CH 12

12-1 An undergraduate heat transfer lab has an experiment to illustrate the effects of different boundary
conditions on heat transfer from a flat plate. A test section is installed into a wind tunnel. The test section
consists of 100 thin strip heaters placed on a 2-m long flat plate. Each heater, 20-mm long and 250-mm
wide, is located so that there is no space between adjoining heaters and is electrically and thermally
insulated from the adjacent heaters; the backside of the plate is heavily insulated. The power to each heater
can be individually controlled. The free-stream air temperature is 25 ºC and has a velocity of 4 m/s. By
controlling the power to each strip, two different boundary conditions can be modeled. Ignoring radiation,
determine for strips 1, 5, 25, 100, and 200:
a. the heat transfer rate when the power is adjusted in each heater to maintain a uniform plate
temperature of 50 ºC (in W)
b. the wall temperature on strip number 25 when the power is adjusted in each heater to maintain a
uniform heat flux (equal to that on strip 25 from the first part) over the entire plate (in ºC).
Approach:
Each heater is individually controlled so that the
temperature or heat flux distribution can be set. To
find the total power dissipated by the heaters, we need
to calculate the heat transfer coefficient for the two
different boundary conditions.
Assumptions:
1. Air is at one atmosphere.
2. Radiation is ignored.
Solution:
a) The power from the plate can be calculated from Q = hA (Tw − T∞ )
We ignore radiation. The heat transfer coefficient is determined from the appropriate correlation. We need the
Reynolds number: Re = ρ V∞ L µ . The properties are evaluated (Appendix A-7) at the film temperature
Tfilm = (T∞ + Tw)/2 = (25 + 50)/2 = 37.5 oC, ρ = 1.137 kg/m3, k = 0.0270 W/mK, µ = 1.89 × 10-5 N⋅s/m2, Pr =
0.706
Re =
(1.137 kg m3 ) ( 4m s )( 2m ) (1Ns2 kgm ) = 481, 000
1.89 ×10-5 N ⋅ s/m 2
This is laminar flow. The average heat transfer coefficient on a flat plate is obtained from
Nu k 410 ( 0.0270 W/mK )
NuL = 0.664 Re1/L 2 Pr1/ 3 = 0.664 ( 481000 ) ( 0.706 ) = 410 → h = L =
1/ 2 1/ 3
= 5.54W/m 2 K
L 2m
Q = ( 5.54 W m 2 K ) (100 )( 0.02m )( 0.25m )( 50 -25 ) o C = 69.2W Answer
b) For the uniform heat flux situation, we need to calculate the local heat transfer coefficient on strip number 25
in part (a) to calculate the local heat flux. Then, assuming that heat flux is uniform over the whole plate, we
calculate the heat transfer coefficient with the correlation for the uniform heat flux boundary condition. We
assume that the heat transfer coefficient at the center of the strip is representative of the whole strip.
(1.137 )( 4 ) ⎡⎣ 24( 0.02 ) + 0.02 / 2⎤⎦
Rex = = 118, 000
1.89 ×10-5
The local heat transfer coefficient with a uniform wall temperature is:
h x (102 )( 0.0270 )
Nu x = x = 0.332 Re1/x 2 Pr1/ 3 = 0.332 (118, 000 ) ( 0.706 ) = 102 → hx =
1/ 2 1/ 3
= 5.59W/m 2 K
k 24 ( 0.02 ) + 0.02 / 2
So the heat flux on strip number 25 is q ′′ = h (Tw − T∞ ) = ( 5.59 W m 2 K )( 50 -25 ) o C =140 W/m 2
The heat transfer coefficient for a laminar flow with a uniform heat flux boundary condition is
(139 )( 0.0270 )
Nu x = 0.453 Re1/x 2 Pr1/ 3 = 0.453 (118, 000 ) ( 0.706 ) = 139 →
1/ 2 1/ 3
hx = = 7.63 W/m 2 K
24 ( 0.02 ) + 0.02 / 2
So the surface temperature is from q ′′ = h (Tw − T∞ )
Tw = T∞ + q ′′/h = 25 o C + (140 W m 2 ) / ( 7.63W/m 2 K ) = 43.3 o C Answer
12- 1
12-2 Many schemes have been proposed to supply arid regions with fresh water. One plan involves towing
icebergs from the polar regions to dry regions that need fresh water. Consider an iceberg, 1000-m long and
500-m wide, that is towed through 10 ºC water at a velocity of 1km/hr. The density of ice is 917 kg/m3 and
the heat of fusion is 333.4 kJ/kg. Determine:
a. the average rate at which the flat bottom of the iceberg will melt (in mm/hr)
b. how much ice will melt if the voyage is 1500 km long (in kg).
Approach:
We define a control volume to encompass the
iceberg. Heat transfer from the sea water melts the
ice. An energy balance will equate the heat transfer
with the change in the iceberg’s internal energy.
The heat transfer rate is calculated once a heat
transfer coefficient is estimated.
Assumptions:
1. The system is closed with no work or potential
or kinetic energy effects.
2. The bottom of the iceberg acts as a flat plate.
Solution:
a) The average melting rate dH/dt is determined from an energy balance on a closed system that is defined as the
iceberg. Assuming the bottom of the iceberg acts as a flat plate, there is no work, no mass flow across the control
volume boundary, and negligible potential and kinetic energy changes. Conservation of energy gives us
dU d d
Q= = ( musf ) = ( ρ LWH usf )
dt dt dt
dH Q hL ( LW )(T∞ − Tice ) hL (T∞ − Tice )
= = =
dt ρ usf LW ρ usg LW ρ usf
The average heat transfer coefficient requires the Reynolds number, so for water from Appendix A-6 at the Tfilm =
(T∞ + Tice)/2 = (10 + 0)/2 = 5 oC, ρ = 999.9 kg/m3, k = 0.578 W/mK, µ = 15.0 × 10-4 N⋅s/m2, Pr = 10.91
ρ V∞ L ( 999.9 kg m ) (1km hr )(1000m 1km )(1hr 3600s )(1000m )
3
ReL = = = 1.85 ×108

µ (15.0 ×10 N ⋅ s m )( kgm 1N ⋅ s )
-4 2 2
This is highly turbulent, so ignoring the very short laminar contribution, the average heat transfer coefficient is
NuL = 0.037 ReL0.8 Pr1/ 3 = 0.037 (1.85 ×108 ) (10.91)
0.8 1/ 3
= 337, 000
NuL k ( 337, 000 )( 0.578 W mK )
hL = = =195W/m 2 K
L 1000m
dH
=
(195 W m 2 K ) (10 − 0 ) o C (1J/s 1W )
dt ( 999.9 kg m3 ) ( 333.4 kJ kg )( 1000J 1kJ )
m mm mm
= 5.85 ×10-6 = 5.85 ×10-3 5 = 21.1 Answer
s s hr
b) The total ice melted for the voyage
⎛ dH ⎞ ⎛ dH ⎞ ⎛ Lv ⎞
∆m = ρ LW ⎜ ⎟t = ρ LW ⎜ ⎟⎜ ⎟
⎝ dt ⎠ ⎝ dt ⎠ ⎝ V∞ ⎠
⎛ kg ⎞ ⎛ m ⎞ ⎛ 1,500,000m ⎞ ⎛ 1km/1000m ⎞
= ⎜ 999.9 3 ⎟ (1000m )( 500m ) ⎜ 5.85×10-6 ⎟ ⎜ ⎟⎜ ⎟
⎝ m ⎠ ⎝ s ⎠ ⎝ 1km/hr ⎠ ⎝ 1hr/3600s ⎠
=1.58×1010 kg Answer
12- 2
12-3 Two brothers have rooms side by side in a flat-roofed mobile home. The older brother continually
complains that his room is colder than that of the younger brother. The older brother decides to add more
heating to his room. As shown below, the first room is 4-m long and the second one is 3-m long; each is 4-
m deep into the plane of the page. The roof thickness is 0.25 m with a thermal conductivity of 1.2 W/m·K.
The outside wind is parallel (out of the plane of the page) to the roof at a velocity of 20 km/hr at -10 ºC, the
inside temperature is to be maintained at 21 ºC, and the inside heat transfer coefficient is 7.5 W/m2·K.
Determine the heat loss from the roof of each of the two rooms (in W).
Approach:
The basic heat transfer rate equation ( Q = ∆T/Rtot) can
be used to calculate the heat transfer rate. The average
heat transfer coefficient for each of the two rooms
must be determined.
Assumptions:
1. The heat transfer is one dimensional.
3. The roof can be treated as a flat plate.
Solution:
∆T TIN − T f
The heat transfer can be calculated with Q=
=
Rtot 1 t 1
+ +
hi Ai kA ho Ao
We need the average heat transfer coefficients for the two rooms, which requires the Reynolds number with air
properties from Appendix A-7 evaluated at Tfilm = (-10 + 21)/2 = 5.5 oC = 278.5 K, ρ = 1.267 kg/m3, k = 0.0245
W/mK,
µ = 1.69 × 10-5 N⋅s/m2, Pr = 0.714
ρ VL1 (1.267kg/m ) ( 5.56m/s )( 4m ) ( N ⋅ sm kgm )
3 2
ReL1 = = =1.667 ×106

µ 1.69 ×10-5 N ⋅ s/m 2
Similarly, ReL 2 = 2.918 ×106
This is turbulent flow, and ignoring the minor laminar contribution for Room A
NuL = 0.037 ReL0.8 Pr1/ 3 = 0.037 (1.667 ×106 ) ( 0.714 )
0.8 1/ 3
= 3140
Nuk ( 3140 )( 0.0245W/mK )
hA = = =19.2W/m 2 K
L1 4m
For Room B, we need to evaluate the average h from L1 = 4m to L2 = 7 m
L2
1 1 1 k
hB = ∫ hx dA =
A W ( L2 − L1 ) ∫ hx Wd = ∫
L2 − L1 L1 x
( 0.0296 Rex0.8 Pr1/ 3 )dx
0.037k
hB = ⎡ ReL0.8 − ReL0.8 ⎤ Pr1/ 3
L2 − L1 ⎣ 2 1 ⎦
0.037 ( 0.0245 ) ⎡
( 2.918 × 106 ) − (1.667 × 106 ) ⎤⎥ ( 0.714 ) = 14.5W m 2 K
0.8 0.8 1/ 3
= ⎢
( )
7 − 4 ⎣ ⎦
⎡⎣ 21 − ( −10 ) ⎤⎦ K
QA = = 1260W Answer
1 0.25m 1
+ +
( 7.5 W m2 K ) ( 4m )( 4m ) (1.2W mK )( 4m )( 4m ) (19.2W m2 K ) ( 4m )( 4m )
⎡⎣ 21 − ( −10 ) ⎤⎦
QB = = 906W Answer
1 0.25 1
+ +
( 7.5)( 3)( 4 ) (1.2 )( 3)( 4 ) (14.5 )( 3)( 4 )
12- 3
12-4 Rolling mills are used to reduce the thickness of steel plates to create thin steel strips. The metal must be at
a high temperature so that the power (force) required to reduce the metal thickness is not excessive and so
that the desired material properties are obtained. Consider a 304 stainless steel strip 3-mm thick leaving a
rolling mill at 1000 ºC at a speed of 20 m/s. A length of 50 m is exposed to air at 35 ºC. Convective heat
transfer occurs on both the top and bottom surfaces of the strip. Ignoring radiation and axial conduction in
the steel, estimate the temperature of the strip when it reaches 50 m from the roller (in ºC).
Approach:
We will use conservation of energy on the strip,
similar to what was done in Section 12.5 for a heat
exchanger with a constant temperature fluid. Note
that this “flow” of 304 stainless steel is analogous to
the flow of a fluid. When we analyze this metal
working process, we will obtain the same solution as
we obtained for the constant temperature heat
exchanger.
Assumptions:
1. The heat transfer coefficient is constant.
4. The flow can be approximated as flow over a flat
plate.
Solution:
The equation for the outlet temperature is Tout = T f + (Tin − T f )exp ( − hA mc p )
We have assumed that h is uniform over the length of the metal strip and that radiation can be ignored
We assume that flow over a flat plate is a reasonable model of the flow. The average heat transfer coefficient
requires the Reynolds number with air properties from Appendix A-7 evaluated at (approximately) Tfilm = (1000 +
35)/2 = 517.5 oC = 790.5 K, so ρ = 0.4465 kg/m3, µ = 3.60 × 10-5 N⋅s/m2, k = 0.0573 W/mK, Pr = 0.688
ρ VL ( 0.4465 kg/m ) ( 20m/s )( 50m )
3
Re = = = 1.24 ×107
µ 3.60 ×10-5 N ⋅ s/m 2
This is turbulent flow, so ignoring the minor laminar contribution
Nu = 0.037 ReL0.8 Pr1/ 3 = 0.037 (1.24 ×107 ) ( 0.688 )
0.8 1/ 3
=15, 450
Nuk (15450 )( 0.0573W/mK )
h= = = 17.7W/m 2 K
L 50m
From Appendix A-2 for 304 SS, at 1200 K, cp = 640 J/kg K, ρ = 7900 kg/m3
⎡ - (17.7W/m 2 K ) ( 50m ) W ⎤
Tout = 35 C + (1000 -35 ) K exp
o
⎢ ⎥ = 997 o C Answer
⎢⎣ ( 7900kg/m ) ( 0.003m ) W ( 20m/s ) 640 J/kgk ⎥⎦
3
Comments:
No radiation is not a good assumption because of the high metal temperature.
The cooling rate with air is very low. Water is more typically used.
12- 4
12-5 The failure rate of computer chips increases with increasing operating temperature. Consider a 15-mm by
15-mm chip that is cooled on its top surface by a 5 m/s flow of 25 ºC air. Any heat transfer from its bottom
surface to the circuit board is ignored. Because of the chip construction, the electrical power dissipated in the
chip results in a uniform heat flux over the surface of the chip. The maximum temperature that any part of
the chip can experience is 80 ºC. Determine:
a. the maximum allowable chip power (in W)
b. the maximum allowable chip power if this chip is the fifth in a column of identical chips all mounted
flush to the surface with no space between the chips (in W).
Approach:
With uniform heat flux on each chip, the maximum
temperature will occur at the down steam end. The
local heat transfer coefficient should be used in the
rate equation ( Q = hA ∆T) to determine the maximum
permissible power.
Assumptions:
2. The flow can be treated as flow over a flat plate.
Solution:
The maximum allowable power to keep all parts of the chip less than 80 oC can be calculated with
Q = hx A ( (Ts − T f ) where hx is the local heat transfer coefficient. We assume flow over a flat plate with a uniform
heat flux is a reasonable model of the flow. The Reynolds number must be evaluated at Tfilm ~ (25 + 80)/2 = 52.5
o
C = 325.5 K. So from Appendix A-7: ρ = 1.084 kg/m3, k = 0.0281 W/mK, µ = 1.96 × 10-5 N⋅s/m2, Pr = 0.703
ρ VL (1.084kg/m ) ( 5m/s )( 0.015m ) ( N ⋅ s kgm )
3 2
a) Re = = = 4150
µ 1.96 ×10-5 N ⋅ s/m 2
This is laminar flow, so
Nu x = 0.453 Re1/x 2 Pr1/ 3 = 0.453 ( 4150 ) ( 0.703)
1/ 2 1/ 3
= 25.9
Nuk ( 25.9 )( 0.0281W/mK )
hx = = = 48.6W/m 2 K
x 0.015m
Q = ( 48.6W/m 2 K ) ( 0.015m )( 0.015m ) [80 − 25]K = 0.60W Answer
b) For the fifth chip, L = 5(0.015 m) = 0.075 m
Rex = 20750
Nux = 58.0
( 58.0 )( 0.0281)
hx = = 21.7W/m 2 K
0.075
Q = (21.7) (0.015) (0.015) (80 – 25) = 0.27 W Answer
Comments:
Because the heat transfer coefficient decreases with length, there is a dramatic decrease in allowable power if the
chip is in the fifth place rather than the first. This illustrates well why you should be aware of whether a local or
an average heat transfer coefficient is needed in a problem.
12- 5
12-6 The walls of a house are constructed of an exterior sheathing, insulation, framing timber, and drywall, and
their composite resistance is estimated to be 4.15 m2·K/W. A winter wind blows parallel to the 3-m high, 18-
m long wall. The wind velocity is 30 km/hr, and its temperature is -5 ºC. The heat transfer coefficient at the
interior of the house is 5 W/m2·K. For an inside air temperature of 21 ºC, determine the heat transfer rate
through the wall (in W).
Approach:
The basic heat transfer rate equation ( Q = ∆T/Rtot) is
used to calculate the heat transfer rate. The total
thermal resistance is composed of the inside
convective resistance, the wall resistance, and the
external convective resistance, which must be
determined from appropriate heat transfer coefficient
correlations.
Assumptions:
2. The external flow can be treated as flow over a flat
plate.
3. The air is at one atmosphere.
Solution:
∆T Tin − T f
The heat transfer rate equation is Q = =
Rtot Rconv ,in + Rwall + Rconv , out
1 1
where Rconv ,in = = = 0.00370 K/W
hi A ( 5 W m 2 K ) ( 3m )(18m )
′′
Rwall 4.15m 2 K/W
Rwall = = = 0.0769K/W
A ( 3m )(18m )
1
Rconv ,out =
ho A
We have forced convection over an assumed flat plate and need the Reynolds number. For Tfilm we assume
the temperature of the outside surface is Ts ~ 0 oC, Tfilm = (0-5)/2 = -2.5oC = 270.5 K and the air properties (from
Appendix A-7 by interpolation) are: ρ = 1.305 kg/m3, µ = 1.63 × 10-5 N⋅s/m2, k = 0.0238 W/mK, Pr = 0.716
ρ VL (1.305 kg m ) ( 8.33m/s )(18m ) ( Ns kgm )
3 2
ReL = = = 1.20 × 107

µ (1.63×10−5 N ⋅ s m2 )
This is turbulent flow. Ignoring the minor laminar contribution and using the correlation for the average heat
transfer coefficient over the total length:
Nu L = 0.037 ReL0.8 Pr1/ 3 = 0.037 (1.20 × 107 ) ( 0.716 )
0.8 1/ 3
= 15,300
Nuk (15300 )( 0.0238 W/mK )
h= = = 20.2 W/m 2 K
L 18m
1
Rconv ,out = = 0.00092 K W
( 20.2W/m K )( 3m )(18m )
2
⎡⎣ 21 − ( −5 ) ⎤⎦ K
Q= = 319 W Answer
[0.00370 + 0.0769 + 0.00092] K/W
12- 6
12-7 One wall of an older office building (6-m high and 30-m long) is all glass 7-mm thick. Wind blows parallel
to it at 20 km/hr and 5 ºC. The inside surface temperature of the glass is 20 ºC. Determine:
a. the heat transfer rate from the glass (in W)
b. the heat transfer rate if the wind velocity is tripled (in W).
Approach:
The basic heat transfer rate equation ( Q = ∆T/Rtot)
coefficient over the glass.
Assumptions:
3. Ignore the laminar contribution to the heat transfer
coefficient.
Solution:
a) The heat transfer rate equation is:
∆T Tw − T f Tw − T f
Q= = =
Rtot Rglass + Rconv t kA + 1 hA
From appendix A , for plate glass k = 1.4 W/mK.
The heat transfer coefficient requires the Reynolds number with air properties from Appendix A-7 evaluated at the
film temperature. Assuming Ts ~ 10oC, Tfilm = (10 + 5)/2 = 7.5oC = 280.5K, so that ρ = 1.258 kg/m3, µ = 1.70 ×
10-5 N⋅s/m2, k = 0.02465 W/mK, Pr = 0.714.
ρ VL (1.258kg m ) ( 5.56 m/s )( 30 m ) ( Ns kg m )
3 2
Re = = = 1.23 × 107
µ 1.70 × 10-5 N ⋅ s/m 2
This is turbulent flow, and ignoring the minor laminar contribution, the average heat transfer coefficient for flow
over a flat plate is:
NuL = 0.037 ReL0.8 Pr1/3 = 0.037 (1.23 × 107)0.8 (0.714)1/3 = 15,600
Nuk (15600 )( 0.02465 W/mK ) W
h= = = 12.8 2
L 30 m mK
( 20 − 5 )K
Q= = 32,500 W Answer
0.007 m 1
+
(1.4 W mK ) ( 6m )( 30 m ) (12.8 W m 2 K ) ( 6 m )( 30 m )
b) If V2 = 3 V1 , ReL,2 = 3 ReL,1
NuL,2 = (ReL,2 / ReL,1)0.8 NuL,1 = 37,600
h2 = 30.9 W/m2K
Q = 72,300 W Answer
Comments:
This is a substantial energy loss. The use of a double pane window would probably more than halve this energy
loss.
12- 7
12-8 The roof of a minivan can be approximated as a flat plate, 2-m wide and 3.5-m long. The sun beats down
on the roof such that the net solar radiation absorbed is 350 W/m2. If the ambient air is 32 ºC, the car is
moving at 100 km/hr, and the inside surface of the roof is heavily insulated, determine the steady-state
temperature of the roof (in ºC). Ignore radiation emitted from the roof.
Approach:
An energy balance on the roof will equate the
incoming solar energy with the outgoing convective
heat transfer. The convective heat transfer coefficient
must be evaluated to calculate the convective heat
transfer.
Assumptions:
1. The control volume is steady.
2. Potential and kinetic energy effects are negligible.
3. No work is done on or by the control volume.
4. The flow is over a flat plate.
5. Laminar flow can be ignored.
Solution:
We define a control volume around the roof and assume steady, negligible potential and kinetic energy effects,
and no work. Applying conservation of energy, we obtain:
Q=0
Recognizing that this heat transfer is the net heat transfer:
Qsolar − Qconv = 0
′′ A − hA (Ts − T f ) = 0
qsolar
′′ h
Ts = T f + qsolar
The convective heat transfer coefficient requires the Reynolds number with properties evaluated at
T film = (Ts − T f ) 2 . Assuming Ts ≈ 45 o C , T film = (32 + 45) / 2 ≈ 38.5 o C , by interpolation the air properties from
Appendix A-7 are: ρ = 1.133 kg/m3; µ = 1.90×10-5 N·s/m2; k = 0.0271 W/mK; Pr = 0.705
ρ VL (1.133kg m ) ( 27.8 m s )( 3.5 m ) ( Nis kg im )
3 2
ReL = = = 5.80 × 106

µ -5
1.90×10 Nis m
2
This is turbulent flow. Assuming flow over a flat plate and ignoring the minor laminar contribution, the average
heat transfer coefficient is:
Nu = 0.037 ReL0.8 Pr1 3 = 0.037 ( 5.80 × 106 ) ( 0.705)
0.8 13
= 8480
Nu ik ( 8480 )( 0.0271W miK )
h= = =65.7 W m 2 iK
L 3.5 m
350 W m 2
Ts = 32 o C+ =37.3 o C Answer
65.7 W m 2 iK
Comments:
Our assumed surface temperature is close enough, or we could iterate one time for an improved answer.
12- 8
12-9 Power transformers change the voltage of electricity, but the devices are not 100% efficient. Dissipated
heat must be removed from transformers so that they do not reach a temperature that could damage them.
Consider a transformer that dissipates 30 W. It is 10-cm wide and 20-cm long with 8 fins 2-cm tall, 2-mm
thick, and 20-cm long evenly distributed across the surface of the transformer. Air at 25 ºC is blown
parallel along the length of the fins. Assume the fins have a fin efficiency of 100%. Ignoring radiation,
and for a base temperature less than 65 ºC, determine the minimum air velocity required (in m/s).
Approach:
The heat transfer rate equation ( Q = hA∆T) can
be used to determine the required heat transfer
coefficient, h. Once that is known, the
appropriate correlation can be used to calculate
the velocity.
Assumptions:
3. The fin efficiency is 100%.
4. The flow can be approximated as that over a
flat plate.
Solution:
Q
The heat transfer rate equation is: Q = hA (Ts − T f ) ⇒ h =
A (Ts − T f )
2
where A = LW + 2NHL = (0.2m)(0.1m) + 2(8)(0.02m)(0.2m) = 0.084 m
30W
h= = 8.93 W/m 2 K
( 0.084m2 ) ( 65- 25) K
This is the minimum h needed to ensure Ts < 65oC. We assume flow over a flat plate is applicable to this
situation, and evaluate air properties from Appendix A-7 at Tfilm = (65 + 25)/2 = 45 ºC, ρ = 1.110 kg/m2, µ =
1.93 × 10-5 N⋅s/m2, k = 0.0276 W/mK, Pr = 0.704.
An average heat transfer coefficient is needed. We assume the flow is laminar. This will need to be
checked for an isothermal flat plate.
2
⎡ hL ⎤
NuL = hk L = 0.664 Re1/L 2 Pr1/ 3 → ReL = ⎢ 1/ 3 ⎥
⎣ 0.664 k Pr ⎦
(8.93W/m2 K ) ( 0.2m )
2
⎡ ⎤
ReL = ⎢ ⎥ = 12, 000
⎢⎣ ( 0.664 )( 0.0276W/mK )( 0.704 ) ⎥⎦
13
Assuming transition to turbulence occurs at a Reynolds number of about 500,000, this flow is laminar, so our
assumption is valid. Therefore,
ReL =
ρ VL
→ V=
µ ReL
=
(1.93×10−5 N ⋅ s/m2 ) (12, 000 ) =1.04 m/s Answer
µ ρL (1.110 kg/m 2 ) ( 0.2 m ) ( Ns2 kgm )
12- 9
12-10 Solar-powered planes have been designed to be able to stay aloft for very long times. Proposed uses
include meteorology and surveillance. Photovoltaic (PV) cells are mounted on the top surface of the wing.
The PV panel is 1.5 m wide and 8-m long. The solar energy absorbed but not converted to electricity
equals about 850 W/m2, and the PV cell conversion efficiency decreases with increasing temperature.
Determine the temperature of the trailing edge of the panel when the plane flies at 110 km/hr at 5000 m
where the pressure is 54 kPa and temperature is 256 K.
Approach:
The heat transfer rate equation ( Q =hA ∆T) can be
used. The heat transfer coefficient must be
determined.
Assumptions:
1. Laminar effects at the leading edge are ignored.
2. Flow behaves as flow over a plate plate.
3. Air is an ideal gas.
Solution:
An energy balance on the PV cells gives: Qsolar = Qconv ⇒ ′′ A = hA (TPV − T f
qsolar )
′′ /h
TPV = T f + qsolar
We assume that flow over a flat plate with a uniform heat flux wall condition is appropriate. Because we want the
PV cell temperature at the trailing edge of the panel, the local heat transfer coefficient is used.
The fluid properties from Appendix A-7 are evaluated at Tfilm. We assume TPV ~ 274 K, so Tfilm = (274 + 256)/2 =
265 K, ρ = 0.710 kg/m3, µ = 1.60 × 10-5 N⋅s/m2, k = 0.0235 W/mK, Pr = 0.718
ρ VL ( 0.710 kg m ) ( 30.6 m s )(1.5m ) ( N ⋅ s kgm )
3 2
Re = = = 2.04 ×106
µ 1.60 ×10-5 N ⋅ s/m 2
This is turbulent flow, so for the local heat transfer coefficient
Nu x = 0.0308 Rex0.8 Pr1/ 3 = 0.0308 ( 2.04 ×106 ) ( 0.718 )
0.8 1/ 3
= 3070
Nuk ( 3070 )( 0.0235 W mK )
h= = = 48.2W/m 2 K
L 1.5m
850W/m 2
TPV = 256K + = 273.6K Answer
48.2W/m 2 K
12- 10
12-11 For a quick solution to an overheating problem, brass rods, 6-mm in diameter and 5-cm long, are attached
to a surface of a power supply. Air at 20 ºC and 5 m/s is blown perpendicular to the tubes. If the base
temperature must not exceed 75 ºC, how much power can be dissipated by one rod (in W)?
Approach:
This is a problem involving conduction heat transfer
in a pin fin. The convective heat transfer must be
evaluated first, and then the fin heat transfer
evaluated.
Assumptions:
2. The fin is one dimensional.
3. The flow is as over a circular tube.
Solution:
A pin fin was evaluated in Chapter 11. Heat transfer from it can be calculated with:
Q = (Tb − T f ) hPk Ax tanh ( mL )
where m = hp/k Ax . This is for an adiabatic tip. To account for convection from the tip, we use a corrected
length. L* = L + Ax/p = L + D/4.
We assume the forced convection heat transfer coefficient for flow over a cylinder is applicable and is
uniform over the fin. The Reynolds number is determined with air properties from Appendix A-7 evaluated at
Tfilm = (Tb + Tf)/2 = (75 + 20)/2 = 47.5 oC = 320.5 K, ρ = 1.101 kg/m3, µ = 1.94 × 10-5 N⋅s/m2, k = 0.0278 W/mK,
Pr = 0.704.
(This temperature is assumed to be a reasonable “average” over the length of the fin.)
ρ VD (1.101kg m ) ( 5 m s )( 0.006m ) (1Ns kgm )
3 2
Re = = = 1700
µ (1.94 ×10-5 N ⋅ s/m2 )
From Table 12-1, Nu = 0.683 Re0.466 Pr1/3
Nuk (19.5 )( 0.0278W/mK )
Nu = 0.683 (1700)0.466 (0.704)1/3 = 19.5 → h= = = 90.2W/m 2 K
D 0.006m
From Appendix A-2 for brass, k ≈110 W/mK
mL * =
( 90.2W/m K )π ( 0.006m )
2
⎛ 0.006 ⎞
⎜ 0.05 + ⎟ m = 1.204
(110W/mK )(π 4 )( 0.006m )
2
⎝ 4 ⎠
0.5
⎡⎛ W ⎞ ⎛ W ⎞⎛ π ⎞ 2⎤
Q = ( 75 − 20 ) K ⎢⎜ 90.2 2 ⎟π ( 0.006m ) ⎜110 ⎟ ⎜ ⎟ ( 0.006m ) ⎥ tanh (1.204 )
⎣⎝ m K ⎠ ⎝ mK ⎠⎝ ⎠
4 ⎦
= 3.34W Answer
12- 11
12-12 If the brass rods in P 12-11 are replaced by rectangular aluminum alloy fins, 10-cm wide (in the direction
of air flow), 2-mm thick, and 5-cm long, determine how much power can be dissipated by one fin (in W).
Approach:
This is a problem involving conduction heat transfer
in a rectangular fin. The convective heat transfer
coefficient must be evaluated first, and then the fin
heat transfer evaluated.
Assumptions:
2. The fin is one dimensional.
3. The flow is as over a flat plate.
Solution:
A fin with a constant cross-sectional area was evaluated in Chapter 11. Heat transfer from it can be calculated
with
Q = (Tb − T f ) h pkAx tanh ( mL )
where m = h p/k Ax . This is for an adiabatic tip. To account for convection from the tip, we use a corrected
length L* = L + Ax/p = L + t/2. (Heat transfer from the two end edges is ignored.)
We assume forced convection heat transfer over a flat plate is applicable and is uniform over the fin. The
Reynolds number is determined with air properties from Appendix A-7 evaluated at a film temperature that is
reasonable over the length of the fin:
Tfilm = (75 + 20)/2 = 47.5 oC = 320.5K: ρ = 1.101 kg/m3, µ = 1.94 × 10-5 N⋅s/m2, k = 0.0278 W/mK, Pr = 0.704.
ρ VW (1.101kg m ) ( 5 m s )( 0.1m ) ( N ⋅ s kgm )
3 2
Rew = = = 28, 400

