Professional Documents
Culture Documents
TRANSFER
PROBLEMS
Sensitivities To
Governing Variables
by M. Kemal Atesmen
2009 by ASME, Three Park Avenue, New York, NY 10016, USA (www.asme.org)
All rights reserved. Printed in the United States of America. Except as permitted
under the United States Copyright Act of 1976, no part of this publication may be
reproduced or distributed in any form or by any means, or stored in a database or
retrieval system, without the prior written permission of the publisher.
INFORMATION CONTAINED IN THIS WORK HAS BEEN OBTAINED BY THE
AMERICAN SOCIETY OF MECHANICAL ENGINEERS FROM SOURCES BELIEVED
TO BE RELIABLE. HOWEVER, NEITHER ASME NOR ITS AUTHORS OR EDITORS
GUARANTEE THE ACCURACY OR COMPLETENESS OF ANY INFORMATION
PUBLISHED IN THIS WORK. NEITHER ASME NOR ITS AUTHORS AND EDITORS
SHALL BE RESPONSIBLE FOR ANY ERRORS, OMISSIONS, OR DAMAGES ARISING
OUT OF THE USE OF THIS INFORMATION. THE WORK IS PUBLISHED WITH
THE UNDERSTANDING THAT ASME AND ITS AUTHORS AND EDITORS ARE
SUPPLYING INFORMATION BUT ARE NOT ATTEMPTING TO RENDER
ENGINEERING OR OTHER PROFESSIONAL SERVICES. IF SUCH ENGINEERING
OR PROFESSIONAL SERVICES ARE REQUIRED, THE ASSISTANCE OF AN
APPROPRIATE PROFESSIONAL SHOULD BE SOUGHT.
ASME shall not be responsible for statements or opinions advanced in papers
or ... printed in its publications (B7.1.3). Statement from the Bylaws.
For authorization to photocopy material for internal or personal use under those
circumstances not falling within the fair use provisions of the Copyright Act, contact
the Copyright Clearance Center (CCC), 222 Rosewood Drive, Danvers, MA 01923,
tel: 978-750-8400, www.copyright.com.
2008047423
TABLE OF CONTENTS
Introduction ................................................................................................... 1
Chapter 1 Heat Loss from Walls in a Typical House ............................. 5
Chapter 2 Conduction Heat Transfer in a Printed Circuit Board .... 13
Chapter 3 Heat Transfer from Combustion Chamber Walls.............. 25
Chapter 4 Heat Transfer from a Human Body During
Solar Tanning ............................................................................ 33
Chapter 5 Efficiency of Rectangular Fins .............................................. 41
Chapter 6 Heat Transfer from a Hot Drawn Bar .................................. 51
Chapter 7 Maximum Current in an Open-Air Electrical Wire .......... 65
Chapter 8 Evaporation of Liquid Nitrogen in a
Cryogenic Bottle ...................................................................... 77
Chapter 9 Thermal Stress in a Pipe ........................................................ 85
Chapter 10 Heat Transfer in a Pipe with Uniform Heat
Generation in its Walls ......................................................... 93
Chapter 11 Heat Transfer in an Active Infrared Sensor .................. 103
Chapter 12 Cooling of a Chip ................................................................. 113
iii
iv
I NTRODUCTION
Introduction
CHAPTER
HEAT LOSS
FROM WALLS
IN A TYPICAL HOUSE
(1-1)
Most walls are constructed from three types of materials: inner wall
board, insulation and outer wall board. The heat transfer from these
wall layers will occur by conduction, and is presented by the following
rate Eqs., (1-2) through (1-4):
Q/A = (kinner wall/tinner wall) (Tinner wall inside surface Tinner wall outside surface) (1-2)
Q/A = (kinsulation/tinsulation) (Tinner wall outside surface Touter wall inside surface) (1-3)
Q/A = (kouter wall/touter wall) (Touter wall inside surface Touter wall outside surface) (1-4)
The heat transfer from the outer surface of the outer wall to the
atmosphere is by convection and can be expressed by the following
rate Eq. (1-5):
Q/A = hout (Tout Touter wall outer surface)
(1-5)
Eliminating all the wall temperatures from Eqs. (1-1) through (1-5),
the heat loss from a house wall can be rewritten as:
Q/A = (Tin Tout)/[(1/hin) + (tinner wall/kinner wall) + (tinsulation/kinsulation)
(1-6)
+ (touter wall/kouter wall) + (1/hout)]
The denominator in Eq. (1-6) represents all the thermal resistances
between the inside of the house and the atmosphere, and they are
in series.
In the construction industry, wall materials are rated with their
R-value, namely the thermal conduction resistance of one-inch
material. R-value dimensions are given as (hr-ft2-F/BTU)(1/in). The
sensitivity analysis will be done in the English system of units rather
than the International System (SI units). The governing Eq. (1-6)
for heat loss from a house wall can be rewritten in terms of R-values
as follows:
Q/A = (Tin Tout)/[(1/hin) + Rinner wall tinner wall + Rinsulation tinsulation
+ Router wall touter wall + (1/hout)]
(1-7)
1.8
Q/A, Btu/hr-ft2
1.79
1.78
1.77
1.76
1.75
10
15
20
25
Figure 1-1 Wall heat loss versus outside convection heat transfer coefcient
Q/A, Btu/hr-ft2
1.8
1.78
1.76
1.74
1.72
Figure 1-2 Wall heat loss versus inside convection heat transfer coefcient
material is denitely the way to go, depending upon the cost and
benet analysis results. The thickness of the insulation material is also
very crucial. Thicker insulation material is denitely the best choice,
depending upon the cost and benet analysis results.
Q/A, Btu/hr-ft2
2
1.8
1.6
1.4
1.2
3.5
4.5
hr-ft2-F/Btu-in
Q/A, Btu/hr-ft2
4
6
8
Insulation Thickness, in
10
12
The effects on heat loss of inner and outer wall board R-values
and thicknesses are similar to the effects of insulation R-value and
thickness, but to a lesser extent. Sensitivities of heat loss to all the
governing variables around the nominal values given above will be
analyzed later.
Sensitivity of heat loss to the outside convection heat transfer
coefcient can be determined in a closed form by differentiating the
heat loss Eq. (1-7) with respect to hout:
(Q/A)/hout = (Tin Tout)/{h2out [(1/hin) + Rinner wall tinner wall
+ Rinsulation tinsulation + Router wall touter wall + (1/hout)]2}
(1-8)
(Q/A) / hout, F
0.09
0.06
0.03
10
15
5
Outside Convection Heat Transfer Coefficient,
Btu/hr-ft2-F
20
Figure 1-5 Sensitivity of house wall heat loss per unit area to the outside
(Q/A) / Rinsulation,
(BTU/hr-ft2)2(in/F)
0
0.5
1
1.5
2
2.5
Figure 1-6 Sensitivity of house wall heat loss per unit area to insulation
material R-value
10
0.5
1.5
4
8
6
Insulation Thickness, in
10
12
Figure 1-7 Sensitivity of house wall heat loss per unit area to insulation
thickness
11
Table 1-1 House wall heat loss change per unit area due to a
10% change in variables nominal values
Variable
Nominal
Value
House Wall
Heat Loss
Change Due To
A 10%
Decrease In
Nominal Value
Tin Tout
36F
10%
+10%
Rinsulation
+7.467%
6.497%
House Wall
Heat Loss
Change Due To
A 10%
Increase In
Nominal Value
tinsulation
4 in
+7.467%
6.497%
Router wall
5 hr-ft2-F/BTU-in
+2.545%
2.545%
touter wall
1 in
+2.545%
2.545%
Rinner wall
+0.424%
0.424%
tinner wall
1 in
+0.424%
0.424%
hin
5 BTU/hr-ft -F
0.110%
+0.090%
hout
10 BTU/hr-ft -F
0.055%
+0.045%
least sensitive to both the inside and outside heat transfer coefcient
changes. Wall heat loss sensitivity to both the inside and outside
heat transfer coefcient changes is an order of magnitude less than
sensitivity to temperature potential changes.
12
CHAPTER
CONDUCTION
HEAT TRANSFER
IN A PRINTED
CIRCUIT BOARD
(2-1)
where the subscript refers to the layers of the PCB. Using the
conduction rate equation in rectangular coordinates for a PCB with
a width of W, a length of L, layer thicknesses ti and layer thermal
conductivities ki, Eq. (2-1) can be rewritten as:
13
(2-2)
(2-4)
(2-5)
(2-6)
(2-7)
kthrough-thickness = ti/(ti/ki).
(2-8)
where
14
15
kin-plane, W/m-C
400
300
200
100
0
20
40
60
80
% Conductor Layers Thickness
100
kin-plane / kinsulator
0.8
0.6
0.4
0.2
0
kin-plane / kinsulator
0
20
40
60
80
% Conductor Layers Thickness
100
to kinsulator
16
Nominal
Value
In-Plane Thermal
Conductivity Change
Due To A 10%
Decrease In
Nominal Value
In-Plane Thermal
Conductivity Change
Due To A 10%
Increase In
Nominal Value
kconductor
377 W/m-C
9.99%
+9.99%
tconductor
250 m
9.98%
+9.98%
tinsulator
250 m
+9.98%
9.98%
kinsulator
0.3 W/m-C
0.01%
+0.01%
Variable
17
kthrough-thickness, W/m-C
20
15
10
5
0
20
40
60
80
% Conductor Layers Thickness
100
20
kthrough-thickness /
kinsulator
16
12
8
4
0
50
% Conductor Layers Thickness
100
kthrough-thickness /
kconductor
18
0.05
kthrough-thickness /
kconductor
0
0.05
0.1
0.15
0.2
20
40
60
80
% Conductor Layers Thickness
100
kthrough-thickness /
kinsulator
19
Variable
tinsulator
250 m
+11.09%
9.08%
kinsulator
0.3 W/m-C
9.99%
+9.99%
tconductor
250 m
9.08%
+11.09%
kconductor
377 W/m-C
0.01%
+0.01%
kin-plane, W/m-C
400
300
200
100
0
0
20
40
60
80
% Conductor Layers Thickness
100
to total thickness
20
kin-plane / kconductor
0.8
0.6
0.4
0.2
0
kin-plane / kinsulator
0
20
40
60
80
% Conductor Layers Thickness
100
to kinsulator
21
Nominal
Value
In-Plane Thermal
Conductivity
Change Due To
A 10% Decrease
In Nominal Value
In-Plane Thermal
Conductivity
Change Due To
A 10% Increase
In Nominal Value
tinsulator
8.5 m
+33.3%
33.3%
kconductor
377 W/m-C
6.5%
+6.5%
tconductor
1.5 m
5.9%
+5.9%
36 W/m-C
3.5%
+3.5
Variable
kinsulator
kthrough-thickness, W/m-C
400
300
200
100
0
20
40
60
80
100
22
3.5
3
kthrough-thickness /
kinsulator
2.5
2
1.5
1
0.5
0
20
40
60
80
% Conductor Layers Thickness
100
kthrough-thickness /
kconductor
10
kthrough-thickness /
kconductor
0
10
20
30
40
50
20
60
40
80
% Conductor Layers Thickness
kthrough-thickness /
kinsulator
100
23
Variable
Nominal
Value
kinsulator
36 W/m-C
9.85%
+9.18%
tinsulator
8.5 m
+9.76%
8.17%
tconductor
1.5 m
1.55%
+1.59%
kconductor
377 W/m-C
0.18%
+0.15%
24
CHAPTER
3
FROM COMBUSTION
HEAT TRANSFER
CHAMBER WALLS
(3-1)
Radiation heat transfer per unit area from hot gases which are
assumed to behave as gray bodies to the hot side of a wall can be
written as:
(Q/A)radiation = hrg (Tg Twh) = = g (Tg4 Twh4)
(3-2)
And so the total heat transfer from the hot gases to a combustion
chamber wall is:
25
(3-4)
(3-5)
(3-6)
(3-7)
26
(3-8)
(3-9)
27
Iteration Function, K
2000
1000
1000
400
600
800
1000
1200
1400
Twh, K
2000
1600
1200
800
400
0
500
1000
1500
Gas Temperature, C
28
2000
1000
900
800
700
200
400
600
800
Cold Medium Temperature, C
1000
Figure 3-3 Hot side wall temperature versus cold medium temperature
860
840
820
800
10
20
30
40
Wall Thermal Conductivity, W/m-K
50
Figure 3-4 Hot side wall temperature versus wall thermal conductivity
29
840
830
820
810
800
0.02
0.04
0.06
Wall Thickness, m
0.08
0.1
1000
900
800
700
600
100
200
300
400
500
W/m2-K
Figure 3-6 Hot side wall temperature versus hot gas side convection heat
transfer coefcient
30
1000
800
600
400
200
100
200
300
400
500
Figure 3-7 Hot side wall temperature versus cold medium side convection
950
900
850
800
750
700
0.2
0.4
0.6
0.8
Figure 3-8 Hot side wall temperature versus hot gas emissivity
31
Nominal
Value
Change In Hot
Side Wall
Temperature For
A 10% Decrease
In Nominal Value
Change In Hot
Side Wall
Temperature For
A 10% Increase
In Nominal Value
Tg
1273 K
11.493%
+11.981%
hc
50 W/m2-K
+2.070%
1.991%
hcg
100 W/m2-K
1.262%
+1.136%
0.2
0.912%
+0.860%
Tc
373 K
0.295%
+0.295%
20 W/m-K
+0.056%
0.046%
0.01 m
0.051%
+0.051%
Variable
kwall
L
and they are applicable around the nominal values assumed for
this study.
