8-1

Chapter 8
INTERNAL FORCED CONVECTION

8-26 Steam is condensed by cooling water flowing inside copper tubes. The average heat transfer
coefficient and the number of tubes needed are to be determined.
Assumptions 1 Steady operating conditions exist. 2 The surface temperature of the pipe is constant. 3
The thermal resistance of the pipe is negligible.
Properties The properties of water at the average Steam, 30C
temperature of (10+24)/2=17C are (Table A-9)
  998.7 kg/m 3
24C
c p  4183.8 J/kg.C Water
10C D = 1.2 cm
Also, the heat of vaporization of water at 30C is 4 m/s
h fg  2431 kJ/kg . L=5m
Analysis The mass flow rate of water and the surface
area are
 D 2 
Vavg  (998.7 kg/m 3 )  (0.012 m) (4 m/s) = 0.4518 kg/s
2
m AcVavg   
 4  4
 
The rate of heat transfer for one tube is
Q m
c p (Te  Ti )  (0.4518 kg/s)( 4183.8 J/kg.C)( 24  10C)  26,460 W

The logarithmic mean temperature difference and the surface area are
Te  Ti 24  10
Tlm    11.63C
 T  Te   30  24 
ln s 

ln 
 Ts  Ti   30  10 

As  DL   (0.012 m)(5 m) = 0.1885 m 2

The average heat transfer coefficient is determined from
Q 26,460 W  1 kW 
Q hAs Tlm 
 h      12.1 kW/m .C
2
As Tlm (0.1885 m )(11.63C)  1000 W 
2

The total rate of heat transfer is determined from

Qtotal  m
cond h fg  (0.15 kg/s)( 2431 kJ/kg)  364.65 kW

Then the number of tubes becomes

Qtotal 364,650 W
N tube    13.8
Q 26,460 W
 8-2

8-57 The flow rate through a specified water pipe is given. The pressure drop and the pumping power
requirements are to be determined.
Assumptions 1 The flow is steady and incompressible. 2 The entrance effects are negligible, and thus
the flow is fully developed. 3 The pipe involves no components such as bends, valves, and connectors. 4
The piping section involves no work devices such as pumps and turbines.
Properties The density and dynamic viscosity of water are given to be  = 999.1 kg/m3 and  =
1.13810-3 kg/ms, respectively. The roughness of stainless steel is 0.002 mm (Table 8-3).
Analysis First, we calculate the mean velocity and the
Reynolds number to determine the flow regime:
Water
V V 0.007 m 3 / s D = 4 cm
Vavg     5.570 m / s
Ac D / 4
2
 (0.04 m) / 4
2
7 L/s
Vavg D 3
(999.1 kg/m )(5.570 m/s)(0.04 m)
Re    1.956  10 5 L = 25 m
 1.138  10 3 kg/m  s
which is greater than 10,000. Therefore, the flow is turbulent. The relative roughness of the pipe is
2 10 6 m
/D  5 10 5
0.04 m
The friction factor can be determined from the Moody chart, but to avoid the reading error, we
determine it from the Colebrook equation using an equation solver (or an iterative scheme),
1  /D 2.51  1  5  10 5 2.51 
 2.0 log     2.0 log  
f  3.7 Re f  f  3.7 1.956  10 5 f 
  
It gives f = 0.0161. Then the pressure drop and the required power input become
L Vavg   1 kPa 
2
25 m (999.1 kg/m 3 )(5.570 m/s)2  1 kN 
P  f  0.0161  156.0 kPa
D 2 0.04 m 2  1000 kg  m/s 2  1 kN/m 2 
 
 1 kW 
Wpump,u  V
P  (0.007 m 3 / s)(156.0 kPa)   1.09 kW
 1 kPa  m /s 
3

Therefore, useful power input in the amount of 1.09 kW is needed to overcome the frictional losses in
the pipe.
Discussion The friction factor could also be determined easily from the explicit Haaland relation. It
would give f = 0.01589, which is sufficiently close to 0.0161. Also, the friction factor corresponding to 
= 0 in this case is 0.01557, which indicates that stainless steel pipes can be assumed to be smooth with
an error of about 3%. Also, the power input determined is the mechanical power that needs to be
imparted to the fluid. The shaft power will be more than this due to pump inefficiency; the electrical
power input will be even more due to motor inefficiency.
Note that for turbulent flow, the entry length is Lh, turb  10D  10(0.04 m)  0.4 m . Therefore, the
assumption for fully developed flow is valid for this 25-m long pipe.
 8-3