µ 1.94 ×10-5 N ⋅ s/m 2
This is laminar flow, so for the length average heat transfer coefficient:
hW
Nuw = 0.664 Re1/w 2 Pr1/ 3 = 0.664 (28400)1/2 (0.704)1/3 = 99.5
k
Nuk ( 99.5 )( 0.0278 W/mK )
h= = = 27.7W/m 2 K
W 0.10m
From Appendix A-2 for aluminum, k ≈ 177 W/mK
( 27.7 W m K ) ( 2 )( 0.10m )
2
mL * = ( 0.05 + 0.002/2 ) m = 0.638

(177W mK )( 0.002m )( 0.10m )
0.5
⎡⎛ W ⎞ ⎛ W ⎞ ⎤
Q = ( 75 − 20 ) K ⎢⎜ 27.7 2 ⎟ 2 ( 0.10m ) ⎜177 ⎟ ( 0.002m )( 0.10m ) ⎥ tanh ( 0.638 )
⎣⎝ m K⎠ ⎝ mK ⎠ ⎦
=13.7W Answer
12- 12
12-13 In an electric hair drier, air at 25 ºC flows with a velocity of 5 m/s perpendicular to a nichrome heating
element. The heating element is 1-mm in diameter and 40-cm long with a resistance of 1.38 Ω/m. The
wire temperature cannot exceed 430 ºC so that the wire will not lose strength and sag. Determine:
a. the total power dissipated (in W)
b. the electric current in the wire (in A).
Approach:
This is forced convection perpendicular to a cylinder.
With the heat transfer coefficient, geometry and
temperatures, we can calculate the maximum power to
not exceed the temperature limit. Once the power is
known, Ohm’s Law can be used to calculate current.
Assumptions:
3. The flow is perpendicular to a cylinder.
Solution:
a) We assume radiation can be ignored, so that the heat transfer from the wire is Q = hA (Ts − T∞ )
To obtain h, we need the Reynolds number which requires the fluid properties. From Appendix A-7 at Tfilm = (T∞
+ Ts)/2 = (25 + 430)/2 = 227.5 oC = 500.5K, ρ = 0.7053 kg/m3, k = 0.04038 W/mK, µ = 2.671 × 10-5 N⋅s/m2, Pr
= 0.680
ρ V∞ D ( 0.7053kg m ) ( 5 m s )( 0.001m ) (1Ns kgm )
3 2
Re = = =132
µ 2.671×10-5 N ⋅ s m 2
Nu = 0.193 Re0.618 Pr1/ 3 = 0.193 (132 ) ( 0.680 )
0.8 1/ 3
From Table 12-1 = 8.44
Nuk ( 8.44 )( 0.04038W/mK )
h= = = 341W/m 2 K
D 0.001m
Q = ( 341W m 2 K ) π ( 0.001m )( 0.4m )( 430 − 25 ) C =173W
o
Answer
2
b) Electrical power is Q = I R where I = current and R is resistance
173W
I = Q/R = = 17.7A Answer
(1.38 Ω m )( 0.4m )
Comments:
Because of the high wall temperature, the assumption that radiation can be ignored should be checked.
12- 13
12-14 An existing electric powerline is being examined to determine if a higher current can be used. You are
asked to calculate the maximum power dissipation per meter of length that is permissible by Joulean
heating such that the inside surface of the cable insulation does not exceed 77 ºC. The copper wire in the
cable is 2-cm in diameter and the insulation is 0.1-cm thick with a thermal conductivity of 0.08 W/m·K.
The wind velocity perpendicular to the wire is 5 km/hr and the air temperature is 27 ºC. Neglecting
radiation, determine the allowable power (heat generation rate) per unit length (W/m). [J12-001]
Approach:
We must take into account two thermal resistances
(conduction and convection) to determine the heat
transfer rate from the inner surface of the insulation to
the air. The heat transfer coefficient is calculated with
the appropriate correlation.
Assumptions:
Solution:
∆T Ti − T∞
The heat transfer rate is calculated with Q= =
Rtot Rtot
1 1n( D2 /D1 ) 1 1n( D2 /D1 )
where Rtot = + = +
hA 2π kL hπ D2 L 2π kL
Combining these two equations and solving for heat transfer per unit length
Q Ti − T∞
=
1 + ( 2
L 1n D − D1 )
hπ D2 2π k
To obtain h, we need the Reynolds number. Properties should be evaluated at the film temperature, but we do not
know it, so we estimate Ts ~ 67°C
Tfilm = (Ts + T∞)/2 = (27 + 67)/2 = 47 °C = 320 K
From Appendix A-7 for air at 320 K (by interpolation)
ρ = 1.110 kg/m3, k = 0.0278 W/mK, µ = 1.94 × 10-5 N-s/m2, Pr = 0.704
kg km 1000m 1hr
ρ V∞ D (1.110 m3 ) (5 hr ) ( 1km ) ( 3600s ) (0.022m)
Re = = = 1750
µ 1.94×10-5 Ν-s/m 2
From Table 12-1 for the heat transfer coefficient:
Nu = 0.683Re0.466Pr1/3 = 0.683(1750)0.466(0.704)1/3 = 19.7
Nuk (19.7 )( 0.0278 W mK )
h= = =24.9 W m 2 K
D 0.022m
Q ( 77 − 27 ) K W
= = 64.9 Answer
L 1 1n (0.022 / 0.02) m
+
( 24.9 W m 2 K) π (0.022m ) 2π ( 0.08 W/mK )
Check on Ts
Q = hA (Ts - T∞) = hπD2L(Ts - T∞)
(Q / L) 64.9W/m
Ts = T∞ + = 27o C + = 64.7o C
hπ D2 ( 24.9 W m 2 K ) π (0.022m)
This assumed Ts is close enough to the estimated temperature, so we do not need to reevaluate the film
temperature.
12- 14
12-15 The insulation on a 15-cm steam pipe deteriorates over time and is to be removed and replaced. The outer
surface of the steam pipe is at 110 ºC. Air at -6 ºC blows perpendicular to the pipe at 40 km/hr. Determine:
a. the heat transfer rate per unit length of pipe if it is left bare (in W/m)
b. the heat transfer rate per unit length of pipe if 4-cm insulation (k = 0.04 W/m·K) is applied to the
pipe (in W/m).
Approach:
This is forced convection perpendicular to a circular
cylinder, with and without insulation. The basic heat
transfer rate equation can be used to find the heat
transfer per unit length.
Assumptions:
Solution:
The basic heat transfer rate equation is:
∆T Ts − T f Ts − T f
Q= = =
Rtot Rconv + Rins 1 hπ DL + 1n (D2 / D1 ) 2π kL
Q Ts − T f
or =
L 1 hπ D + 1n (D2 /D1 ) 2π k
a) With no insulation, D2 = D1 , D = D1, and Rins = 0. Evaluation of the forced convection heat transfer
coefficient requires the Reynolds number with fluid properties evaluation at the film temperature, Tfilm , (-
6+110)/2 = 52oC = 325 K. By interpolation in Appendix A-7: ρ = 1.086 kg/m3, k = 0.0281 W/mK, µ = 1.96 ×
10-5 N⋅s/m2 , Pr = 0.703
ρ VD (1.086 kg/m ) (11.1m/s )( 0.15 m ) ( Ns kg m )
3 2
Re = = = 92,300
µ 1.96 × 10-5 N ⋅ s/m 2
From Table 12-1, Nu = 0.193 Re0.618 Pr1/3
Nu = 0.027 (92300)0.805 (0.703)1/3 = 238
Nuk ( 238 )( 0.0281W/mK )
h= = = 44.7 W/m 2 K
D 0.15 m
Q ⎡⎣(110 - (-6 ) ⎤⎦ o C W
= = 2440 (without insulation) Answer
L ⎡1 ( 44.7 W m K ) π ( 0.15 m ) ⎤
2 m
⎣ ⎦
b) With insulation, the film temperature will be much lower and the fluid properties different. We should iterate
by guessing a temperature on the outside of the insulation, evaluating the properties, and proceeding as before.
Let’s guess Ts = 20oC, Tfilm = (20 – 6)/2 = 7oC = 280 K ρ = 1.261 kg/m3, k = 0.02465 W/mK, µ = 1.70 × 10-5 ,
Pr = 0.714
(1.261)(11.1) ⎡⎣0.15 + 2 ( 0.04 )⎤⎦
Re = = 189, 400
1.70 × 10−5
Nu = 0.027 (189400)0.805 (0.714)1/3 = 427 h = (427)(0.02465) / (0.23) = 45.8 W/m2K
⎡⎣110 - ( -6 ) ⎤⎦ C
o
Q W
= = 67.0
L 1 1n ( 23/15 ) m
+
(45.8 W m K ) π ( 0.23 m ) 2 π ( 0.04 W/mK )
2
We can calculate the outside wall temperature Q = hA (Ts – Tf) = hπDL (Ts – Tf)
Ts = T f +
(Q / L ) = −6 + 67.0 W/m
= − 4 oC Answer
hπ D ( 45.8 W m K )π ( 0.23m )
2
Comments:
We should recalculate with improved fluid properties if we want a more accurate result.
12- 15
12-16 A very long cylinder 25-mm in diameter is placed in a large oven whose walls are maintained at 400 ºC.
Air at 77 ºC flows perpendicular to the cylinder at a velocity of 2.5 m/s. The emissivity of the cylinder is
0.65 Determine the steady-state temperature of the cylinder (in ºC).
Approach:
Two heat transfer modes – convection and radiation –
must be taken into account. Radiation heats the
cylinder, and convection cools it. At steady state
these two heat transfer rates are equal. We must
calculate the heat transfer coefficient to obtain the
convective heat transfer.
Assumptions:
2. The cylinder is “small” relative to the larger oven.
Solution:
At steady state, we apply conservation of energy to the cylinder, and assume no work and no potential or kinetic
energy effects to obtain: Q = 0 → Qrad − Qconv = 0
Qrad = εσ A (Tw4 − Ts4 ) = εσπ DL (Tw4 − Ts4 )
Qconv = h A (Ts − T f ) = hπ DL (Ts − T f)
Therefore εσ = (Tw4 − Ts4 ) = h(T − T )
s f
To evaluate h, we need Re with fluid properties from Appendix A-7 evaluated at Tfilm.
Assuming Ts ≈227 oC = 500 K, Tfilm = (350 + 500)/2 = 425 K,
so by interpolation: ρ = 0.831 kg/m3, µ = 2.385×10-5 N⋅s/m2, k = 0.0354 W/mK, Pr = 0.686
ρ VD ( 0.831kg/m ) ( 2.5m/s )( 0.025m ) ( N ⋅ s kgm )
3 2
Re = = = 2180
µ ( 2.385×10-5 N ⋅ s/m2 )
From Table 12-1,
Nu = 0.683 Re0.466 Pr1/ 3 = 0.683 ( 2180 ) ( 0.686 )
0.466 1/ 3
= 21.6
Nuk ( 21.6 )( 0.0354W/mK )
h= = = 30.6W/m 2 K
D 0.025m
Substituting into energy equation
⎛ W ⎞ ⎛ W ⎞
( 0.65 ) ⎜ 5.67 ×10-8 2 4 ⎟ ⎣⎡( 673K ) − Ts4 ⎦⎤ = ⎜ 30.6 2 ⎟ (Ts − 350K )
4
⎝ m K ⎠ ⎝ m K⎠
Solving this equation, Ts = 513.4 K = 240.4 oC Answer
This is close to the assumed Ts, so we do not need to re-evaluate the fluid properties.
12- 16
12-17 After the extrusion of a long solid plastic rod, it is cooled by a crossflow of 30 ºC air at a velocity of 20
m/s. The rod, whose diameter is 3.5-cm, initially has a uniform temperature of 200 ºC. The plastic’s
properties are: ρ = 2300 kg/m3, cp = 850 J/kg·K, and k = 1.2 W/m·K. Determine:
a. the time required for the surface temperature of the rod to drop to 100 ºC (in s)
b. the centerline temperature at the same time (in ºC).
Approach:
This is a transient conduction problem in an infinite cylinder. We
ignore radiation, and must calculate the heat transfer coefficient.
Assumptions:
3. The one-term approximation for transient conduction is valid.
4. The heat transfer coefficient is constant during the cooling process.
Solution:
a) Transient conduction in an infinite cylinder using the one-term approximation is governed by
T ( ri t ) − T f
= C1 exp ( − λ12 τ ) J 0 ( λ1 r/r0 ) τ > ∼ 0.2
Ti − T f
where τ = α t/r02 and C1 and λ1 require Bi = hr0/k.
⎡ T ( r1 t ) − T f ⎤
1
Solving the equation for τ: τ =−
1n ⎢ ⎥
λ12 ⎢ C1 J 0 (Ti − T f ) ⎥
⎣ ⎦
To calculate h and Bi, we need the Reynolds number with fluid properties from Appendix A-7 evaluated at Tfilm.
Note that Tfilm decreases with time, so we use Tfilm ~ (150 + 30)/2 = 90 oC = 363 K so ρ = 0.972 kg/m3, µ = 2.13 ×
10-5 N⋅s/m3,
K = 0.0310 W/mK, Pr = 0.695.
ρ VD ( 0.972kg/m ) ( 20m/s )( 0.035m ) ( N ⋅ s kgm )
3 2
Re = = = 31,900
µ 2.13 ×10-5 N ⋅ s/m 2
Nu = 0.193 Re0.618 Pr1/ 3 = 0.193 ( 31,900 ) ( 0.695)
0.618 1/ 3
From Table 12-1, =104
h=
Nuk (104 )( 0.0310W/mK )
= = 92.0W/m 2 K → Bi =
( 92.0W/m K ) ( 0.035/2m ) =1.342
2
D 0.035m 1.2W/mk
From Table 11-2 by interpolation λ1 = 1.373, C1 = 1.252. From Table 11-3, J0 (λ1) = 0.5813
1 ⎡ (100 -35) ⎤
τ =− 1n ⎢ ⎥ = 0.326
(1.373) ⎣⎢ (1.252 )( 200 − 35 )( 0.5813) ⎦⎥
2
Because τ > 0.2, the one-term approximation is okay.

τ r2 k (1.2W/mk )(1J/s 1W )
t= 0 , α= = = 6.14 ×10-7 m 2 /s
α ρ C p ( 2300kg/m3 ) ( 850J/kgk )
( 0.326 )( 0.035/2m )
2
t= =162 s Answer
6.14 ×10-7 m 2 /s
b) For the centerline temperature at the same time, J0 (0) = 1
T = T f + (Ti − T f )C1 exp ( − λ12 τ )
= 35 oC + (200 – 35) K (1.252) exp [- (1.373)2 (0.326)] = 146.7 oC Answer
Comments:
Probably a better way to solve this problem is to divide the process into several time intervals and evaluate the
heat transfer coefficient and wall temperature during each interval. This would be done until the target wall
temperature of 100 oC was obtained.
12- 17
12-18 Lead shot is made by dropping molten lead (ρ = 10,600 kg/m3) from a drop tower. Each pellet, a sphere 2-
mm in diameter, is cooled as it passes through air at 10 ºC. Assume the shot falls at its terminal velocity.
The lead must be solidified from its molten state at 327 ºC to a solid state before it reaches the pool of
water at the bottom of the drop tower. The enthalpy of fusion for lead is 24.5 kJ/kg. Ignoring radiation,
determine the minimum required height of the drop tower (in m).
Approach:
This is a transient cooling problem. The lead remains
at a uniform temperature as it solidifies, and we will
use an energy balance to develop the governing
equation. This heat transfer coefficient is needed to
calculate the convective heat transfer. That requires a
velocity which will need to be calculated, too.
Assumptions:
1. The shot falls at its terminal velocity.
3. Radiation is ignore, and the heat transfer
coefficient is constant.
4. Buoyancy force is ignored.
Solution:
For a closed system defined as the lead sphere, conservation of energy is used assuming no work and no potential
or kinetic energy effects:
dU
Q= ⇒ ∫ Qdt = ∫ dU ⇒ Qt = ∆U = m ( u2 − u1 ) = − musf ⇒ hA (T f − Ts )t = − musf
dt
We assume the convective heat transfer rate is constant; heat transfer from the control volume is negative. The
time, t, required to completely solidify the lead is used with the terminal velocity to obtain the height, H, required:
H=Vt
We assume the sphere moves at the terminal velocity for the whole time, so that
− V musf − V ρ L ( 4π R 3 3 )usf − V ρ L Rusf − Vρ L Dusf
H= = = =
hA (T f − Ts ) h 4π R 2 (T f − Ts ) 3h(T f − Ts ) 6 h (T f − Ts )
The terminal velocity is determined from a force balance between drag and gravity (buoyancy is negligible).
0.5
⎛ π D2 ⎞ ⎛ ρa V 2 ⎞ ⎛ π D3 ⎞ ⎡ 4 ρ L gD ⎤
Fdrag = Fgrav ⇒ CD ⎜ ⎟⎜ ⎟ = ρL ⎜ ⎟g ⇒ V =⎢ ⎥
⎝ 4 ⎠⎝ 2 ⎠ ⎝ 6 ⎠ ⎣ 3 ρ a CD ⎦
The drag coefficient (given on Fig. 10-8) is a function of Reynolds number, so evaluating air properties
from Appendix A-7 at 10 oC: ρ = 1.247 kg/m3, µ = 1.72×10-5 N⋅s/m2, k = 0.0249 W/mK, Pr = 0.712 (at 327 oC,
µ = 3.018×10-5 N⋅s/m2). The Reynolds number is Re = ρ V D µ , which requires the velocity that we do not
know, so we iterate. We guess V , calculate Re, obtain CD, and then calculate V . When the guessed V
matches the calculated value, then we have a solution.
⎡ 4 10600 ⎛ ( 9.81m s 2 ) ( 0.002m ) ⎞ ⎤ (1.247 kg m3 ) ( V m s )( 0.002m ) =145 V
1/ 2
⎛ ⎞⎜ 14.91
V =⎢ ⎜ ⎟ ⎟ ⎥ = [ m/s ] Re =
⎢ 3 ⎝ 1.247 ⎠ ⎜ CD ⎟⎥ CD1/ 2 1.72 ×10-5 N ⋅ s/m 2
⎣ ⎝ ⎠⎦
Estimate V = 25 m/s, Re = 3630, CD ≈ 0.4, V = 23.6 m/s → V = 23.6 m/s, Re = 3420, CD ≈ 0.4, V = 23.6
m/s
The heat transfer coefficient is determined from
1/ 4
⎤ ( 0.712 )0.4 ⎡⎢ 1.72 ×10 ⎤⎥

-5
Nu = 2 + ⎡⎣ 0.4 Re1/ 2 + 0.06 Re 2 / 3 ⎤⎦ Pr 0.4 [ µ /µ s ] = 2 + ⎡0.4 ( 3420 ) + 0.06 ( 3420 )
1/ 4 1/ 2 2/3
= 29.9
⎣ ⎦ ⎣ 3.018 ×10 ⎦
-5
Nuk ( 29.9 )( 0.0249W/mK )

h= = = 372W/m3 K
D 0.002m
12- 18
− ( 23.6m/s ) (10600kg/m3 ) ( 0.002m )( 24500 J kg )
H= = 34.6m Answer
3 ( 372W/m 2 K ) (10 − 327 ) K (1J/s 1W )
12- 19
12-19 False temperature readings can be obtained from temperature sensors if they are used incorrectly or if the
effects of radiation are not taken into account. (For example, if a thermometer were used in direct sunlight
or in shade to measure air temperature, significantly different readings would be obtained.) Consider a
thermocouple that is a 1.5-mm sphere, which is used to measure the temperature of an air stream in a large
duct. The air velocity is 4 m/s. The walls of the duct are at 150 ºC. The thermocouple indicates an air
temperature of 300 ºC and has an emissivity of 0.5. Determine the actual air temperature (in ºC).
Approach:
Two heat transfer modes-convection and radiation
must be taken into account. Radiation cools the
thermocouple, and convection heats it. At steady
state, the two heat transfer rates are equal. We must
calculate the heat transfer coefficient before we can
obtain the convective heat transfer.
Assumptions:
2. The thermocouple is small relative to the duct size.
Solution:
At steady state, we apply conservation of energy to the thermocouple, and assume no work and no potential or
kinetic energy effects to obtain: Q = 0 → Qconv − Qrad = 0 . With Qconv = hA (T f − Ts ) and
Qrad = εσ A (Ts4 − Tw4 ) so that h (Tf – Ts) = εσ (Ts4 – Tw4). Solving for the fluid temperature:
εσ
T f = Ts +
h
(Ts4 − Tw4 )
To evaluate h for a sphere, we use Eq. 12-11, which uses air properties from Appendix A-7 evaluated at the bulk
fluid temperature. We assume Tbulk ~ 300 oC, so that ρ = 0.616 kg/m3, µ = 2.926 × 10-5 N⋅s/m2, k = 0.0450
W/mK, Pr = 0.680. We assume the thermocouple is a sphere, so that
ρ VD ( 0.616 kg m ) ( 4m/s )( 0.0015m ) ( N ⋅ s kgm )
3 2
Re = = =126
µ 2.926 ×10-5 N ⋅ s/m 2
Nu D = 2 + ⎡⎣ 0.4 Re1/D 2 + 0.06 ReD2 / 3 ⎤⎦ Pr 0.4 ( µ f /µs )
1/ 4
For a sphere
Assume Tsurface ≈ 320 oC, µs ~ 2.994 × 10-5 N⋅s/m2
Nu = 2 + ⎡ 0.4 (126 ) + 0.06 (126 ) ⎤ ( 0.680 ) ( 2.926 ×10-5 2.994 ×10-5 ) = 7.11
1/ 2 2/3 0.4 1/ 4
⎣ ⎦
Nuk ( 7.11)( 0.0450W/mK )
h= = = 213W/m 2 K
D 0.0015m
Therefore:
( 0.5 ) ( 5.67 x10-8 W m 2 K 4 ) ⎡
⎣(
573K ) − ( 423K ) ⎤ = 583K = 310 o C
4 4
T f = 573K + Answer
213W/m 2 K ⎦
Comments:
For a better estimate, the fluid properties should be re-evaluated at this new fluid temperature and the calculations
repeated.
12- 20
12-20 Fluid velocities are often measured with hot wire anemometers. In such a device, the temperature of a
small diameter cylinder is maintained constant by varying the electric current through it in response to
varying fluid velocity; a wheatstone bridge is used to control the current. A typical hot wire is constructed
of a 0.2-mm diameter polished platinum wire 10-mm long. Air at 23 ºC flows over the hot wire maintained
at 200 ºC. The electrical resistivity of platinum is 17 µΩ-cm. Determine the electric current required for a
velocity of: a) 1 and m/s b) 10 m/s.
Approach:
This is forced convection perpendicular to a cylinder.
With the heat transfer coefficient, geometry, and
temperatures, we can calculate the power to reach the
operating temperature. Once the power it known,
Ohm’s Law can be used to calculate the current.
Assumptions:
Solution:
a) We assume radiation can be ignored so that the heat transfer from the wire is Q = hA (Ts – Tf) where A
= π DL.
From Ohm’s Law, Q = I2R where I is the current and R is the electrical resistance, R = ρE L/Ax , where Ax = π
D2/4 and ρE is the electrical resistivity. Substituting these expressions into the main equation and solving for I:
1/ 2
⎡ h π 2 D 3 (Ts − T f ) ⎤
I =⎢ ⎥
⎢⎣ 4 ρE ⎥⎦
To obtain h, we need the Reynolds number with air properties from Appendix A-7 evaluated at Tfilm = (23 +
200)/2 = 111.5 oC : by interpolation of properties in the appendix ρ = 0.918 kg/m3, µ = 2.23 × 10-5 N⋅s/m2, k =
0.0327 W/mK, Pr = 0.691
ρ VD ( 0.918 kg/m ) (1m/s )( 0.0002m )
3
Re = = = 8.23
µ ( 2.23×10-5 N×s/m2 )( kgm Ns2 )
From Table 12-1
Nu = 0.911Re0.385Pr1/3 = 0.911(8.23)0.385(0.691)1/3 = 1.813
Nuk (1.813)( 0.0327 W/mK ) W
h= = = 296 2
D 0.0002 m mK
⎡ ( 296 W/m 2 K ) π 2 ( 0.0002 m )3 ( 200 − 23) K (100cm m ) ⎤
1/2
I =⎢ ⎥ =2.47A Answer
⎢ 4(17×10-6 Ω-cm ) ⎥
⎣ ⎦
b) For V = 10 m s , Re = 82.3
Nu = 0.683Re0.466Pr1/3 = 0.683(82.3)0.466(0.691)1/3 = 4.72
( 4.72 )( 0.0327 )
h= = 771W/m 2 K
0.0002
1/ 2
⎡ ( 771) π 2 ( 0.0002 )3 ( 200 − 23)(100 ) ⎤
I =⎢ ⎥ = 3.98A Answer
⎢⎣ 4 (17 ×10−6 ) ⎥⎦
12- 21
12-21 After a heat treating process, a 2024-T6 aluminum sphere 20-mm in diameter is removed from an oven
which is at 85 ºC. The sphere is placed in an air stream at 27 ºC that has a velocity of 10 m/s. Determine
the time required for the sphere’s temperature to cool to 40 ºC (in s).
Approach:
This is a transient conduction problem. We need to
determine if lumped systems approach will work or if
a one-dimensional transient solution is required. The
heat transfer coefficient is determined from the
appropriate correlation.
Assumptions:
1. All properties are constant.
Solution:
To determine if the lumped systems approach is valid, we need to calculate Bi = h Lchar/k
Lchar = V/A = ro/3 = (0.01m/3) = 0.00333m
For aluminum from Appendix A-2, k = 177 W/mK, cp = 875 J/kgK, ρ = 2770 kg/m3
The heat transfer coefficient requires the Reynolds number. The sphere heat transfer coefficient correlation uses
properties at the bulk temperature: so from Appendix A-7 at 300K, ρ = 1.774 kg/m3, k = 0.02624 W/mK, µ =
1.846 × 10-5 N⋅s/m2, Pr = 0.708 (at 350 K, µ = 2.075 × 10-5 N⋅s/m2)
ρ VD (1.1774 kg/m ) (10m/s )( 0.02m ) ( Ns kgm )
3 2
Re = = = 12,800
µ 1.846 ×10-5 N×s/m 2
The heat transfer coefficient is determined from:
Nu = 2 + ⎡⎣ 0.4 Re1/ 2 + 0.06 Re 2 / 3 ⎤⎦ Pr 0.4 [ µ /µs ]
1/ 4
= 2 + ⎡0.4(12800 ) + 0.06 (12800 ) ⎤ ( 0.708 ) ⎡⎣1.846 ×10−5 2.075 ×10−5 ⎤⎦