Hot side wall temperature is most sensitive to variations in hot
gas temperature. Next in order of sensitivity are the convection heat
transfer coefcients on both sides of the wall. Changes to emissivity
of hot gases affect the dependent variable at the same level as the
convection heat transfer coefcients. Next in order of sensitivity is
the cold medium temperature. Hot side wall temperature is least
sensitive to variations in wall thermal conductivity and wall thickness.
This variable order of sensitivity is applicable around the nominal
values assumed for this case, due to the nonlinear relationship
between the dependent variable and the independent variables.
32
CHAPTER
HEAT TRANSFER
FROM A HUMAN
BODY DURING
SOLAR TANNING
33
For the present sensitivity analysis, the heat transfer rate equations
that will be used, and the nominal values that will be assumed for
energies outlined in Eq. (4-1), are given below.
Qsolar radiation absorbed = 851 W/m2
(4-2)
(4-3)
(4-4)
where h is the heat transfer coefcient between the skin surface that
is being tanned and the environment. In the present analysis, h is
assumed to be 28.4 W/m2-K and Tenvironment is 30C.
Qconduction to body = (kbody/tbody)(Tskin Tbody)
(4-5)
(4-6)
(4-7)
(4-8)
34
variables are analyzed in the region of the nominal values given above.
The governing Eq. (4-1) takes the following form, and can be solved
for Tskin by trial and error.
C1T4skin + C2Tskin = C3
(4-9)
35
Variable
Nominal
Value
Skin
Temperature,
Tskin, Change
Due To A 10%
Decrease In
Nominal Value
Tenvironment
30C
6.03%
+6.03%
5.11%
+5.13%
851 W/m
Skin
Temperature,
Tskin, Change
Due To A 10%
Increase In
Nominal Value
28.4 W/m -K
+2.93%
2.51%
Qperspiration
337.5 W/m
+2.03%
2.03%
0.8
+0.53
0.51
85 W/m2
0.51%
+0.51%
37C
0.45%
+0.45%
+0.27%
0.27%
Emissivity of
skin surface,
Qatmospheric radiation absorbed
Tbody
Qbasal metabolism
45 W/m
Skin-to-body
conduction
length, tbody
0.1 m
0.12%
+0.10%
0.2 W/m-K
+0.11%
0.11%
ktissue
36
70
65
Perspiration=0.5
liters/hr
Perspiration=1.0
liter/hr
Perspiration=1.5
liter/hr
60
55
50
45
40
35
30
100
50
150
200
for Qsolar radiation absorbed = 851 W/m2 and for three different
perspiration rates
Tskin / h, m2-K2/W
0
0.2
Perspiration=0.5
liters/hr
Perspiration=1.0
liters/hr
Perspiration=1.5
liters/hr
0.4
0.6
0.8
1
50
100
150
200
37
45
Perspiration=0.5
liters/hr
Perspiration=1.0
liter/hr
Perspiration=1.5
liter/hr
40
35
30
25
50
100
150
200
for Qsolar radiation absorbed = 425.5 W/m2 and for three different
perspiration rates
Tskin / h, m2-K2/W
0.2
0.1
0
Perspiration=0.5
liters/hr
Perspiration=1.0
liters/hr
Perspiration=1.5
liters/hr
0.1
0.2
0.3
0.4
0.5
50
100
150
200
38
60
Tskin, C
55
50
45
40
35
30
20
24
28
32
Tenvironment, C
36
40
Tskin, C
50
45
40
35
0.5
0.6
0.7
0.8
0.9
Emissivity Of Human Skin Surface,
39
Tskin / , C
30
28
26
24
0.5
0.6
0.7
0.8
0.9
Emissivity Of Human Skin Surface,
40
CHAPTER
EFFICIENCY OF
RECTANGULAR
FINS
(5-1)
41
(5-2)
(5-4)
(5-5)
(5-6)
42
(5-7)
(5-8)
1
Fin Efficiency
0.9
kcu=377.2
W/m-C
0.8
0.7
kal=206
W/m-C
0.6
0.5
0.4
100
200
300
400
L=0.0508 m
t=0.002 m
W/m2-C
43
h/ h, m2-C/W
0.001
kcu=377.2
W/m-C
kal=206
W/m-C
0.002
0.003
L=0.0508 m
t=0.002 m
0.004
0.005
0
100
200
300
400
Convection Heat Transfer Coefficient, W/m2-C
44
Fin Efficiency,
1
h=5
W/m2-C
0.8
h=100
W/m2-C
0.6
0.4
100
200
300
400
Fin Thermal Conductivity, W/m-C
500
L=0.0508 m
t=0.002 m
/ k, m-C/W
0.0084
0.0063
h=5
W/m2-C
0.0042
h=100
W/m2-C
0.0021
0
L=0.0508 m
t=0.002 m
0
100
200
300
400
500
45
Fin Efficiency,
1
0.9
h=5
W/m2-C
0.8
h=100
W/m2-C
0.7
0.6
t=0.002 m
k=377.2 W/m-C
0
0.02
0.04
0.06
0.08
0.1
Fin Length, m
46
8
h=5 W/m2-C & @
k=377.2 W/m-C
h=100 W/m2-C & @
k=377.2 W/m-C
h=5 W/m2-C & @
k=206 W/m-C
h=100 W/m2-C & @
k=206 W/m-C
7
/ L, 1/m
6
5
4
3
2
1
0
t=0.002 m
0
0.05
Fin Length, m
0.1
Fin Efficiency,
0.9
h=5
W/m2-C
h=100
W/m2-C
0.8
0.7
0.6
k=377.2 W/m-C
L=0.0508 m
0.5
0.4
0.001
0.002
0.003
Fin Thickness, m
0.004
0.005
47
9
8
/ t, 1/m
6
5
4
3
2
1
0
L=0.0508 m
0
0.001
0.002
0.003
0.004
0.005
Fin Thickness, m
48
Variable
Nominal
Value
Rectangular
Fin Efciency
Change Due To
A 10% Decrease
In Nominal Value
0.0508 m
+0.218%
0.239%
377.2 W/m-C
0.129%
+0.106%
0.002 m
0.124%
+0.102%
5 W/m -C
+0.117%
0.116%
Rectangular
Fin Efciency
Change Due To
A 10% Increase
In Nominal Value
Variable
Nominal
Value
Rectangular
Fin Efciency
Change Due To
A 10% Decrease
In Nominal Value
Rectangular
Fin Efciency
Change Due To
A 10% Increase
In Nominal Value
0.0508 m
+3.436%
3.478%
377.2 W/m-C
1.916%
+1.639%
0.002 m
1.844%
+1.574%
100 W/m -C
+1.806%
1.729%
49
CHAPTER
HEAT
TRANSFER FROM
A HOT DRAWN BAR
51
(6-2)
(6-4)
52
(6-5)
and write the differential equation for location i along the bar as
d2Ti/dx2 [(hconvection + hradiation) P/kA](Ti -Tenvironment)
= (cpU/k)dTi/dx
(6-6)
(6-7)
(6-8)
53
D = 0.01 m
= 8000 kg/m3
cp = 450 J/kg-K
k = 40 W/m-K
= 0.5
U = 0.02 m/s
Tenvironment = 298 K
The last independent variable is the convention heat transfer
coefcient between the surface of the bar and the environment.
The convection heat transfer coefcient can be determined from a
drawn bar temperature requirement at a distance from the die. In
the present analysis, Tx=10 is specied to be 373 K. At approximately
x=10 m the radiation heat transfer contribution almost diminishes.
The convection heat transfer coefcient that meets the Tx=10 = 373 K
requirement is determined from the above solution to be:
hconvection = 46.17 W/m2-K
which is in the turbulent region of forced cooling air over the
cylindrical bar, i.e., ReD = 1083 where ReD= VairD/air. The empirical
relationship for the convection heat transfer coefcient for air owing
over cylinders is given in Reference [10] by F. Kreith as:
hconvectionD/kair = 0.615 (VairD/air)0.466 for 40 < ReD < 4000
(6-9)
where Vair is mean air speed over the cylinder (2.18 m/s in this case),
kair is air thermal conductivity, and air is air kinematic viscosity.
Air thermophysical properties are calculated at lm temperature,
namely the average of bar surface temperature and environmental
temperature.
A comparison of convection and radiation heat transfer coefcients
as a function of distance from the die is given in Figure 6-1.
Heat transfer due to radiation is at the same order of magnitude
around the die. As the bar travels away from the die, radiation heat
transfer diminishes rapidly.
54
50
hconvection
40
30
20
10
0
hradiation
0
4
6
8
Distance From Die, m
10
55
Temperature, K
1400
1200
1000
Without Radiation
With Radiation
800
600
400
200
0
2
6
4
8
Distance From Die, m
10
Figure 6-2 Hot drawn bar temperature with and without radiation heat
transfer effects
1400
hconvection=5
W/m2-K
hconvection=20
W/m2-K
hconvection=40
W/m2-K
hconvection=60
W/m2-K
Temperature, K
1200
1000
800
600
400
200
8
6
Distance From Die, m
10
Figure 6-3 Hot drawn bar temperature for different convection heat trans-
56
T/ hconvection, m2-K2/W
0
10
20
30
40
50
10
20
30
40
50
hconvection, W/m2-K
fer coefcient
1000
800
600
400
200
10
20
30
40
hconvection, W/m2-K
50
60
transfer coefcient
57
400
380
360
340
7000
7500
8000
Bar Density,
8500
9000
kg/m3
58
400
380
360
340
400
420
440
460
Bar Specific Heat, J/kg-K
480
500
Figure 6-7 Bar temperature @ x=10 m versus bar specic heat at constant
pressure
373.04
373.03
373.02
373.01
373
20
30
40
50
60
59
700
600
500
400
300
200
0.01
0.02
0.03
Bar Velocity, m/s
0.04
0.05
1000
800
600
400
200
0.01
0.02
0.03
Bar Diameter, m
0.04
0.05
60
380
375
370
365
360
15
20
25
Environment Temperature, C
30
The slope of the curve is 0.9231 K/C under the assumed nominal
conditions.
When the nominal values of the independent variables given above
are varied +-10%, the results shown in Table 6-1 are obtained. The
sensitivity analysis is conducted by neglecting radiation heat transfer
at x=10 meters.
The convection heat transfer coefcient, bar diameter, bar velocity,
bar density and bar-specic heat at constant pressure have the same
order of magnitude sensitivity on bar temperature at x=10 m. Changes
in the temperature of the environment affect bar temperature at x=10
m, at an order of magnitude less. Changes to the thermal conductivity
of the bar have the least effect on bar temperature at x=10 m. The
sensitivity magnitudes and order that are shown in Table 6-1 are only
valid around the nominal values that are assumed for the independent
variables for this analysis. Bar velocity, bar density and bar-specic
heat at constant pressure have the same sensitivity effects on the
temperature of the bar as can be seen in Eq. (6-7).
Another interesting sensitivity analysis can be performed around
x=0.5 m, where both the convection and the radiation heat transfers
are in effect.