8-63 A circuit board is cooled by passing cool air through a channel drilled into the board. The
maximum total power of the electronic components is to be determined.
Assumptions 1 Steady operating conditions exist. 2 The heat flux at the top surface of the channel is
uniform, and heat transfer through other surfaces is negligible. 3 The inner surfaces of the channel are
smooth. 4 Air is an ideal gas with constant properties. 5 The pressure of air in the channel is 1 atm. 5
Flow is fully developed in the channel.
Properties The properties of air at 1 atm and estimated average temperature of 25C based on the
problem statement are (Table A-15)
  1.184 kg/m 3
k  0.02551 W/m.C
Electronic components,
  1.562  10 -5 m 2 /s 50C
c p  1007 J/kg.C
Pr  0.7296
Analysis The cross-sectional and heat Air
L = 20 cm
transfer surface areas are 15C
4 m/s Air channel
Ac  (0.002 m)( 0.14 m)  0.00028 m 2 0.2 cm  14 cm

As  (0.14 m)( 0.2 m)  0.028 m 2

To determine heat transfer coefficient, we first need to find the Reynolds number,
4 Ac 4(0.00028 m 2 )
Dh    0.003944 m
P 2(0.002 m + 0.14 m)
Vavg Dh (4 m/s)(0.003944 m)
Re    1010
 1.562 10 5 m 2 /s
which is less than 2300. Therefore, the flow is laminar and the thermal entry length is
Lt  0.05 Re Pr Dh  0.05(1010)(0.7296)(0.003944 m) = 0.1453 m < 0.20 m
Therefore, we have developing flow through most of the channel. However, we take the conservative
approach and assume fully developed flow, and from Table 8-1 we read Nu = 8.24. Then the heat
transfer coefficient becomes
k 0.02551 W/m.C
h Nu  (8.24)  53.30 W/m2 .C
Dh 0.003944 m
Also,
 VAc  (1.184 kg/m 3 )( 4 m/s)(0.00028 m 2 )  0.001326 kg/s
m
Heat flux at the exit can be written as q h(Ts  Te ) where Ts  50C at the exit. Then the heat transfer
rate can be expressed as Q qAs  hAs (Ts  Te ) , and the exit temperature of the air can be determined from
hAs (Ts  Te )  mc p (Te  Ti )
(53.30 W/m .C)( 0.028 m )(50C  Te )  (0.001326 kg/s)(1007 J/kg.C)(Te  15C)
2 2

Te  33.5C
Then the maximum total power of the electronic components that can safely be mounted on this circuit
board becomes
Qmax  m
c p (Te  Ti )  (0.001326 kg/s)(1007 J/kg.C)(33.5  15C)  24.7 W

Discussion The bulk mean temperature of air is (15 + 33.5)/2 = 24.3C. This is very close to the
assumed temperature of 25C. Therefore, there is no need to repeat calculations.
 8-4

8-66 A computer is cooled by a fan blowing air through its case. The flow rate of the air, the fraction of
the temperature rise of air that is due to heat generated by the fan, and the highest allowable inlet air
temperature are to be determined.
Assumptions 1 Steady flow conditions exist. 2 Heat flux is uniformly distributed. 3 Air is an ideal gas
with constant properties. 4 The pressure of air is 1 atm.
Properties We assume the bulk mean temperature for air
to be 25C. The properties of air at 1 atm and this
temperature are (Table A-15)
  1.184 kg/m 3
k  0.02551 W/m.C
  1.562  10 -5 m 2 /s Cooling
air
c p  1007 J/kg.C
Pr  0.7296
Analysis (a) Noting that the electric energy consumed by the fan is
converted to thermal energy, the mass flow rate of air is
Q Welect, fan (8  12  10) W
Q m
c p (Te  Ti )  m
   0.01053 kg/s
c p (Te  Ti ) (1007 J/kg.C)(10C)

(b) The fraction of temperature rise of air that is due to the heat generated by the fan and its motor is
Q 10 W
Q mc p T  T    0.943C
mc p (0.01053 kg/s)(1007 J/kg.C)
0.943C
f =  0.0943  9.43%
10C
(c) The mean velocity of air is
m (0.01053 / 8) kg/s
m AcVavg  Vavg    3.088 m/s
Ac (1.184 kg/m 3 )(0.003 m)(0.12 m)
4 Ac 4(0.003 m)(0.12 m)
and Dh    0.00585 m
P 2(0.003 m  0.12 m)
Therefore,
Vavg Dh (3.088 m/s)(0.00585 m)
Re    1157
 1.562  10 5 m 2 /s
which is less than 2300. Therefore, the flow is laminar. Assuming fully developed flow, the Nusselt
number is determined from Table 8-4 corresponding to a/b = 12/0.3 = 40 to be Nu = 8.24. Then,
k 0.02551 W/m.C
h Nu  (8.24)  35.9 W/m2 .C
Dh 0.00585 m
The highest component surface temperature will occur at the exit of the duct. Assuming uniform heat
flux, the air temperature at the exit is determined from
q [(8  12  10) W]/[8  2(0.12  0.15 + 0.003  0.15) m 2 ]
q h(Ts,max  Te )  Te  Ts,max   70C   60.0C
h 35.9 W/m2 .C
The highest allowable inlet temperature then becomes
Te  Ti  10C  Ti  Te  10C  60.0C  10C  50.0C
Discussion Although the Reynolds number is less than 2300, the flow in this case will most likely be
turbulent because of the electronic components that that protrude into flow. Therefore, the heat transfer
coefficient determined above is probably conservative.
 8-5