1/ 2 2/3 0.4 1/ 4
= 67.9
⎣ ⎦
Nuk ( 67.9 )( 0.02624 W/mK )
h= = = 89.1W/m 2 K
D 0.02 m
Bi =
(89.1W/m K ) ( 0.00333m ) = 0.00168
2
< 0.1
177 W/mK
so the lumped systems approach is valid:
T − Tf ⎡ −hA ⎤ ⎡ −ht ⎤
= exp ⎢ t ⎥ = exp ⎢ ⎥
Ti − T f ⎣⎢ mc p ⎦⎥ ⎣ ρ Lchar cP ⎦
Solving for time
− ρ Lchar c p ⎛ T − T f ⎞ − ( 2770kg m ) ( 0.00333 m )( 875 J/kgK ) ⎡ 40 − 27 ⎤
3
t= ln ⎜ ⎟ = ln ⎢ ⎥
h ⎜ Ti − T f ⎟
⎝ ⎠ (89.1W/m2 K ) (1J/s 1W ) ⎣ 85 − 27 ⎦
= 136 s Answer
12- 22
12-22 In an oil refinery, a steam pipe (k = 15 W/m·K) with an inside diameter of 10-cm and an outside diameter
of 11-cm is covered with 3.5-cm of insulation (k = 0.03 W/m·K). The steam is at 300 ºC. Air and
surrounding surfaces are at 17 ºC, and the air flows perpendicular to the pipe at a velocity of 2 m/s. The
heat transfer coefficient of the steam is 100 W/m2·K. Determine the heat transfer rate per unit length of
pipe (in W/m).
Approach:
The heat transfer rate equation ( Q = ∆T/Rtot) can be
used directly. The heat transfer coefficient on the
outside must be determined.
Assumptions:
Solution:
The governing equation is
∆T Tsteam -T f
Q= =
Rtot 1 1n( D2 / D1 ) 1n ( D3 /D2 ) 1
+ + +
hsteam At 2π ksteel L 2π kins L ho Ao
For flow perpendicular to the pipe, the Reynolds number with air properties from Appendix A-7 at Tfilm is required
to calculate h. Assuming Tsurface ≈ 27oC, Tfilm ≈ (17 + 27)/2 = 22oC = 295K: ρ = 1.197 kg/m2, k = 0.0258 W/mK,
µ = 1.81 × 10-5 N⋅s/m2, Pr = 0.709
ρ VD (1.197kg/m ) ( 2m/s ) ⎡⎣0.11 + 2 ( 0.035 ) ⎤⎦ m
3
Re = = = 23,800
µ (1.81×10-5 N ⋅ s/m2 )( ms2 Nkg )
From Table 12-1
Nu = 0.193Re0.618Pr1/3 = (0.193)(23800)0.618(0.709)1/3 = 87.2
Nuk ( 87.2 )( 0.0258W/mK )
h= = = 12.5W/m 2 K
D ( 0.18m )
1 1 0.00318
= =
hsteam Asteam (100W/m 2 K )π (1.10m ) L L
1n ( D 2 /Ds ) 1n ( 0.11/0.10 )
0.00101
= =
2π k steel L 2π (15W/mK ) L L
1n ( D3 /D2 ) 1n ( 0.18/0.11) 2.613
= =
2π kins L 2π ( 0.03W/mK ) L L
1 1 0.1415
= =
ho Ao (12.5 )π ( 0.18 ) L L
Q ( 300 −17 )K
= =103 W/m Answer
L ( 0.00318 + 0.00101+ 2.613+ 0.1415 ) Km/W
Comments:
The insulation dominants the total thermal resistance. Inaccuracies in the other resistances would have little
effect.
12- 23
12-23 If you have ever changed a hot incandescent light bulb, then you know that much of the power going into
the bulb is converted to heat (about 90%) rather than to light (about 10%). (Fluorescent light bulbs are
much more efficient.) All the heat is dissipated from the glass bulb. Consider an 8-cm diameter 100-W
light bulb cooled by air at 30 ºC. Both convection and radiation (ε = 0.85) cool the glass. Assuming the
surroundings are at 30 ºC for radiation purposes, determine the temperature of the glass bulb if air at 2 m/s
flows across it (in ºC).
Approach:
An energy balance on the light bulbs is needed. The
power into the glass (90% of 100 W) is removed by
convection and radiation. The heat transfer coefficient
must be evaluated.
Assumptions:
2. Radiation is from a small body to a large area.
3. The light bulb is a sphere.
Solution:
A steady energy balance on the glass gives Qin = Qrad + Qconv
Qin = εσ A (T − T
s
4
f
4
) + hA (T − T )
s f
The light bulb is assumed to be a sphere, and the surroundings are very large. The heat transfer coefficient
correlation for a sphere is based on air properties from Appendix A-7 evaluated at the bulk temperature, 300K
ρ = 1.1774 kg/m2, µ = 1.846 × 10-5 N⋅s/m2, k = 0.02624W/mK, Pr = 0.708 (assume Ts = 450K, µs = 2.484 × 10-5
N⋅s/m2).
ρ VD (1.1774kg/m ) ( 2m/s )( 0.08m )
3
Re = = = 10, 200
µ (1.846 ×10−5 N ⋅ s/m2 )( kgm Ns2 )
The heat transfer coefficient on a sphere is determined from
Nu = 2 + ⎡⎣ 0.4 Re1/ 2 + 0.06 Re 2 / 3 ⎤⎦ Pr 0.4 [ µ /µ s ]
1/ 4
= 2 + ⎡ 0.4 (10200 ) + 0.06 (10200 ) ⎤ ( 0.708 ) (1.846 ×10−5 2.484 ×10−5 )

1/ 2 2/3 0.4 1/ 4
= 58.5
⎣ ⎦
Nuk ( 58.5 )( 0.02624 W mK )
h= = = 19.2 W/m 2 K
D 0.08 m
0.9 (100W ) = ( 0.85 ) ( 5.67 ×10−8 W m 2 K 4 ) ( 4π )( 0.04m ) ⎡Ts4 − ( 300K ) ⎤ + ( 22.6 W m 2 K ) ( 4π )( 0.04m ) (Ts − 300 ) K
2 4 2
⎣ ⎦
90=9.69 × 10-10 (Ts4 − 8.1×109 ) + 0.386 (Ts − 300 )
Solving the equation for the surface temperature:
Ts = 450.3K = 177.3oC Answer
12- 24
12-24 After preheating long 316 stainless steel rods 50-mm in diameter to a uniform temperature of 1000 ºC, they
must be conveyed to another location in the plant for additional processing. The conveyor moves the rods
perpendicular to the direction of travel at a velocity of 3 m/s. The rod emissivity is 0.5, and the air and
surrounding temperatures are at 27 ºC. The centerline temperature of the rod must be greater than 900 ºC
for the next processing step. Determine:
a. the convective heat transfer coefficient at the start of the travel (in W/m2·K)
b. the radiation heat transfer coefficient at the start of the travel between one processing station and
another (in W/m2·K)
c. the allowable time for transit between the two stations assuming the total heat transfer coefficient
(radiation and convection) is the sum of those calculated in parts (a) and (b) (in s).
Approach:
The convective heat transfer coefficient for forced
convection perpendicular to a cylinder and the
radiative heat transfer coefficient both can be
evaluated with the given information. Because the
diameter is small relative to the cylinder length, we
will analyze it as an infinite cylinder. The Biot
number should be checked (with the total heat
transfer coefficient) to determine if lumped systems
approach is applicable. If not, the one-term
approximation for one-dimensional transient
conduction in an infinite cylinder will be used.
Assumptions:
Solution:
a) To obtain the convective heat transfer coefficient, we need the Reynolds number with properties evaluated at
Tfilm = (Ts + Tf)/2. Because we want h at the start of travel, Ts = 1000 oC, Tfilm ~ 787 K. From Appendix A-7 by
interpolation, ρ = 0.449 kg/m3, µ = 3.59 × 10-5 N⋅s/m2, k = 0.0571 W/mK, Pr = 0.688.
ρ VD ( 0.449kg m ) ( 3m s )( 0.050m )
3
Re = = = 1880
µ 3.59 ×10-5 N ⋅ s/m 2
From Table 12-1, Nu = 0.683Re0.466Pr1/3 = 0.683 (1880)0.466 (0.688)1/3 = 20.2
Nuk ( 20.2 )( 0.0571W/mK )
h= = = 23.1W/m 2 K
D 0.050m
b) Assuming the surroundings are very long and are at Tf , the radiative heat transfer coefficient is defined as:
hr = εσ (Ts + T f ) (Ts2 + T f2 )
With Ts = 1000 + 273 = 1273 K and Tf = 27 + 273 = 300 K
⎛ W ⎞
hr = ( 0.5 ) ⎜ 5.67 ×10-8 2 4 ⎟ (1273K + 300K ) ⎡(1273K ) + ( 300K ) ⎤ = 76.3W/m 3 K
2 2
⎝ m K ⎠ ⎣ ⎦
( h+hr ) Lchar
c) The Biot number, with a combined coefficient, is: Bi = where LChar = V/A = D/4
k
From Appendix A-2, for 316 stainless steel at 1000 K, ρ = 8238 kg/m3, k = 24.2 W/mK, cp = 602 J/kg K,
( 23.1 + 76.3) W/m 2 K ( 0.05m/4 )
Bi = = 0.051 < 0.1
24.2W/mK
The lumped systems approach is valid, and the equation governing it is:
T − Tf ⎛ hA ⎞ ⎛ ht ⎞ − ρ Lchar c p ⎛ T − T f ⎞
= exp ⎜ − t ⎟ = exp ⎜ − ⎟⎟ t= 1n ⎜
⎜ T − T ⎟⎟
→
⎜ ⎟ ⎜
Ti − T f ⎝ ρV c p ⎠ ⎝ ρ Lchar c p ⎠ h ⎝ i f ⎠
− ( 8238kg m3 ) ( 0.05 m 4 )( 602 J kgK ) ⎛ 900 - 27 ⎞

t= 1n ⎜ ⎟ = 67.6 s Answer
( 23.1+ 76.3) W/m 2
K ⎝ 1000 - 27 ⎠
12- 25
12-25 Ventilation ducts are often uninsulated when they run through attics and other uninhabited spaces.
Consider an air flow at 70 ºC and 15 m3/min which enters a 20-m long 30-cm square duct. The duct runs
through a space that is at 10 ºC. Ignoring the temperature drop across the metal duct, determine:
a. the outlet temperature of the air (in ºC)
b. the heat transfer rate from the hot air (in W).
Approach:
This is a heat exchanger a constant wall temperature,
so we can use the equation given in Section 12.5.
The appropriate heat transfer coefficient correlation
is used once the Reynolds number is determined.
Assumptions:
2. The duct wall temperature is the same as the air
temperature outside of the duct.
3. The system is steady with negligible potential
and kinetic energy effects, no work, and air an
ideal gas with constant specific heat.
Solution:
a) An equation for a constant wall temperature heat exchanger with only one convective thermal resistance is:
Tout = T f + (Tin − T f ) exp ( − hA mc p )
The mass flow rate is evaluated with the air density at the inlet ρ = 1.029 kg/m3
⎛ 15 m3 ⎞ ⎛ 1min ⎞
m = ρV = (1.029 kg/m3 ) ⎜ ⎟⎜ ⎟ = 0.257 kg/s
⎝ min ⎠ ⎝ 60 s ⎠
For the heat transfer coefficient, we evaluate the air properties from Appendix A-7 at the average air temperature.
Because we do not know the exit temperature, assume Tout ~ 60oC, Tavg = (70 + 60)/2 ≅ 65oC, so ρ = 1.044
kg/m3, µ = 2.02 × 10-5 N⋅s/m2, k = 0.0291 W/mK, Pr = 0.700, cp = 1.009 kJ/kgK. This is a square duct, so the
hydraulic diameter is needed
4 Ax 4 ( 0.30m )( 0.30m )
DH = = = 0.30m
Pwetted 4 ( 0.30m )
ρ VDH (1.044kg/m )(15 m min ) ( 0.30m )(1min 60s )

3 3
Re = = = 43,100
µ ( 2.02 ×10−5 N ⋅ s/m2 ) ( 0.30m )2 ( kgm Ns2 )
This is turbulent flow, so using the Dittus-Boelter equation, with an exponent of 0.3 on Pr since the air is being
cooled:
Nu = 0.023 Re0.8 Pr 0.3 = 0.023 ( 43,100 ) ( 0.700 )
0.8 0.3
= 105
NuK (105 )( 0.0291W/mK )
h= = = 10.2 W/m 2 K
DH 0.30m
⎡ - (10.2W/m 2 K ) ( 4 )( 0.30m )( 20m ) ⎤
Tout = 20 C + ( 70 - 20 ) C exp ⎢
o
o
⎥ = 39.4o C Answer
⎢⎣ ( 0.257kg/s )(1.009 kJ/kgK )(1000 J kJ ) ⎥⎦
b) The heat transfer rate can be determined from conservation of energy. Assuming steady, negligible potential
and kinetic energy effects, no work, and air an ideal gas with constant specific heat:
⎛ kg ⎞ ⎛ kJ ⎞
Q = mc p (Tin − Tout ) = ⎜ 0.257 ⎟ ⎜1.009 ⎟ ( 70-39.4 ) K=7.93kW Answer
⎝ s ⎠⎝ kgK ⎠
Comments:
A better estimate can be obtained by re-evaluating the air properties at the average temperature.
12- 26
12-26 Because of a shallow ocean floor and deep draft, an oil tanker must use an offshore oil depot to unload.
The depot is connected to a shore installation by a 1100-m long 45-cm pipe. In the winter, the ocean water
temperature is 5 ºC. The oil (properties equivalent to unused engine oil), initially at a temperature of 20 ºC,
is pumped from the tanker at a flow rate of 0.08 m3/min. Ignoring the water and pipe thermal resistances,
determine:
a. the outlet temperature of the oil (in ºC)
b. the heat transfer rate (in W).
Approach:
This is a heat exchanger with a constant wall temperature,
so we can use the analysis given in Section 12.5. The
appropriate heat transfer coefficient correlation is used
once the Reynolds number is determined.
Assumptions:
1. The water and pipe thermal resistances are ignored.
2. The system is steady with no work, negligible potential
and kinetic energy effects, and oil is an ideal liquid.
Solution:
An equation for a constant wall temperature heat exchanger with one convective thermal resistance is:
⎛ −hA ⎞
Tout = T f + (Tin − T f ) exp ⎜
⎜ mc ⎟⎟
⎝ p⎠
The heat transfer coefficient requires the Reynolds number with properties evaluated at the average oil
temperature. Because we do not know Tout, we assume Tout ~ 15oC, so Tavg = (15 + 20)/2 = 17.5oC. For unused
engine oil from Appendix A-6
ρ = 890 kg/m3, µ =9990×10-4 N⋅s/m2, k = 0.145 W/mK, Pr = 12900, cp = 1.868 kJ/kgK.
ρ VD m 4m
Re = but V = → Re =
µ ρ (π / 4 ) D 2 πµ D
4 ( 890kg/m3 )( 0.08m3 / min ) (1min 60s )
Re = = 3.36
π ( 9990 × 10−4 N ⋅ s/m 2 ) ( 0.45m ) ( ms 2 Nkg )
This is laminar flow, so check the entrance length:
Lent = 0.037 Re Pr D = 0.037 (3.36) (12900) (0.45 m) = 722 m
More than half of the 1100-m pipe is in the entrance region, so we take the entrance length into account, noting
that Gz = RePrD/L
0.0668Gz 0.0668 ( 3.36 ) (12900 )( 0.45 /1100 )
Nu = 3.66 + = 3.66 + = 4.59
1 + 0.04 G z 1 + 0.04 ⎣⎡( 3.36 )(12900 )( 0.45 /1100 ) ⎦⎤
2/3 2/3
Nuk ( 4.59 )( 0.145 W/mK )

h= = = 1.48 W/m 2 K
D 0.45m
⎡ - (1.48 W/m 2 K ) π ( 0.45m )(1100m ) ⎤
Tout = 5 C + ( 20 -5 ) C exp
o
o
⎢ ⎥
⎢⎣ ( 890 kg m3 )( 0.08m3 min ) (1.868k J kgK )(1min 60s )(1000 J kJ ) ⎥⎦
= 10.3o C Answer
b) For the heat transfer rate, using the energy equation:

Q = mcP (Tout − Tin ) = ρVcP (Tout − Tin )
= ( 890 kg m3 )( 0.08 m3 min ) (1min 60s )(1.868 kJ kgK )(10.3-20 ) K
=-21,500 W Answer
12- 27
12-27 A steam condenser downstream of a turbine in a Rankine cycle power plant, has 5,000 tubes, each with an
internal diameter of 0.75-in. The steam condenses at 120 ºF on the outside of the tubes. The total cooling
water flow rate is 3,500 lbm/s, enters the tubes at 54 ºF, and leaves at 85 ºF. Because the condensation heat
transfer coefficient is very high, ignore the steam (and tube wall) thermal resistances. Determine:
a. the heat transfer rate (in Btu/hr)
b. the tube length required (in ft).
Approach:
With the flow rating and inlet and outlet temperatures
given for the cooling water, the heat transfer rate can
be calculated with conservation of energy. Because
this is a constant fluid temperature heat exchanger, the
analysis given in Section 12.5 is used to calculate the
tube length. This requires the heat transfer coefficient.
Assumptions:
1. The system is steady with no work and negligible
potential and kinetic energy effects.
2. The liquid water is an ideal liquid with constant
properties.
Solution:
a) Applying the conservation of energy to the water, and assuming steady, no work, negligible potential and
kinetic energy effects, and an ideal liquid with constant specific heat:
Q = mcP (Tout − Tin )
For water at an average temperature of (54 + 85)/2 = 69.5 ºF, cp = 1.00 Btu/lbmR
Q = ( 3500 lbm s )(1.00 Btu lbmR )( 85-54 ) R ( 3600s hr ) =3.906×108 Btu hr
b) For a heat exchanger with one constant temperature fluid and Rtot = 1/hA (we ignore the condensation and wall
thermal resistances): Tout = Tin + (Tin – Tf) exp (-hA/ m cp) where A = NπDL and m = mtot . Solving for L:
− mc p ⎡ Tout − T f ⎤
L= ln ⎢ ⎥
N π Dh ⎢⎣ Tin − T f ⎥⎦
We need h, which requires the Reynolds number for flow in a single tube with the cooling water properties from
Appendix B-6 evaluated at Tavg = 69.5oF, µ = 65.8 × 10-5 lbm/fts, k = 0.347 Btu/hrft oF, Pr = 6.82
ρ VD mtot 4m
Re = but V = ⇒ Re =
µ N ρ (π / 4 ) D 2
N πµ D
4 ( 3500lbm/s )
Re = = 21, 700
( 5000 )π ( 65.8 ×10−5 lbm/fts ) ( 0.0625 ft )
This is turbulent flow so using the Dittus-Boelter correlation:
Nu = 0.023Re0.8Pr0.4 = 0.023(26700)0.8(6.82)0.4 = 146
Nuk (146 ) ( 0.347 Btu/hrft F )
0
h= = = 810 Btu/hrft 0 F
D 0.0625 ft
− ( 3500 lbm/s ) (1.00 Btu lbm o F ) ( 3600s 1hr ) ⎡ 85 − 120 ⎤
L= 1n ⎢ ⎥ = 10.0 ft Answer
( 5000 )π ( 0.0625 ft ) (810 Btu/hrft 2 o F ) ⎣ 54 −120 ⎦
12- 28
12-28 An air compressor used in a large car body shop is located in an inside equipment room. Fresh air at 5 ºC,
96 kPa is conveyed to the compressor from outside through a 30-cm circular duct that is 15-m long. The
duct runs along the ceiling of the facility where the temperature is 34 ºC. If the air flow rate is 0.35 m3/s,
determine:
a. the temperature of the air when it reaches the compressor (in ºC)
b. the heat transfer rate (in W).
Approach:
This is a heat exchanger with a constant wall
temperature. We can use the analysis given in Section
12.5 (with Rtot = 1/hA) to calculate the outlet
temperature. To do so, we need to determine the
convective heat transfer coefficient. Once TOUT is
determined, the heat transfer rate is calculated with
conservation of energy.
Assumptions:
3. Air is an ideal gas with constant specific heat.
Solution:
a) An equation for a constant wall temperature heat exchanger, with Rtot = 1/hA, is
Tout = Tf + (Tin – Tf) exp (- hA/ m cp)
The mass flow rate is m = ρ V . Assuming air is an ideal gas
PM ( 96 kN m ) ( 28.97 kg kmol )(1kJ kN ⋅ m )
2
ρ= = = 1.20 kg/m3
RT (8.314kJ kmolK )( 5 + 273)K
m = (1.20kg/m3 )( 0.35m3 /s ) = 0.42kg/s
The heat transfer coefficient requires the Reynolds number with properties from Appendix A-7evaluated at
Tavg. Assume, Tout ≈10 oC, Tavg ≈ 7.5 oC ≈ 280.5 K, µ = 1.70×10-5 N⋅s/m2, k = 0.0246 W/mK, Pr = 0.714, cp =
1.005 kJ/kgK
ρ VD m 4m
µ ρ (π / 4 ) D 2 πµ D
4( 0.42kg/s )
Re = = 105, 000
π (1.70 ×10−5 N ⋅ s/m 2 )( 0.30m )
This is turbulent flow, so using the Dittus-Boelter equation
Nu = 0.023 Re0.8 Pr 0.4 = 0.023(105, 000 ) ( 0.714 )
0.8 0.4
= 209
Nuk ( 209 )( 0.0246 W/mK )
h= = = 17.1W/m 2 K
D 0.30m
⎡ - (17.1W/m 2 K ) π ( 0.30m )(15m )(1J 1Ws ) ⎤
Tout = 34o C + ( 5-34 ) K exp ⎢ ⎥ = 17.6 o C Answer
⎢⎣ ( 0.42kg/s )(1.005kJ/kgK )(1000 J kJ ) ⎥⎦
b) The heat transfer rate is determined from the energy equation. Assuming steady, no work, negligible potential
and kinetic energy effects, and ideal gas with constant specific heat so that ∆h = cp ∆T :
Q = mc p (Tout − Tin ) = ( 0.42kg/s )(1.005k J kgK )(17.6 − 5 ) K (1000Ws 1kJ ) = 5320 W Answer
12- 29
12-29 A proposed cooling technique for high power computer chips is to machine micro-channels into the
backside of the silicon chip. Because Nusselt number is defined as Nu = hk/Lchar, for a given value of Nu,
as the characteristic length decreases the heat transfer coefficient increases. Consider a 1.5-cm by 1.5-cm
computer chip that dissipates 50 W. Water at 20 ºC is used as the coolant, and its outlet temperature is
limited to 25 ºC. Heat transfer is primarily from the base of the channel (that is, ignore heat transfer to the
water from the channel sides). Assume the distance between the sides of two channels is 0.1-mm.
Determine:
a. the number of 0.25-mm deep and 0.25-mm wide micro-channels that are on a chip and the average
surface temperature of the base of the micro-channels (in ºC)
b. the number of 1-mm deep and 1-mm wide micro-channels that are on a chip and the average surface
temperature of the base of the micro-channels (in ºC).
Approach:
If we assume the chip is at a
uniform temperature, Ts, we can
use the analysis given in Section
12.5 for a heat exchanger with a
constant temperature wall. We
need to determine the water flow
rate and the heat transfer
coefficient. The water flow rate
can be determined from
conservation of energy.
Assumptions:
1. The system is steady with no
work and negligible potential
and kinetic energy effects.
2. Water is an ideal liquid with
constant specific heat.
Solution:
a) For a constant wall temperature heat exchanger (with Rtot = 1/hA), the water exit temperature can be calculated
from
Tout = Ts + (Tin – Ts) exp (-hA/ m cp)
Tout − Tin exp ( − hA/mc p )
Solving for Ts =
1- exp ( − hA/mc p )
Therefore, we must evaluate A, h, and m .
The heat transfer area depends on the number of channels. The number of channels, N, on a chip is based
solely on geometry.
W−D 0.015m − 0.0001m
W = ( NLc ) + ( N + 1) D ⇒ N= = = 42.5
Lc + D 0.00025m + 0.0001m
This must be an integer, so N = 42.
Area for heat transfer is only the base of the channel
A = NLcL = (42) (0.00025 m) (0.015 m) = 1.575 × 10-4 m2
The water flow rate is determined from conservation of energy heat plus the heat transfer rate and temperatures.
Assuming steady, no work, no potential or kinetic energy effects, and an ideal liquid with constant specific heat so
that ∆h = cp ∆T :
Q
Q = mc p (Tout − Tin ) ⇒ m=
c p (Tout − Tin )
For evaluating cp and eventually the heat transfer coefficient, we need water properties from Appendix A-6 at Tavg
= (20 + 25)/2 = 22.5 oC, so by interpolation: ρ = 997.6 kg/m3, cp = 4.180 kJ/kgK, k = 0.607 W/mK, µ = 9.29 ×
10-4 N⋅s/m2, Pr = 6.40
( 50W )(1J Ws )
m= = 0.00239 kg/s
( 4.180kJ/kgK )(1000 J kJ )( 25- 20 )K
12- 30
ρ VDH 4 Ax 4 HLc 4( 0.00025m )( 0.000025m )
The Reynolds number is Re = where DH = = = = 0.00025m
µ Pwetted 2( H + Lc ) 2( 0.00025m + 0.00025m )
The velocity is for the flow in one channel
m/N ( 0.00239kg/s )
V = = = 0.91m/s
ρ A x ( 42 ) ( 997.6 kg/m3 ) ( 0.00025m )2
Re =
( 997.6kg/m ) ( 0.91m/s )( 0.00025m ) = 245
3
9.25 ×10-4 N ⋅ s/m 2

This is laminar flow, so check the entrance length.
Lent = 0.037 Re Pr D = (0.037) (245) (6.40) (0.00025 m) = 0.0145 m
Because the entrance length is about the same length as the channel, we will take into account entrance conditions.
Assuming Ts ~ 25oC, µs = 8.72 × 10-4 N⋅s/m2
⎡ ( 245 )( 6.40 )( 0.00025 ) ⎤
1/ 3 0.14
⎛ 9.29 ×10−4 ⎞
Nu =1.86 Gz1/ 3 ( µ µ s )
0.14
= 1.86 ⎢ ⎥ ⎜ -4 ⎟
= 5.57
⎣ 0.015 ⎦ ⎝ 8.76×10 ⎠
Nuk ( 5.57 )( 0.607W/mK )
h= = = 13,500W/m 2 K
DH 0.00025m
hA (13500W/m K )(1.575 x10 m ) (1J Ws )
2 −4 2
= = 0.213
mc p ( 0.00239kg/s )( 4.180 kJ/kgK )(1000 J kJ )
( 25+273) - ( 20+273) exp ( -0.213)
Ts = =319.0K=46.0o C Answer
1- exp ( -0.213)
b) For a channel with Lc = H = 0.001 m
N = 13, A = 1.95 × 10-4 m2, DH = 0.001 m
V = 0.184 m/s, Re = 199
Lent= 0.047 m, Nu = 8.24, h = 5,000 W/m2K
Ts = 346.7 K = 73.7 oC Answer
Comments:
Increasing the size of the channels has a detrimental effect on the wall temperature. Even though the heat transfer
area increased, the decrease in the heat transfer coefficient was larger, so that the hA product decreased, and it’s
the hA product that is most important.
12- 31
12-30 Consider turbulent flow of a fluid through a tube maintained at constant temperature. The mass flow rate is
0.32 kg/sec, and the heat transfer coefficient is 250 W/m2·K. Now the free-stream velocity of the fluid is
doubled. Assume the flow regime remains unchanged. Estimate:
a. the percent change in the pressure drop of the fluid between the old and new flow rates
b. the percent change in the local heat flux between the fluid and the walls of the channel
c. the percent change in the total heat transfer rate over the length of the channel if the heat transfer
area is 3.7 m2, specific heat is 2200 J/kg·K, inlet temperature is 25 C, and wall temperature is 75 C.
Approach:
This is a comparison study of changes in performance
(∆P, h, Q ) when a turbulent flow is varied inside a
tube. Basic expressions for ∆P, f, h, and Q are used.
Assumptions:
1. The system is steady with no work or potential or
kinetic energy effects.
Solution:
L V2
a) Let V2 = 2 V1 ⇒ m2 = 2m1 . An expression for pressure drop is ∆P = f ρ . The tube geometric and
D 2
∆P2 f 2 ( L / D ) ( ρ V2 2 ) f 2 ⎛ V2 ⎞
2 2
fluid properties remain fixed. Therefore, = = ⎜ ⎟