61
Table 6-1 Effects of 10% change in nominal values of variables to bar temperature @ x=10 m
Nominal
Value
Bar Temperature
@ x=10 m Change
Due To A 10%
Decrease In
Nominal Value
Bar Temperature
@ x=10 m Change
Due To A 10%
Increase In
Nominal Value
46.17 W/m2-K
+5.879%
4.549%
0.01 m
4.986%
+5.280%
0.02 m/s
4.986%
+5.279%
8000 kg/m
4.986%
+5.279%
cp
450 J/kg-K
4.986%
+5.279%
25C (298 K)
0.619%
+0.619%
40 W/m-K
0.0007%
+0.0007%
Variable
hconvection
Tenvironment
k
Nominal
Value
Bar Temperature
@ x=0.5 m Change
Due To A 10%
Decrease In
Nominal Value
Bar Temperature
@ x=0.5 m Change
Due To A 10%
Increase In
Nominal Value
0.01 m
1.386%
+1.183%
0.02 m/s
Variable
1.386%
+1.183%
8000 kg/m
1.386%
+1.183%
cp
450 J/kg-K
1.386%
+1.183%
46.17 W/m -K
+0.788%
0.783%
hconvection
Tenvironment
k
62
0.5
+0.500%
0.482%
25C (298 K)
0.046%
+0.046%
40 W/m-K
0.0003%
+0.0003%
63
CHAPTER
MAXIMUM
CURRENT IN
AN OPEN-AIR
ELECTRICAL WIRE
(7-1)
65
The convection heat transfer from the outer surface of the wire
insulator to the environment can be written by the rate equation in
cylindrical coordinates:
Q = 2rwLh (Tinsulation outer radius Tenv)
(7-2)
The heat transfer mechanisms in Eqs. (7-1) and (7-2) are in a series
thermal resistance path. The energy balance for this heat transfer
problem, heat generated by the conductor equals heat lost to the
environment, can be written as follows, by eliminating Tinsulation outer radius
from Eqs. (7-1) and (7-2):
I2R = (Tc Tenv)/{[ln(rw/rc)/(2Lkins)] + (1/2rwLh)}
(7-3)
66
(7-4)
0.8128
20
1.0236
18
1.2900
16
1.6280
14
2.0520
12
2.5900
10
3.2640
4.1150
5.1890
6.5430
7.3406
8.2550
1/0
9.2710
2/0
10.4140
3/0
11.6840
4/0
67
40
30
20
AWG20
10
AWG6
4
AWG2
6
AWG1/0 AWG3/0
8
10
12
(7-5)
68
Maximum Current
Allowed, A
400
NEC
Current
Allowed, A
300
200
Maximum
Current
Allowed, A
100
0
0
10
12
Figure 7-2 Maximum wire current allowed versus wire conductor diameter
rw = 3rc
kins = 0.3 W/m-C
h = 9 W/m2-C
Under the above conditions the maximum wire current calculated
is compared to National Electric Code (NEC) recommendations in
Figure 7-2.
The calculated maximum wire currents are very much in line with
the NEC-recommended values up to AWG6. The variable assumptions
given above start to predict higher maximum current values for large
diameter wires, and therefore have to be adjusted for large diameter
wire applications.
The sensitivity of maximum wire current to environmental
temperature, Imax/Tenv, is given in Figure 7-3.
Sensitivities in Figure 7-3 are for a wire temperature rating of
60C. As the environmental temperature approaches the wire rating
temperature, the maximum wire current allowed becomes more
sensitive to environmental temperature changes.
69
0
2
4
Tenv=0C
Tenv=20C
Tenv=40C
Tenv=50C
6
8
10
12
10
12
70
10000
8000
6000
4000
2000
0
0.000 0.002 0.004 0.006 0.008 0.010 0.012
Ins Thickness=
0.5*Wire Dia
Ins Thickness=
0.75*Wire Dia
Ins Thickness=
1.0*Wire Dia
Ins Thickness=
1.25*Wire Dia
Ins Thickness=
1.5*Wire Dia
8000
AWG20
AWG16
AWG12
AWG8
AWG4
AWG1
AWG2/0
AWG4/0
7000
6000
5000
4000
3000
2000
1000
0.5*d
0.75*d
1.25*d
1.5*d
Insulation Thickness
71
300
AWG20
AWG16
AWG12
AWG8
AWG4
AWG1
AWG2/0
AWG4/0
250
200
150
100
50
0
0.2
0.25
0.3
0.35
0.4
Imax/ h, A-m2-K/W
500
AWG20
AWG16
AWG12
AWG8
AWG4
AWG1
AWG2/0
AWG4/0
400
300
200
100
0
12
16
20
2
72
Imax/ h, A-m2-K/W
600
500
h=7 W/m2-K
400
h=8 W/m2-K
300
h=9 W/m2-K
h=12 W/m2-K
200
h=15 W/m2-K
h=20 W/m2-K
100
0
0
0.01
0.012
Wire Diameter, m
73
0
AWG 20
AWG 16
AWG 12
AWG 8
AWG 4
AWG 1
AWG 2/0
AWG 4/0
50
100
150
200
1.00E-08
1.50E-08
2.00E-08
2.50E-08
3.00E-08
74
Variable
Conductor
Diameter, 2rc
4.115 mm
13.9%
+14.5%
Conductor
Resistivity,
1.60E-08
-m
+5.4%
4.7%
Temperature Rating Of
Wire Tc Temperature
Of Environment, Tenv
30C
5.1%
+4.9%
Convection Heat
Transfer Coefcient, h
9 W/m2-C
4.32%
+4.01%
Insulation Thickness,
rw rc
2rc
2.35%
+2.17%
Thermal Conductivity
Of Wire insulation, kins
0.3 W/m-C
0.93%
+0.78%
of the wire conductor diameter rather than the accuracy of the wire
insulation thermal conductivity. It is important to remember that the
maximum wire current change order shown in Table 7-2 is good only
in this region of the application, due to non-linear behavior of the
sensitivities.
75
CHAPTER
EVAPORATION
OF LIQUID
NITROGEN IN A
CRYOGENIC BOTTLE
H
(8-1)
77
(8-2)
where
ln is natural logarithm of argument (r2/r1),
r1 = inner radius of inner tube in m,
r2 = outer radius of inner tube in m,
L = height of the tube in m, and
kss = thermal conductivity of stainless steel inner tube in W/m-K.
R32 is the radiation heat transfer thermal resistance between the
inner surface of the outer insulation tube, radial location (3), and the
outer surface of the inner tube, radial location (2); in other words,
through the vacuum gap.
R32 = 1/(2r3Lhr)
(8-3)
where
r3 = inner radius of outer insulation tube in m,
hr = (T34 T24)/(T3 T2) in W/m2-K,
= emissivity of inner surface of outer insulation tube,
= Stefan-Boltzmann constant, namely 5.67 108 W/m2-K4
T3 = temperature of inner surface of outer insulation tube in K,
T2 = temperature of outer surface of inner tube in K
R43 is the conduction heat transfer thermal resistance between the
outer surface of the outer insulation tube, radial location (4) and the
inner surface of the outer insulation tube, radial location (3).
78
R43 = ln(r4/r3)/(2Lkinsulation)
(8-4)
where
r4 = outer radius of outer insulation tube in m and
kinsulation = thermal conductivity of insulation material of the outer
tube in W/m-K.
Renvironment-4 is the convection heat transfer thermal resistance between
the environment and the outer surface of outer insulation tube, radial
location (4). Radiation heat transfer between the environment and
the outer surface of outer insulation tube is assumed to be negligible.
Renvironment-4 = 1/(2r4ho)
(8-5)
where
ho = convection heat transfer coefcient between the environment
and the outer surface of the outer tube in W/m2-K.
Since hr depends on unknown temperatures T3 and T2, an iterative
method is used to calculate these temperatures, and thereafter hr
(i.e., the radiation heat transfer coefcient between the inner surface
of the outer insulation tube and the outer surface of the inner tube, or
through the vacuum gap). All calculations are done using the Kelvin
temperature scale due to fourth power calculations in the radiation
heat transfer coefcient. The thermophysical properties used in these
calculations are assumed to be constants.
Nominal values of the variables used in this problem are as follows:
Tenvironment = 293 K
Tliquid nitrogen = 77 K
r1 = 0.1 m
r2 = 0.106 m, namely inner tube stainless steel wall thickness is 6 mm
r3 = 0.116 m, namely vacuum gap is 0.01 m
r4 = 0.166 m, namely outer tube insulation thickness is 0.05 m
L = 0.8 m
kss = 10 W/m-K
79
(8-6)
Temperature, K
180
160
140
120
100
21
19
17
15
13
11
80
Number Of Iterations
insulation tube
80
Evaporation Time, hr
1500
1000
500
0
0
0.002
0.004
0.006
0.008
0.01
81
t/ kinsulation, hr-m-K/W
0.0E+00
1.0E+06
2.0E+06
3.0E+06
0
0.001
0.002
0.003
0.004
0.005
t/
100
200
300
400
0.2
0.4
0.6
0.8
82
Variable
Nominal
Value
Change in Time
Change in Time
For 5 Liters Of
For 5 Liters Of
Liquid Nitrogen
Liquid Nitrogen
To Evaporate For To Evaporate For
A 10% Decrease
A 10% Increase
In Nominal Value In Nominal Value
0.8 m
+11.111%
9.091%
0.002 W/m-K
+9.323%
7.694%
r1
0.1 m
+8.062%
6.931%
r4 r3 (outer
tube insulation
thickness)
0.05 m
7.212%
+6.947%
0.5
+1.671%
1.434%
Tenvironment
293 K
+1.064%
1.043%
r3 r2
(vacuum gap)
0.01 m
+0.751%
0.740%
r2 r1 (stainless
steel inner tube
wall thickness)
0.006 m
+0.449%
0.445%
ho
10 W/m2-K
+0.0315%
0.0258%
kss
10 W/m-K
+0.00006%
0.000049%
L
kinsulation
of importance, and they are applicable only around the nominal values
assumed for this study.
The time required for ve liters of liquid nitrogen to evaporate
is most sensitive to cryogenic tube height, insulation layer thermal
conductivity, inside radius of the inner tube, and outer tube insulation
thickness. The second tier of sensitivities are an order of magnitude
less; emissivity of inner surface of the outer insulation tube,
temperature of environment, vacuum gap thickness, and stainless
steel inner tube wall thickness. Time for ve liters of liquid nitrogen to
83
84
CHAPTER
9
STRESS IN A PIPE
THERMAL
(9-1)
If the temperatures at the inner surface, Ti, and the outer surface,
To, of the pipe wall are known, Eq. (9-1) can be solved by using the
following boundary conditions:
T = Ti at R = Ri
(9-2)
85
and
T = To at R = Ro
(9-3)
(9-4)
(9-5)
(9-6)
-T}
(9-7)
where
R, , and Z are the thermal stress distributions at the pipe wall in
MPa (Mega Pascals),
is the coefcient of thermal expansion of the pipe wall material in
m/m-C,
E is modulus of elasticity for the pipe wall material in MPa, and
is Poisson's ratio for pipe wall material.
86
(9-8)
The conduction heat transfer per unit length through the pipe wall
can be written as:
Q = 2k (Ti To)/ln(Ro/Ri)
(9-9)
The convection heat transfer per unit length from the outer surface
of the pipe wall to the environment is:
Q = 2Ro ho (To Tenvironment)
(9-10)
87
88
120
100
80
60
Tenvironment= 50C
40
20
0
20
40
60
80
89
86
84
82
80
78
76
0.2
0.25
0.3
0.35
0.4
0.45
0.5
Figure 9-2 Thermal stress at Ro versus pipe outside radius for a constant
120
80
40
0
0
0.05
0.1
0.15
0.2
0.25
0.3
Figure 9-3 Thermal stress at Ro versus pipe wall thickness for a pipe outside
90
100
80
60
40
20
0
200
400
600
800
1000
Figure 9-4 Thermal stress at Ro versus the heat transfer coefcient at inside
160
120
80
40
0
200
400
600
800
1000
91
Variable
Pipe Wall
Thickness, Ro-Ri
Nominal
Value
Change In
Change In
Thermal Stress Thermal Stress
At Ro For A 10% At Ro For A 10%
Decrease In
Increase In
Nominal Value
Nominal Value
0.1 m
+10.049%
13.011%
2 105 m/m-C
10%
+10%
210000 MPa
10%
+10%
100C
10%
+10%
20 W/m-C
+7.311%
6.378%
ho
100 W/m2-C
5.502%
+5.002%
0.3
4.110%
+4.478%
hi
500 W/m -C
1.717%
+1.450%
Ro
0.3 m
+0.920%
1.248%
Tuid-Tenvironment
92
CHAPTER
10
HEAT
TRANSFER IN A
PIPE WITH UNIFORM
HEAT GENERATION IN
ITS WALLS
U
(10-1)
(10-2)
T = To at R = Ro
(10-3)
and
93
(10-5)
and
(2RoL) k dT/dR = (2RoL) ho (To Tenvironment) at R = Ro (10-6)
where hi is the convection heat transfer coefcient at inside surface
of the pipe in W/m2-C, Tuid is mean uid temperature in C in the pipe,
ho is the convection heat transfer coefcient at outside surface of the
pipe in W/m2-C, and Tenvironment is environmental temperature in C.
L is the length of the pipe in meters and it cancels out from the
energy balance equations.
