/m2 䡠 K (30,000 ⫻ 2 ⫻ ␲ ⫻ 0.025 m)

Comments:
 Recognize that / is the tube length per pass, in which case the total length per tube is
9.0 m. The entire length of tubing in the condenser is 1 ⫻ / ⫻ 2 ⫽ 30,000 ⫻ 4.51 m ⫻
2 ⫽ 271,000 m or 271 km.
 Over time, the performance of the heat exchanger would be degraded by fouling on both
the inner and outer tube surfaces. A representative maintenance schedule would call for
taking the heat exchanger off-line and cleaning the tubes when fouling factors reached
values of 5⬙ƒ,L ⫽ 5⬙ƒ,R ⫽ 10⫺4 m2 䡠 K/W. To determine the effect of fouling on perfor-
mance, the ␧⫺NTU method may be used to calculate the total heat rate as a function of
the fouling factor, with 5⬙ƒ,R assumed to equal 5⬙ƒ,L. The following results are obtained:
2.2

2.0
P• F,1 = 1 kg/s
1.8 7K = 50°C
T x 10–9 (W)

7F,L = 20°C
1.6

1.4

1.2

1.0
0.0 0.2 0.4 0.6 0.8 1.0
5
I
× 104 (m2•K/W)
11.5 䊏 Heat Exchanger Design and Performance Calculations 735

To maintain the requirement of T ⫽ 2 ⫻ 109 W with the maximum allowable fouling


and the restriction of PF,1 ⫽ 1 kg/s, the tube length or the number of tubes would have to
be increased. Keeping the length per pass at / ⫽ 4.51 m, 1 ⫽ 48,300 tubes would be
needed to transfer 2 ⫻ 109 W for 5⬙ƒ,L ⫽ 5⬙ƒ,R ⫽ 10⫺4 m2 䡠 K/W. The corresponding increase
 
in the total flow rate to P F ⫽ 1P F,1 ⫽ 48,300 kg/s would have the beneficial effect of
reducing the water outlet temperature to 7F,R ⫽ 29.9⬚C, thereby ameliorating potentially
harmful effects associated with discharge into the environment. The additional tube
length associated with increasing the number of tubes to 1 ⫽ 48,300 is 165 km, which
would result in a significant increase in the capital cost of the condenser.
 The steam plant generates 1250 MW of electricity with a wholesale value of $0.05
per kW 䡠 h. If the plant is shut down for 48 hours to clean the condenser tubes, the loss
in revenue for the plant’s owner is 48 h ⫻ 1250 ⫻ 106 W ⫻ $0.05/(1 ⫻ 103 W 䡠 h) ⫽
$3 million.
 Assuming a smooth surface condition within each tube, the friction factor may be
determined from Equation 8.21, I ⫽ (0.790 ln (59,567) ⫺ 1.64)⫺2 ⫽ 0.020. The pres-
sure drop within one tube of length / ⫽ 9 m may be determined from Equation 8.22a,
where XP ⫽ 4PÚ /(␳␲'2) ⫽ (4 ⫻ 1 kg/s)/(997 kg/m3 ⫻ ␲ ⫻ 0.0252 m2) ⫽ 2.04 m/s.
␳X2P 997 kg/m3(2.04 m/s)2
⌬S ⫽ I / ⫽ 0.020 9.0 m ⫽ 15,300 N/m2
2' 2(0.025 m)

Therefore, the power required to pump the cooling water through the 48,300 tubes
may be found by using Equation 8.22b and is
⌬SPÚ 15,300 N/m2 ⫻ 48,300 kg/s
3⫽ ␳ ⫽ ⫽ 742,000 W ⫽ 0.742 MW
997 kg/m3

The cooling water pump is driven by an electric motor. If the combined efficiency of
the pump and motor is 87%, the annual cost to overcome friction losses in the con-
denser tubes is 24 h/day ⫻ 365 days/yr ⫻ 0.742 ⫻ 106 W ⫻ $0.05/1 ⫻ 103 W 䡠 h/0.87 ⫽
$374,000.
 Optimal condenser designs are based on the desired thermal performance and environ-
mental considerations as well as on the capital cost, operating cost, and maintenance
cost associated with the device.

EXAMPLE 11.6

A geothermal power plant utilizes pressurized, deep groundwater at 7* ⫽ 147⬚C as the heat
source for an RUJDQLF5DQNLQHF\FOH, the operation of which is described further in Comment 2.
An evaporator, consisting of a vertically oriented shell-and-tube heat exchanger with one
shell pass and one tube pass, transfers heat between the tube side groundwater and the coun-
terflowing shell-side organic fluid of the power cycle. The organic fluid enters the shell side
of the evaporator as a subcooled liquid at 7FL ⫽ 27⬚C, and exits the evaporator as a saturated
vapor of TXDOLW\ ;5R ⫽ 1 and temperature 7FR ⫽ 7sat ⫽ 122⬚C. Within the evaporator, heat
transfer occurs between liquid groundwater and the organic fluid in Stage A with
8A ⫽ 900 W/m2 䡠 K, and between liquid groundwater and boiling organic fluid in Stage B