∆P1 f1 ( L / D) ( ρ V1 2 2 ) f1 ⎝ V1 ⎠
0.184 0.184
For turbulent flow in a tube f= =
Re0.2 ( ρ D / µ )0.2 V 0.2
Substituting this into the ∆P expression and canceling like terms
1.8
∆P2 ⎛V ⎞
= ⎜ 2 ⎟ = (2)1.8 = 3.48 Answer
∆P1 ⎝ V1 ⎠
q2′′ h2 (Tw −T f ) h2
b) The change in the local heat flux = =
q1′′ h1 (Tw −T ) h1
f
0.8
k ⎛ ρ VD ⎞
Using the Dittus-Boelter equation h = ( 0.023 ) ⎜ ⎟ Pr = C V
0.4 0.8
D ⎝ µ ⎠
Substituting this into the local heat flux comparison and canceling like terms
0.8
q2′′ ⎛ V2 ⎞
=⎜ ⎟ = (2) =1.74
0.8
Answer
q1′′ ⎝ V1 ⎠
c) For the total transfer rate, we can use conservation of energy. Assuming steady, no work, negligible potential
energy and kinetic energy effects, and ideal fluid so that ∆h = cp ∆T
Q2 m2 c p (Tout ,2 −Tin , 2 )
=
Q1 m1 c p (Tout ,1 −Tin ,1 )
We can obtain the outlet temperatures for this constant wall temperature heat transfer situation from
⎡ -(250 W m 2 K )(3.7m 2 ) ⎤
Tout = Tw − (Tw − Tin ) exp ( −hA mc p ) → Tout ,1 = 75o C − (75 − 25)o C exp ⎢ ⎥ = 61.6 C
o
⎣ (0.32kg/s)(2200 J/kgK) ⎦
⎡ − (1.74)(250)(3.7) ⎤
Tout ,2 = 75 − (75 − 25)exp ⎢ ⎥ = 59.1 C
o
⎣ 2(0.32)(2200) ⎦
Q2 m2 c p (Tout ,2 −Tin ,2 ) ⎛ 59.1− 25 ⎞
Therefore, = = 2⎜ ⎟ =1.86 Answer
Q1 m1 c p (Tout ,1 −Tin ,1 ) ⎝ 61.6 − 25 ⎠
12- 32
12-31 Glycerin is pumped through a 1.5 cm diameter tube that is 5 m long. The inlet temperature is 32 °C, the
required outlet temperature is 22 °C, and the flow rate is 100 kg/hr. Determine the wall temperature
required to obtain this outlet temperature (in °C).
Approach:
This is a constant wall temperature heat exchanter, so
we can use Eq. 12- 33. The appropriate heat transfer
coefficient correlation is used once the Reynolds
number is determined.
Assumptions:
Solution:
An equation for a constant wall temperature heat exchanger is
Tout = Tw − (Tw − Tin ) exp ⎡⎣ − hA mc p ⎤⎦
Solving for Tw
Tout − Tin exp ⎡⎣ − hA mc p ⎦⎤
Tw = with T in K
1 − exp ⎡⎣ − hA mc p ⎤⎦
To calculate h, we need Re. Fluid properties are obtained from Appendix A-6 at the average bulk temperature
(32+22)/2 = 27 ºC: cp = 2.427 kJ/kgK, k = 0.286 W/mK, µ = 7990 × 10-4 N-s/m2, Pr = 6780
ρ VD 4m 4m
µ ρπ D 2
πµ D
4(100kg/hr)(1hr 3600s )
Re = = 2.95
π(7990×10-4 Ns/m 2 )(0.015m)
Because this is laminar flow, we must check entrance length
Lent ,t = 0.037 Re Pr D = 0.037 (2.95) (6780) (0.015m) = 11.1m
This is much longer than the tube length, so we use a correlation that takes into account the entrance length.
Using Eq, 12- 42 and evaluating the viscosity at an assumed wall temperature of 7 ºC, µw= 53400 × 10−4 Ns/m2
Nu =1.86 Gz1/ 3 ( µ / µ w ) 0.14 =1.86( Re Pr D L ) ( µ / µw )
1/ 3 0.14
1/ 3 0.14
⎡ (2.95)(6780)(0.015) ⎤ ⎛ 7990 × 10 −4 ⎞
=1.86 ⎢ ⎥ ⎜ −4 ⎟
= 5.58
⎣ 5 ⎦ ⎝ 53400 × 10 ⎠
Nuk ⎛ 5.58)(0.286W/mk ⎞
h= =⎜ ⎟ = 106 W/m K
2
D ⎝ 0.015m ⎠
⎡ -(106W/m 2 K) (π) (0.015m) (5m) ⎤
(273+ 22) − (273+ 32) exp ⎢ ⎥
Tw = ⎣ (100 kg hr )(1hr 3600s )( 2.427 kJ kgK )(1000 J kJ ) ⎦
⎡ -(145W/m 2 K) (π) (0.015m) (5m) ⎤
1- exp ⎢ ⎥
⎣ (100 kg hr )(1hr 3600s )( 2.427 kJ kgK )(1000 J kJ ) ⎦
= 272.7K = -0.3 ºC Answer
Comments:
Because the viscosity changes so dramatically with temperature, we will iterate one time. At 273 K, ,
µw= 106, 000 × 10−4 Ns/m2, Nu = 5.07, h = 96.7 W/m2K, Tw = 270.1 K = -2.9 ºC
12- 33
12-32 In a pharmaceutical application, the product is subjected to a final sterilization by heating it from 32 ºC to
80 ºC. A flow of 60 cm3/s is passed through a 10-mm tube that is heated with a uniform heat flux produced
by wrapping the tube with an electric resistance heater. If product properties can be approximated by those
of ethylene glycol, and the tube is 25-m long, determine:
a. the required power (in W)
b. the wall temperature at the tube exit (in ºC).
Approach:
The required heat transfer rate can be obtained from
conservation of energy and the given information.
The wall temperature at the tube outlet can be
calculated using the basic rate equation ( q ′′ = h∆T).
Assumptions:
2. The liquid is ideal with constant specific heat.
Solution:
a) For conservation of energy on the liquid, assume steady, no work, no potential or kinetic energy effects, and an
ideal liquid with constant specific heat so that ∆h = cp ∆T
Q = mc p (Tout − Tin )
The properties required to calculate the heat transfer coefficient are evaluated from Appendix A-6 at Tavg = (32 +
80)/2 = 56 oC ~ 330 K, ρ = 1089.5 kg/m3, cp = 2.549 kJ/kgK, µ = 56.1×10-4 N⋅s/m2, k = 0.260 W/mK, Pr = 55.0
Q = (1089.5kg m3 )( 60cm3 s ) (1m 100cm ) ( 2.549kJ kgK )( 80 − 32 ) K ( 1000W 1kJ/s ) = 8000W
3
b) For the wall temperature at the tube outlet q ′′ = h (Ts − Tout ) ⇒ Ts = Tout + q ′′ h
Ts = Tout + ( Q/A ) h = Tout + Q π DLh
To calculate the heat transfer coefficient, we need the Reynolds number:
ρ VD m 4m
Re = ⇒ V = ⇒ Re =
µ ρ (π /4 ) D 2
πµ D
4 (1089.5 kg m3 )( 60 cm3 s ) (1m 100cm )
3
Re = = 1480
π ( 56.1×10-4 Ns/m 2 ) ( 0.01m )( kgm Ns 2 )
This is laminar flow, so we will check the entrance length
Lent,t ~ 0.053 Re Pr D = 0.053 (1480) (55.0) (0.01 m) = 43.2 m
This is significantly longer than the tube length, so entrance effects must be taken into account.
We do not have a correlation that gives the heat transfer coefficient for laminar, developing flow with a
constant wall heat flux boundary condition. So we will use the Seider and Tate correlation. Assuming Ts ~ 350
K,
µs = 34.2×10-4 N⋅s/m2
⎡ (1480 )( 55.0 )( 0.01m ) ⎤
1/ 3 0.14
⎛ 56.1× 10−4 ⎞
Nu =1.86 Gz1/ 3 ( µ /µ s )
0.14
=1.86 ⎢ ⎥ ⎜ −4 ⎟
= 6.37
⎣ 25m ⎦ ⎝ 34.2 × 10 ⎠
Nuk ( 6.37 )( 0.260W/mK )
h= = =166W/m 2 K
D 0.01m
8000W
Ts = 80 o C + =141.3 o C
π ( 0.01m )( 25m ) (166W/m 2 K )
Re-evaluating µs = at 373 K (highest temperatures for which we have properties µs = 21.5 ×10−4 N⋅s/m2
Nu 6.80 , h 177W/m 2 K , Ts = 137.5 o C Answer
Comments:
The wall temperature will be lower than this, since µs should be smaller than with what we have available.
12- 34
12-33 Air enters a compressor operating at steady state with a volumetric flow rate of 37 m3/min at 105 kPa and
30 °C and exits with a pressure of 690 kPa and temperature of 240 °C. The compressor is cooled with 40
kg/min of water that circulates in a water jacket enclosing the compressor. The water jacket can be
approximated as 25 3-cm diameter, 2-m long pipes. The water enters at 20 °C and the wall temperature of
the pipes can be approximated as being constant at 135 °C. Assume fully developed flow. Determine:
a. heat transfer rate from the compressor to the water (in kW)
b. the mass flowrate of air (in kg/sec)
c. the power input to the compressor (in kW).
Approach:
The heat transfer rate from the compressor to the
water jacket can be determined from a combination of
an energy balance on the water and the use of the
constant wall temperature heat exchanger equation
given in Section 12.5. Assuming one-dimensional
flow the air mass flow rate can be determined from its
definition. Finally, an energy balance on the
compressor is used to determine the required input
power.
Assumptions:
1. The system is steady with negligible potential and
2. Air and water are ideal fluids.
3. No work is done in the heat exchanger.
Solution:
a) Using control volume CVII defined in the schematic, we apply conservation of energy to the water. Assume
steady, no work, negligible potential and kinetic energy effects, and an ideal liquid with constant specific heat so
that ∆h = cp ∆T: QII = mc c p (TD − TC )
The water outlet temperature, TD, is obtained with the analysis for a constant wall temperature heat exchanger
with Rtot = 1/hA: TD = Twall + (Tc – Twall) exp (- hA/ m cp). The heat transfer coefficient is calculated with
the Reynolds number for flow in one tube evaluated at Tavg = (TC + TD)/2. From Appendix A-7 and assuming TD ≈
30 oC, Tavg = (20 + 30)/2 = 25 oC, so µ = 8.72 × 10-4 N⋅s/m2, k = 0.611 W/mK, cp = 4.178 kJ/kgK, Pr = 5.97.
ρ VD m/N 4m 4 ( 40kg/min )(1min 60s )
Re = ⇒ V= → Re = = = 1300
µ ρ (π /4 ) D 2
πµ DN π ( 8.72 ×10-4 N ⋅ s/m 2 ) ( 0.03m )( 25 )
This is laminar flow, so assuming fully developed flow for constant wall temperature boundary condition
Nuk ( 3.66 )( 0.611W/mK )
Nu = 3.66 → h = = = 74.5W/m 2 K
D 0.03m
⎡ - ( 74.5W m 2 K ) π ( 0.03m )( 2m )( 25 ) ⎤
TD = 135 C + ( 20 -135 ) K exp ⎢
o
⎥ = 33.6 o C
⎢⎣ ( 40 kg min )(1min 60s )( 4.178kJ kgK )(1000 J kJ ) ⎥⎦
QII = ( 40 kg min )(1min 60s )( 4.178 kJ kgK )( 33.6 − 20 ) K = 37.9kW Answer
b) For the air mass flow rate
m = ρ VA = ρV =
PM
V=
(105 kN m 2 ) ( 28.97 kg kmol ) ( 37 m3 min ) (1min 60s )
0.745
kg
Answer
RT (8.314 kg kmolK )( 30 + 273)K s
c) Applying the energy equation to CVI, and assuming steady, negligible potential and kinetic energy effects,
ideal gas with constant specific heat so ∆h = cp ∆T:
QI − W + m ( hA − hB ) = 0 → W = QI + m c p (TA − TB )
Note that QI = − QII → W = − QII + mc p (TA − TB ) At the average air temperature, cp = 1013 J/kgK
W = − 37.9k W + ( 0.745 kg s )(1013J kgK )( 30 − 240 ) K (1kJ 1000J ) = − 196k W Answer
12- 35
12-34 The condenser downstream of the turbine in a large Rankine cycle power plant is constructed of 30,000 25-
mm tubes. The steam condenses at 50 ºC with a heat transfer coefficient of 9,000 W/m2·K on the outside
the tubes. The cooling water enters the tube side of the condenser at 20 ºC at a flow rate of 17,000 kg/s.
For a 1000-MW (net) power output and a cycle thermal efficiency of 42%, determine:
a. the cooling rate required (in MW)
b. the outlet temperature of the cooling water (in ºC)
c. the length of tubing required (in m).
Approach:
The Rankine cycle thermal efficiency definition is
used to calculate the cooling rate. The energy
equation applied to the cooling water is used to
calculate the outlet temperature. Condensing steam is
a constant fluid temperature, so the analysis in Section
12.5 is used to calculate the required length.
Assumptions:
1. The system is steady, with no work or potential or
2. Water is an ideal liquid with constant specific heat.
Solution:
a) The definition of Rankine cycle thermal efficiency is ηcycle = Wnet Qin
Also Wnet = Qin − Qout → Qin = Wnet + Qout Substituting this into the definition of ηcycle:
Qout = ⎡⎣(1−ηcycle ) ηcycle ⎤⎦ Wnet = ⎣⎡(1− 0.42 ) 0.42 ⎦⎤ (1000 MW ) =1381MW Answer
b) The outlet cooling water temperature is calculated with conservation of energy applied to a control volume
around the cooling water. Assuming steady, no work, no potential or kinetic energy effects, and an ideal liquid
with constant specific heat so ∆h = cp ∆T: Q = mc p (Tout − Tin ) → Tout = Tin + Q/mc p
Evaluating water properties from Appendix A-6 at the average temperature, assume Tout ≈ 40 oC,
Tavg ≈ 30 oC cp = 4.176 kJ/kgK, k = 0.618 W/mK, µ = 7.79 × 10-4 N⋅s/m2, Pr = 5.26
(1,381,000kW )(1kJ/s 1kW )
Tout = 20 o C + = 39.5 o C Answer
(17000 kg s )( 4.176 kJ kgK )
c) For a heat exchange with one constant fluid temperature Tout = T f + (Tin − T f )exp ( − 1/mc p Rtot )
1 1n ( D 2 /D1 ) 1
where Rtot = + + . Because no information is given about the tube material or wall thickness,
hi A1 2π kL h0 A0
1 1 1 ⎡1 1 ⎤
wall resistance must be ignored, and Ai = A0 = NπDL. → Rtot = + = ⎢ + ⎥
hi N π DL ho N π DL N π DL ⎣ hi ho ⎦
− mc p ⎡ 1 1 ⎤ ⎛ Tout − T f ⎞
Solving for the tube length: L= ⎢ + ⎥1n ⎜ ⎟⎟
N π D ⎣ hi ho ⎦ ⎜⎝ Tin − T f ⎠
ρ VD m/N 4m
The Reynolds number in one tube is Re = ⇒ V = ⇒ Re =
µ ρ (π /4 ) D 2 Nπ µD
4 (17000 kg s ) ( Ns 2 kgm )
Re = = 37050
( 30000 )( π ) ( 7.79×10-4 Ns/m 2 ) ( 0.025m )
Turbulent flow, so using Dittus-Boelter equation Nu = 0.023Re0.8Pr0.4 = (0.023) (37050)0.8 (5.26)0.4 = 202
Nuk ( 202 )( 0.618W/mK )
h= = = 4990W/m 2 K
D 0.025m
(17000 kg s )( 4.176 kJ kgK )(1000 J kJ ) ⎡ 1 1 ⎤ ⎡ 39.5-50 ⎤
L=− ⎢ + ⎥1n ⎢ ⎥ = 9.85m Answer
( 30000 )π ( 0.025m ) 2 2
⎣ 4990W m K 9000W m K ⎦ ⎣ 20 -50 ⎦
12- 36
12-35 Air at a mass flow rate of 0.0015 lbm/s and an inlet temperature of 80 ºF enters a rectangular duct 3.5-ft
long, 0.15-in. high, and 0.60-in. wide. A uniform heat flux of 50 Btu/hr·ft2 is imposed on the duct surface.
Determine:
a. the outlet temperature of the air (in ºF)
b. the highest wall temperature and its location (in ºF and ft).
Approach:
The outlet temperature can be determined from an
energy balance on the air flow combined with the
given information. The highest wall temperature
will occur at the end of the duct where the fluid
temperature is highest and can be found by using the
rate equation ( Q = q″A = hA∆T)
Assumptions:
1. Air is an ideal gas at one atmosphere.
2. The system is steady, with no work or potential
or kinetic energy effects.
Solution:
a) For an energy balance on the air flow, assume steady, no work, no potential or kinetic energy effects, and an
ideal gas with constant specific heat so that ∆h = cp ∆T:
Q = mc p (Tout − Tin ) → Tout = Tin + Q/mc p = Tin + q ′′ A/mc p
The fluid properties from Appendix B-7 are evaluated at Tavg, so assuming Tout ≈ 100 oF, Tavg ≈ 90 oF;
ρ = 0.00725 lbm/ft3, cp = 0.240 Btu/lbm oF, k = 0.0152 Btu/hr ft oF, µ = 1.267 × 10-5 lbm/ft⋅s, Pr = 0.72
Tout = 80 o F +
( -50Btu hrft 2 ) ( 2 )( 0.0125 + 0.05)ft ( 3.5ft ) = 96.9 o F Answer
( 0.0015lbm s )( 3600s 1hr ) ( 0.240Btu lbm o F )
b) Using the rate equation Q = q ′′ A = hA (Tw − Tout ) → Tw = Tout + q ′′/h
The Reynolds number is Re = ρ VDH µ . For the non-circular duct, we need the hydraulic diameter
4 Ax 4( 0.0125ft )( 0.05ft )
DH = = = 0.020ft
Pwetted 2( 0.0125 + 0.05 )ft
m 0.0015lbm/s ft
The velocity is V = = = 33.1
ρ Ax ( 0.0725lbm ft )( 0.0125ft )( 0.05ft )
3
s
Re =
( 0.0725lbm ft )( 33.1ft s )( 0.02ft ) = 3790
3
1.267 ×10-5 lbm/fts

( f 8)( Re −1000 ) Pr
Turbulent flow, so using the Gnielinski correlation Nu =
1 + 12.7( f 8 ) ( Pr 2 / 3 − 1)
1/ 2
f = [ 0.7901n Re −1.64] = ⎣⎡0.791n ( 3790 ) − 1.64 ⎦⎤ = 0.0422

−2 −2
( 0.0422 8 )( 3790 − 1000 )( 0.72 )

Nu = = 13.0
1 + 12.7 ( 0.0422 8 ) ⎡( 0.72 ) − 1⎤
1/ 2 2/3
⎣ ⎦
Nuk (13.0 ) ( 0.0152 Btu/hrft F )
o
Btu
h= = = 9.89
DH 0.02ft hrft o F
50Btu/hrft 2
Tw = 96.9 o F + 2 o
=102.0 o F Answer
9.89 Btu/hrft F
12- 37
12-36 Water enters at 40 ºF and flows at a rate of 0.25 ft3/s inside a 20-ft long annulus, whose inner and outer
radii at 1 in. and 2 in., respectively. The inner surface is maintained at 150 ºF and the outer surface is
heavily insulated. Determine:
a. the outlet temperature (in ºF)
b. the heat transfer rate (in Btu/hr).
Approach:
This is a constant wall temperature heat exchanger.
Using the analysis from section 12.5, we can
determine the outlet temperature. Once that is known,
conservation of energy is used to calculate the heat
transfer rate.
Assumptions:
2. The water properties are constant.
Solution:
a) For a constant wall temperature heat exchanger with Rtot = 1/hA, the exit temperature can be obtained from:
Tout = Ts + (Tin – Ts) exp (-hA/ m cp)
The heat transfer coefficient requires the Reynolds number with fluid properties evaluation at the average
temperature. Because we do not know the outlet temperature, we assume Tout ~ 120oF, Tavg = (40 + 120)/2 =
80oF, and the properties from Appendix B-6 are ρ = 62.2 lbm/ft3, cp = 1.00 Btu/lbmoF, k = 0.353 Btu/hr ft oF, µ =
57.8 × 10-5 lbm/ft⋅s, Pr = 5.89
ρ V DH 4 Ax 4(π / 4 ) ( D22 − D12 )
Re = where DH = = = D2 − D1 = 0.0833ft
µ Pwetted π ( D2 + D1 )
V V 4 ( 0.25ft 3 /s )
V = = = = 15.3ft/s
A (π 4 ) ( D22 − D12 ) π ( 0.167 2 − 0.08332 ) ft 2
Re =
( 62.2 lbm/ft ) (15.3ft/s )( 0.0833ft ) = 137, 000
3
57.8×10-5 lbm/ft s
This is turbulent flow, so using the Dittus-Boelter correlation:
Nu = 0.023Re0.8Pr0.4 = (0.023) (137000)0.8 (5.89)0.4 = 601
Nuk ( 601) ( 0.353Btu/hrft F )
o
Btu
h= = = 2550
DH 0.0833ft hr ft 2 o F
⎡ - ( 2550Btu hrft 2 R ) π ( 0.0833ft )( 20 ft ) ⎤
Tout =150o F + (40 - 150)o F exp ⎢ ⎥
⎢⎣ ( 62.2 lbm ft )( 0.25ft s ) (1.00Btu lbmR )( 3600s 1hr ) ⎥⎦
3 3
= 63.4 o F Answer
Because the calculated Tout is significantly different than the assumed Tout, the fluid properties should be
re-evaluated and Tout calculated again.
b) The heat transfer rate can be determined from conservation of energy applied to the water. Assuming steady,
no work, no potential or kinetic energy effects, and an ideal liquid with constant specific heat so that ∆h = cp ∆T
Q = m cp ∆T
⎛ lbm ⎞ ⎛ ft 3 ⎞ ⎛ Btu ⎞ ⎛ 3600s ⎞
⎟ ( 63.4 - 40 ) F ⎜
o
= ⎜ 62.2 3 ⎟ ⎜ 0.25 ⎟ ⎜1.00 6
⎟ =1.310 × 10 Btu/hr Answer
⎝ ft ⎠ ⎝ s ⎠⎝ lbmR ⎠ ⎝ 1hr ⎠
12- 38
12-37 Unused engine oil is to be heated from 20 ºC to 65 ºC using condensing steam at 100 ºC. The oil flows
inside a 1-cm diameter tube at a flow rate of 0.1 kg/s. The resistance of the condensing steam and the tube
wall can be ignored. Determine the length of tube required (in m).
Approach:
This is a constant fluid temperature heat exchanger.
Using the analysis given in Section 12.xx we can
determine the required length. The heat transfer
coefficient must be determined.
Assumptions:
1. The condensing and wall thermal resistances are
ignored.
Solution:
For a heat exchanger with one constant temperature fluid and ignoring the condensing coefficient and the wall
resistance, then Rtot = 1/hA . Therefore, Tout = Tst + (Tin − Tst ) exp ( − hA mc p ) where A = πDL. Substituting this
into the main equation and solving for L:
⎛ T −T ⎞
− mc p 1n ⎜ out st ⎟
L= ⎝ Tin − Tst ⎠
π Dh
The heat transfer coefficient requires the Reynolds number with oil properties from Appendix A-6 evaluated at the
average temperature (65 + 20)/2 = 42.5 oC; by interpolation µ = 1970×10-4 N⋅s/m2, k = 0.144 W/mK, cp = 1.972
kJ/kgK, Pr = 2683
ρV D m 4m 4 ( 0.1kg/s ) ( Ns 2 kgm )
Re = → V= → Re = = = 64.6
µ ρ (π /4 ) D 2 π µ D π (1970 ×10-4 N ⋅ s/m 2 ) ( 0.01m )
This is laminar flow, so check the entrance length
Lent = 0.037 RePr D = 0.037 ( 64.6 )( 2683)( 0.01m ) = 64.2m
We assume the tube length will be shorter than this so we use the Seider-Tate correlation
Nu =1.86 ( Gr ) ( µ /µ s ) =1.86 ( RePr D L ) ( µ /µ s )
1/ 3 0.14 1/ 3 0.14
But this requires the tube length. Therefore, an iterative solution is required.
The viscosity at the wall temperature of about 100 ºC is approximately 180 × 10-4 N⋅s/m2. Assume a tube length
of 10 m.
⎡ ( 64.6 )( 2683)( 0.01m ) ⎤ Nuk (14.5 )( 0.144W/mK )
1/ 3 0.14
⎛ 1970 × 10-4 ⎞
Nu =1.86 ⎢ ⎥ ⎜ -4 ⎟
= 14.5 → h = = = 209W/m 2 K
⎣ 10m ⎦ ⎝ 180 × 10 ⎠ D 0.01m
⎡ 65-100 ⎤
− ( 0.1kg/s )(1.972kJ/kgK )1n ⎢ (1000J kJ )
⎣ 20 -100 ⎥⎦
L= = 24.8 m
π ( 0.01m ) ( 209W/m 2 K ) (1J/s 1W )
This is longer than our assumed length, so using this L
⎡ ( 64.6 )( 2683)( 0.01) ⎤ ⎛ 1970 × 10−4 ⎞ (10.7 )( 0.144 )
1/ 3 0.14
Nu =1.86 ⎢ ⎥ ⎜ −4 ⎟
= 10.7 → h= = 154W/m 2 K
⎣ 24.8 ⎦ ⎝ 180 × 10 ⎠ 0.01
⎡ 65-100 ⎤
− ( 0.1)(1.972 )1n ⎢ (1000 )
⎣ 20 − 100 ⎥⎦
L= = 33.7m
π ( 0.01)(154 )
Additional iterations give
Nu = 9.66, h =139W/m3 K L = 37.3m
Nu = 93.33, h =134 W/m K 2
L = 38.6m Answer
12- 39
12-38 In a small ship with limited space, water must be heated from 10 ºC to 50 ºC with condensing steam at 100
ºC. The water flow rate is 1.5 kg/s. Either one 4-cm diameter tube, two 3-cm tubes, or three 2-cm tubes
can be used in parallel. Determine which configuration will yield the shortest length.
Approach:
Whether there is 1, 2 or 3 tubes, the flow is evenly
distributed among the tubes. This is a constant fluid
temperature heat exchanger. Using the analysis given
in Section 12.5 and the appropriate flow in each tube
(which depends on the diameter), the required length
can be determined. The heat transfer coefficient for
each configuration must be evaluated.
Assumptions:
1. The condensing and wall thermal resistances are
ignored.
Solution:
For a heat exchanger with one constant temperature fluid and ignoring the condensing heat transfer coefficient and
the wall resistance, then Rtot = 1/hA: Tout = Tin + (Tin − Tst ) exp ( − hA mc p ) where A = π DL for one tube.
Substituting this into the equation, recognizing that the flow in one tube is m /N:
⎛ T − Tst ⎞
− mc p 1n ⎜ out ⎟
L= ⎝ Tin − Tst ⎠
N π DH
The heat transfer coefficient requires the Reynolds number (for flow in one tube) with fluid properties from
Appendix A-6 evaluated at the average temperature, (10 + 50)/2 = 300C, so cp = 4.176 kJ/kgK, µ = 7.79 × 10-4
N⋅s/m2, k = 0.618 W/mK, Pr = 5.26
ρV D m/N 4m
Re = → V= → Re =
µ πµ D N πµ D
For D = 4 cm and N = 1
4 (1.5kg/s ) ( Ns 2 kgm )
Re = = 61,300
(1) π ( 0.04m ) ( 7.79 ×10-4 N ⋅ s/m 2 )
Similarly, for
D = 3 cm and N = 2, Re = 40860
D = 2 cm and N = 3, Re = 40860
These are all turbulent flow, so using the Dittus-Boelter correlation:
For D = 4 cm Nu = 0.023Re0.8Pr0.4 = 0.023(61300)0.8 (5.26)0.4 = 302
Nuk ( 302 )( 0.618W/mK )
h= = = 4, 670W/m 2 K
D 0.04m
Similarly for
D = 3 cm Nu = 218 h = 4500 W/m2 K
D = 2 cm Nu = 218 h = 6750 W/m2 K
For D = 4 cm
⎡ 50 − 100 ⎤
− (1.5kg/s )( 4176J/kgK ) ln ⎢ (1000 J kJ )
⎣ 10 − 100 ⎥⎦
L= = 6.27m
(1)π ( 0.04m ) ( 4670W/m 2 K )
Similarly for
D = 3 cm and N = 2, L = 8.68 m
D = 2 cm and N = 3, L = 8.68 m
The single 4 cm tube yields the shortest length. Answer
12- 40
12-39 A thick, stainless steel (AISI 316) pipe with inside and outside diameters of 20 mm and 40 mm,
respectively, is heated electrically to provide a uniform heat generation rate of 107 W/m3. This pipe is
encased within a larger concentric tube with an inside diameter of 50 mm, whose outer surface is heavily
insulated. Pressurized water flows through the annular region between the two tubes with a flow rate of 0.6
kg/s. The water inlet temperature is 20 ºC. Determine:
a. the required pipe length if the desired outlet temperature is 40 ºC (in m)
b. highest surface temperature and its location (in ºC and m).
Approach:
This is an internal forced convection problem. With
the known heat generation rate and water flow, we can
determine the length using conservation of energy.
The highest surface temperature can be determined
from the heat transfer rate and the heat transfer
coefficient.
Assumptions:
3. Water is an ideal liquid with constant specific heat.
Solution:
a) Assume a steady flow with no work, negligible potential and kinetic energy effects, and water is an ideal liquid
with constant specific heat so that ∆h = cp∆T. Applying these to the conservation of energy and mass:
Q + m (h in – hout) = 0 → q''' Vwall + m cp (Tin – Tout) = 0
π m c p (Tin − Tout )
q ''' ⎡ Do2 − Di2 ⎤⎦ L + m c p (Tin − Tout ) = 0 → L=
4⎣ q ''' (π 4 ) ⎡⎣ Do2 − Di2 ⎤⎦
Evaluating the water specific heat at the average temperature Tavg = (Tin – Tout)/2 = (20 + 40)/2 = 30°C.
From Appendix A-6. cp = 4.176 kJ/kgK
(0.6kg/s) (4.176 kJ/kgK) (40 − 20)K (1000J 1kJ )
L= =5.32m Answer
(107 W m3 )(π 4 ) ⎡⎣(0.04m)2 − (0.02m)2 ⎤⎦ (1J Ws )
b) The location of the highest surface temperature is at the tube outlet. There the bulk water temperature is 40ºC,
so from Appendix A-6: ρ = 992.2 kg/m3, k = 0.631 W/mK, µ = 6.34 × 10-4 N-s/m2, Pr = 4.19
q ''' ⎡⎣ Do − Di ⎤⎦
2 2
q ′′ q ′′′Vwall
The heat flux of that location is q′′ = h(Ts- Tout) → Ts = Tout + = Tout + = Tout +
h hAs h 4 Do L
Because this is flow in an annulus, we need the hydraulic diameter for Re and Nu
4 Ax
Dh = = Do − Di = 50 mm – 40 mm = 10mm = 0.01m
pwetted
m m 0.6kg/s m
V = = = = 0.855
ρ Ax ρ (π 4 ) ⎡⎣ D 2 − Do2 ⎤⎦ 2
( 992.2 kg m ) (π 4 ) ⎣⎢( 0.05m ) −(0.04m) ⎦⎥
3 ⎡ 2 ⎤ s
ρ VDh ( 992.2 kg m ) ( 0.855 m s )( 0.01m ) (1Ns kgm )