Obtaining dT/dR at R = Ri from Eq. (10-4), the governing energy
balance Eq. (10-5) at R = Ri becomes:
hi (Tuid Ti) = 0.5 Q Ri
[k/Ro ln(Ro/Ri)][(To Ti) + (Q/4k)(Ro2 Ri2)] (10-7)
Obtaining dT/dR at R = Ro from Eq. (10-4), the governing energy
balance Eq. (10-6) at R = Ro becomes:
ho (To Tenvironment)
= 0.5 Q Ro [k/Ro ln(Ro /Ri)][(To Ti) + (Q/4k)(Ro2 Ri2)] (10-8)
94
95
250000
200000
150000
100000
50000
0
0.09
0.095
0.1
0.105
0.11
radius
200000
150000
100000
50000
0
0.1
0.105
0.11
0.115
0.12
radius
96
24000
23000
22000
21000
20000
20
40
60
80
100
thermal conductivity
97
60000
40000
20000
0
20000
40000
20
40
60
80
100
500000
400000
300000
200000
100000
0
100000
100
20
40
60
80
Convection Heat Transfer Coefficient At Pipe Wall
Outside Surface, W/m2-C
98
150000
100000
50000
0
50000
100000
150000
20
40
60
80
Mean Fluid Temperature Inside The Pipe, C
100
99
60000
40000
20000
0
20000
40000
40
20
20
40
60
Environment Temperature, C
temperature
250000
200000
150000
100000
50000
0
50000
100000
150000
30
60
Pipe Inner Surface Temperature, C
90
surface temperature
100
Nominal
Value
Change In
Required
Thermal Energy
Generation For
A 10% Decrease
In Nominal Value
Change In
Required
Thermal Energy
Generation For
A 10% Increase
In Nominal Value
Ti
40C
66.49%
+66.49%
Tuid
50C
+60.93%
60.93%
Ro
0.11 m
+104.35%
@ 5% decrease
in nominal value
45.38%
Ri
0.10 m
41.75%
+84.65%
@ 5% increase
in nominal value
ho
10 W/m2-C
22.08%
+22.06%
30 W/m -C
+12.19%
12.19%
10C
+4.44%
4.44%
15 W/m-C
0.13%
+0.11%
Variable
hi
Tenvironment
k
101
CHAPTER
HEAT
11
TRANSFER
IN AN ACTIVE
INFRARED SENSOR
I
103
(11-1)
(11-2)
(11-3)
(11-4)
(11-5)
The net radiation heat transfer between the object and the sensor
is determined by combining Eqs. (11-4) and (11-5):
Qnet radiation = Asensor object (T4object T4sensor)
(11-6)
104
transfer can be written as a rate equation, where the two heat transfer
mechanisms act in series:
Qconduction + convection = (1/RT) (Tsensor Tenvironment)
(11-7)
Energy balance for the sensor can be written as a first order and
non-linear differential equation:
cpV(dTsensor/d) = I2R (Tsensor Tenvironment)/Rtotal
+ Aobjectsensor (T4object T4sensor) (11-8)
where
cpV is the sensor thermal capacitance in W-s/K
dTsensor/d is the time rate of change of sensor temperature in K/s
I2R is control power supplied to regulate the surface temperature of
the sensor in W
Tenvironment is the environment temperature in K
Rtotal is the total heat transfer resistance between the sensor and
the environment due to conduction and convection heat transfer
in K/W
A is the sensor area in m2
= 5.6710-8 W/m2-K4 is the Stefan-Boltzmann constant
object is the surface emissivity of the object being measured
sensor is the surface emissivity of the sensor
Tobject is the temperature of the object being measured in K
Tsensor is the temperature of the sensor surface in K
For the steady-state sensor temperature, Eq. (11-8) can be
rewritten by eliminating the left-hand side of the equation, which
means that the change in internal energy of the sensor with respect to
time becomes negligible. Tsensor is calculated from the following quartic
equation, by trial and error:
T4sensor + C1 Tsensor = C2
(11-9)
C1 = 1/(RT Aobjectsensor)
(11-10)
where
105
and
C2 = (I2R/Aobjectsensor) + (Tenvironment/Rtotal Aobjectsensor) + T4object
(11-11)
The transient solutions to Eq. (11-8) are obtained from the following
explicit finite difference equation, using small time intervals :
cpV(Tsensor @ (i+1) Tsensor @ i /) = I2R (Tsensor @ i Tenvironment)/Rtotal
(11-12)
+ Aobjectsensor (T4object T4sensor @ i)
For the present sensitivity analysis, nominal values of the
independent variables are assumed to be as follows:
cpV = 0.014 W-s/K
I2R = 0.1 W
Tenvironment = 20C (293 K)
Rtotal = 100 K/W
A = 0.0001 m2 (a 1 cm 1 cm sensor surface area)
object = 0.9
sensor = 0.8
The sensitivities are analyzed for three different object temperatures,
namely 100C (373 K), 500C (773 K) and 1000C (1273 K).
Sensor temperature as a function of time for three different
object temperatures is given in Figures 11-1, 11-2, and 11-3. As the
temperature of the object increases, the sensor time constant, or
the time the sensor reaches 63.2% of its steady-state temperature,
decreases. Also as the sensor emissivity increases, the sensor time
constant decreases.
As the object temperature approaches the environmental
temperature, the sensor temperature deviates from the object
temperature because fixed nominal values are used for the analysis
of this heat transfer problem. The control circuit of the sensor
has to respond and change control energy, I2R, in order to achieve
accurate results. The details of the control circuit are explained in the
Reference by J. Fraden [3].
106
200
Tsensor, C
150
Emissivity
Sensor=0.9
Emissivity
Sensor=0.8
Emissivity
Sensor=0.7
100
50
0
0
2
Time, s
Figure 11-1 Sensor temperature versus time for different sensor emissivities
200
Tsensor, C
150
Emissivity
Sensor=0.9
Emissivity
Sensor=0.8
Emissivity
Sensor=0.7
100
50
0
0
Time, s
Figure 11-2 Sensor temperature versus time for different sensor emissivities
107
Tsensor, C
40
35
30
25
20
15
10
5
0
Emissivity
Sensor=0.9
Emissivity
Sensor=0.8
Emissivity
Sensor=0.7
0
2
Time, s
Figure 11-3 Sensor temperature versus time for different sensor emissivities
Figure 11-4 Sensor temperature versus time for different total heat transfer
108
Variable
Sensor Thermal
Capacitance, cpV
Nominal
Value
Thermal Time
Thermal Time
Constant Change Constant Change
Due To A 10%
Due To A 10%
Decrease In
Increase In
Nominal Value
Nominal Value
0.014 W-s/K
20.39%
+10.38%
Tobject
1000C
(1273 K)
+16.73%
15.38%
0.0001 m2
+8.12%
7.44%
object
0.9
+8.12%
7.44%
sensor
0.8
+8.12%
7.44%
Rtotal
100 K/W
3.04%
+2.31%
Tenvironment
20C
(293 K)
0.15%
+0.15%
0.1 W
+0.049%
0.049%
I 2R
109
Variable
Nominal
Value
Steady-State
Steady-State
Sensor
Sensor
Temperature
Temperature
Change Due To
Change Due To
A 10% Decrease A 10% Increase
In Nominal Value In Nominal Value
Tobject
1000C
(1273 K)
18.33%
+18.22%
Rtotal
100 K/W
4.13%
+3.48%
0.0001 m
4.07%
+3.44%
object
0.9
4.07%
+3.44%
sensor
0.8
4.07%
+3.44%
20C
(293 K)
0.11%
+0.11%
0.1 W
0.054%
+0.054%
0.014 W-s/K
0%
0%
Tenvironment
I 2R
Sensor Thermal
Capacitance, cpV
110
Variable
Tenvironment
Sensor Thermal
Capacitance, cpV
RT
Thermal Time
Constant
Change Due
To A 10%
Decrease In
Nominal Value
Thermal Time
Constant
Change Due
To A 10%
Increase In
Nominal Value
20C
(293 K)
14.97%
+13.16%
0.014 W-s/K
10.33%
+10.39%
100 K/W
9.90%
+9.83
Nominal
Value
0.0001 m
0.45%
+0.45%
object
0.9
0.45%
+0.45%
sensor
0.8
0.45%
+0.45%
0.1 W
0.028%
+0.028%
100C
(373 K)
+0.023%
0.023%
IR
Tobject
and the environment, the sensor area, and surface emissivities of the
sensor and the object. Sensor thermal capacitance does not affect
the steady-state sensor temperature, as expected. The steady-state
sensor temperature experiences non-linear sensitivity behaviors
from all the variables except the environmental temperature and
the I2R power input to the sensor, in the present region of operation.
The steady-state sensor temperature experiences similar magnitude
sensitivity behaviors from the sensor area, sensor emissivity, and
object emissivity variables, as expected.
A similar sensitivity analysis is performed for a low object
temperature case, namely Tobject = 100C, and the results are given in
Tables 11-3 and 11-4.
111
Variable
Nominal
Value
Steady-State
Sensor
Temperature
Change Due To
A 10% Decrease
In Nominal
Value
Steady-State
Sensor
Temperature
Change Due To
A 10% Increase
In Nominal
Value
Tenvironment
20C
(293 K)
5.58%
+5.57%
RT
100 K/W
3.99%
+3.95%
0.1 W
2.79%
+2.79%
100C
(373 K)
2.27%
+2.46%
0.0001 m2
+1.18%
1.19%
IR
Tobject
A
object
0.9
+1.18%
1.19%
sensor
0.8
+1.18%
1.19%
0.014 W-s/K
0%
0%
Sensor Thermal
Capacitance, cpV
112
CHAPTER
COOLING
OF A CHIP
12
(12-1)
where
cpV is the chip thermal capacitance in W-s/C
dTchip/d is the time rate of change of chip temperature in C/s
P is the energy generated in the chip in W
Tenvironment is the environment temperature in C
RTotal is the total heat transfer resistance between the chip, its heat
sink and the environment due to conduction and convection heat
transfer in C/W
Tchip is the temperature of the chip in C.
Since the governing equation is linear in temperature, centigrade
dimension is used instead of Kelvin. Total heat transfer resistance
between the chip, its heat sink and the environment has two
parallel components. One component is convection heat transfer
resistance between the chip and the environment. The other is the
heat loss in series circuit from the chip through the adhesion layer
by conduction, from the adhesion layer through the heat sink by
conduction, and from the heat sink to the environment by convection.
The heat transfer resistance between the surface of the chip and the
environment can be written as:
Rchip to environment = 1/(hAchip for convection)
(12-2)
The series circuit heat transfer resistance between the chip and
the environment going through the adhesive and the heat sink can be
written as:
Rchip through heat sink = [Ladhesive/(kadhesiveAadhesive for conduction)]
+ [Lheat sink/(kheat sinkAheat sink for conduction)]
(12-3)
+ [1/(hAheat sink for convection)]
114
Cooling Of A Chip
(12-4)
The solution to the governing first order differential Eq. (12-1) can
be written as follows by using the initial condition of Tchip = Tenvironment:
Tchip = Tenvironment + PRTotal [1 exp(/cpVRTotal)]
(12-5)
where cpVRTotal is the thermal time constant for the chip, which is the
product of its thermal capacitance and its total thermal resistance.
Under steady-state conditions, the solution in Eq. (12-5) provides
the allowable chip power dissipation as follows:
P = (Tchip Tenvironment)/RTotal
(12-6)
For the present sensitivity analysis, the nominal values of the above
independent variables are assumed to be as follows:
cpV = 0.0197 W-s/C
(assuming a silicon dioxide chip with 10 mm 10 mm 0.1 mm
dimensions)
h = 200 W/m2-C (convection heat transfer coefficient at the chip
and heat sink surfaces)
Tenvironment = 30C
Tchip = 90C
Achip for convection = 0.000104 m2
Ladhesive = 0.00001 m
Aadhesive for conduction = Aheat sink for conduction = 0.0001 m2
Lheat sink = 0.01 m
Aheat sink for convection = 0.0005 m2 (assuming a copper alloy heat sink
with 10 mm 10 mm 10 mm dimensions)
kadhesive = 10 W/m-C (assuming silver epoxy adhesive)
kheat sink = 400 W/m-C
115
200
RTotal, C/W
150
100
50
0
0
100
200
300
400
Figure 12-1 Total chip heat transfer resistance versus the convection heat
transfer coefficient
Total heat transfer resistance between the chip, its sink and the
environment is given as a function of the convection heat transfer
coefficient in Figure 12-1. As the convection mechanism goes into
forced convection, and especially forced convection in liquids,
35
30
25
20
15
10
5
0
0
200
400
600
800
1000
W/m2-C
Figure 12-2 Chip power dissipation versus the convection heat transfer
116
Cooling Of A Chip
Variable
Tchip Tenvironment
h
Aheat sink for convection
Achip for convection
Nominal
Value
Steady-State
Chip Power
Dissipation
Capability
Change Due To
A 10% Decrease
In Nominal
Value
Steady-State
Chip Power
Dissipation
Capability
Change Due To
A 10% Increase
In Nominal
Value
60C
10%
+10%
9.81%
+9.77%
8.05%
+8.01%
1.76%
+1.76%
200 W/m-C
2
0.0005 m
0.000104 m
0.0001 m2
0.23%
+0.19%
kheat sink
400 W/m-C
0.22%
+0.18%
Lheat sink
0.01 m
+0.20%
0.20%
kadhesive
10 W/m-C
0.009%
+0.007%
Ladhesive
0.00001 m
+0.008%
0.008%
117
3.5
3
2.5
2
1.5
1
0.5
0
0
100
200
300
400
Figure 12-3 Chip thermal time constant versus the convection heat transfer
coefficient
118
Cooling Of A Chip
Variable
h
cpV
Nominal
Value
Chip Thermal
Time Constant
Change Due To
A 10% Decrease
In Nominal
Value
Chip Thermal
Time Constant
Change Due To
A 10% Increase
In Nominal
Value
200 W/m-C
+10.88%
8.90%
0.0197 W-s/C
10%
+10%
0.0005 m
+8.76%
7.42%
0.000104 m2
+1.79%
1.79%
0.0001 m2
+0.233%
0.190%
kheat sink
400 W/m-C
+0.223%
0.183%
Lheat sink
0.01 m
0.201%
+0.201%
kadhesive
10 W/m-C
+0.009%
0.007%
Ladhesive
0.00001 m
0.008%
+0.008%
thermal time constant of the chip. The results are given in descending
order, from the most sensitive variable to the least.