8-101 Flow of hot air through uninsulated square ducts of a heating system in the attic is considered.
The exit temperature and the rate of heat loss are to be determined.
Assumptions 1 Steady operating conditions exist. 2 The inner surfaces of the duct are smooth. 3 Air is
an ideal gas with constant properties. 4 The pressure of air is 1 atm.
Properties We assume the bulk mean temperature for air to be 75C since the mean temperature of air at
the inlet will not drop significantly because the surfaces are at 70C. The properties of air at 1 atm and
this temperature are (Table A-15)
  1.014 kg/m 3
k  0.02917 W/m.C
  2.046  10 -5 m 2 /s
c p  1007.5 J/kg.C Te

Pr  0.7166
Analysis The characteristic length that is the hydraulic
diameter, the mean velocity of air, and the Reynolds
number are
L = 10 m
4A 4a 2 Ts = 70C
Dh  c   a  0.15 m
P 4a Air
V 0.20 m 3 /s 75C
Vavg    8.889 m/s 0.2 m3/min
Ac (0.15 m) 2
Vavg Dh (8.889 m/s)(0.15 m)
Re    65,168
 2.046  10 5 m 2 /s
which is greater than 10,000. Therefore, the flow is turbulent and the entry lengths in this case are
roughly
Lh  Lt  10Dh  10(0.15 m)  1.5 m
which is much shorter than the total length of the duct. Therefore, we can assume fully developed
turbulent flow in the entire duct, and determine the Nusselt number from
hDh
Nu   0.023 Re0.8 Pr 0.3  0.023(65,168) 0.8 (0.7166) 0.3  147.8
k
Heat transfer coefficient is
k 0.02917 W/m.C
h Nu  (147.8)  28.73 W/m2 .C
Dh 0.15 m
Next we determine the exit temperature of air,
As  4aL  4(0.15 m)(10 m) = 6 m 2
m V (1.014 kg/m 3 )(0.20 m 3 /s) = 0.2028 kg/s
( 28.73)( 6)

c p )
 hA /( m
Te  Ts  (Ts  Ti )e  70  (70  75)e ( 0.2028)(1007.5)
 72.15C
Then the logarithmic mean temperature difference and the rate of heat loss from the air becomes
Te  Ti 72.15  75
Tlm    3.377C
 T  Te   70  72.15 
ln s 

ln 
 Ts  Ti   70  75 
Q hAs Tlm  (28.73 W/m2 .C)(6 m 2 )(3.377C)  582 W
Note that the temperature of air drops by about 3C as it flows in the duct as a result of heat loss.
 8-6

8-107 Water flows through a concentric annulus tube with constant inner surface temperature and
insulated outer surface, the length of the annulus tube is to be determined.
Assumptions 1 Steady operating conditions. 2 Constant properties. 3 Constant inner tube surface
temperature. 4 Insulated outer tube surface. 5 Fully developed flow.
Properties The properties of water at Tb = (Ti + Te)/2 = 50°C: cp = 4181 J/kg∙K, k = 0.644 W/m∙K, =
0.547  10−3 kg/m∙s, and Pr = 3.55 (Table A-15).

Analysis The Reynolds number is

Vavg ( Do  Di ) m( Do  Di ) 4m
Re   
 ( / 4)( Do2  Di2 )   ( D o  Di ) 
4(0.05 kg/s)

 (0.025 m  0.1 m)( 0.547  10 3 kg/m  s)
 931
Since Re < 2300, the flow through the annulus is laminar. Assuming fully developed flow, the Nusselt
number for the inner tube surface is (from Table 8-4)
hi Dh
Nu i   7.37 for Di / Do  0.25
k
Hence, the convection heat transfer coefficient is
 0.644 W/m K 
hi  7.37   63.28 W/m  K
2
 0.075 m 
The length of the concentric annulus tube is
c p
m Ts  Te (0.05 kg/s)( 4181 J/kg  K) 120  80
L ln  ln  38.5 m
Di hi Ts  Ti  (0.025 m)(63.28 W/m  K) 120  20
2

Discussion Similar to regular tubes, the total rate of heat transfer in the annulus tube can be determined
using Q mc p (Te  Ti ) .