3 2
Re = = = 13, 400
µ 6.34×10-4 N-s/m 2
Nuk ( 81.7 ) (0.631W/mK) W
Nu = 0.023Re0.8Pr0.4 = 0.023 (13400)0.8 (4.19)0.4 = 81.7 → h= = = 5160 2
Dh 0.01m m K
Ts = 40o C +
(10 W m ) ⎡⎣(0.04m) −(0.02m) ⎤⎦ = 40 C+2.73 C=42.7 C
7 3 2 2
o o o
Answer
( 5160 W/m K ) 4 ( 0.04m )( 5.32m )
2
12- 41
12-40 Pasteurization is the sterilization of milk to ensure no diseases are transmitted with the milk. Consider a
flow of 1.5 kg/s of milk whose temperature must be raised from 35 ºC to 75 ºC in a 2-cm tube. The wall
temperature is 100 ºC. Milk properties are: ρ = 1030 kg/m3, µ = 2.12 × 10-3 Ns/m2, cp = 3850 J/kgK, and k
= 0.6 W/mK. Determine the required tube length (in m).
Approach:
With the given information, the analysis in Section
12.5 for a constant wall temperature heat exchanger
can be used to calculate the required length. The
heat transfer coefficient must be evaluated.
Assumptions:
1. Only the milk thermal resistance is taken into
account.
Solution:
For a constant wall temperature heat exchanger, the governing equation from Section 12.5 is:
⎛ hA ⎞
Tout = Tin + (Tin − Tw ) exp ⎜ −
⎜ mc ⎟⎟
⎝ p ⎠
where A = π DL . Substituting this in and solving for L:

mc p ⎛ Tout − Tw ⎞
L=− ln ⎜ ⎟
hπ D ⎝ Tin − Tw ⎠
To calculate the heat transfer coefficient, we need the Reynolds number:
ρ VD m 4m
Re = → V= → Re =
µ ρ Ax πµ D
Using the given fluid properties
4 (1.5 kg s ) ( Nis 2 kg im )
Re = = 45, 040
π ( 2.12 × 10-3 Nis m 2 ) ( 0.02 m )
µ c p ( 2.12×10 Nis m ) ( 3850 J kgiK )
-3 2
Pr = = = 13.6
k 0.6 W miK
Nu = 0.023Re0.8 Pr 0.4 = 0.023 ( 45040 ) (13.6 )
0.8 0.4
= 345
Nu ik ( 345 )( 0.6 W miK )
h= = =10,360 W m 2 iK
L 0.02 m
(1.5 kg s )( 3850 J kg iK ) ⎛ 75 − 100 ⎞
L=− ln ⎜ ⎟ = 8.48m Answer
(10360 W m i K ) π ( 0.02 m )
2
⎝ 35 − 100 ⎠
12- 42
12-41 Pressurized liquid water enters a 2-cm diameter, 6-m long tube at 20 ºC at a flow rate of 0.5 kg/s. The tube
surface temperature is constant, and the total power transferred to the water is 150 kW. Determine the
surface temperature (in ºC).
Approach:
This is a constant wall temperature heat exchanger.
Using the analysis given in Section 12.5, we can
determine the required surface temperature.
Assumptions:
2. Liquid water is an ideal liquid with constant
specific heat.
Solution:
For a constant wall temperature heat exchanger, (with Rtot = 1 hA ) the exit temperature can be calculated from:
Tout = Ts + (Tin − Ts ) exp ( − hA mc p )
Tout − Tin exp ( − hA mc p )
Solving for Ts =
1 − exp ( − hA mc p )
The outlet temperature can be obtained from an energy balance on the water. Assuming steady, no work, no
potential or kinetic energy effects, and an ideal liquid with constant specific heat so that ∆h = cp ∆T:
Q = mc p (Tout − Tin ) → Tout = Tin + Q/mc p
All the properties should be evaluated at the average water temperature. From Appendix A-6 and assuming Tout
≈100 oC, Tavg = (20 + 100)/2 = 60 oC, cp = 4.181 kJ/kg K, ρ = 983.2 kg/m3, k = 0.653 W/mK, µ = 4.52×10-4
N⋅s/m2, Pr = 2.89
150kW (1000J/s 1kW )
Tout = 20 o C + = 91.8 o C
( 0.5kg/s )( 4.181kJ/kgK )(1000 J kJ )
To calculate h, we need the Reynolds number
ρ VD m m 4m
Re = ⇒ V= = ⇒ Re =
µ ρ A ρ (π /4 ) D 2 π µD
4 ( 0.5kg/s ) ( Ns 2 kgm )
Re = = 70, 400
π ( 4.52 ×10-4 N ⋅ s/m 2 ) ( 0.02m )
Nu = 0.023Re0.8Pr0.4 = 0.023(70400)0.8(2.89)0.4 = 266
Nuk ( 266 )( 0.653W/mk )
h= = = 8670W/m 2 K
D 0.02m
hA ( 8670W/m K ) π ( 0.02m )( 6m )(1J/s 1W )
2
= = 1.564
mc p ( 0.5kg/s )( 4.181 kJ/kgK )(1000 J kJ )
( 91.8 + 273) − ( 20 + 273)exp ( −1.564 )
Ts = = 383.8K = 110.8 o C Answer
1 − exp ( −1.564 )
12- 43
12-42 Parts of the Alaskan oil pipeline (1-m in diameter) are buried 3-m below the surface of the earth (k = 0.65
W/m·K) and covered with 20-cm of insulation (k = 0.05 W/m·K). Pumping stations are 60-km apart. To
decrease pumping power the oil is heated to about 100 ºC (to reduce its viscosity) before it enters the
pipeline at a pumping station. In the winter the surface temperature of the earth is -30 ºC. Assume the oil
properties can be approximated with those of unused oil given in the appendix. For a flow rate of 0.5 m3/s,
and using properties evaluated at the average temperature of the oil, determine:
a. the oil temperature when it reaches the next pumping station (in ºC)
b. the heat transfer required at the pumping station to raise the oil temperature back to 100 ºC (in W)
c. the pumping power required (in W)
d. the pumping power if the inlet oil temperature is 50 ºC instead of 100 ºC (in W).
Approach:
This is a heat exchanger with one fluid with constant
temperature, so that analysis given in Section 12.5 can
be used to calculate the outlet temperature. Once that
temperature is known, conservation of energy is used
to calculate the heat transfer rate. The pumping power
also is calculated with conservation of energy, but first
the pressure drop is needed.
Assumptions:
2. The heat transfer coefficient is uniform.
4. Oil is an ideal liquid with constant specific heat.
Solution:
a) For a constant wall temperature heat exchanger, the exit temperature is calculated with
Tout = Ts + (Tin − Ts ) exp ( − 1 mc p Rtot )
Three resistances contribute to Rtot.
1 ln ( D2 D1 ) 1
Rtot = + +
k ground S 2π kins L hA
First evaluating the ground resistance, we assume two-dimensional conduction in the ground with constant kground,
and from Table 11-1 for this configuration z ~D
S = 2π L/cosh −1 ( 2 z/D ) = 2π ( 60000m ) /cosh -1 ⎡⎣ 2( 3m ) 1m ⎤⎦ = 152,100m
Evaluating the oil properties from Appendix A-6 at Tavg ≈ 370 K, ρ = 841.8 kg/m3, cp = 2.206 kJ/kgK, µ = 186 ×
10-4 N⋅s/m2, k = 0.137 W/mK, Pr = 300
m = ρ VA = ( 841.8 kg m3 )( 0.5m3 /s ) = 421kg/s
4m 4( 421kg/s )
For a circular tube Re = = = 28,800
π µ D π (186 × 10-4 N ⋅ s/m 2 ) (1m )
Nu = 0.023 Re0.8 Pr 0.3 = 0.023 ( 28,800 ) ( 300 )
0.8 0.3
= 470
Nuk ( 470 )( 0.137W/mK )
h= = = 64.5W/m 2 K
D 1m
1 1n (1.4/1) 1
Rtot = + + = 2.80 ×10-5 K/W
( 0.65 W mK )(152,100m ) 2π ( 0.05W mK )( 60000m ) ( 64.5 W m K )π (1m )( 60000m )
2
Tout = −30 0 C + ⎡⎣100 − ( − 30 ) ⎤⎦ K exp ⎡⎣ −1/ ( 421kg s )( 2.206 kJ kgK ) ( 2.80 ×10-5 K W ) (1000 J kJ ) ⎤⎦
= 95.1 oC = 368.1 K Answer
Using this temperature, Tavg = 370.5 K. This is close enough to our assumed value so no iteration is required.
12- 44
b) Applying conservation of energy to the oil, and assuming steady, no work, negligible potential or kinetic
energy effects, and ideal liquid with constant specific heat so that ∆h = cp ∆T
Q = mc p (Tin − Tout ) = ( 421kg s )( 2.206 kJ kgK )(100 − 95.1) K (1000 J kJ ) = 4.55 ×106 W = 4550kW Answer
c) For pumping power, we define a control volume around the pump. Apply conservation of energy, and assume
steady, adiabatic, negligible potential and kinetic energy effects, isothermal, and incompressible:
W = V ∆P
L ρV2
Pressure drop is ∆P = f
D 2
V 4V 4 ( 0.5 m3 s )
V= = = = 0.637m/s
Ax π D 2 π (1m )
2
For the friction factor, we need the Reynolds number, which is the same as we used above.
−2
f = ⎡⎣ 0.790 1n ( 28,800 ) − 1.64 ⎤⎦ = 0.0239
kg ⎞ ( 0.637 m s ) ⎛ Ns 2 ⎞
2
⎛ 60000m ⎞ ⎛ N
∆P = ( 0.0239 ) ⎜ ⎟⎜ 841.8 3 ⎟ ⎜ ⎟ = 244, 700 2
⎝ 1m ⎠ ⎝ m ⎠ 2 ⎝ kgm ⎠ m
Finally, pumping power is
⎛ m3 ⎞ ⎛ N ⎞ ⎛ 1J ⎞⎛ 1W ⎞
W = − ⎜ 0.5 ⎟ ⎜ 244700 2 ⎟ ⎜ ⎟⎜ ⎟ = − 122,300W = − 122.3kW Answer
⎝ s ⎠⎝ m ⎠ ⎝ 1N ⋅ m ⎠⎝ 1J/s ⎠
d) Using oil properties from Appendix A-6 for Tin ≈ 50 oC → use Tavg ≈ 320 K, ρ = 871.8 kg/m3 µ =
1410 × 10-4 N⋅s/m2
4 ( 421kg/s )
Re = = 3800
π (1410 × 10-4 N ⋅ s/m 2 ) (1m )
f = 0.0421
2
⎛ 60000m ⎞ ⎛ 0.637 ⎞
∆P = ( 0.0421) ⎜ ⎟ ( 871.8 ) ⎜ ⎟ = 447, 000N/m
2
⎝ 1 ⎠ ⎝ 2 ⎠
W = - (0.5) (447,000) = - 223,500 W = - 223.5 kW Answer
Comments:
Because of the steep change in viscosity with changing temperature, the pumping power increases dramatically
when the oil temperature is decreased.
12- 45
12-43 Pressurized liquid water flowing inside a tube at a rate of 1 kg/s is to be heated from 25 ºC to 90 ºC using
condensing steam. The 304 stainless steel tube has an inside diameter of 25-mm, a wall thickness of 1-mm,
and a length of 6-m. The condensing coefficient on the outside of the tube is 6,500 W/m2·K. Determine:
a. the steam temperature and pressure required (in ºC and kPa)
b. the condensation rate of the steam assuming the steam enters as a saturated vapor and exits as a
saturated liquid (in kg/s).
Approach:
determine the required steam temperature.
Assumptions:
Solution:
For a heat exchanger with one constant temperature fluid, Tout = Tst + (Tin − Tst ) exp ( − 1 mc p Rtot )
Tout − Tin exp ( −1/mc p Rtot ) 1 1n ( Do /Di ) 1
Solving for Tst = where Rtot = + +
1 − exp ( −1/mc p Rtot ) hi Ai 2π kL ho Ao
From Appendix A-2, for 304 stainless steel, k = 14.9 W/mK. The inside heat transfer coefficient requires Re with
properties from Appendix A-6 evaluated at the average temperature, Tavg = (90 + 25)/2 = 56.5 oC, so by
interpolation µ = 4.70 × 10-4 N⋅s/m2, k = 0.650 W/mK, Pr = 3.02, cp = 4.179 kJ/kg K
ρ V Di m 4m 4 (1kg s ) ( N ⋅ s 2 kgm )
Re = → V = → Re = = = 108, 400
µ ρ (π /4 ) Di2 π µ Di π ( 4.70 ×10-4 N ⋅ s/m 2 ) ( 0.025m )
This is turbulent flow, so using the Gnielinski correction
( f 8 )( Re −1000 ) Pr
Nu =
1 + 12.7( f 8 ) ( Pr 2 / 3 − 1)
1/ 2
f = [ 0.7901n Re − 1.64] = ⎡⎣0.7901n (108400 ) − 1.64 ⎤⎦ = 0.0177

−2 −2
( 0.0177/8)(108360 −1000 )( 3.02 ) Nuk ( 435 )( 0.650W/mk )

Nu = = 435 → h = = =11,300W/m 2 K
1 + 12.7 ( 0.0177/8 ) ⎡( 3.02 ) −1⎤ Di 0.025m
1/ 2 2/3
⎣ ⎦
1 1n ( 27/25 ) 1
Rtot = + +
(11,300 W m K )π ( 0.025m )( 6m ) 2π (14.9W mK )( 6m ) ( 6500 W m K )π ( 0.027m )( 6m )
2 2
= 0.000188 + 0.000137 + 0.000302 = 0.000627 K/W

1 1 (1J/s 1W )
= = 0.382
mc p Rtot (1kg s )( 4.179kJ kgK )( 0.000627 K W )(1000 J kJ )
( 90 + 273) - ( 25 + 273) exp ( - 0.382 )
Tst = = 502.9K = 230 o C Answer
1- exp ( - 0.382 )
From the saturated water table, Appendix A-10, at 230 oC, P = 2795 kPa Answer
b) For the condensation rate, we apply conservation of energy to a control volume around the steam and liquid
water. Assume steady, adiabatic, no work, no potential or kinetic energy effects, and an ideal liquid water:
mst h fg = mw c p , w (Tout − Tin )
mw c p , w (Tout − Tin ) (1kg s )( 4.179 kJ kgK )( 90 − 25 ) K kg
mst = = = 0.150 Answer
h fg 1813.8 kJ kg s
with hfg from the steam table.
12- 46
12-44 The oil from a large Diesel engine flows through an oil cooler before it is returned to the engine. Consider
a flow rate of 0.1 kg/s that must be cooled from 90 ºC to 40 ºC by passing through a thin-walled tube with a
diameter of 12.7 mm. Air at 30 ºC is in crossflow outside the tubes with a velocity of 10 m/s. Determine
the required tube length (in m).
Approach:
determine the required length. The outside heat
transfer coefficient must be determined.
Assumptions:
3. Ignore wall resistance.
Solution:
For a heat exchanger with one constant temperature fluid: Tout = T f + (Tin − Tout ) exp ⎡⎣ − 1/mc p Rtot ⎤⎦
⎛ Tout − T f ⎞⎛ 1 1 ⎞
− mc p 1n ⎜ ⎟⎟ ⎜ + ⎟
⎜ T −T
where Rtot =
1
+
1
=
1 ⎛ 1
+
1 ⎞
L= ⎝ in f ⎠ ⎝ hoil hair ⎠
⎜ ⎟ Solving for L
hoil π DL hair π DL π DL ⎝ hoil hair ⎠ πD
The oil heat transfer coefficient requires the Reynolds number with fluid properties at Tavg = (90 + 40)/2 = 65
o
C=338 K. From Appendix A-6 at 340 K µ = 531×10-4 N⋅s/m3, k = 0.139 W/mK, Pr = 793, cp = 2.076 kJ/kg K
ρ VD m 4m 4 ( 0.1kg s ) ( N ⋅ s 2 kgm )
Re = → V = → Re = = = 189
µ ρ (π /4 ) D 2 π µ D π ( 531× 10−4 N ⋅ s/m 2 ) ( 0.0127m )
This is laminar flow, so the entrance length is Lent,t ≈ 0.037 Re Pr D = 0.037 (189) (793) (0.0127 m) = 70.4 m
We assume the tube length will be shorter than this so we will use the Seider-Tate equation:
Nu = 1.86 (Gz)1/3 (µ/µs)0.14 = 1.86 (Re Pr D/L)1/3 (µ/µs)0.14
This requires the tube length, so an iterative solution is needed. Assuming Ts ≈ 330 K, µs = 836×10-4 N⋅s/m2 and
a tube length of 40 m.
⎡ (189 )( 793)( 0.0127 ) ⎤ ⎛ 531× 10−4 ⎞ Nuk ( 6.33)( 0.139W/mK )
1/ 3 0.14
Nu =1.86 ⎢ ⎥ ⎜ −4 ⎟
= 6.33 → hoil = = = 69.2W/m 2 K
⎣ 40 ⎦ ⎝ 836 × 10 ⎠ D 0.0127
The air side heat transfer coefficient requires the Reynolds number with properties from Appendix A-7 at
Tfilm = (Ts + Tf)/2= (340 + 273 + 30)/2 = 322 K: ρ = 1.096 kg/m3, µ = 1.95 × 10-5 N⋅s/m2, k = 0.0279 W/mK, Pr
= 0.703
ρ VD (1.096kg m ) (10 m s )( 0.0127m ) ( N ⋅ s kgm )
3 2
Re = = = 7140
µ (1.95 ×10-5 N ⋅ s/m2 )
From Table 12-1 for crossflow over a cylinder Nu = 0.193 Re0.618 Pr1/3 = 0.193 (7140)0.618 (0.703)1/3 = 41.3
hair = Nuk D = ( 41.3)( 0.0279 ) 0.0127 = 90.8W/m 2 K
⎛ kg ⎞ ⎛ kJ ⎞ ⎛ J ⎞ ⎡ 40 -30 ⎤ ⎛ 1 1 ⎞ m2 K
- ⎜ 0.1 ⎟ ⎜ 2.076 ⎟ ⎜ 1000 ⎟1n ⎢ ⎜ + ⎟
⎝ s ⎠⎝ kgK ⎠ ⎝ kJ ⎠ ⎣ 90 -30 ⎦⎥ ⎝ 69.2 90.8 ⎠ W
L= = 237m
π ( 0.0127m )
This is much longer than the calculated entrance length, so we assume fully developed laminar flow with constant
properties for the constant wall temperature boundary condition.
( 3.66 )( 0.139 )
Nu = 3.66 → h = = 40.1W/m 2 K
0.0127
The air side h is the same as before, so calculating the new length:
L = 335 m Answer
Comments:
This is an impractical length. Multiple tubes in parallel should be used.
12- 47
12-45 For some applications, enhanced cooling capabilities are obtained by attaching a heat generating system to
a cold plate, which is maintained at a cold temperature by passing water through it. Consider the copper
cold plate (shown below) that has heat generating equipment attached to its top and bottom surfaces. Each
of the six channels is 6-mm square and 100-mm long, and the walls of each channel is 4-mm thick. If
chilled water at 10 ºC is pumped through the channels at a velocity of 0.5 m/s, and the surfaces of the cold
plate must stay below 45 ºC, determine:
a. the maximum allowable power (top and bottom surfaces) to the cold plate (in W)
b. the water outlet temperature (in ºC).
Approach: Assumptions:
Heat enters the cold plate from the top and bottom surfaces. An 1. The heat transfer is one dimensional.
energy balance on the water can give the total heat transfer but 2. The system is steady with no work and
the exit temperature is unknown. With the top and bottom no potential or kinetic energy effects.
surfaces at a constant temperature, we have a heat exchanger with 3. Water is an ideal liquid with constant
one constant temperature fluid, so the analysis from Section 12.5 specific heat.
is applicable. Examining the figure, a unit cell (shown) can be
analyzed that is representative of all 12 unit cells. We have heat
transfer from a fin. The heat transfer coefficient must be
determined once one unit cell is analyzed, we multiply by 12 to
determine the total performance of the cold plate.
Solution:
An energy balance is applied to the water flow. Defining a control volume that encompasses only the water, and
assuming steady, no work, negligible potential and kinetic energy effects, and ideal liquid with constant specific
heat so that ∆h = cp ∆T: Q = mc p (Tout − Tin )
The outlet temperature can be determined by applying the analysis from Section 12.5 for a heat exchanger with a
constant wall temperature. Tout = Tb + (Tin − Tb ) exp ⎡⎣ − 1/mc p Rtot ⎤⎦
The resistances that must be taken into account are the conduction through the plane wall with thickness δ and
through the fin, as described by the unit cell. Therefore, the total resistance is:
δ 1 NA
Rtot = + where η0 =1 − f (1 −η f )
kcu Asur ηo h Atot Atot
For a unit cell, N = 1
Asur = (S + W)L = (0.004 m + 0.006 m) (0.1 m) = 0.001 m2
Af = 2(W/2)L = (0.006 m) (0.1 m) = 0.0006 m2
Ab = WL = 0.0006 m2
Atot = Ab + NAf = 0.0006 m2 + (1)(0.0006 m2) = 0.0012 m2
The convective resistance and the fin efficiency require the heat transfer coefficient. From Appendix A-6
assuming Tout ≈ 30 oC, Tavg ≈ (10 + 30)/2 = 20 oC, ρ = 998.2 kg/m3, cp = 4.182 kJ/kg K, k = 0.603 W/mK,
µ = 9.85 × 10-4 N⋅s/m2, Pr = 6.83. In addition for copper, k = 401 W/mK. For the heat transfer coefficient, we
DH = 4 Ax /Pwetted = 4( W ) / ⎡⎣ 2 ( W + W ) ⎤⎦ = W = 0.006m
2
need the Reynolds number Re = ρ VDH µ with
Re =
( 998.2 kg m3 ) ( 0.5m s )( 0.006m )
= 3040
9.85 ×10-4 N ⋅ s/m 2
This is turbulent flow, so using the Gnielinski correlation
f = [ 0.7901n Re − 1.64] = ⎡⎣0.791n ( 3040 ) −1.64 ⎤⎦ = 0.0454
−2 −2
12- 48
( f 8)( Re −1000 ) Pr ( 0.0454/8 )( 3040 −1000 )( 6.83)
Nu = = = 22.7
1 + 12.7 ( f 8 ) ( Pr −1) 1 + 12.7 ( 0.0454/8 ) ( 6.832 / 3 − 1)
1/ 2 2/3 1/ 2
Nuk ( 22.7 ) ( 0.603 W/m K )

3
h= = = 2280W/m 2 K
DH 0.006m
The fin is a straight rectangular fin with an adiabatic tip, so fin efficiency is
tanh ( mW 2 )
1/2 1/ 2
mW W ⎡ hp ⎤ W ⎡ h2L ⎤
ηf = where = ⎢ ⎥ = ⎢ ⎥
mW 2 2 2 ⎣ k Ax ⎦ 2 ⎣kS L⎦
0.006m ⎡ ( 2280 W m K ) 2 ⎤
1/ 2
2
= ⎢ ⎥ = 0.160
2 ⎢⎣ ( 401W mK )( 0.004m ) ⎥⎦
η f = tanh ( 0.160 ) /0.160 = 0.992
(1)( 0.0006 )
ηo = 1 − (1 − 0.992 ) = 0.996
0.0012
For one unit cell
0.004m 1
Rtot = + = 0.377m 2 K/W
( 401W/mK ) ( 0.001m 2
) ( 0.996 ) ( 2280W/m K )( 0.0012m )
2 2
m = ρ V Ax = ( 998.2 kg m3 ) ( 0.5m/s )( 0.006m )( 0.006m ) = 0.0180kg/s

Tout = 45o C + (10 - 45 ) K exp ⎡⎣ -1/ ( 0.018kg/s )( 4.182 kJ/kgK )(1000 J kJ ) ( 0.377m 2 K /W ) ⎤⎦ =11.2 o C Answer
We have a total of 12 unit cells, so the total heat transfer rate is
Q = (12 )( 0.018 kg s )( 4.182 kJ kgK )(11.2 −10 ) K (1000 J kJ ) =1095W Answer
Comments:
Because the assumed Tout was higher than the calculated Tout, the fluid properties probably should be reevaluated
and the calculations repeated.
12- 49
12-46 The total thermal resistance between the outside and inside of a home consists of the external convective
resistance, the wall resistance, and the internal convective resistance. Adding additional insulation reduces
the heat transfer but it also changes the inside wall temperature. Compare the average natural convection
heat transfer coefficient on a 2.5-m tall wall for two situations:
a. inside air temperature of 22 ºC and wall temperature of 10 ºC
b. inside air temperature of 22 ºC and wall temperature of 17 ºC.
Approach:
This is a natural convection problem. We assume that
the wall can be approximated as a vertical flat plate.
Assumptions:
Solution:
The natural convection heat transfer coefficient requires the Rayleigh number with properties from Appendix A-7
evaluated at Tfilm = (Tw + Tf)/2.
1) Tfilm = (10 + 22)/2 = 16 0C = 289 K, so ρ = 1.221 kg/m3 ; µ = 1.77×10-5 N⋅s/m2 ; k = 0.0254 W/mk, Pr =
0.711
2) Tfilm = (17 + 22)/2 = 19.5 0C = 292.5 K, so ρ = 1.207 kg/m3, µ = 1.79×105 N⋅s/m2; k = 0.0256 W/mK; Pr =
0.710
1) Ra =
g β ρ 2 (Tw − f )δ 3
Pr
=
( 9.81m s ) (1 289K ) (1.221kg m ) ( 22 −10 )K ( 2.5m ) ( 0.711) = 2.15×10
2 3 2 3
10
µ2 (1.77 ×10 N ⋅ s/m )