Changes in the convection heat transfer coefficient, chip thermal
capacitance and heat sink surface area for convection heat transfer
affect the chip thermal time constant the most. Sensitivities of the
chip thermal time constant to the rest of the independent variables
follow the same order as in Table 12-1. Changes in independent
variables affecting the conduction heat transfer contribute the least to
chip thermal time constant sensitivities.
119
CHAPTER
13
A CHIP UTILIZING
COOLING OF
(13-1)
121
where
cpV is the chip thermal capacitance in W-s/C
dTchip /d is the time rate of change of chip temperature in C/s
P is the heat generated in the chip in W
Tenvironment is the environment temperature in C
RTotal is the total heat transfer resistance between the chip, its
heat sink with fins and the environment, due to conduction and
convection heat transfer in C/W
Tchip is the temperature of the chip in C
Since the governing equation is linear in temperature, centigrade
dimension is used instead of Kelvin. Total heat transfer resistance
between the chip, its sink and the environment has two parallel paths.
One path is the direct convection heat transfer between the chip and
the environment. The other is the heat loss in series from the chip to
the adhesion layer by conduction, from the adhesion layer to the heat
sink body by conduction, and from the heat sink and its fins to the
environment by convection. The convection heat transfer from the
heat sink and its fins to the environment has two parallel components;
one is the convection heat transfer from the sink base surface to the
environment, and the other is the convection heat transfer from the
fins surfaces to the environment.
The convection heat transfer resistance between the heat sinks
un-finned surfaces and the environment can be written as follows:
Rheat sink convection from un-finned surfaces = 1/(hAheat sink convection un-finned surfaces) (13-2)
The convection heat transfer between heat sink fin surfaces and
the environment can be written by combining Eqs. (5-5) and (5-6):
Qfin total = NhAfin (Tfin base Tenvironment)
(13-3)
The convection heat transfer resistance between the heat sinks fin
surfaces and the environment can be written as follows:
Rheat sink convection from fin surfaces = 1/(NhAfin)
122
(13-4)
(13-6)
The total heat transfer resistance between the chip, its heat
sink with fins and the environment can be written by combining
Eqs. (13-5) and (13-6) in a parallel thermal circuit:
RTotal = {1/[(1/Rchip to environment) + (1/Rchip through heat sink)]}
(13-7)
The solution to the governing first order differential Eq. (13-1) can
be written as follows, using the initial condition of Tchip = Tenvironment
and assuming temperature-independent thermophysical properties:
Tchip = Tenvironment + PRT [1 exp(/cpVRTotal)]
(13-8)
where cpVRTotal is the thermal time constant for the chip, which is the
product of its thermal capacitance and its total thermal resistance.
123
For the present sensitivity analysis, the nominal values of the above
independent variables are assumed to be as follows:
cpV = 2.52 W-s/C
(assuming a silicon dioxide chip in the shape of a rectangular box
with 0.03 m 0.03 m 0.002 m dimensions)
h = 50 W/m2-C
Tenvironment = 30C
Tchip = 90C
Achip for convection = 0.00114 m2 which can be detailed as [0.03 0.03
+ 2 (0.03 + 0.03) 0.002]
Ladhesive = 0.002 m
Aadhesive for conduction = Aheat sink for conduction = 0.0009 m2 which can be
detailed as (0.03 0.03)
Lheat sink = 0.005 m
N = 7 (seven rectangular fins which are nominally 0.002 m thick
and 0.03 m in length)
Aheat sink convection un-finned surfaces = 0.00108 m2 which can be detailed as
[with 0.002 m thick seven fins, namely 0.03 (0.03 7 0.002)
+ 2 (0.03 + 0.03) 0.005]
Afin = 0.00186 m2 which can be detailed as [for a 0.002 m thick, 0.03 m
long, and 0.03 m wide fin, namely 2 0.03 (0.03 + 0.5 0.002)]
kadhesive = 100 W/m-C (assuming silver epoxy adhesive)
kheat sink = 300 W/m-C
It is assumed that the heat sink and the fins are machined from the
same block material, and the minimum machinable spacing between
the fins is 0.002 m. For seven fins, the resulting fin thickness is 0.002 m.
The heat transfer efficiency for a rectangular fin can be calculated
from Eq. (5-6) given in Chapter 5, = tanh(mLc)/(mLc) where
m = [2h(w + t)/kwt ]1/2, Lc = L + 0.5t. For cases where the fin width,
w, is much greater than its thickness t, m becomes m = ( 2h /kt )1/2.
The rectangular fin efficiency for this nominal case is 0.95.
Using the above nominal values, and changing the fin length
and the number of fins on the heat sink, the chip heat dissipation
characteristics given in Figure 13-1 are obtained. As fin length
increases, so does chip power dissipation. The chip power dissipation
124
100
80
Lfin=0.01 m
Lfin=0.03 m
Lfin=0.05 m
Lfin=0.07 m
Lfin=0.09 m
Lfin=0.15 m
60
40
20
0
0
2
4
6
Number Of Rectangular Fins
Figure 13-1 Chip power dissipation versus number of fins for different fin
lengths
asymptotes around Lfin = 0.15 meters for an eight-fin heat sink. With
increasing fin length, the efficiency of the fin decreases, even if the
convection heat transfer area increases. Adding the maximum amount
of fins (eight in this case due to machinability constraints), with
lengths of up to 0.15 m per fin, can enhance the chip heat dissipation
by as much as 12-fold, as compared to a non-finned heat sink.
The sensitivity of chip heat dissipation to fin length is given
in Figure 13-2. The sensitivity approaches zero as the fin length
increases. By increasing the fin length, the chip power dissipation
improves less and less, and choosing the right fin length becomes a
cost-benefit issue. For example, using Figure 13-1, if the desired chip
power dissipation is 40 W, choosing a four-fin heat sink design, with a
fin length of 0.05 m, will suffice.
Chip power dissipation versus the convection heat transfer
coefficient is shown in Figure 13-3. Cases with seven fins and with
no fins are compared for a fixed fin length of 0.03 meters. Chip
power dissipation varies linearly with the convection heat transfer
coefficient, since conduction heat transfer resistances through the
125
1400
P/ Lfin, W/m
1200
1000
N=4
N=5
N=6
N=7
N=8
800
600
400
200
0
0.01
0.04
0.07
0.1
0.13
Figure 13-2 Chip heat dissipation sensitivity to fin length versus fin length
160
140
120
100
N=7
N=0
80
60
40
20
0
0
50
100
150
200
h, W/m2-C
Figure 13-3 Chip power dissipation versus convection heat transfer coefficient
with no fins and with seven fins for fin length = 0.03 m
126
40
30
20
10
0
0
50
100
150
Convection Heat transfer Coefficient, W/m2-C
200
Figure 13-4 Chip thermal time constant versus convection heat transfer
coefficient
adhesive and through the heat sink are negligible for this case
study. The slope of the chip power dissipation versus the convection
heat transfer coefficient line for a seven-fin design, (P/h), is
0.71 m2-C. The slope of the chip power dissipation versus the
convection heat transfer coefficient line for the finless design is
0.13 m2-C. It is apparent that the present nominal heat sink design,
with seven fins, enhances the chip power dissipation by over five-fold.
The thermal time constant for the chip is the product of thermal
capacitance and total thermal resistance, cpVRT, and it is given as a
function of the convection heat transfer coefficient in Figure 13-4.
The thermal time constant for the chip is a strong function of the
convection heat transfer coefficient at low values of forced convection
heat transfer regime; i.e., h < 50 W/m2-C.
In this finned heat sink design, the heat sink base and the fins
are assumed to be the same material. The thermal conductivity of
the heat sink and fin material also starts to affect the chip power
dissipation at lower values of kheat sink , i.e., kheat sink < 100 W/m-C. The
chip power dissipation versus heat sink and fin material thermal
conductivity is given in Figure 13-5 for a seven-fin design with a fin
127
Power Dissipation, W
45
40
35
30
25
20
0
100
200
300
Heat Sink & Fin Thermal Conductivity, W/m-C
400
Figure 13-5 Chip power dissipation versus heat sink and fin thermal
0.7
P/ ksink, m-C
0.6
0.5
0.4
0.3
0.2
0.1
0
0
100
200
300
Heat Sink & Fin Thermal Conductivity, W/m-C
400
Figure 13-6 Chip power dissipation sensitivity to heat sink and fin thermal
128
Variable
Tchip Tenvironment
Nominal
Value
Steady-State
Chip Power
Dissipation
Capability
Change Due To
A 10% Decrease
In Nominal Value
Steady-State
Chip Power
Dissipation
Capability
Change Due To
A 10% Increase
In Nominal Value
60C
10%
+10%
50 W/m-C
9.41%
+9.29%
kheat sink
300 W/m-C
0.574%
+0.476%
Lheat sink
0.005 m
0.289%
+0.287%
kadhesive
100 W/m-C
0.148%
+0.122%
Ladhesive
0.002 m
+0.134%
0.134%
129
CHAPTER
HEAT
14
TRANSFER
ANALYSIS FOR
COOKING IN A POT
F
131
The pot loses energy to food from its bottom and its side by natural
convection heat transfer. The natural convection heat transfer
mechanism between the pot and the food is analyzed up to a food
temperature of 105C at sea level atmospheric conditions; boiling heat
transfer regime is not considered. The pot also loses energy to the
environment from its bottom and its side by natural convection and
radiation.
The food receives energy from the bottom and the side of the
pot by natural convection. Temperature gradients in the food are
neglected, and the natural convection heat transfer coefficient
between the pot and the food is calculated using an average food
temperature. Conduction and radiation heat transfer mechanisms
between the pot and the food are neglected. Also, the natural
convection and radiation heat transfers from the top of the food to
the environment are neglected. Neglecting these secondary heat
transfer mechanisms introduces an initial error of about 7% in the net
heat transferred to the food, but this error diminishes fast to zero as
cooking time increases.
Food is assumed to have the same thermophysical properties as
water. Temperature-dependent variations for all of the thermophysical
properties are considered. The thermophysical properties are
calculated at a film temperature, which is the average of the surface
and the medium temperatures.
Energy balance for the food can be written as follows:
Change in internal energy of the food with respect to time =
Energy input by convection from the bottom of the pot to the
food + Energy input by convection from the sides of the pot to
the food
Energy balance for the pot can be written as follows:
Change in internal energy of the pot with respect to time =
Energy input from the heater to the bottom of the pot Energy
output by convection from the bottom of the pot to the food
Energy output by convection from the sides of the pot to
132
(14-1)
(cpV)pot(dTpot/d)
= Qin hpot bottom to food natural convection Apot bottom (Tpot Tfood)
hpot sides to food natural convection Apot sides (Tpot Tfood)
hpot bottom to environment natural convection Apot bottom (Tpot Tenvironment)
hpot sides to environment natural convection Apot sides (Tpot Tenvironment)
hpot to environment radiation (Apot bottom + Apot sides) (Tpot Tenvironment) (14-2)
All the natural convection heat transfer coefficients can be
obtained from empirical relationships in literature for the appropriate
geometries; for example, for natural convection heat transfer from
heated vertical plates and horizontal plates, see References [6] and
[10]. The natural convection heat transfer coefficient from the bottom
of the pot to the food is:
(hL/k) = 0.5 RaL0.25
(14-3)
(14-4)
(14-5)
133
(14-6)
134
air are obtained from References [6] and [10]. The radiation heat
transfer coefficient is obtained from the following equation:
hradiation = (T4pot T4environment)/(Tpot Tenvironment)
(14-7)
135
120
100
80
Tfood, C
Tpot, C
Experiment, C
60
40
20
0
Time, min
Figure 14-1 Food and pot modeling temperatures, and food experiment
1600
1200
800
hpot bottom to
food, W/m2-C
400
0
0
Time, min
Figure 14-2 Pot bottom to food and pot sides to food natural convection
136
Nominal
Value
Change in Time
for the Food in
the Pot to Reach
105C Due To
A 10% Decrease
In Nominal Value
Change in Time
for the Food in
the Pot to Reach
105C Due To
A 10% Increase
In Nominal Value
750 W
+11.16%
9.09%
Thermal
Capacity
of Food
Varies
with time,
W-hr/C
8.36%
+8.36%
Apot bottom
0.01267 m2
2.71%
+2.84%
Thermal
Capacity
of Pot
0.1286 W-hr/C
1.65%
+1.65%
0.02027 m2
0.84%
+0.84%
Heat Loss
from Pot to
Environment
Varies
with
time, W
0.30%
+0.30%
Varies with
time, W/m2-C
+0.17%
0.15%
Varies with
time, W/m2-C
+0.13%
0.12%
Variable
Qin
Apot sides
natural convection
137
138
CHAPTER
HEAT
TRANSFER AND
INSULATING
A WATER PIPE
15
139
(15-1)
(15-2)
(15-3)
140
141
or
kcopper 2 (Tr1 Tr2)/ln(r2/r1) kinsulation 2 (Tr2 Tr2+r)/ln(r2 + r/r2)
= {coppercp copper2[r2 0.25(r2 r1)] 0.5(r2 r1)
+ insulationcp insulation2[r2 + 0.25(r2 + r)] 0.5r} (Tr2+1 Tr2)/
(15-5)
where
kinsulation = thermal conductivity of insulation material in W/m-C
insulation = density of insulation material in kg/m3
cp insulation = specific heat of insulation material in J/kg-C
The outer node energy balance per meter of pipethe node at
the interface between the insulation layer outer surface and the
environment is:
Energy gained due to conduction into the radial element
Energy lost to environment due to convection out from the radial
element = Change in stored thermal energy in nodal volume
[This nodal volume starts at 0.5r from the outer radius of the
insulated pipe and goes out to the outer radius, r3.]