-5 2 2
From Table 12-3, for a vertical wall

Nu = 0.1 Ra1/ 3 = 0.1( 2.15 ×1010 )
1/ 3
= 278
Nuk ( 278 )( 0.0254W/mK )
h= = = 2.83W/m 2 K Answer
H 2.5m
( 9.81)(1/292.5 )(1.207 ) ( 22 −17 )( 2.5) ( 0.710 )
2 3
2) Ra = = 8.46 ×109
(1.79 ×10 ) -5 2
9 1/3
Nu = 0.1 (8.46 × 19 ) = 204
( 204 )( 0.0257 )
h= = 2.09W/m 2 K (~ 26% smaller) Answer
2.5
Comments:
As can be seen, a small change in the driving temperature difference reduces the inside convective heat transfer
coefficient. So adding insulation helps in two ways: increased conduction resistance and increased convection
resistance.
12- 50
12-47 Consider again the light bulb in Problem P 12-23. For all the same conditions, determine the temperature
of the glass bulb if it is cooled by natural convection (in ºC).
Approach:
An energy balance on the light bulbs is needed. The
power into the glass (90% of 100W) is removed by
convection and radiation. The natural convection heat
transfer coefficient must be evaluated.
Assumptions:
2. Radiation is from a small body to a large area.
3. The light bulb is a sphere.
Solution:
A steady energy balance on the glass gives
Qin = Q rad + Qconv = εσ A (Ts4 − T f4 ) + hA (Ts − T f )
The light bulb is assumed to be a sphere and the surroundings are very large. The natural convection heat transfer
coefficient correlation for a sphere is
0.589 Ra1/ 4
Nu = 2 + 4/9
⎡1 + ( 0.469/Pr )9 /16 ⎤
⎣ ⎦
g βρ 2 (Ts − T f ) D 3 Pr
where Ra =
µ2
Evaluate the fluid properties from Appendix A-7 at Tfilm = (Ts + Tf)/2; because we seek Ts, we must first estimate
Ts = 200oC. Tfilm = (200 + 27)/2 = 113.5oC = 386.5K, ρ = 0.913kg/m3, µ = 2.23 × 10-5 N⋅s/m2, k = 0.0327
W/mK, Pr = 0.691
Ra =
( 9.81m/s )(1 386.5K ) ( 0.913 kg/m ) ( 200 − 27 ) K ( 0.08m ) ( 0.693) = 2.61×10
2 3 2 3
6
( 2.23×10 N ⋅ s/m )
−5 2 2
0.589 ( 2.61×10 ) 6 1/ 4
Nu = 2 + = 20.2
4/9
⎡1 + ( 0.469 0.691)9 /16 ⎤
⎣ ⎦
Nuk ( 20.2 )( 0.0327 W mK )
h= = = 8.26 W/m 2 K
D 0.08 m
Qin = 0.9 (100 W) = 90 W
Qrad = ( 0.85 ) ( 5.67 ×10−8 W m 2 K 4 ) ( 4π )( 0.04 m ) ⎡Ts4 − ( 300K ) ⎤ = 9.690 ×10−10 ⎡⎣Ts4 − 8.1×109 ⎤⎦
2 4
⎣ ⎦
Qconv = (8.26W/m2K) (4π) (0.04m)2 (Ts – 300)K = 0.1661(Ts – 300)
Combining these expressions
90 = 9.69 × 10-10 ⎡⎣Ts4 − 8.1×109 ⎤⎦ + 0.1661(Ts − 300 )
Solving Ts ~ 506.1 K = 233 oC Answer
This is reasonably closed to our assumed value of Ts so we will not iterate.
12- 51
12-48 In any design process decisions have to be made about placement of components. Cost, performance, and
maintainability are some of the criteria used. Consider the placement of a 1.2 W, 60-mm by 60-mm
electric component in a larger device. The component’s surface temperature must not exceed 85 ºC. The
air is quiescent at 25 ºC. Neglecting radiation, determine if the component can be located facing downward
or facing upward. (In other words, what is the component’s surface temperature if it is facing upward or
downward?)
Approach:
This is a natural convection heat transfer problem.
The maximum power transferred from a 85°C surface
to 25°C air should be calculated and compared to the
1.2W specification. Appropriate heat transfer
coefficient correlations for the two geometrics must be
used.
Assumptions:
1. Radiation is neglected.
2. The plates match the geometry for the correlations
in Table 12-3.
Solution:
Neglecting radiation, the heat transfer rate is calculated with Q = hA (Tw - T∞)
We need the Rayleigh number to calculate h. Evaluating air properties from Appendix A-7 at Tfilm = (85 + 25)/2 =
55oC = 328K, (by interpolation) ρ = 1.076 kg/m3, k = 0.0283 W/mK, µ = 1.99 × 10-5 N⋅s/m2,
ν = 18.5 × 10-6 m2/s, Pr = 0.702
g β (Tw − T∞ )δ 3
Ra = Pr
ν 2
From flat plates, the characteristic length is

δ = A/p = (0.06m) (0.06) / (4) (0.06m) = 0.015m
Ra =
( 9.81m/s ) (1/ 328K )(85− 25) K ( 0.015m ) ( 0.702 ) = 12400
2 3
(18.5×10 m /s )
−6 2 2
From Table 12-3 for a horizontal plate

1) Hot plate facing upward Nu = 0.54Ra1/4
= 0.54(12400)1/4 = 5.70
Nuk ( 5.70 )( 0.0283W/mK ) W
h= = = 10.8 2
δ 0.015m m K
Q = (10.8 W m 2 K ) ( 0.06m ) ( 85− 25 ) K = 2.33W
2
Answer
2) Hot plate facing downward

Nu = 0.27 Ra1/4 = 0.27 (12400)1/4 = 2.85
h = (2.85) (0.0283)/0.015 = 5.38 W/m2K
Q = (5.38) (0.06) (0.06) (85 – 25) = 1.16 W Answer
The safe choice is to use the location with the component facing upward. However, because facing downward is
close to satisfying the design specification, it might usable if other design considerations had to taken into
account.
Comment:
A second approach would be to estimate the wall temperature for the 1.2W power input. This would require an
iterative solution.
12- 52
12-49 To lower the viscosity of an oil before it is used in a process, an electric resistance heater 1.5-cm in
diameter and 30-cm long is immersed horizontally in a vat of unused engine oil which is at 20 ºC. If the
heater surface should not rise above 150 ºC so that the oil does not smoke, determine the maximum power
that can be dissipated in the heater (in W).
Approach:
This is a natural convection problem. Because of the
large length compared to the diameter, we will ignore
heat transfer from the two ends. The basic heat
( )
transfer rate equation Q = hA∆T is used to calculate
the maximum power. The heat transfer coefficient
must be evaluated.
Assumptions:
1. The system is steady.
2. All properties are constant and are evaluated at the
appropriate temperatures.
3. The ends of the heater are ignored.
Solution:
The heat transfer rate is calculated with: Q = hA (Ts − T f ) . Ignoring the ends, A = π DL .
From Table 12-3, the heat transfer coefficient correlation is:
2
⎧ ⎫
⎪ 0.387 Ra1 6 ⎪
Nu = ⎨0.6 + ⎬
9 16 8 27
⎪ ⎡1 + ( 0.559 Pr ) ⎤ ⎪
⎩ ⎣ ⎦ ⎭
g β (Ts − T f ) D 3 Pr
where Ra = . Evaluating the oil properties from Appendix A-6 at
ν2
T film = (150 + 20) / 2 = 85 o C=358 K , ν= 32.1×10-6 m2/s; k = 0.138 W/mK; Pr = 425, β = 0.7×10-3/K
( 9.81m s ) ( 0.7×10-3 K ) (150-20 ) K ( 0.015 m ) ( 425 )

3
Ra = = 1.24 × 106
( 32.1×10-6 m2 s )
2
2
⎧ ⎫
0.387 (1.24 × 106 )
16
⎪ ⎪
Nu = ⎨0.6 + ⎬ = 21.0
9 16 8 27
⎪ ⎡1 + ( 0.559 425 ) ⎤ ⎪
⎩ ⎣ ⎦ ⎭
Nu ik ( 21.0 )( 0.138 W miK )
h= = =193 W m 2 iK
D 0.015 m
Q = (193 W m 2 K ) π ( 0.015 m )( 0.30 m )(150-20 ) K=355 W Answer
Comments:
A factor of safety might be used to derate the power to ensure the surface temperature remains below 150 ºC. In
addition, as the oil temperature increases, lower power would be used to ensure the surface temperature remains
below 150 ºC.
12- 53
12-50 An electric resistance heater, 10-mm in diameter and 300-mm long, is rated at 550 W. If the heater is
horizontally positioned in a large tank of water that is at 20 °C, estimate the surface temperature of the
heater (in °C).
Approach:
We will ignore heat transfer from this ends of this
cylinder and assume that the tank is large enough to
not influence this heat transfer coefficient. Because
the driving force is the temperature difference and we
do not know it, an iterative solution is required.
Assumptions:
1. The ends of the heater are ignored.
2. Only natural convection occurs.
Solution:
The surface temperature can be determined from the basic heat transfer rate equation:
Q = hA (Tw − T∞ ) → Tw = T∞ + Q hA
We ignore the cylinder ends, so that A = πDL. The natural convection heat transfer coefficient for a horizontal
cylinder is used:
2
⎧ ⎫
⎪ 0.387 Ra1/ 6 ⎪ g β (Tw − T∞ ) D 3
Nu = ⎨0.6 + 8 / 27 ⎬
where Ra = Pr
⎪⎩ ⎡⎣1 + (0.559 Pr )9 /16 ⎤⎦ ⎪ ν2
⎭
The properties should be evaluated at the film temperature Tfilm =( Tw - T∞)/2, so as first approximation, estimate
Tw ~ 40oC. Therefore Tfilm = 30oC, and the properties from Appendix A-6 are β = 3.06 × 10-4/K, µ = 7.79 × 10-4
N⋅s/m2, ρ = 995.6 kg/m3, Pr = 5.26 , ν = 0.782 × 10-6 m2/s , k = 0.618 W/mK
Ra =
( 9.81m/s )( 3.06 ×10 /K ) ( 40 − 20 ) K ( 0.01m ) ( 5.26 ) = 516, 000
2 −4 3
( 0.782 ×10 m /s )
-6 2 2
2
⎧ 0.387 ( 516, 000 ) ⎫
Nuk (14.6 )( 0.618 W/mK )
1/ 6
⎪ ⎪
Nu = ⎨0.6 + 8 / 27 ⎬ = 14.6 → h = = = 902W/m 2 K
⎪⎩ ⎡⎣1 + (0.559 / 5.26) ⎤⎦ ⎪
9 /16 D 0.01m
⎭
550W
Tw = 20o C + = 84.7 o C
( 902W/m K )π ( 0.01m )( 0.3m )
2
This temperature is significantly about what we first estimated, so we iterate

Tfilm = (84.7 + 20)/2 = 52.3 oC → at 50 oC the fluid properties are: β = 4.95 × 10-4/k, µ = 4.88 × 15-4
N⋅s/m , ρ = 985.7 kg/m3, Pr = 3.15, k = 0.648 W/mK, ν = 4.95 × 10-6 m2/s
2
( 9.81) ( 4.95 ×10−4 ) (84.7 − 20 )( 0.01) ( 3.15 )

3
Ra = = 4.04 × 106
( 0.495×10 ) −6 2
2
⎧ ⎫
0.387 ( 4.04 × 106 )
1/ 6
⎪ ⎪
Nu = ⎨0.6 + 8 / 27 ⎬
= 25.4 h = (25.4) (0.648)/0.01 = 1650 W/m2K
⎪ ⎡1 + ( 0.559/ 3.15 )9 /16 ⎤ ⎪
⎩ ⎣ ⎦ ⎭
550
Tw = 20 + = 55.4 o C
(1650 ) π ( 0.01) ( 0.30 )
One more iteration, based on trends, let’s assume Tw ~ 70oC ⇒ Tfilm ~ 45oC, β = 4.23 × 10-4/K, µ = 5.77 × 10-4
m2/s, ρ = 990.2 kg/m3, Pr = 4.23, k = 0.637 W/mK, ν = 0.583 × 10-6 m2/s →
Ra = 2.58×106, Nu = 22.8 , h = 1450 W/m2K → Tw = 60.2 oC
Another iteration gives Tw ~ 62oC. Answer
12- 54
12-51 A passive solar heating technique is to use a massive masonry wall (a Trombe wall) to absorb solar energy
and then to release it slowly when the air temperature surrounding the wall is lower than that of the wall.
Consider a long 3-m tall wall, well insulated on its backside, that has a net radiant solar energy flux into the
wall of 150 W/m2. The air temperature is 21 ºC. Assuming that the temperature of the wall changes very
slowly and the wall operation can be approximated as quasi-steady, determine the average surface
temperature of the wall (in ºC).
Approach:
This is a natural convection problem that can be
solved with the basic rate equation ( Q = q″A = hA
∆T). The heat transfer coefficient must be calculated.
Assumptions:
1. The system is quasi-steady.
Solution:
Assuming a steady state system, the convective rate equation is: Q = q ′′A = hA (Ts − T f )
Solving for Tf : Ts = T f + q ′′/h
Using an average heat transfer coefficient will give us the average wall temperature. We need to evaluate the
Rayleigh number at Tfilm = (Ts + Tf)/2. Because Ts is what we seek, we must first estimate Ts, evaluate h, and then
calculate Ts. If the estimate and calculated values are close, then we are done; if not, we should iterate.
From Appendix A-7 and assuming Ts ~ 73 oC → Tfilm = (21 + 73)/2 = 47 oC = 320 K, ρ = 1.103 kg/m3,
µ = 1.94 × 10-5 N⋅s/m2, k = 0.0278 W/mK, Pr = 0.703
2
⎛ m ⎞⎛ 1 ⎞⎛ kg ⎞
1.103 3 ⎟ ( 73 − 21) K ( 3m ) ( 0.703)
3
9.81 2 ⎟ ⎜
g βρ (Ts − T f ) H Pr ⎝
2 3 ⎜ ⎟
s ⎠ ⎝ 320K ⎠ ⎝
⎜
m ⎠
Ra = = = 9.78 ×1010
µ (1.94 ×10 N ⋅ s/m )
2 -5 2 2
Nu = 0.1 Ra 1/ 3 = 0.1( 9.78 ×1010 )

1/ 3
From Table 12-3, = 461
Nuk ( 461)( 0.0278W/mK )
h= = = 4.27W/m 2 K
H 3m
150W/m 2
Ts = 21o C + 2
= 56.1 o C
4.27W/m K
Assuming Ts ~ 56.4 oC gives Tfilm ~ 312 K so that ρ = 1.131 kg/m3, µ = 1.90 × 10-5 N⋅s/m2, k = 0.0272 W/mK,
Pr = 0.704
( 9.81)(1/312 )(1.131) ( 56.4 − 21)( 3) ( 0.704 )
2 3
Ra = = 7.50 x1010
(1.90 ×10 )
-5 2
Nu = 0.1 (7.50 × 1010)1/3 = 421

( 421)( 0.0272 )
h= = 3.82W/m 2 K
3
Ts = 21 + 150/3.82 = 60.2 0 C Answer
This is close enough, and another iteration is not needed.
12- 55
12-52 A steam-heated cooking vat in a food processing plant has a bottom that is 1.5 m by 1.5 m. The vat is filled
with water, initially at 25 ºC, and the bottom is heated with condensing steam at 105 ºC. Determine:
a. the initial heat transfer rate from the bottom of the vat to the water (in W)
b. how long it would take for the water temperature to rise to 30 ºC if the water depth is 60-cm (in
min).
Approach:
We will assume that the heat transfer coefficient from
an upward facing heated plate is applicable. The time
to heat the water from 25 0C to 30 0C can be
determined with conservation of energy.
Assumptions:
1. The heat transfer coefficient correlation for an
upward facing heated plate is applicable.
2. The system is closed with no work or potential or
3. The condensing resistance is ignored.
4. The heat transfer rate is constant.
Solution:
a) The initial heat transfer rate is calculated with Q = hA (Tw − T∞ )
We assume the condensing heat transfer coefficient is very high and the conduction resistance through the vat
bottom is very small, so that the wall temperature is the same as the condensing temperature.
The heat transfer coefficient calculation requires the Rayleigh number with properties evaluated at
Tfilm = (T∞ + Tw)/2 = (25 + 105)/2 = 65 oC. From Appendix A-6 ρ = 980.5 kg/m3, k = 0.658 W/mK, µ = 4.21 ×
10-4 N⋅s/m2, ν = 0.429 × 10-6 m2/s, β = 5.53 × 10-4/K, Pr = 2.67
The characteristic length is δ = A/p = (1.5 m) (1.5 m) / (4 × 15 m) = 0.375 m
Ra =
g β (Tw − T∞ )δ 3 Pr
=
( 9.81m/s )( 5.53×10 /K ) (105 − 25)K ( 0.375m ) ( 2.67 ) = 3.32 ×10
2 -4 3
11
ν2 ( 0.429 ×10 m /s )
-6 2
From Table 12-3, for an upward facing heated plateNu = 0.15Ra1/3

We are outside the range of the correlation’s applicability, but with nothing else available we are forced to use
this.
Nu = 0.15 (3.32 × 1011)1/3 = 1040
Nuk (1040 )( 0.658 W/mK )
h= = =1820W/m 2 K
δ 0.375m
Q = (1820 W/m2K) (1.5 m) (1.5 m) (105 – 25) K = 328,000 W Answer
b) The time to reach 30 oC can be determined by conservation of energy applied to a closed system. Assume no
work and no potential or kinetic energy changes:
dU d
Q= = ( mu )
dt dt
For an ideal liquid, du = c dT, c ~ cp ~ cv
dT
Q = mc p
dt
Assuming the heat transfer rate is approximately constant, we integrate this expression:
mc p ∆T
∫ Qdt = ∫ mc p dT → t = Q
At the average bulk temperature (25 + 30)/2 = 27.5 oC, cp = 4.177 kJ/kg K, ρ = 996 kg/m3
t=
( 996 kg m3 ) (1.5m )(1.5m )( 0.6m )( 4.177 kJ kgK )(1000 J kJ )( 5K ) 85.6s =1.43 min Answer
( 328,000W )(1J/s 1W )
12- 56
12-53 You devise a transient heat transfer experiment to measure the natural convection heat transfer coefficient
on a 2024-T6 aluminum sphere. Initially, the 3-cm sphere is at a uniform temperature of 90 ºC as measured
by a thermocouple inserted into the sphere’s center. You plunge the sphere into 10 ºC water and record the
center temperature as it decreases with time. The center temperature reaches 80 ºC after 0.83 s, 50 ºC after
5.66 s, and 40 ºC after 9.78 s. Determine:
a. the average heat transfer coefficient as the sphere changes temperature from 90 to 80 ºC (in W/m2·K)
b. the average heat transfer coefficient as the sphere changes temperature from 50 to 40 ºC (in
W/m2·K)
c. whether or not this approach is valid.
Approach:
This is a transient heat transfer problem. Because the
sphere is aluminum and natural convection has a
relatively low heat transfer coefficient h, we will
assume a lumped systems analysis is applicable to this
problem, and the recorded sphere temperatures are
uniform. We also assume h is relatively constant over
the time intervals.
Assumptions:
1. The lumped systems analysis is valid.
2. The heat transfer coefficient is constant over the
time intervals.
Solution:
T − Tf ⎛ − hAt ⎞
= exp ⎜
⎜ ρVc ⎟⎟
The lumped systems equation is
Ti − T f ⎝ p ⎠
− ρV c p ⎛ T − T f ⎞
Solving for h: h = ln ⎜
At ⎜ T − T ⎟⎟
⎝ i f ⎠
For a sphere V = 4π R 3 3 = π D 3 6 and A = 4π R 2 = π D 2 . Combining these with the main equation:

− ρ Dc p ⎛ T − Tf ⎞
h= ln ⎜
6t ⎜ T − T ⎟⎟
⎝ i f ⎠
From Appendix A-2 for 2023-T6 aluminum, ρ= 2770 kg/m3, k = 177 W/mK, cp = 875 J/kgK
a) For the average heat transfer coefficient as the sphere changes temperature from 90 to 80 ºC
−( 2770 kg m3 ) ( 0.03m )( 875J/kgK ) ⎡ 80 -10 ⎤
h= ln ⎢ ⎥ =1950 W/m K
2
Answer
6 ( 0.83s ) ⎣ 90 − 10 ⎦
b) For this time interval, we must reset Ti = 50 oC T = 40 oC, and the time interval is t = 9.785 – 5.66s = 4.125
− ( 2770 )( 0.03)( 875 ) ⎡ 40 -10 ⎤
h= ln ⎢ ⎥ = 846W/m K
2
Answer
6 ( 4.12 ) ⎣ 50 -10 ⎦
c) To determine if this approach is valid, check the Blot number.
h( D/6 ) (1950W/m K ) ( 0.03m )
2
hL
Bi = char = = = 0.055 < ∼ 0.1
k k 6 (177W/mk )
So the approach is valid. The second condition does not need to be checked explicitly because that h is smaller
than the one used.
12- 57
12-54 Farmer Brown installs an electric resistance heater in the watering trough for his cows so that the water will
not freeze during the long cold winter. He does not want a cow to burn its tongue if it accidentally touches
the heater. He places a 25-mm diameter, 30-cm long, 100-W heater horizontally in the water which is
maintained at 5 ºC. Determine:
a. the surface temperature of the heater (in ºC)
b. the surface temperature of the heater if the water trough develops a leak, all the water drains out, and
the air temperature is -15 ºC (in ºC).
Approach:
This is a natural convection problem. We ignore the heat
transfer from the heater ends and use the rate equation
( Q = hA ∆T ) to calculate the surface temperature. The
appropriate heat transfer coefficient must be evaluated.
Assumptions:
1. Ignore heat transfer from the ends.
3. There is no radiation in air.
Solution:
a) The surface temperature can be determined from the basic convective heat transfer rate equation (ignoring heat
transfer from the ends): Q = hπ DL (Ts − T f ) → Ts = T f + Q hπ DL
The natural convection heat transfer coefficient for a horizontal cylinder is
2
⎧ ⎫
⎪ 0.387 Ra1/ 6 ⎪ g βρ 2 (Ts − T f ) D 3 Pr
Nu = ⎨0.6 + ⎬ where Ra =
⎪
8 / 27
⎡1 + ( 0.559/Pr )9 /16 ⎤ ⎪ µ2
⎩ ⎣ ⎦ ⎭
Evaluating the properties of water from Appendix A-6 at Tfilm = (Ts – Tf)/2, we first assume Ts ≈ 30 oC →
Tfilm ≈ = 290 K, ρ = 999 kg/m3, µ = 10.8 × 10-4, N⋅s/m2, k = 0.598 W/mK, Pr = 7.56, β = 1.74 × 10-4/K
( 9.81m/s )(1.74 ×10 /K )( 999kg/m3 ) ( 30 − 5 ) K ( 0.025m ) ( 7.56 )
2 -4 2 3
Ra = = 4.31×106
(10.8×10 -4
N ⋅ s/m 2 2
)
2
⎧ ⎫
0.387 ( 4.31×106 )
1/ 6
⎪ ⎪ Nuk ( 27.5 )( 0.598W/mK )
Nu = ⎨0.6 + ⎬ = 27.5 → h= = = 657W/m 2 K
9 /16 8 / 27 D 0.025m
⎪ ⎡1 + ( 0.559/7.56 ) ⎤ ⎪
⎩ ⎣ ⎦ ⎭
Ts = 5 o C +100W ⎡⎣( 657 W m 2 K ) π ( 0.025m )( 0.30m ) ⎤⎦ =11.5 o C
The first estimated Ts is significantly larger than the calculated temperature. The fluid properties should be re-
evaluated and a new Ts calculated. Using Ts = 11.5 oC, and properties evaluated at Tfilm ~ 280 K.
Ra = 2.33×105 → Nu = 12.1 → h = 282W/m 2 K → Ts = 20.1 o C
Guessing Ts ≈ 16 oC and Tfilm ≈ 285K: Ra =1.13×106 → Nu = 18.8 → h = 443W/m 2 K → Ts = 14.6 0 C Answer
This is close enough, so additional iterations are not required.
b) Now, with air we also assume no radiation. Following the same procedure as above, guess Ts ~ 500 oC,
Tfilm ~ 515K, ρ = 0.685 kg/m3, µ = 2.75 × 10-5 N⋅s/m2, k = 0.0417 W/mK, Pr = 0.680
Ra = ( 9.81)(1/515 )( 0.685 ) ( 500 + 15 )( 0.025 ) ( 0.680 )
2 3
( 2.75×10 ) -5 2
= 64, 600
Nu = 6.91 → h = 11.6W/m K 2
→ Ts = −15 C +100 ⎡⎣(11.6 )π ( 0.025 )( 0.30 ) ⎤⎦ = 353 o C
o
Iterating a second time with Ts ≈ 353 oC, and evaluating properties at Tfilm ≈ 480 K
Ra = 77650 → Nu = 7.25 → h =11.4W/m 2 K → Ts = 357 o C Answer
Comment:
At this temperature, radiation is significant, so more calculations with only convection are not justified.
12- 58
12-55 The manufacturer of the electric resistance heater described in Problem P 12-54 wants to expand her sales
and considers using the heater for fuel oil tanks, too. Concern about a possible fire hazard if the oil
anywhere in an oil tank reaches too high a temperature makes her contact a consulting engineer for an
analysis. If the oil has properties of unused oil and the oil is at 0 ºC, determine the surface temperature of
the heater (in ºC).
Approach:
This is a natural convection problem. We will ignore
heat transfer from the two ends. The basic convective
heat transfer equation ( (Q = hA ∆T ) is used to
calculate the surface temperature. The appropriate
heat transfer coefficient must be evaluated.
Assumptions:
1. Ignore heat transfer from the ends.
Solution:
Ignoring heat transfer from the two ends, the basic convective heat transfer rate equation is used to calculate the
surface temperature: Q = hπ DL (Ts − T f ) → Ts = T f + Q hπ DL
The natural convection heat transfer coefficient for a horizontal cylinder is:
2
⎧ ⎫
⎪ 0.387 Ra1/ 6 ⎪ g β ρ 2 (Ts − T f ) D 3 Pr
Nu = ⎨0.6 + 8 / 27 ⎬
where Ra =
⎪ ⎡1 + ( 0.559/Pr )9 /16 ⎤ ⎪ µ2
⎩ ⎣ ⎦ ⎭
Evaluating the oil properties from Appendix A-6 at Tfilm = (Ts + Tf)/2, we first must estimate Ts ≈ 35 oC →
Tfilm ≈ 290.5 K, ρ = 890 kg/m3, µ = 9990×10-4 N⋅s/m2, k = 0.145 W/mK, Pr = 12900, β = 0.7 × 10-3 /K
( 9.81m s )( 0.7 ×10 /K )( 890kg m3 ) ( 35 − 0 ) K ( 0.025m ) (12900 )
2 -3 2 3
Ra = = 38,500
( 9990 ×10 N ⋅ s/m 2 )
−4 2
2
⎧ ⎫
0.387( 38,500 )
1/ 6
⎪ ⎪
Nu = ⎨0.6 + ⎬ = 8.07
9 /16 8 / 27
⎪ ⎡1 + ( 0.559/12900 ) ⎤ ⎪
⎩ ⎣ ⎦ ⎭
Nuk ( 8.07 )( 0.145W/mK )
h= = = 46.8W/m 2 K
D 0.025m
100W
Ts = 0o C + = 90 o C
( 46.8W/m K ) π ( 0.025m )( 0.30m )
2
Because this calculated Ts is significantly different than the estimated Ts, we must iterate. Guessing Ts = (35 +
90)/2 = 62.5 oC ⇒ Tfilm ~ 305K, ρ = 881 kg/m3, µ = 3700 × 10-4 N⋅s/m2, k = 0.145 W/mK, Pr = 4900, β = 0.7
× 10-3/K
( 9.81) ( 0.7 ×10-3 ) (881) ( 62.5 − 0 )( 0.025 ) ( 4900 )
2 3
Ra = =1.86 ×105
( 3700 ×10 ) -4 2
Nu =12.4 → h = 72.0W/m 2 K → Ts = 59 o C Answer

Close enough so no more iterations are required.
12- 59
12-56 The coils in electric power transformers mounted on telephone poles in every neighborhood are cooled by
oil. If the coils reach too high a temperature, the transformer can fail. To prevent this problem, the
transformer is externally cooled by air. The worst-case scenario occurs on hot, still summer days.
Consider a transformer that is 55-cm in diameter and 1.5-m tall on a day when the temperature is 40 ºC.
Assume the heat transfer coefficients on the ends are the same as on the cylinder sides. If 250 W must be
dissipated, determine:
a. the surface temperature of the transformer (in ºC) when ignoring radiation
b. the surface temperature if radiation is included with an emissivity of 0.6 (in ºC).
Approach:
This is a natural convection problem. Using the basic
heat transfer rate equation ( Q = hA ∆T), the surface
temperature can be determined. An iterative solution
must be used to find the natural convective heat
transfer coefficient.
Assumptions:
2. The same heat transfer coefficient is used on sides
and ends.
3. The transformer is “small” relative to its
surroundings.
Solution:
a) The rate equation is Q = hA∆T = hA (Ts − T f ) . Solving for the surface temperature Ts = T f + Q hA
where A = π DH + 2π D /4 = π (0.55 m) (1.5 m) + 2π (0.55 m )/4 = 3.070 m
2 2 2
We assume that h can be approximated by that on a vertical wall, and this also is assumed to be valid for
the two ends. The Rayleigh number is evaluated with properties from Appendix A-7 at Tfilm = (Ts + Tf)/2. We
assume Ts ≈ 65 oC, so Tfilm ≈ (65 + 40)/2 = 52.5 oC ≈ 325 K, and
ρ = 1.086 kg/m3, µ = 1.96 × 10-5 N⋅s/m2, k = 0.0282 W/mK, Pr = 0.703
g β ρ 2 (Ts − T f ) H 3 Pr ( 9.81m s ) (1 325K ) (1.086 kg m ) ( 65 − 40 )(1.5m ) ( 0.703) = 5.50 ×10
2 3 2 3
Ra = = 9
µ2 (1.96 ×10 N ⋅ s/m ) -5 2 2
From Table 12-3, Nu = 0.1 Ra = 0.1( 5.50 ×10 ) =176

1/ 3 9 1/ 3
Nuk (176 )( 0.0281W/mK )

h= = = 3.31W/m 2 K
H 1.5m
250W
Ts = 40 0 C + = 64.7 0 C Answer
( 3.31W/m K )( 3.070m )
2 2
This is reasonably close to our assumed Ts, so iteration is not required.

b) If we include radiation, then the energy balance is
Q = Qconv + Qrad = hA (Ts − T f ) + εσ A (Ts4 − T f4 )
The surface temperature will be lower, so h should be re-evaluated at a lower Ts. Assume Ts ≈ 55 oC = 328 K
Tfilm ≈ 47.5 oC ≈ 320 K, ρ = 1.103 kg/m3, µ = 1.94 × 10-5 N⋅s/m2, k = 0.0278 W/mK, Pr = 0.703
( 9.81)(1 320 )(1.103) ( 55 − 40 )(1.5 ) ( 0.703)
2 3
Ra = = 3.53×109
(1.94 ×10 N ⋅ s/m )
-5 2 2
(152 )( 0.0278)
Nu = 0.1( 3.53 ×109 )
1/ 3
= 152 ⇒ h= = 2.82W/m 2 K
1.5
Substituting into the energy balance
250W = ( 2.82 W m 2 K )( 3.07m 2 ) (Ts -313) C + ( 0.6 ) 5.67×10-8 W m 2 K
o
( 4
) (3.07m ) ⎡⎣T
2
s
4
- ( 313K ) ⎤
4
⎦
Solving for Ts ~ 324.3 K ~ 51.3 oC Answer
12- 60
12-57 To improve the heat transfer from the transformer described in Problem P 12-56, 16 fins made of plain
carbon steel are attached. Each fin has the same length as the transformer, is 4-mm thick, and extends from
the surface 100-mm. Using only natural convection, determine the surface temperature of the transformer
(in ºC).
Approach:
This is a natural convection problem from fins. Using
the convective heat transfer rate equation applied to
the finned surface and to the ends, the surface
temperature can be determined. An iterative solution
may be required, since the calculation of the natural
convection heat transfer coefficient depends on the
surface temperature.
Assumptions:
Solution:
The rate equation for convective heat transfer from a finned surface and from the ends is:
Q =η0 h Atot (Ts − T f ) + 2 h Aend (Ts − T f )
⎡ 1 ⎤
Solving for the surface temperature Ts = T f + Q ⎢ ⎥
⎣η0 h Atot + 2 h Aend ⎦
We assume that the same heat transfer coefficient is applicable to the ends as to the finned surface. Assuming a
straight, rectangular profile fin with an adiabatic tip, and using a corrected length:
N Af
η0 = 1 −
Atot
(1−η f ) and Atot = Ab + N Af
N = 16 Af = 2 L*H = 2(L + t/2)H = (2) (0.10 + 0.004/2) m (1.5 m) = 0.31 m2
Ab = π DH - Nt H = π (0.55 m) (1.5 m) – 16 (0.004 m) (1.5 m) = 2.50 m2
Atot = 2.50 + 16 (0.31) = 7.40 m2 Aend = π D2/4 = π (0.55 m)2/4 = 0.24 m2
tan h ( mL* )
1/ 2 /2
⎛ hp ⎞ ⎛ h2 H ⎞
The fin efficiency is ηf = where m = ⎜ ⎟ = ⎜ ⎟
mL* ⎝ k Ax ⎠ ⎝ kt H ⎠
We assume that h can be approximated by that on a vertical wall, and this is also valid for the ends. The Rayleigh
number is evaluated with properties from Appendix A-7 at Tfilm = (Ts + Tf)/2. We assume Ts ≈ 54 oC →
Tfilm ≈ 320 K, ρ = 1.103 kg/m3, µ = 1.94 × 10-5 N⋅s/m2, k = 0.0278 W/mK, Pr = 0.703
2 ⎛ 1 ⎞
g β ρ 2 (Ts − T f ) H 3 Pr ( 9.81m s ) ⎜⎝ 320 ⎟⎠ (1.103kg m ) ( 54 − 40 ) K (1.5m ) ( 0.703)
3 2 3
Ra = = = 3.29 ×109
µ2 (1.94 ×10 N ⋅ s/m 2 )
-5 2
Nuk (149 )( 0.0278W/mK )

Nu = 0.1 Ra1/ 3 = 0.1( 3.29 ×109 )
1/ 3
From Table 12-3 =149 → h = = = 2.76W/m3 K
H 1.5m
⎡ ( 2.76 W m 2 K ) 2 ⎤
1/ 2
For plain carbon steel, k = 60.5 W/mK, mL = ⎢

*
⎥ ( 0.10 + 0.004/2 )m = 0.487
⎢⎣ ( 60.5W mK )( 0.004m ) ⎥⎦
tanh ( 0.487 ) 16 ( 0.31m 2 )

ηf = = 0.928 η0 =1 − (1 − 0.928 ) = 0.951
0.487 7.40m 2
250W ⎡ 1 ⎤
Ts = 40 o C + ⎢ ⎥ = 52.0 o C Answer
( 2.76W m K ) ⎢⎣ ( 0.951) ( 7.40m ) + 2 ( 0.24m ) ⎥⎦
2 2 2
This is close enough to our estimate, so we do not need to iterate.

Comments:
Without the fins or radiation, the surface temperature was 64.7 ºC. This problem illustrates well the effectiveness
of fins in increasing heat transfer.
12- 61
12-58 Two 1.5-m in diameter, 2-m tall tanks connected to a common piping header are used to store propane for
use in an isolated cabin in the mountains. Unknown to the owner, the spring in the pressure relief valve on
the system weakens and allows the pressure in the two tanks to drop to atmospheric. The temperature of
the propane falls to -42 ºC when the pressure inside the two tanks reaches one atmosphere. The still
ambient air is at 15 ºC. Heat transfer from the air to the propane causes it to vaporize, and the vapor is
vented from the tank. Properties of propane are vf = 0.001755 m3/kg, vg = 0.4127 m3/kg, hfg = 425 kJ/kg,
and hg = 493 kJ/kg. Ignoring the wall resistance of the tank and radiation, determine how long it will take
for the tank to empty (in days).
Approach:
The rate of the conservation of energy and mass are
applied to the control volume shown. These equations
are integrated with respect to time. The heat transfer
coefficient is required to calculate the heat transfer rate.
Assumptions:
2. Radiation and wall resistance is ignored.
3. The system has no work, and potential and kinetic
energy effects are negligible.
Solution:
For the control volume shown, assume no work, constant pressure, and no potential or kinetic energy effects.
Conservation of mass and energy give: − mout = dm/dt and Q − mout hout = dU dt where hout is the
enthalpy of the leaving vapor and is equal to hg (saturated vapor). Integrating both equations with respect to time:
− ∫ mout dt = ∫ dm ⇒ − ∫ mout dt = m2 − m1 ∫ Qdt − ∫ mout hg dt = ∫ dU
With hg constant and assuming Q is constant Qt + ( m2 − m1 ) hg = m2 u2 − m1 u1 . Solving for time:
t = ⎡⎣ m2 u2 − m1u1 − ( m2 − m1 ) hg ⎤⎦ Q
The initial internal energy is saturated liquid, u1 = uf. The final internal energy is saturated vapor, u2 = ug.
From the given information we can determine the internal energies and masses:
m1 = V/ν1 = V/νf V = 2πD2H/4 = 2π (1.5 m)2 (2 m)/4 = 7.07 m3
m1 = (7.07 m ) / (0.001755 m3/kg) = 4030 kg
3
and m2 = V/νg = (7.07 m3) / (0.4127 m3/kg) = 17.1 kg
The final mass is negligible compared to the initial mass, so one will ignore m2; that is, use m2 ~ 0. The initial
internal energy is
u1 =u f = h f − Pv f =( hg − h fg ) − Pv f = ( 493− 424 ) kJ/kg −(101.3kN m 2 )( 0.001755 m3 kg )(1kJ 1kN ⋅ m ) = 68.8kJ kg
This is a natural convection problem (ignore radiation). We assume the heat transfer coefficient can be calculated
for a vertical wall, the convection from the top of the tanks can use the same heat transfer coefficient, and there is
no heat transfer from the tank bottoms (they sit on the ground). Evaluating air properties from Appendix A-7 at
Tfilm = (T f + TP ) 2 260 K , ρ = 1.359 kg/m3, µ = 1.56 × 10-5 N⋅s/m2, k = 0.0230 W/mK, Pr = 0.719
g βρ 2 (T f − TP ) H 3 Pr ( 9.81m/s ) (1/260K ) (1.359 kg m ) (15 + 42 ) K( 2m ) ( 0.719 ) = 9.39 ×10

2 3 2 3
Ra = = 10
µ2 (1.56 ×10 N ⋅ s/m )

-5 2 2
Nuk ( 454 )( 0.0230W/mK )

From Table 12-3 Nu = 0.1Ra1/3 = 0.1(9.39 × 1010)1/3 = 454 → h = = = 5.23W/m 2 K
H 2m
The surface area is (insulated bottoms) A = 2π DH + 2π D 2 /4 = 2π ⎡⎣(1.5m )( 2m ) + (1.5m ) 2/4 ⎤⎦ = 23.6m 2
The heat transfer rate is Q = hA (T f − TP ) = ( 5.23W m 2 K )( 23.6m 2 ) ⎡⎣15 − ( − 42 ) ⎤⎦ K = 7040W
The time requires for the tanks to empty
−( 4030kg )( 68.8kJ kg ) + ( 4030kg )( 493kJ/kg )
t= = 2.43 ×105 s = 67.5 hrs = 2.81 days Answer
( 7040W )(1kW 1000W )(1kJ/s 1kW )
Comments: Note that radiation heat transfer will have an influence and will accelerate vaporization.
12- 62
12-59 An experiment is performed to determine the heat transfer coefficient on a horizontal circular cylinder.
Radiation effects are minimized by polishing the cylinder’s surface. The 30-cm long, 2.5-cm diameter
cylinder has well insulated ends. Measurements show that 30 W are dissipated when the cylinder surface
temperature is 95 ºC and the surrounding air and surfaces are at 20 ºC. Determine:
a. the natural convection heat transfer coefficient from the data
b. the natural convection heat transfer coefficient if radiation is taken into account and the surface
emissivity is estimated to be 0.07 (in W/m2·K)
c. compare the calculated heat transfer coefficient to one calculated with the appropriate correlation (in
W/m2·K). Comment on the accuracy of the experimental results.
Approach:
The convective heat transfer rate equation ( Q = hA∆T)
is used to determine the convective heat transfer
coefficient. If radiation is included, the heat transfer
rate due to convection is reduced, so h increases.
Assumptions:
Solution:
a) The rate equation is: Q = hA (Ts – Tf). Solving for h and noting that the ends are well insulated
Q 30W
h= = =17.0W/m 2 K Answer
A(Ts − T f ) π ( 0.025m )( 0.30m )( 95 − 20 ) K
b) If radiation is included Qtot = Qconv + Qrad = hA (Ts - T f ) + εσ A (Ts4 − T f4 )
Qtot − εσ A(Ts4 − T f4 )
Solving for h h=
A(Ts − T f )
30W − ( 0.07 ) ( 5.67 ×10 −8 W m 2 K 4 )π ( 0.025m )( 0.30m ) ⎡( 95 + 273) − ( 20 + 273) ⎤ K 4
4 4
= ⎣ ⎦
π ( 0.025m )( 0.30m )( 95 − 20 ) K
=16.4 W/m 2 K Answer
c) To calculate h from a correlation, use properties from Appendix A-7 Tfilm = (95 + 20)/2 = 57.5oC = 330.5 K, ρ
= 1.068 kg/m3, k = 0.0285 W/mK, µ = 1.98 × 10-5 N⋅s/m2, Pr = 0.701
g β ρ 2 (Ts − T f ) D 3 Pr
Ra = 2
µ
=
( 9.81m/s ) (1/330.5K ) (1.068 kg/m ) ( 95 − 20 )K ( 0.025m ) ( 0.701) = 70,900
2 3 2 3
(1.98×10 N ⋅ s/m )−5 2 2
From Table 12-3

⎧ ⎫ ⎧ ⎫
2
( )
1/ 6
⎪⎪ 0.387 Ra 1/ 6 ⎪
⎪ ⎪ 0.387 70900 ⎪
Nu = ⎨0.6 + 8 / 27 ⎬
= ⎨0.6 + ⎬ = 7.10
9 /16 8 / 27
⎪ ⎡1 + ( 0.559/Pr ) ⎤ ⎪ ⎪ ⎡1 + ( 0.559/0.701) ⎤ ⎪
9 /16
⎪⎩ ⎣⎢ ⎦⎥ ⎪⎭ ⎩ ⎣ ⎦ ⎭
( 7.10 ) ( 0.0285 W/m 2 K )
h= = 8.1W/m 2 K Answer
0.025m
Comments:
This is poor accuracy. Radiation plays a small role, reducing the experimental heat transfer coefficient by about
3.5%. However, that is minor compared to the factor of 2 differences between the experimental h and the h
calculated with the correlation. The experimental is flawed and the measurement sensors must be checked as well
as the experimental procedures. For example, the smallest air draft can raise the experimental h significantly, so
care must be exercised to enclose the test section with panels to minimize drafts.
12- 63
12-60 Arrays of vertical fins are often attached to equipment to aid passive (i.e., natural convection) cooling of
the device. Consider the assembly shown below that is located in air at 20 ºC. Each fin has a length of 25
mm, a thickness of 1.5 mm, and a height of 100 mm. Assume the fin has a fin efficiency of 100%, the base
temperature is 75 ºC, and each fin operates as if it is independent of all other surfaces nearby. For a fin
spacing of 8 mm, determine the heat transfer rate from an array of fins that covers 150 mm of wall (in W).
Approach:
This is a natural convection between fins.
We assume that it is similar to natural
convection in vertical parallel plate
channels, and the heat transfer coefficient
calculated from the appropriate correlation
is applicable to all the surfaces. The
number of fins must be determined, and
then the basic rate equation is used to
calculate the heat transfer rate.
Assumptions:
Solution:
The rate equation for a finned surface is Q =η0 h Atot (Tb − T f )
Because ηf = 1, the overall surface efficiency also is unity, that is, η0 = 1.
The total area is Atot = N 2 LH + WL
W 0.150m
For the number of fins, N: W = N (S +t) → N= = = 15.8
S + t 0.008m + 0.0015m
Because there must be an integer number of fins, N = 15
Atot = (15) (2) (0.10 m) (0.025 m) + (0.15 m) (0.1 m) = 0.090 m2
For the average heat transfer coefficient, we assume isothermal surfaces and use the Bar-Cohen and Rohsenow
correlation.
−1/ 2
⎧ ⎫
⎪ ⎪
⎡ q ′′ ⎤ S ⎪ 576 2.87 ⎪
Nus = ⎢ ⎥ = ⎨ + 1/ 2 ⎬
T − T
⎣⎢ b f ⎦⎥ k ⎪⎡ ⎛ S ⎞⎤
2
⎡ ⎛ S ⎞⎤ ⎪
⎪ ⎢ Ras ⎜⎝ L ⎟⎠ ⎥ ⎢ Ras ⎜ L ⎟ ⎥ ⎪
⎝ ⎠⎦ ⎭
⎩⎣ ⎦ ⎣
g β ρ 2 (T f − T f ) S 3 Pr
Ras =
µ2
From Appendix A-7 at Tfilm = (74 + 20)/2 = 46.5 oC = 320.5 K, ρ = 1.101 kg/m3, µ = 1.94×10-5 N⋅s/m2, k =
0.0278 W/mK, Pr = 0.704
( 9.81m s ) (1/320.5K ) (1.101kg m ) ( 75 − 20 )K ( 0.008m ) ( 0.704 ) =1950
2 3 2 3
Ras =
(1.94 ×10 N ⋅ s/m )-5 2 2
−1/ 2
⎧ ⎫
⎪ 576 2.87 ⎪
Nus = ⎨ + 1/ 2 ⎬
=1.99
⎪⎩ ⎡⎣1950 ( 0.008 0.10 ) ⎤⎦ ⎡⎣1950 ( 0.008 0.10 ) ⎤⎦ ⎪⎭
2
Nuk (Tb − T f ) = (1.99 )( 0.0278W/mK )( 75 − 20 )K = 380 W/m

q ′′ = 2
S 0.008m
Q = q ′′ Atot = ( 380 W m 2 )( 0.090m 2 ) = 34.2W Answer
12- 64
12-61 A window 30-cm tall and 45-cm wide is centered in an oven door that is 50-cm tall and 75-cm wide.
During operation when the room temperature is 24 ºC, the window reaches a temperature of 45 ºC, and the
door surface reaches 33 ºC. Assume that both the door and window have an emissivity of 1.0 and the
surroundings also are at 24 ºC. Estimate:
a. the heat transfer from the door and window (in W)
b. the heat transfer if the door did not have a window (in W).
Approach:
We will need to use approximations for part A,
because of the natural convection over two surfaces at
different temperatures. Heat loss is due to radiation
and convection. We will assume that the window is
flush with the door, and we have the same natural
convection heat transfer coefficient over both door
and window.
Assumptions:
1. The heat transfer coefficient for he window and
door is the same.
Solution:
a) The total heat loss flow the door and window combination is:
Qtot = QDOOR + QWINDOW
QDOOR = εσ AD (T24 − T f4 ) + h AD (T2 − T f ) where AD = L2 H 2 − L1 N1
QWINDOW = εσ Aw (T − T 1
4
f
4
) + h A (T − T )
w 1 f where Aw = L1 H1
The natural convection heat transfer coefficient requires the Rayleigh number with properties from
Appendix A-7 evaluated at Tfilm = (Ts + Tf)/2. Because the door and window are at different temperatures, we
assume Ts ~ (Ts + T2)/2 = (45 + 33)/2 = 39 oC = 312 K, so that Tfilm = (312 + 297)/2 = 304.5 K. The properties are:
ρ= 1.159 kg/m3, µ = 1.887 × 10-5 N⋅s/m2, k = 0.0266 W/mK, Pr = 0.707
g βρ 2 (Ts − T f ) H 23 Pr ( 9.81m s ) (1 304.5K ) (1.159 kg m ) ( 312 − 297 )K ( 0.5m ) ( 0.707 ) =1.61×10

2 3 2 3
Ra = = 8
µ2 (1.887 ×10 N ⋅ s/m )

-5 2 2
From Table 12-3, the vertical plate correlation is:

Nuk ( 66.5 )( 0.0266W/mK )
Nu = 0.59 Ra / 4 = 0.59 (1.61×108 )
1/ 4
= 66.5 → h= = = 3.54W/m 2 K
H2 0.5m
Aw = (0.45 m) (0.30 m) = 0.135 m2
AD = (0.50 m) (0.75 m) = 0.135 m2 = 0.240 m2
⎛ W ⎞ ⎛ W⎞
QDOOR = (1) ⎜ 5.67 ×10-8 2 4 ⎟ ( 0.24m 2 ) ⎡( 306K ) − ( 297K ) ⎤ + ⎜ 3.54 2 ⎟ ( 0.24m 2 ) ( 306 − 297 ) K
4 4
⎝ m K ⎠ ⎣ ⎦ ⎝ m ⎠
= 13.4 + 7.6 = 21.0 W
QWINDOW = (1) ( 5.67 x10-8 ) ( 0.135 ) ⎡( 318 ) − ( 297 ) ⎤ + ( 3.54 )( 0.135 )( 318 − 297 ) = 18.7 + 10.0 = 28.7W
4 4
⎣ ⎦
Qtot = 21.0 + 28.7 = 49.7 W Answer
b) For a door with no window, assuming the heat transfer coefficient is the same as in part (a)
Q = (1) ( 5.67 ×10-8 ) ( 0.5 )( 0.75 ) ⎡( 306 ) − ( 297 ) ⎤ + ( 3.54 )( 0.5 )( 0.75 )( 306 − 297 )
4 4
⎣ ⎦
= 21.0 + 11.9 = 32.9W Answer
12- 65
12-62 The heat loss situation described in Problem P 12-15 changes when the wind stops and the air is calm. For
this new condition, determine: a) the heat transfer rate per unit length of pipe (in W/m) with no insulation,
and b) the heat transfer rate per unit length of pipe if 4-cm insulation (k = 0.04 W/m·K) is applied to the
pipe (in W/m).
Approach:
This is a natural convection problem on a circular
cylinder. The basic heat transfer rate equation ( Q =
∆T/Rtot) can be used to find the heat transfer per unit
length. The heat transfer coefficient must be
evaluated.
Assumptions:
Solution:
The basic heat transfer rate equation is
∆T Ts − T f Ts − T f Q Ts − T f
Q= = = or =
Rtot Rconv + Rcond 1 hπ DL + 1n ( D2 / D1 ) 2π kL L 1 hπ D + 1n ( D2 / D1 ) 2π k
a) With no insulation D2 = D1, D = D1, and Rcond = 0. Evaluation of the natural convection h requires the
Rayleigh number with fluid properties evaluated of Tfilm = (-6 + 110)/2 = 52oC = 325K. From Appendix A-7,
ρ = 1.086 kg/m3, µ = 1.96 × 10-5 N⋅s/m2, k = 0.0281W/mK, Pr = 0.703
Ra =
g βρ 2 (Ts − T f ) D 3 Pr
=
( 9.81m/s ) (1 325 K ) (1.086 kg/m ) ⎡⎣110 −( −6 )⎤⎦ ( 0.15 m ) ( 0.703)
2 3 2 3
= 2.55 × 107
µ2 (1.96 ×10 N ⋅ s/m )
−5 2 2
From Table 12-3 for a horizontal cylinder

2 2
⎧ ⎫ ⎧ ⎫
0.387( 2.55 ×107 )
1/ 6
⎪ 0.387 Ra1/ 6 ⎪ ⎪ ⎪
Nu = ⎨0.6 + ⎬ = ⎨0.6 + ⎬ = 37.3
9 /16 8 / 27 9 /16 8 / 27
⎪ ⎡1 + ( 0.559/ Pr ) ⎤ ⎪ ⎪ ⎡1 + ( 0.559 0.703) ⎤ ⎪
⎩ ⎣ ⎦ ⎭ ⎩ ⎣ ⎦ ⎭
Nuk ( 37.3)( 0.0281W/mK )
h= = = 6.99 W/m 2 K
D 0.15 m
Q ⎡⎣110 − ( −6 ) ⎤⎦ K W
= = 382 (without insulation) Answer
L 1/ ⎡( 6.99 W/m K ) π ( 0.15 m ) ⎤
2 m
⎣ ⎦
b) Once we add the insulation, the outside insulation temperature will be low. Assuming Ts ~ 0oC ⇒ Tfilm =
270K,
ρ = 1.307 kg/m3, µ = 1.63 × 10-5 N⋅s/m2, k = 0.0239W/mK, Pr = 0.716
( 9.81)(1/ 270 )(1.307 ) ⎡⎣110 −( −6 ) ⎤⎦ ( 0.23) ( 0.716 )
2 3
Ra = = 2.36 × 108
(1.63×10 )−5 2
( 73.8 )( 0.0239 )
Nu = 73.8 ⇒ h= = 7.67 W/m 2 K
0.23
Q ⎡⎣110 − ( −6 ) ⎤⎦
= = 61.7 W/m
L 1 ⎡⎣( 7.67 ) π ( 0.23) ⎤⎦ + 1n ( 23/15 ) ⎡⎣ 2 π ( 0.04 ) ⎤⎦
The outside wall temperature can be calculated with Q = hπDL (Ts – Tf) ⇒ Ts = Tf + Ts = T f + Q / L ( ) ( hπ D )
( 61.7W/m )
Ts = − 60 C + = 5.1o C Answer
( 7.67W/m 2
K ) π ( 0.23 m )
This is close to what we assumed, so no iteration is required.
12- 66
12-63 A power amplifier is mounted vertically in air that is at 27 ºC. The case is made of anodized aluminum
with a surface area of 3800 mm2 and a height of 40 mm. If the amplifier operates at 127 ºC, estimate the
total power dissipation (natural convection and radiation) from the unit (in W). Assume a surface
emissivity of 0.76.
Approach:
This is a combined natural convection plus radiation
heat transfer problem. With the surface temperature
known, this is a straight forward calculation. The
appropriate heat transfer coefficient correlation must
be chosen, and we assume the surface is small relative
to its surroundings.
Assumptions:
2. The power amp is “small” relative to the large
surrounding area.
Solution:
The total heat transfer from the surface by convection and radiation is:
Q = Qconv + Qrad = hA (Tw − T∞ ) + εσ A (Tw4 − T∞4 )
To evaluate h, we need the Rayleigh number with properties from Appendix A-7 evaluated at Tfilm = (Tw + T∞)/2 =
(400 + 300)/2 = 350 K, ρ = 0.998 kg/m3, ν = 20.76 × 10-6 m2/s, k = 0.03003 W/mK, Pr = 0.697
Ra =
g β (Tw − T∞ ) H 3 Pr
=
( 9.81m/s ) (1 350K )(127 − 27 )K ( 0.04m ) ( 0.70 ) = 2.90 ×10
2 3
5
ν2 ( 20.76 ×10 m /s )
-6 2 2
From Table 12-3, Nu = 0.59 Ra1/4 = 0.59 (2.90 × 105)1/4 = 13.7

Nuk (13.7 )( 0.03003W/mK )
h= = = 10.3W/m 2 K
H 0.04m
⎛ W ⎞ ⎛ W ⎞
Q = ⎜ 10.3 2 ⎟ ( 0.0038m 2 ) ( 400 − 300 ) K+ ( 0.76 ) ⎜ 5.67 × 10-8 2 4 ⎟ ( 0.0038m3 ) ⎡( 400K ) − ( 300K ) ⎤
4 4
⎝ m K ⎠ ⎝ m K ⎠ ⎣ ⎦
Q = 3.91W + 2.87W = 6.78W Answer
12- 67
12-64 In car paint shops and other drying applications, radiant heaters are often used because the radiant thermal
energy heats the surface directly with minimal heating of the surrounding air. Consider a vertical flat panel
1-m tall and 4-m long with an emissivity of 0.85 mounted on the wall of a large room. The panel is
maintained at a uniform temperature of 330 ºC and the walls and air in the room are maintained at 25 ºC.
Determine the heat transfer rate from the panel to the room (in W).
Approach:
Power input to the heater must equal radiation plus
convection from panel. We assume the panel is
located in a very large room for radiation purposes,
and we use Q = hA ∆T for the convection. The
natural convection heat transfer coefficient must be
evaluated.
Assumptions:
2. The heater is small relative to the large room.
Solution:
An energy balance on the heater gives Qin = Qrad + Qconv = εσ A (Ts4 − T f4 ) + hA (Ts − T f )
where A = HL.
The natural convection heat transfer coefficient requires the Rayleigh number with properties from
Appendix A-7 evaluated at Tfilm = (330 + 25)/2 = 177.5 oC = 450.5 K, ρ = 0.784 kg/m3, µ = 2.484 × 10-5 N⋅s/m2,
k = 0.03707 W/mK, Pr = 0.683
2
⎛ m ⎞⎛ 1 ⎞⎛ kg ⎞
⎟ ⎜ 0.784 3 ⎟ ( 603 − 298 ) K (1m ) ( 0.683)
3
9.81 2 ⎟ ⎜
g βρ 2 (Ts − T f ) H 3 Pr ⎜⎝ s ⎠ ⎝ 450.5K ⎠ ⎝ m ⎠
Ra = = = 4.52 ×109
µ2 ( 2.484 ×10 N ⋅ s/m )
-5 2 2
From Table 12-3 for a vertical wall

Nu = 0.1 Ra1/3 = 0.1 (4.52 × 109)1/3 = 165
Nuk (165 )( 0.03707W/mK )
h= = = 6.13W/m 2 K
H 1m
⎛ W ⎞
Qrad = ( 0.85 ) ⎜ 5.67 ×10-8 2 4 ⎟ (1m )( 4m ) ⎡( 603K ) − ( 298K ) ⎤ = 23,970W
4 4
⎝ m K ⎠ ⎣ ⎦
⎛ W ⎞
Qconv = ⎜ 6.13 2 ⎟ (1m )( 4m )( 603 − 298 ) K = 7,480W
⎝ m K⎠
Qin = 23,970W + 7,480W = 31,450W Answer
12- 68
12-65 A home hobbyist builds a kiln to fire her ceramic pots. Plans obtained from the Internet state that because
of the thick fireclay bricks used to construct the kiln, insulation on the outside surface are not required.
When she uses the 1-m by 1-m by 1-m kiln for the first time in a room at 30 ºC, the kiln’s outside wall
temperature is 90 ºC. Assuming that there is heat loss from the four sides and the top only, and that these
surfaces have an emissivity of 0.8, determine:
a. the total heat loss from the kiln (in W)
b. the total heat loss from the kiln if 4-cm thick insulation with k = 0.04 W/m·K and ε = 0.1 (assume
the outside brick temperature remains at 90 ºC
c. the simple payback time for the insulation if the insulation costs $400, the cost of natural gas is
$0.50/105 kJ, the furnace has an efficiency of 84%, and the furnace operates 2000 hr/yr.
Approach:
This is a combined convection and radiation problem
in part (a), and conduction, convection, and radiation
in part (b). The natural convection heat transfer
coefficient must be evaluated for the two different
situations.
Assumptions:
2. The kiln is “small” relative to the large room size.
Solution:
a) The heat transfer rate is calculated with the convective rate equation for the sides and top and the radiation rate
equation assuming the kiln is in a very large room:
Q = ( hs As + hT AT )(Ts − T f ) + εσ Atot (Ts4 − T f4 )
The heat transfer coefficient on the sides is assumed to be that on a vertical wall; for the top, we use a correlation
for a heated plate facing upward. From Appendix A-7 at Tfilm ≈ (90 + 30)/2 = 60oC = 333 K, ρ = 1.060 kg/m3,
µ = 2.00 × 10-5 N⋅s/m2, k = 0.0288 W/mK, Pr = 0.702.
For vertical wall (sides)
Ra =
( 9.81m s ) (1/333K ) (1.060kg/m ) ( 90 − 30 )K (1m ) ( 0.702 ) = 3.49 ×10
2 3 2 3
9
( 2.00 ×10 N ⋅ s/m )