or
kinsulation 2 (Tr3-r Tr3)/ln(r3/r3 r) ho 2r3(Tr3 Tenvironment)
(15-6)
= insulationcp insulation2r3 0.5r (Tr3+1 Tr3)/
where
ho = Convection heat transfer coefficient between the outer surface
of insulation and the environment in W/m2-C
r3 = Outer radius of insulated pipe in meters
Nominal values of independent variables used in the present
sensitivity analysis are as follows:
r1 = 0.023 m
r2 = 0.025 m
r3 = 0.035 m
142
Twater = 12C
Tenvironment = 20C
water = 1000 kg/m3
cp water = 4200 J/kg-C
kcopper = 50 W/m-C (Note: water pipe is assumed to be copper alloy)
copper = 8800 kg/m3
cp copper = 400 J/kg-C
kinsulation = 0.04 W/m-C
insulation = 26 kg/m3
cp insulation = 835 J/kg-C
ho = 10 W/m2C
rcopper pipe = 0.001 m
rinsulation 0.001 m
= 0.2 s
The dependent variable in the calculations is the time it takes
the water in the pipe to reach 0C. After water reaches 0C at the
inner radius of the copper pipe, it will take extra time for it to freeze
completely. This freezing time can be calculated from the energy
required to compensate for the total latent heat of fusion, Hfg,
required for the water in the pipe:
Hfg water = 334800 W-s/kg
After the temperature of the inner node reaches 0C, the time
required for complete freezing is as follows:
for complete freezing
= waterr12 Hfg /[kcopper2(0 Tr2=time @ r1=0C)/ln(r2/r1)]
(15-7)
(15-8)
143
10
8
k insulation=0.02
W/m-C
k insulation=0.04
W/m-C
k insulation=0.06
W/m-C
6
4
2
0
0
10
20
30
Insulation Thickness, mm
40
Figure 15-1 Time for water to start freezing versus insulation thickness for
12
10
8
6
4
2
0
50
40
30
20
10
Environment Temperature, C
temperature
144
Nominal
Value
Change In Time
For Water To
Start Freezing
For A 10%
Decrease In
Nominal Value
Change In Time
For Water To
Start Freezing
For A 10%
Increase In
Nominal Value
water
1000 kg/m3
8.645%
+8.645%
cp water
4200 J/kg-C
8.645%
+8.645%
kinsulation
0.04 W/m-C
+8.282%
6.774%
Twater
12C
8.115%
+7.816%
Tenvironment
20C
7.363%
+8.669%
Insulation
thickness, r3 r2
0.01 m
5.766%
+5.640%
ho
10 W/m2-C
+2.829%
2.314%
copper
8800 kg/m3
1.315%
+1.315%
cp copper
400 J/kg-C
1.315%
+1.315%
Copper pipe
thickness, r2 r1
0.002 m
+0.243%
0.240%
insulation
26 kg/m3
0.041%
+0.041%
cp insulation
835 J/kg-C
0.041%
+0.041%
kcopper
50 W/m-C
+0.003%
0.003%
Variable
145
146
CHAPTER
QUENCHING
OF STEEL
BALLS IN AIR
FLOW
16
147
(16-1)
(16-2)
148
(16-4)
where
is emissivity of steel ball surface and
is Stefan-Boltzmann constant, namely 5.67 108 W/m2-K4.
During the calculations, instantaneous thermophysical properties
of air, kair and air, are calculated at mean temperatures of Tsphere and
Tair, namely at (Tsphere + Tair)/2.
Steel ball properties , cp, , and D are assumed to be constants.
The nominal values of governing independent variables for the
present sensitivity calculations are as follows:
D = 0.01 m,
= 7800 kg/m3
cp = 430 J/kg-C
ksteel = 40 W/m-C
= 0.2
Um = 5 m/s
149
Tair = 27C
Tsphere initial = 800C
dt = 1 s
The present case is analyzed below using different time intervals.
As the time interval is reduced to around one second, the resulting
temperatures-time proles are overlaid, and therefore dt = 1 second is
used for the results presented in the present sensitivity calculations. For
the nominal case, the Biot number is 0.022 for a hmax of 174 W/m2-C. Steel
ball temperature versus time is given in Figure 16-1. For the nominal
case, the steel ball reaches 36.8% of (Tsphere initial Tair), the thermal
response time for the quenching process in 57.0 s, and it reaches 50C
in 153.4 s.
Initially radiation heat transfer is 18% of the total heat transfer.
Radiation heat transfer effect diminishes fast and convection heat
transfer dominates the cooling process, as shown in Figure 16-2.
The effect of mean air velocity is analyzed without violating
the convection heat transfer coefcients range of application in
Eq. (16-3), 17 < ReD < 70000, and the Biot number being less than
0.1. The time for the steel ball to reach 50C and the time for it to
reach 36.8% of (Tsphere initial Tair) are shown as a function of mean
Temperature, C
800
600
400
200
0
50
100
Time, s
150
150
200
h, W/m2-C
200
150
h convection
h radiation
100
50
0
50
100
150
Time, s
200
250
Figure 16-2 Convection and radiation heat transfer coefcients versus time
air velocity in Figure 16-3. Changes in mean air velocity below 5 m/s
affect the quenching process time signicantly.
Figure 16-4 shows that for the present analyses, the internal
conduction resistance for the steel ball is much less than the
400
Time To
Reach
50C, s
Time, s
300
Time To
Reach
36.8% of
(Tsphere
initial-Tair), s
200
100
0
5
10
15
Mean Air Velocity, m/s
20
Figure 16-3 Time for the steel ball to reach 50C and to reach 36.8% of
151
Biot Number
0.06
0.04
0.02
10
15
20
Nominal
Value
t50C
Change For
A 10% Decrease
In Nominal
Value
t50C
Change For
A 10% Increase
In Nominal
Value
0.01 m
14.25%
+14.79%
kair
varies with
temperature
+11.07%
9.18%
7800 kg/m3
10.51%
+10.44%
cp
430 J/kg-C
10.51%
+10.44%
Um
5 m/s
+6.53%
5.66%
air
varies with
temperature
6.17%
+5.90%
773C
3.70%
+3.02%
0.2
+0.47%
0.47%
Variable
152
153
CHAPTER
QUENCHING 17
OF STEEL
BALLS IN OIL
n Chapter 16, the quenching medium analyzed was air. When water,
oil and other similar mediums are used for quenching, heat transfer
gets more complicated because the quenching medium goes into
different heat transfer regimes, such as lm boiling, transition
between lm boiling and nucleate boiling, nucleate boiling, and free
convection, as the surface temperature of the material being
quenched decreases.
Heat transfer in the boiling regime depends on material surface,
quenching medium and material surface combination, and on the
quenching medium's saturated liquid and vapor thermophysical
characteristics. The best way to achieve heat transfer coefcients
for boiling heat transfer in a quenching process is by experimental
measurements (see Reference by Lee, W. J., Kim, Y. and,
E. D. Case [12]).
In order to assume a uniform temperature in the steel ball, namely
Biot number < 0.1, (see Reference by J. P. Holman [5]), oil quenching
instead of water quenching is considered. Oil quenching heat transfer
coefcients are in the order of 1000 W/m2-C
155
156
htotal, W/m2-C
3000
nucleate
boiling
2000
free
convection
1000
C
A
transition
200
400
600
Steel Ball Temperature, C
film boiling
(vapor
blanket)
800
Figure 17-1 Total heat transfer coefcient versus steel ball temperature for
oil quenching
for both boiling and radiation heat transfer mechanisms. For oil
quenching, the total heat transfer coefcient as a function of steel ball
surface temperature is shown in Figure 17-1.
In temperatures above 750C, point C in Figure 17-1, the steel
ball encounters lm boiling heat transfer and is covered with a vapor
blanket. The temperatures between points B and C, 600C to 750C,
constitute the transition region from boiling heat transfer to lm
boiling heat transfer. The temperatures between points A and B, 200C
to 600C, are in the boiling heat transfer regime. Temperatures below
200C are considered to be in the free convection heat transfer regime.
Steel ball properties , cp, ksteel, and D are assumed to be constants
during the oil quenching process. The nominal values of governing
independent variables for the present sensitivity calculations are as
follows:
D = 0.012 m,
= 7800 kg/m3
cp = 430 J/kg-C
157
ksteel = 60 W/m-C
Toil = 70C
Tsphere initial = 900C
dt = 0.1 s
For the present calculations, a time interval of 0.2 seconds is used
because the thermal time constant for quenching in oil has dropped
an order of magnitude as compared to quenching in air (see Chapter
16). Calculations are also repeated with a 0.1 second time interval,
which improves the results by only one percent. The Biot number
for this case, hR/ksteel, is 0.1 for an average heat transfer coefcient
of 1000 W/m2-C, where R = D/2. Steel ball temperatures versus time
are shown in Figure 17-2 for nominal heat transfer coefcients, and
for a 10% variation about the nominal heat transfer coefcient
values given in Figure 17-1. Figure 17.2 shows the initial seconds
of the quenching process to emphasize the effects of heat transfer
coefcient variations. Boiling heat transfer phases occur during the
initial seconds of the oil quenching process. For the nominal case,
after 8.6 seconds, the natural convection heat transfer regime starts.
Temperature, C
800
Nominal h
(Figure 17-1)
600
10% higher h
than nominal
400
200
10% lower h
than nominal
10
Time, s
Figure 17-2 Steel ball temperature versus time for oil quenching for three
158
Variable
h
Variable
h
Nominal
Value
Change in
Time to Reach
100C For A
10% Decrease
In h Nominal
Value
Change in
Time to Reach
100C For A
10% Increase
In h Nominal
Value
Figure 17-1
+11.3%
9.6%
Nominal
Value
Change in
Time to Reach
36.8% of
(Tsphere initial Toil)
For A 10%
Decrease In
h Nominal
Value
Change in
Time to Reach
36.8% of
(Tsphere initial Toil)
For A 10%
Increase In
h Nominal
Value
Figure 17-1
+12.3%
9.6%
159
CHAPTER
COOKING
TIME FOR
TURKEY IN
AN OVEN
118
8
(18-1)
(18-2)
161
(18-3)
(18-4)
where
= (k/cp) is the thermal diffusivity of turkey in m2/s,
k is the thermal conductivity of turkey in W/m-C,
is the density of turkey in kg/m3,
cp is the specific heat of turkey at constant pressure in
J/kg-C, and
h is the convection heat transfer coefficient between the surface
of the turkey and the oven environment in W/m2-C.
The solution to Eqs. (18-1) through (18-4) can be found in
References [1], [6], [11], and [16], and it is as follows:
(Toven Tturkey @ r)/(Toven Tturkey initial) =
(18-5)
(18-6)
162
(18-8)
(18-9)
(18-10)
163
164
Dimensionless Temperature
At Center Of Turkey
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0.1
Bi=0.13
Bi=0.5
Bi=1
Bi=2
Bi=3
Bi=5
Bi=7
Bi=10
0.2
0.3
0.4
Fourier Number, Fo
0.5
Bi=16
165
Cooking Time, hr
15
14
13
12
11
10
9
8
7
6
5
4
3
2
USDA recommended
cooking times:
for a 4.5 kg (10 lb)
stuffed turkey 3-3.5 hrs,
for a 7.3 kg (16 lb)
0
5
10
15
20 stuffed turkey 4-4.5 hrs
for a 10 kg (22 lb)
Convection Heat Transfer Coefficient, W/m2-K
stuffed turkey 4.75-5.25
hrs
0
0.1
W=4.5 kg
(10 lb)
W=7.3 kg
(16 lb)
W=10 kg
(22 lb)
0.2
0.3
0.4
0.5
0
10
20
30
Coefficient,W/m2-C
166
0.8
1
1.2
1.4
1.6
1.8
180
190
200
210
Oven Temperature, C
220
167
Nominal
Value
Cooking
Time Change
Due To A 10%
Decrease In
Nominal
Value
Cooking
Time Change
Due To A 10%
Decrease In
Nominal
Value
3.33 kJ/kg-C
10%
+10%
73.9C
8.12%
+8.12%
186.1C
+8.10%
6.41%
Thermal
Conductivity, k
0.5 W/m-C
+7.10%
6.14%
7.3 kg
6.02%
+6.02%
Convection
Heat Transfer
Coefficient, h
10 W/m2-C
+5.13%
4.42%
Density,
1000 kg/m3
4.32%
+4.63%
Variable
Specific Heat @
Constant
Pressure, cp
Weight, W
168
CHAPTER
HEAT
GENERATED IN
PIPE FLOWS DUE
TO FRICTION
19
(19-1)
169
The pressure loss in the pipe is related to the heat generated in the
pipe by the rst law of thermodynamics:
Q = (VmA) P
(19-2)
where
P is the pressure drop in pipe in N/m2,
L is the length of pipe in m,
D is the internal diameter of pipe in m,
is the density of uid in kg/m3,
Vm is the average uid velocity in pipe in m/s,
Q is heat generated due to friction in W, and
A is cross sectional area of pipe in m2.
The rst example analyzes a high-viscosity engine oil, and ow
rates are in the laminar ow region; i.e., ReD < 2000. The friction
factor for laminar ow is given as follows, (see Reference [10]), and it
is independent of pipe surface roughness:
f = 64/ReD
(19-3)
Then the heat generated due to friction, after combining Eqs. (19-1),
(19-2) and (19-3), becomes:
Q = (128/) L (VmA)2/D4
(19-4)
170
0.003
0.002
0.001
0.000
15
25
35
45
Mean Oil Temperature, C
55
temperature
oil decreases as its temperature increases. Similar behavior is seen in
Figure 19-2 for the heat generated due to friction.
The mean temperature increase in oil behaves the same way as shown
in Figure 19-3. Tmean is calculated from Q/(VmAcp), assuming that all
Heat Generated, kW
150
100
50
15
25
35
45
55
171
Temperature Increase, C
5
4
3
2
1
0
15
25
35
Mean Oil Temperature, C
45
55
the heat generated due to uid friction goes to increase the temperature
of oil. cp is the specic heat of oil at constant pressure, in J/kg-C.
Sensitivities of heat, generated by oil owing in a pipe in the
laminar ow region, to the governing independent variables can be
obtained by differentiating Eq. (19-4):
Q/L = (128/) (VmA)2/D4
(19-5)
Q/ = (128/) L (VmA)2/D4
(19-6)
(19-7)
(19-8)
172
Q/ D, W/m
0.E+ 00
1.E+ 07
2.E+ 07
3.E+ 07
0.05
0.07
0.09
0.11
0.13
0.15
173
Q/ (VmA), W-s/m3
1.5E+07
1.0E+07
5.0E+06
0.0E+00
0.01
0.03
0.05
Flow Rate, m3/s
0.07
0.09
Figure 19-5 Sensitivity of heat generated due to friction to pipe ow rate for
Variable
D
VmA
L
@ 27C
174
Nominal
Value
Heat Generated
Due To Friction
Change For A 10%
Decrease In
Nominal Value
Heat Generated
Due To Friction
Change For A 10%
Increase In
Nominal Value
0.1 m
+52.42%
31.70%
1000 L/min
19%
+21%
400 m
10%
+10%
10%
+10%
0.486 N-s/m
(19-9)
1.50E 06
1.30E 06
1.10E 06
9.00E 07
7.00E 07
5.00E 07
15
25
35
45
55
temperature
175
Heat Generated, kW
2.6
2.4
2.2
15
25
35
45
55
Figure 19-7 Heat generated by friction for fully developed and turbulent
water ow in a pipe
Temperature Increase, C
0.038
0.036
0.034
0.032
0.030
0.028
15
25
35
45
55
176
Nominal
Value
Variable
D
0.1 m
+65.95%
36.75%
1000 L/min
25.60%
+30.67%
400 m
10%
+10%
997 kg/m
8.15%
+7.99%
2.02%
+1.86%
VmA
-3
160
120
80
40
0
200
400
600
800
Pipe Inner Surface Roughness, m
1000
177
roughness are analyzed for ReD = 2.47 10+5 using the Moody
Diagram (see Reference [10]). The results are given in Figure 19-9
for steady-state, fully developed, and turbulent water owing in a
pipe. The effect of pipe inner surface roughness to heat generated
due to friction increases linearly above e = 100 m. A pipe with an
inner surface roughness of 100 m generates 38% more heat than
a smooth one.
178
CHAPTER
20
ACTIVE SOLAR
SIZING AN
COLLECTOR
FOR A POOL
S
179
The present heat transfer model for the pool solar collector
system neglects temperature gradients in pool water. The model
is for unsteady-state heat transfer in one-dimensional rectangular
coordinates. There are 25 independent variables that govern the
dependent variable, size of the solar collector, and are considered
for sensitivity analysis.
The present heat transfer model has the following assumptions:
The pool is covered at night for 14 hours and open for swimming
during the day for ten hours. Energy lost from the water in the solar
collector at night and energy lost in piping between the pool and the
solar collector are considered to be lumped into the solar collector
efciency. Average heat transfer properties are used during both
day and night hours. When a pool cover is used at night, there is no
evaporation from the pool's surface.
The following energy balances can be constructed for the water in
the pool for the day-time and night-time hours.
Energy balance during the day for the pool:
Qin solar collector + Qin convection from environment during day Qout evaporation during day
+ Qin net radiation on pool surface Qout conduction to earth
(20-1)
= mw cpw (dTpool/d)
Energy balance during the night for the pool with a cover:
Qout convection to environment at night Qout conduction to earth
Qout radiation from pool cover surface = mw cpw (dTpool/d)
(20-2)
Energy balance during the night for the pool without a cover:
Qout convection to environment at night Qout conduction to earth
Qout radiation from pool surface Qout evaporation at night
= mw cpw (dTpool/d)
(20-3)
180
(20-4)
(20-5)
(20-6)
(20-7)
181
(20-8)
hdiffusion day is the mass transfer coefcient between pool water surface
and air in m/hr, R is the universal gas constant which is equal to
0.08205 m3-atm/kmol-K, pw is the saturation pressure of water vapor on
pool water surface at pool temperature, pa is partial pressure of water
vapor in air at average environmental temperature (calculated from
pa = psaturation, where is relative humidity in the air and psaturation is
the saturation pressure of water vapor in air at average environmental
temperature), L is a characteristic length of pool surface, namely pool
length, ka is thermal conductivity of air in W/m-K, a is density of air in
kg/m3, and cpa is specic heat of air at constant pressure in W-hr/kg-C.
The net solar radiation that is absorbed by water is:
Qin net radiation on pool surface = w G Apool water surface
(20-9)
(20-10)
182
Heat lost during the night from the pool with a cover to the
environment can be written as a series thermal circuit of convection
and conduction heat transfer:
Qout convection to environment at night
= Apool water surface (Tpool Tenvironment during night)/[(1/hnight )
+ (tcover /kcover)] (20-11)
where Tenvironment during night is average environmental temperature at
night in C, hnight is the average convection heat transfer coefcient at
night integrated over the length of the pool, given by the following
empirical relationship (see Reference [6]), in W/m2-K:
(hnight L/ka) = (0.037 ReL0.8 871) Pr0.333
for 0.6 < Pr < 60, 5 105 < ReL < 1 108, and Recritical = 5 105 (20-12)
where ReL = VL/ is Reynolds number, with V being average air
speed in m/hr over the length of the pool L in m and being
kinematic viscosity of air in m2/hr at lm temperature, the average
of water surface and environmental temperature. Pr = /t is the
Prandtl number, where t is thermal diffusivity of air in m2/hr
at lm temperature, the average of surface and environmental
temperature. Recritical is the transitional Reynolds number from
laminar to turbulent ow on a at plate. tcover is thickness of pool
cover in meters and kcover is thermal conductivity of pool cover at
pool temperature.
Heat radiated from the pool cover surface to the night sky is:
Qout radiation from pool cover surface = pool cover surface T4pool cover surface
(20-13)
where pool cover surface is emissivity of pool cover surface and the is
Stefan-Boltzmann constant, namely 5.67 108 W/m2-K4.
Energy balance during the night for the pool without
a cover has convection and radiation heat transfers dened
as follows:
183
(20-15)
184
185
Collector Area, m2
8000
6000
4000
2000
0
0.2
0.4
0.6
Collector Efficiency
0.8
Figure 20-1 Solar collector area versus solar collector efciency at average
186
(Collector Area) / , m2
10000
20000
30000
0.2
0.4
0.6
0.8
Collector Efficiency,
Collector Area, m2
3000
2000
1000
15
17
19
21
23
Figure 20-3 Solar collector area versus pool water initial temperature
187
Collector Area, m2
1250
1220
1190
1160
25
27
29
31
33
35
Average Environment Temperature During The Day, C
Collector Area, m2
2000
1500
1000
500
10
15
20
25
Figure 20-5 Solar collector area versus average earth temperature under
the pool
188
Collector Area, m2
1500
1300
1100
900
700
30
35
40
45
50
Pool Length, m
55
60
Collector Area, m2
2000
1600
1200
800
400
10
15
20
25
30
35
40
Pool Width, m
189
Collector Area, m2
2000
1600
1200
800
400
0
1.2
1.4
1.6
1.8
2.2
2.4
2.6
190
Collector Area, m2
2000
1500
1000
500
200
250
300
350
400
450
500
Figure 20-9 Solar collector area versus average solar insolation on pool
surface
Collector Area, m2
1800
1600
1400
1200
1000
0.5
0.6
0.7
0.8
0.9
Absorptivity Of Pool Water Surface
Figure 20-10 Solar collector area versus absorptivity of pool water surface
191
Collector Area, m2
2800
Without
Cover At
Night
2300
1800
With Cover
At Night
1300
800
0.2
0.4
0.6
0.8
Relative Humidity
higher sensitivity comes from higher evaporation heat loss to the cold
night atmosphere.
If a region with a relative humidity of 10% is considered to build a
solar heated pool, as compared to a region with a relative humidity
of 60%, with everything else being the same and considering a pool
cover at night, the dry region requires 240 m2 more solar collector
area because more heat is lost due to evaporation to the dry
atmosphere.
There are over 25 independent variables that govern this heat
transfer problem. When the nominal values of the variables given
above are varied 10%, the results shown in Table 20-1 are obtained
for a pool that is covered at night. Required solar collector area
sensitivities to a 10% change in the governing variables are given in
descending order of importance, and they are applicable only around
the nominal values assumed for this study.