−5 2 2
From Table 12-3 Nu = 0.1Ra1/3 = 0.1 (3.49 × 109)1/3 = 152

Nuk (152 ) ( 0.0288W/m K )
2
hs = = 4.38W/m 2 K
H 1m
As = 4HW = 4m2
For the top, Lchar = A/P = HD/2(H + D) = 0.25m
RaT = 5.45 × 107
Nu = 0.15Ra1/3 = 0.15 (5.45 × 107)1/3 = 56.9
hT = (56.9) (0.0288)/0.25 = 6.55 W/m2K
AT = 1m2
⎡⎛ W ⎞ ⎛ W ⎞ ⎤ ⎛ W ⎞
Q = ⎢⎜ 4.38 2 ⎟ ( 4m 2 ) + ⎜ 6.55 2 ⎟ (1m 2 ) ⎥ [90 − 30]K + ( 0.8 ) ⎜ 5.67x10 −8 2 4 ⎟( 5m 2 ) ⎡( 90 + 273K ) − ( 30 + 273K ) ⎤
4 4
⎣⎝ mK⎠ ⎝ m K⎠ ⎦ ⎝ mK ⎠ ⎣ ⎦
= [1051 + 393] + 2026 W = 3470 W Answer

b) Draw a circuit diagram to evaluate the heat loss when insulation is added.
From the diagram: Qtot = Qconv + Qrad
Ts − Tins Tins − T f
= + εσ Atot (Tins4 − T f4 )
t 1
kins Atot hs As + hT AT
We want the outside temperature of the insulation but the heat transfer coefficient depends on that temperature, so
assuming Tins ≈ 44oC, Tfilm ≈ 310 K, ρ = 1.139 kg/m3, µ = 1.89 × 10-5 N⋅s/m2, k = 0.0270 W/mK, Pr = 0.706
12- 69
( 9.81)(1/310 )(1.139 ) ( 44 − 30 )(1) ( 0.706 )
2 3
Ras = = 1.14 × 109

(1.89 ×10−5 )
Nus = 0.1 (1.14 × 109)1/3 = 104 → hs = 2.82 W/m2K
RaT = 1.78 x 107
NuT = 0.15 (1.78 × 107)1/3 = 39.2 → hT = 4.23 W/m2K
Now substituting these into the main energy balance
( 363 − Tins )K (Tins − 303 ) K
= + ( 0.1) ( 5.67 × 10 −8 ) ( 5 ) ⎡⎣Tins4 − 3034 ⎤⎦
0.04m 1
( 0.04W/mK ) ( 5m 2 ) ( 2.82 )( 4 ) + ( 4.23)(1)
Solving for Tins = 315.3 K = 42.3oC and Q tot = 239 W Answer
Because the calculated Tins is close to the assumed value, no iteration is required.
c) The energy savings with the insulation is 3470 – 239 = 3231 W
For 2000 hrs/yr in a 84% efficient furnace, this is results in an energy savings of
( 2000hr/yr )( 3231W )(1kW 1000W )(1kJ/s 1kW )( 3600s hr )
= 2.77 ×107 kJ/yr
0.84
Gas costs $0.50/105 kJ, so total fuel expense savings is ( 2.77 × 107 kJ yr )( $0.50 105 kJ ) = $138/yr
If the insulation costs $400, the simple payback is
$400
= 2.9 yrs. Answer
$138/yr
12- 70
12-66 In oil refineries and chemical process plants, insulation on pipes is wrapped in thin aluminum metal sheaths
to protect the insulation from the weather. After weathering and exposure to harsh air borne chemicals
around the plants, the surface of the metal sheath corrodes and the emissivity is about 0.4. Consider a 30-
cm I.D., 40-cm O.D. carbon steel pipe (k = 60 W/m·K) carrying saturated steam at 350 ºC covered with
7.5-cm of fiberglass insulation (k = 0.036 W/m·K). The steam convective heat transfer coefficient is 600
W/m2·K. With an air temperature of 0 ºC, determine the heat transfer rate per unit meter of pipe length
when:
a. a crossflow at 30 km/hr is on the outside of the pipe (in W/m)
b. natural convection is on the outside of the pipe (in W/m) (use an approximation based on the results
of part (a) and justify).
Approach:
Assuming one-dimensional heat transfer, the base heat
transfer rate equation ( Q = ∆T/Rtot) can be used to
calculate the heat transfer rate. The heat transfer
coefficient on the outside must be evaluated, and the
outside surface temperature is needed to determine the
radiation heat transfer.
Assumptions:
Solution:
For steady one-dimensional heat transfer, the total heat transfer from steam to outer surface is:
∆T Tst − T3
Q= =
Rtot 1 1n ( D2 /D1 ) 1n ( D3 /D2 )
+ +
hst π D1 L 2π k s L 2π kins L
where the outside surface temperatures, T3, is unknown. This heat transfer rate must equal that from the surface to
the air Q = Qconv + Qrad = h π D3 L (T3 − T f ) + εσ π D3 L (T34 − T f4 )
The length, L, can be factored out of both equations. We have assumed that the surroundings are very large and
are at the same temperature as the air. We have two equations and two unknowns ( Q/L and T3) that can be solved
simultaneously.
We can recast the radiation contribution in terms of a radiative heat transfer coefficient
T3 − T f
Q= where hr = εσ (T3 + T f ) (T32 + T f2 )
1 1
+
hπ D3 L hr π D3 L
For the forced convection heat transfer coefficient, from Appendix A-7 and assuming T3 → 7 oC, Tfilm = (0 + 7)/2
= 3.5 oC ≈ 275 K, ρ = 1.284 kg/m3, µ = 1.67 × 10-5 N⋅s/m2, k = 0.0243 W/mK, Pr = 0.715
ρ VD (1.284kg m ) ( 30000m hr )(1hr 3600s )( 0.55m )
3
Re = = = 3.48 ×105
µ 1.67 ×10-5 N ⋅ s/m 2
From Table 12-1 Nu = 0.027 Re0.806 Pr1 3 = 0.027 ( 3.48 × 105 ) ( 0.713)
0.806 13
= 706
Nuk ( 706 )( 0.0243W/mK )
Nu = = = 31.2W/m 2 K
D 0.55m
The resistances are
1 1 0.00177
= =
hst π D1 L ( 600W/m K )π ( 0.30m ) L
2
L
1n ( D2 /D1 ) 1n ( 0.40/0.30 ) 0.000763
= =
2π ks L 2π ( 60W/mK ) L L
12- 71
1n ( D3 /D2 ) 1n ( 0.55/0.40 ) 1.408
= =
2π kins L 2π ( 0.036 W/mK ) L L
1 1 0.0185
= =
hπ D3 L ( 31.2W m 2 K )π ( 0.55m ) L L
With a first approximation of T3 = 7 oC = 280 K
hr = ( 0.4 ) ( 5.67 ×10-8 W m 2 K 4 ) ( 280K + 273K ) ⎡( 280K ) + ( 273K ) ⎤ =1.9W/m 2 K
2 2
⎣ ⎦
1 1 0.302
= =
hr π D3 L (1.9 W m 2 K )π ( 0.55m ) L L
Solving the two rate equations, we obtain T3 = 337.9 K. This does not match our assumed value.
Iterating on T3, we obtain
T3 = 326.4 K
hr = 2.46 W/m2K
1 0.2351
=
hr π D3 L L
Q/L = 210 W/m Answer
b) For the natural convection case, if we assume hnat 0.1h forced as a first approximation, then solving for the
conv conv
heat loss as above:

T3 = 350 K = 77 oC
hr = 2.79 W/m2K
1 0.2079
=
hr π D3 L L
Q/L = 194 W/m Answer
Comments:
Because the heat loss per unit length does not change much with a factor of ten change in the air side heat transfer
coefficient, h, a more accurate estimate of h is not needed.
12- 72
12-67 The power cables for an electric welding rig are suspended above the floor of a factory so that a tripping
hazard is not created. The cables are 25-m long; the copper is 10-mm in diameter and is covered by a 2-
mm thick rubberized cover (k = 0.26 W/m·K), which is black, has an emissivity of 0.9 and cannot have a
temperature greater than 65 ºC. The cable resistance is 4 × 10-4 ohms. If the cable is suspended in calm air
at 27 ºC, determine the maximum allowable current (in A).
Approach:
Because the cable is long, we assume one-dimensional
heat transfer. This is a combined natural convection and
radiation problem. We will need to estimate the surface
temperature to obtain the natural convection heat transfer
coefficient and the radiation. An iterative solution may be
required. Once the heat transfer rate is determined, the
maximum allowable current can be calculated.
Assumptions:
Solution:
The total heat transfer rate is calculated with Qtot = Qcond = Qconv + Qrad
Assuming the surroundings are large and at the same temperature as the air, and all properties are constant
2π kL(Tmax − Ts )
= hA (Ts - T f ) +εσ A (Ts4 − T f4 )
1n ( D2 /D1 )
The convective heat transfer coefficient must be determined with fluid properties from Appendix A-7 evaluated at
the film temperature, Tfilm = (Ts + Tf)/2. We assume Ts ≈ 45oC = 318K, Tfilm ≈ (27 + 45)/2 = 360C ≈ 310 K, so
ρ = 1.143 kg/m3, k = 0.0268 W/mK, µ = 1.90 × 10-5 N⋅s/m2, Pr = 0.711. When calculating the Rayleigh number,
we will keep these properties constant and only change Ts during the iteration. This is for simplicity only:
g βρ 2 (Ts − T f ) D23 Pr
Ra = 2
µ
⎛ ⎞
( 9.81m s ) ⎜ T +2300 ⎟ (1.143kg m3 ) (Ts −300 )( 0.014m ) ( 0.711)
2 2 2
⎝ ⎠ ⎛ T − 300 ⎞
= s
= (1.385 ×10−5 ) ⎜ s ⎟
(1.90 ×10 −5
N ⋅ s/m )
2 2
⎝ Ts + 300 ⎠
For Ts = 318 K, Ra = 4035, so from Table 12-3
2 2
⎧ ⎫ ⎧ ⎫
0.387 ( 4035 )
1/ 6
⎪ 0.387 Ra1/ 6 ⎪ ⎪ ⎪
Nu = ⎨0.6 + 8 / 27 ⎬
= ⎨ 0.6 + 8 / 27 ⎬ = 3.54
⎪ ⎡1 + ( 0.559/Pr )9 /16 ⎤ ⎪ ⎪ ⎡1 + ( 0.559/0.711)9 /16 ⎤ ⎪
⎩ ⎣ ⎦ ⎭ ⎩ ⎣ ⎦ ⎭
NuK ( 3.54 )( 0.0268W/mK )
h= = = 6.78W/m 2 K
D 0.014m
Therefore, the right hand side of the heat transfer rate equation is:
Q = (6.78 W/m2K) π (0.014 m) (25 m) (318 – 300) K
+ ( 0.9 ) ( 5.67×10-8 W m 2 K 4 ) π ( 0.014m )( 25m ) ⎡( 318K ) − ( 300K ) ⎤ = 134.2 + 119.3 = 253.5W
4 4
⎣ ⎦
2π ( 0.26 W mK )( 25m )
The left hand side is Q= ( 338-318) K=2428 W
ln ( 0.014 0.010 )
These two heat transfer rates do not match, so an iteration is required. Doing so, we obtain:
Ts = 333.6 K =60.6o C Ra = 7350 Nu = 4.06 h = 7.78 W m 2 K Q = 529W
Electric power is Q = I 2 R , where I is current. Solving for current:
( ) = ( 529W 4 × 10−4 Ω )
0.5 0.5
I= Q R = 1150 A Answer
12- 73
12-68 A manufacturer of prefabricated buildings is considering using the same structure (shown below) for walls
and roofs for quickly and cheaply assembled buildings. The inner and outer surfaces are 1.27-cm thick
plywood (k = 0.115 W/m·K), and the air-filled gap is 10-cm wide; the panels are 2.5-m long and 1.25-m
deep out of the plane of the page. For temperatures on the outside of the two plywood sheets of -10 ºC and
15 ºC, determine:
a. the heat transfer rate for both horizontal and vertical orientations (in W)
b. the effect of inserting a baffle at midheight for when the panel would be used in a vertical
orientation (in W).
Approach:
Assuming steady, one-dimensional heat transfer, the
basic rate equation, Q = ∆T/Rtot, can be used to
calculate the heat transfer rate. The natural
convection heat transfer coefficient inside the wall
must be estimated.
Assumptions:
4. Heat transfer from the ends is ignored.
Solution:
Assuming adiabatic ends, and that radiation can be ignored, the rate equation is:
∆T Ts − T1
Q= =
Rtot RWALL ,1 + RWALL , 2 + Rconv
The wall resistances are
t 0.0127m
RWALL ,1 = RWALL , 2 = = = 0.0353K/W
kA ( 0.115W/mK )(1.25m )( 2.50m )
a) The convective resistance in a vertical enclosure requires the natural convective heat transfer coefficient, with
properties evaluated at the average of the two wall temperatures. We first estimate the heat transfer rate by
assuming the convective resistance is equal to twice that contributed by the plywood, and then use this estimate to
calculate the inside wall temperatures
⎡15 − ( −10 ) ⎦⎤ K
Qest = ⎣ =177W
4( 0.035k/W )
The temperature drop across both sheets of plywood is the same, so
Ts − T2, in
Qest = → T2 − T2, in = Qest ( t kA ) = 177 W ( 0.0353K W ) = 6.3K
t kA
So Ts,in ≈ = 10 + 6.3 = -3.7 oC and T2,in ≈15 – 6.3 = 8.7 oC Therefore, from Appendix A-7 at
Tavg = ( −3.7 + 8.7 ) 2 = 2.5 C = 275K, ρ = 1.284 kg/m3, µ = 1.67 × 10-5 N⋅s/m2, k = 0.0243 W/mK, Pr = 0.715
o
( 9.81m/s ) (1 275K ) (1.284 kg m ) (8.7 + 3.7 )( 0.10m ) ( 0.715) =1.87 ×10

g βρ 2 (T2,in − T1,in )S3 Pr 2 3 2 3
Ras = = 6
µ 2
(1.67 ×10 N ⋅ s/m )
-5 2
Nu = 0.073 Ra ( H S ) = 0.073 (1.87 × 10 ) ( 2.50 0.10 ) = 6.06

6 1/ 3
1/ 3 −1/ 9 −1/ 9
From Table 12-6 s
Nuk ( 6.06 )( 0.0243W/mK )

h= = =1.47W/m3 K
S 0.10m
1 1
= = 0.217K/W
hA (1.47W m K ) (1.25m )( 2.50m )
2
Therefore, for vertical configuration:
12- 74
⎡⎣15 − ( −10 ) ⎤⎦ K
Q= = 86.8W
2 ( 0.0353K/W ) + 0.217K/W
Calculating a new T1,in = - 6.9 oC and T2,in = 11.9 oC
Ras = 2.84 ×106 , Nu = 7.23 , h =1.76 W/m 2 K Q = 98.9 W Answer
We could iterate again for an improved estimate.
For the horizontal configuration, from Table 12-6 and using the same Ra as for the vertical configuration
Nu = 0.061 Ra1/ 3 = 0.061( 2.84 ×106 )
1/ 3
= 8.64
Nuk ( 8.64 )( 0.0243)
h= = = 2.10W/m 2 K
S 0.10
1 1
= = 0.152K/W
hA ( 2.10W/m 2 K ) (1.25m )( 2.50m )
⎡⎣15 − ( −10 ) ⎤⎦
Q= =112W Answer
2 ( 0.0353) + 0.152
b) If a baffle is inserted in the vertical wall, the only change in the calculation of the heat transfer coefficient is in
the term (H/S)-1/9
( H/S )baffle ⎛ 1.25 ⎞−1/ 9
−1/ 9
hbaffle
= =⎜ ⎟ = 1.08
( H/S )
−1/ 9
h ⎝ 2.50 ⎠
hbaffle = 1.08 (1.76 W/m2K) = 1.90 W/m2 K
Q = 105 W Answer
12- 75
12-69 A thermal pane window often is constructed of two panes of plate glass separated by a short distance; this
arrangement provides increased thermal resistance compared to a single pane of glass. Consider a 4-ft wide
and 6-ft high window whose inside wall temperature is 75 ºF and outside wall temperature is 45 ºF. It is
constructed of glass that is 0.25-in thick. Determine the heat transfer rate:
a. for a single pane of glass (in Btu/hr)
b. for two panes separated by a 1-in air gap (in Btu/hr)
c. for two panes of glass with if a thin (0.1-in.) sheet of glass is inserted between the other two panes
that are still separated by 1 in. (in Btu/hr).
Approach:
Assuming one-dimensional heat transfer, the heat
transfer rate is determined by the rate equation,
Q =∆T/Rtot. This thermal resistance can be found for
combinations of t/kA and 1/hA. Radiation is ignored.
Assumptions:
1. Air is at one atmosphere between the two panes of
glass.
Solution:
a) The thermal resistance of a single pane of glass is: Rglass = t kA
o
From Appendix B-3, for plate glass k = 0.8 Btu/hr ft F, and
( 0.25/12ft )
Rglass = = 0.001085 hr o F Btu
( 0.8 Btu hrft o F ) ( 6ft )( 4ft )
The heat transfer rate is Q = ∆T/Rglass = (75 – 45)oF/(0.001085 hr oF/Btu) = 27,650 Btu/hr Answer
b) Resistance of two panes of glass and an air gap is: Rtotal = 2 Rglass + 1 hA
For air properties from Appendix B-7 at Tavg = (75+45)/2= 60oF, ρ = 0.077 lbm/ft3, k = 0.0146 Btu/hrftoF,
µ = 1.214×10-5 lbm/ft⋅s,, Pr = 0.72
g βρ 2 (T1,in − T2,in ) S 3 Pr
Ra =
µ2
The temperatures to use are the inside wall temperatures. We could estimate the temperatures from the glass
resistance. Let’s assume for the air gap, that resistance is approximation equal to that contribution by both panes
of glass. Therefore:
( 75-45 ) F
o
T1 − T2 Btu
Qestimate = = = 6912
4 Rglass 4 ( 0.001085hr F/Btu )
o
hr
Now calculate the ∆T across the glass:
Qt
∆T = = Qestimate Rglass = ( 6912 Btu hr ) ( 0.001085 hr o F Btu ) = 7.5 o F
kA
So T1,in = 75 – 7.5 = 67.50F and Ts,in = 45 + 7.5 = 52.5 oF
( 32.2ft/s ) (1 520R ) ( 0.077 lbm ft ) ( 67.5 − 52.5)R (1 12 ft ) ( 0.72 ) =15570
2 3 2 3
Ra =
(1.214 ×10 lbm fts )
−5 2
( 6 [1/12])
−1/ 9
Nu = 0.197 Ra1/ 4 ( H S) = 0.197 (15570 )
−1/ 9 1/ 4
From Table 12-5 = 1.369
Nuk (1.369 ) ( 0.0146 Btu hrft F )

o
h= = = 0.24 Btu hrft 2 o F

S 1/12ft
1 1
= = 0.1738 hr 0 F Btu
hA ( 0.24 Btu hrft 2 o F ) ( 6ft )( 4ft )
12- 76
( 75 -45 )
o
F Btu
Therefore, Q = =170
2 ( 0.001085 hr F Btu ) +0.1738 hr F Btu
o o
hr
We need to re-estimate h. Because of the symmetric system, the fluid properties are the same, so ∆Tglass =
(170) (0.001085) = 0.185 oF, T1,in = 74.8 oF, T2,in = 45.2 oF, Ra = 30720, Nu = 1.623, h = 0.28 Btu/hr ft2 oF,
1/hA = 0.147 hr oF/Btu, and Q = 201.7 Btu/hr. Answer
c) Adding the thin glass between the two panes changes the Rayleigh number, because the characteristic length is
changed as is the ∆T. We will keep the same fluid properties. Because of symmetry, half the temperature drop
occurs on each side of the thin glass partition, so
( 32.2 )(1 520 )( 0.077 ) ( 75 − 60 ) ⎡⎣(1 − 0.1) ( 2i12 )⎤⎦ ( 0.72 )
2 3
Ra =1419
(1.214 ×10 ) −5 2
From the Table 12-5, Nu = 1

(1)( 0.0146 ) Btu
h= = 0.39
⎡− ⎤
1 0.1 hrft 0 F
⎣ 2 (12) ⎦
1 1 hr 0 F
= = 0.107
hA ( 0.39 Btu hrft 0 F ) ( 6ft )( 4ft ) Btu
∆T
Q=
2 Rglass + 2 Rconv + R partition
( 75 − 45)
o
F Btu
= = 138 Answer
2 ( 0.001085hr 0 F Btu ) + 2 ( 0.107hr 0 F Btu ) +
( 0.1 12 )ft hr
( 0.8Btu hrft F) ( 6ft )( 4ft )
0
Comments:
Using a double pane window reduces the heat loss substantially. Inserting a thin glass partition between the two
other panes of glass further reduces the heat loss by 32%. Another technique to reduce heat loss is to lower the
interpane pressure and to use a gas (such as argon) with a lower thermal conductivity.
12- 77
12-70 Flat plate solar collectors have their best efficiency if they are tilted toward the sun at an angle that equals
the latitude of the location of the collector. Consider a 3-m wide and 2-m high solar collector. The solar
absorber plate, maintained at 65 ºC, is separated from the glass cover plate, which is at 30 ºC, by a distance
of 5-cm. Determine:
a. the heat loss from the collector if it horizontal (in W)
b. the heat loss from the collector if it is tilted at an angle of 33º from the horizontal (in W).
Approach:
Assuming steady, one-dimensional heat transfer, the
basic rate equation, Q = ∆T/Rtot, can be used to
calculate the heat transfer rate. The natural
convective heat transfer coefficient in the enclosure
must be determined.
Assumptions:
Solution:
a) Assuming constant properties, no radiation, and no heat loss from the ends on back, the heat transfer from the
absorber plate is Q = hA (Tp – Tc)
The natural convective heat transfer coefficient is calculated with properties from Appendix A-7 evaluated at
Tavg = (Tp + Tc ) 2 = 47.5 o C = 320.5 K, ρ = 1.103 kg/m3, k = 0.0278 W/mK, µ = 1.94 × 10-5 N⋅s/m2, Pr = 0.704
g βρ 2 (Tp − Tc ) S 3 Pr ( 9.81m s ) (1 320.5K ) (1.103kg ) ( 65 − 30 )K ( 0.05m ) ( 0.704 ) = 3.05 ×10

2 2 3
m3
Ra = = 5
µ2 (1.94 ×10 −5
N ⋅ s/m )
2 2
From Table 12-6, we see that we are near the dividing Ra for two correlations. We use the one in our range but
understand that there may be a larger uncertainty than if we are not near the end of the range of applicability.
Nu = 0.212 Ra0.25 = 0.212 (3.05 × 105)0.25 = 4.98
Nuk ( 4.98 )( 0.0278 W/mK )
h= = = 2.77W/m 2 K
S 0.05m
Q = (2.77 W/m2K) (3 m) (2 m) (65 – 30) K = 582 W Answer
b) For an angle 330 from the horizontal, we can use gcosθ in the Rayleigh number rather than just g.
Note θ = 90 - δ = 57o
Raθ = Ra cosθ = ( 3.05×105 ) cos ( 57o ) =1.66×105
Nu = 0.212 (1.66 × 105)0.25 = 4.28
( 4.28 )( 0.0275 )
h= = 2.38W/m 2 K
0.05
Q = (2.38) (3) (2) (65 – 30) = 500 W Answer
12- 78

CH 12

Uploaded by

Copyright:

Available Formats

CH 12

Uploaded by

Document Information

Original Description:

Original Title

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

CH 12

Uploaded by

Copyright:

Available Formats

12-1 An undergraduate heat transfer lab has an experiment to illustrate the effects of different boundary

ReL = = = 1.85 ×108

ReL1 = = =1.667 ×106

ReL = = = 1.20 × 107

ReL = = = 5.80 × 106

Rew = = = 28, 400

mL * = ( 0.05 + 0.002/2 ) m = 0.638

Because τ > 0.2, the one-term approximation is okay.

⎤ ( 0.712 )0.4 ⎡⎢ 1.72 ×10 ⎤⎥

Nuk ( 29.9 )( 0.0249W/mK )

= 2 + ⎡0.4(12800 ) + 0.06 (12800 ) ⎤ ( 0.708 ) ⎡⎣1.846 ×10−5 2.075 ×10−5 ⎤⎦

= 2 + ⎡ 0.4 (10200 ) + 0.06 (10200 ) ⎤ ( 0.708 ) (1.846 ×10−5 2.484 ×10−5 )

− ( 8238kg m3 ) ( 0.05 m 4 )( 602 J kgK ) ⎛ 900 - 27 ⎞

ρ VDH (1.044kg/m )(15 m min ) ( 0.30m )(1min 60s )

Nuk ( 4.59 )( 0.145 W/mK )

b) For the heat transfer rate, using the energy equation:

9.25 ×10-4 N ⋅ s/m 2

fluid properties remain fixed. Therefore, = = ⎜ ⎟

1.267 ×10-5 lbm/fts

f = [ 0.7901n Re −1.64] = ⎣⎡0.791n ( 3790 ) − 1.64 ⎦⎤ = 0.0422

( 0.0422 8 )( 3790 − 1000 )( 0.72 )

The single 4 cm tube yields the shortest length. Answer

ρ VDh ( 992.2 kg m ) ( 0.855 m s )( 0.01m ) (1Ns kgm )

where A = π DL . Substituting this in and solving for L:

f = [ 0.7901n Re − 1.64] = ⎡⎣0.7901n (108400 ) − 1.64 ⎤⎦ = 0.0177

( 0.0177/8)(108360 −1000 )( 3.02 ) Nuk ( 435 )( 0.650W/mk )

= 0.000188 + 0.000137 + 0.000302 = 0.000627 K/W

Nuk ( 22.7 ) ( 0.603 W/m K )

m = ρ V Ax = ( 998.2 kg m3 ) ( 0.5m/s )( 0.006m )( 0.006m ) = 0.0180kg/s

µ2 (1.77 ×10 N ⋅ s/m )

From Table 12-3, for a vertical wall

From flat plates, the characteristic length is

From Table 12-3 for a horizontal plate

2) Hot plate facing downward

( 9.81m s ) ( 0.7×10-3 K ) (150-20 ) K ( 0.015 m ) ( 425 )

This temperature is significantly about what we first estimated, so we iterate

( 9.81) ( 4.95 ×10−4 ) (84.7 − 20 )( 0.01) ( 3.15 )

Nu = 0.1 Ra 1/ 3 = 0.1( 9.78 ×1010 )

Nu = 0.1 (7.50 × 1010)1/3 = 421

From Table 12-3, for an upward facing heated plateNu = 0.15Ra1/3

For a sphere V = 4π R 3 3 = π D 3 6 and A = 4π R 2 = π D 2 . Combining these with the main equation:

Nu =12.4 → h = 72.0W/m 2 K → Ts = 59 o C Answer

µ2 (1.96 ×10 N ⋅ s/m ) -5 2 2

From Table 12-3, Nu = 0.1 Ra = 0.1( 5.50 ×10 ) =176

Nuk (176 )( 0.0281W/mK )

This is reasonably close to our assumed Ts, so iteration is not required.

Nuk (149 )( 0.0278W/mK )

For plain carbon steel, k = 60.5 W/mK, mL = ⎢

tanh ( 0.487 ) 16 ( 0.31m 2 )

This is close enough to our estimate, so we do not need to iterate.

g βρ 2 (T f − TP ) H 3 Pr ( 9.81m/s ) (1/260K ) (1.359 kg m ) (15 + 42 ) K( 2m ) ( 0.719 ) = 9.39 ×10

µ2 (1.56 ×10 N ⋅ s/m )

Nuk ( 454 )( 0.0230W/mK )

(1.98×10 N ⋅ s/m )−5 2 2

From Table 12-3

Nuk (Tb − T f ) = (1.99 )( 0.0278W/mK )( 75 − 20 )K = 380 W/m

g βρ 2 (Ts − T f ) H 23 Pr ( 9.81m s ) (1 304.5K ) (1.159 kg m ) ( 312 − 297 )K ( 0.5m ) ( 0.707 ) =1.61×10

µ2 (1.887 ×10 N ⋅ s/m )