Required solar collector area is most sensitive to changes in initial
pool water temperature, followed by pool water height and pool water
specic heat at constant pressure. The next set of variables in order
of sensitivity are solar collector efciency, average solar insolation on
collector, average solar insolation on pool water surface, average solar
192
Variable
Nominal
Value
Change In
Required Solar
Collector Area
For A 10%
Decrease In
Nominal Value
Tpool initial
20C
+64.35%
64.35%
H, pool water
height
2m
17.53%
+17.62%
1.162
W-hr/kg-K
17.37%
+17.45%
, solar collector
efciency
0.7
+11.16%
9.06%
q, average solar
insolation on
collector
500 W/m2
+11.16%
9.06%
G, average solar
insolation on
pool water surface
350 W/m2
+10.07%
10.07%
cpw
Change In
Required Solar
Collector Area
For A 10%
Increase In
Nominal Value
w, average
solar radiation
absorptivity at
water surface
0.96
+10.07%
4.19% with
only a possible
4.2% increase
to 1
L, pool length
50 m
9.90%
+9.98%
W, pool width
25 m
9.82%
+9.90%
Tearth
20C
+8.89%
8.81%
30C
+1.85%
1.76%
, relative humidity
0.40
+1.60%
1.59%
tpool wall
0.1 m
+1.26%
1.01%
kpool wall
0.7 W/m-K
1.09%
+1.17%
193
194
CHAPTER
HEAT
TRANSFER
IN A HEAT
EXCHANGER
21
195
(21-1)
(21-2)
Eqs. (21-1) and (21-2) can be combined to eliminate dQ, and then
integrated between the inlet and outlet temperatures of the heat
exchanger to give the following log mean temperature difference
design method equations for a counterow heat exchanger:
Q=
(21-3)
(21-4)
(21-5)
where
UAwall = UhAwall hot uid side = UcAwall cold uid side
(21-6)
and Awall hot uid side represents inside surface area of the inner tube and
Awall cold uid side represents outside surface area of the inner tube. The
overall heat transfer coefcient from the hot uid side is:
Uh = 1/[(1/hh) + Rh foul + (Dh ln(Dc/Dh)/2kss)
+ (Awall hot uid side Rc foul/Awall cold uid side)
+ (Awall hot uid side/Awall cold uid sidehc)]
196
(21-7)
All the variables in these governing Eqs., (21-1) through (21-7), are
dened as follows:
h = Hot oil density in kg/m3
VmhAh = Hot oil ow rate in liters/minute
cph = Hot oil specic heat at constant pressure in J-kg/C
Th in = Hot oil inlet temperature to the heat exchanger in C
Th out = Hot oil outlet temperature from the heat exchanger in C
c = Cold water density in kg/m3
VmcAc = Cold water ow rate in liters/minute
cpc = Cold water specic heat at constant pressure in J-kg/C
Tc in = Cold water inlet temperature to the heat exchanger in C
Tc out = Cold water outlet temperature from the heat exchanger in C
Uh = Overall heat transfer coefcient based on the inside surface of
the inner tube in W/m2-C
Awall hot uid side = Inside surface area of the inner tube, DhL, in m2
Uc = Overall heat transfer coefcient based on the outside surface
of the inner tube in W/m2-C
Awall cold uid side = Outside surface area of the inner tube, (Dh + 2t)L =
DcL, in m2
Dh = Inside diameter of the inner tube in m
Dc = Outside diameter of the inner tube, Dh + 2t, in m, and where t
is the thickness of the inner tube
kss = Wall tube material thermal conductivity in W/m-C
hh = Convection heat transfer coefcient between the hot oil and
the inner tube inside surface in W/m2-C
hc = Convection heat transfer coefcient between the cold water
and the inner tube outside surface in W/m2-C
Rh foul = Fouling resistance for the inside surface of the inner tube in
m2-C/W
Rc foul = Fouling resistance for the outside surface of the inner tube
in m2-C/W
In a heat exchanger, fouling resistances affecting heat transfer
on surfaces of walls are caused by corrosion or by foreign material
deposited over time, and they are determined by experimental
methods during heat exchanger life tests.
197
198
(21-8)
(21-9)
300
200
100
10
15
199
160
120
80
40
0
30
35
40
45
50
Figure 21-2 Heat exchanger length versus hot oil outlet temperature for
300
200
100
20
30
10
Cold Water Inlet Temperature, C
40
Figure 21-3 Heat exchanger length versus cold water inlet temperature for
200
Variable
Th out
Nominal
Value
Heat
Exchanger
Length
Change For A
10% Decrease
In Nominal
Value
Heat
Exchanger
Length
Change For A
10% Increase
In Nominal
Value
40C
+24.98%
20.27%
70C
16.78%
+14.65%
0.14 W/m-C
+10.74%
8.79%
5 liters/minute
10.27%
+10.33%
870 kg/m
10.27%
+10.33%
cph
2000 J-kg/C
10.27%
+10.33%
Tc in
20C
6.25%
+7.21%
Dh
0.02 m
+0.448%
0.377%
25 liters/minute
+0.347%
0.282%
1000 kg/m
+0.347%
0.282%
cpc
41.8 J-kg/C
+0.341%
0.277%
0.0008 m -C/W
0.235%
+0.236%
0.025 m
0.116%
+0.126%
0.0002 m -C/W
0.049%
+0.049%
15 W/m-C
+0.040%
0.033%
0.001 m
+0.013%
0.014%
kc
0.6 W/m-C
+0.0090%
0.0077%
0.00096 N-s/m
0.0057%
+0.0054%
0.1 N-s/m2
0%
0%
Th in
kh
VmhAh
VmcAc
Rh foul
Douter tube
Rc foul
kss
201
202
CHAPTER
I CE
FORMATION
ON A LAKE
22
203
(22-3)
(22-4)
204
(22-6)
(22-7)
(22-8)
(22-9)
205
2000
1500
1000
500
0
0.1
0.2
0.3
0.4
Thickness Of Ice, m
0.5
0.6
the growing ice surface. Then there is no more energy left to support
the latent heat of fusion of water to keep forming the ice layer.
Under this condition, the natural logarithm term in the denominator
of the ice formation rate Eq. (22-9) approaches zero, or:
0.6
0.4
0.2
10
20
30
40
50
206
x* = (1/Z) 1
(22-10)
300
250
200
150
100
50
0
50
40
30
20
10
Temperature Of Atmosphere, C
atmosphere
207
80
60
40
20
0
6
8
10
Temperature Of Lake Water, C
12
14
lake water
208
0
5
10
15
20
25
100
20
40
60
80
hice upper surfaceatmosphere, W/m2-C
Figure 22-5 Sensitivity of time to form 0.2 m thick of ice to the convection
30
20
10
10
15
20
hlake water-growing ice surface, W/m2-C
25
Figure 22-6 Sensitivity of time to form 0.2 m thick of ice to the convection
209
0
50
100
150
200
250
1
1.5
2.5
conductivity of ice
Variable
Nominal
Value
Tatmosphere
20C
Change in
Change in
Time For Ice
Time For Ice
Thickness to
Thickness to
Reach 0.2 m For Reach 0.2 m For
A 10% Decrease A 10% Increase
In Nominal
In Nominal
Value
Value
12.67%
+17.05%
kice
1.88 W/m-C
+10.87%
8.28%
ice
920 kg/m
10%
+10%
Hice
93 W-hr/kg
10%
+10%
30 W/m -C
+5.80%
4.61%
10 W/m -C
4.31%
+4.78%
6C
4.31%
+4.78%
Tlake water
210
2
2
211
CHAPTER
SOLIDIFICATION 23
IN A CASTING
MOLD
xact temperature distribution solutions to unsteady-state onedimensional heat conduction equations in rectangular, cylindrical,
and spherical coordinates, for initial condition and different boundary
conditions, have been provided in literature (see References [1], [6],
[10] and [16]). The reference by Carslaw, H. S. and J. C. Jaeger [1]
extended these exact solutions to moving boundary conditions with
phase change.
In this chapter, sensitivity analysis for the solidication front
of a semi-innite liquid in a semi-innite mold is considered. The
unsteady-state one-dimensional heat conduction equation in
rectangular coordinates for the mold region, solidied cast material
region and liquid cast material region, with constant property
assumptions, is given by:
2T/x2 = (1/) T/
(23-1)
213
(23-2)
(23-3)
The interface between the mold region and the solidied material
at x = 0 has the following energy balance boundary condition:
ksand Tsand/x = ksolid Tsolid/x
(23-4)
(23-5)
214
(23-6)
(23-7)
The temperature of the liquid cast material region far away from
x = 0 always stays at the initial temperature of the liquid cast
material, Tliquid:
T(+, ) = Tliquid
(23-8)
The interface between the liquid cast material region and the
solidied cast material region at x=L has the following boundary
condition, which is the result of an energy balance at the solidication
front that supports the latent heat of fusion for the liquid cast
material:
kliquid Tliquid/x ksolid Tsolid/(x = solid HdL/d)
(23-9)
(23-10)
215
(23-11)
(23-12)
(23-13)
Eq. (23-11) is solved for a by trial and error. For the present
analysis, liquid silver is considered as the cast material, and sand is
considered as the mold material. For solidication to occur there are
two conditions that have to be met. One condition is that the liquid
cast materials temperature cannot exc eed the melting temperature
of the mold material:
Tliquid < Tsand melting point
(23-14)
The second condition is that the liquid cast material has to have
enough heat loss initially through the mold by conduction heat
transfer to be able to start solidifying. Eq. (23-11) provides this
condition by setting a equal to zero:
(Tliquid Tsolidication)/(Tsolidication Tsand)
< [(sandcp sandksand)/(liquidcp liquidkliquid)]0.5 (23-15)
For the present sensitivity analysis, the following nominal values
are assumed for the independent variables that also satisfy Eqs.
(23-14) and (23-15):
Sand values for the mold region:
216
Tsand = 100C
Tsand melting point = 1430C
sand = 2330 kg/m3
cp sand = 712 J/kg-C
ksand = 42 W/m-C
Solidied silver values for the solidied cast region:
Tsolidication = 960C
solid = 10500 kg/m3
cp solid = 235 J/kg-C
ksolid = 400 W/m-C
Solidification Front L, m
0.4
0.3
0.2
0.1
0
10
15
Time, min
20
25
30
217
Nominal
Value
Change in a
For A 10%
Decrease In
Nominal Value
Change in a
For A 10%
Increase In
Nominal Value
Tliquid
1000 C
20.738% for
4% change
+69.208%
Tsolidication
960 C
+58.119%
17.677% for
4.2% change
104400W-s/kg
+9.745%
8.206%
solid
10500 kg/m
+9.641%
8.062%
sand
2330 kg/m
3.192%
+2.933%
cp sand
712 J/kg-C
3.192%
+2.933%
ksand
42 W/m-C
3.192%
+2.933%
cp liquid
318 J/kg-C
1.520%
+1.517%
ksolid
400 W/m-C
1.324%
+1.143%
Variable
Tsand
100 C
+0.773%
0.777%
liquid
9300 kg/m3
0.717%
+0.686%
kliquid
360 W/m-C
0.717%
+0.686%
cp solid
235 J/kg-C
+0.673%
0.652%
218
219
CHAPTER
AVERAGE
TEMPERATURE
RISE IN SLIDING
SURFACES IN
CONTACT
24
(24-1)
(24-2)
221
(24-4)
Eq. (24-3) assumes that all the heat generated during the contact
goes to the semi-innite medium under the sliding contact area, but
in reality this is not the case. A portion, mQ, of the heat generated
during the sliding contact goes into the semi-innite body and the
rest, (1-m)Q, goes into the square sliding contact area. Equating the
average temperatures in Eqs. (24-3) and (24-4), the proportionality
constant m can be determined:
m = k1 (LV)0.5/[1.125 k2 10.5 + k1 (LV)0.5]
(24-5)
222
(24-6)
The heat generated during the contact comes from the mechanical
energy that is dissipated during the contact:
Q = fd W g V/(4 L2 J)
(24-7)
(24-8)
where
1 = thermal diffusivity of semi-innite medium in m2/s,
k1 = thermal conductivity of semi-innite medium in W/m-C,
k2 = thermal conductivity of the square sliding contact area in W/m-C.
The following nominal values for the independent variables are
assumed for the present sensitivity analysis. Also both the
semi-innite medium and the square sliding contact area are assumed
to be stainless steel.
fd = 0.5,
W = 1000 kg,
V = 10 m/s,
1 = 4 106 m2/s,
k1 = 15 W/m-C,
k2 = 15 W/m-C,
L = 0.01 m
223
1000
800
600
400
200
0
10
15
20
25
30
velocity of bodies
224
Taverage / V, C-s/m
400
300
200
100
0
10
15
20
25
30
velocity of bodies
2000
1600
1200
800
400
0
0.E+00
1.E05
2.E05
3.E05
4.E05
5.E05
225
Taverage / 1, C-s/m2
1.2E+08
1.0E+08
8.0E+07
6.0E+07
4.0E+07
2.0E+07
0.E+00
1.E05
2.E05
3.E05
4.E05
5.E05
1600
1200
800
400
0
50
100
150
200
250
300
226
10
30
70
110
150
50
100
150
200
250
Thermal Conductivity, k1, W/m-C
300
227
550
540
530
520
100
0
20
40
60
80
Thermal Conductivity Of Sliding Square Contacting body, k2,
W/m-C
0.2
0.25
0.3
20
40
60
80
100
228
3000
2500
2000
1500
1000
500
0
0.02
0.04
0.06
0.08
0.1
Figure 24-9 Sliding contact area average temperature versus half width, L,
Taverage / L, C/m
0
50000
100000
150000
200000
250000
300000
0.02
0.04
0.06
0.08
0.1
229
Nominal
Value
Change In
Sliding Contact
Area Average
Temperature
For A 10%
Decrease
In Nominal
Value
Change In
Sliding Contact
Area Average
Temperature
For A 10%
Increase
In Nominal
Value
0.01 m
+17.08%
13.30%
15 W/m-C
+11.02%
9.03%
10%
+10%
W, load at the
sliding contact
area
1000 kg
10%
+10%
V, relative velocity
of sliding square
contact area and
the semi-innite
body
10 m/s
5.17%
+4.92%
1, thermal
diffusivity of
semi-innite body
4 106 m2/s
5.10%
+4.84%
15 W/m-C
+0.07%
0.07%
Variable
L, half width of
sliding square
contact area
k1, thermal
conductivity of
semi-innite body
fd, dynamic
coefcient of
friction0.5
k2, thermal
conductivity of
sliding square
contact area
230
231
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
233
[13]
[14]
[15]
[16]
[17]
234