이 윤 우
서울대학교 화학생물공학부
Heat and Mass Transfer
24
SOME DESIGN EQUATIONS FOR
CONVECTION HEAT TRANSFER
How to obtain equations
for predicting heat-transfer coefficients
1. Combination of momentum, energy, and continuity equations for laminar flow
2. von Karman integral method for turbulent flow
3. The analogy between heat and momentum transfer for turbulent flow
4. The dimensional analysis
(22-19) 3
/ 1 2
/
1 Pr
) (Re 332
.
0 x
Nu x =
3 / 1 2
/
1 Pr
) (Re 664
.
0 L
Nu m =
( )
( ) 4 / 5 1 / 3
3 / 5 1
/ 4
Pr Re
0365 .
0
Pr Re
0292 .
0
L m
x x
Nu Nu
=
=
Flow parallel to a flat plate
3 / 1 8
.
0 Pr Re
023 .
= 0 Nu
Laminar flow ( Re < 5 x 10
5)
Turbulent flow ( Re < 5 x 10
5)
Flow in a pipe
Turbulent flow ( Re>2100 )
14 . 3 0
/ 1 3
/ 1 3
/
1 Pr
Re 86 .
1 ⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎟ ⎛
⎠
⎜ ⎞
⎝
= ⎛
L s
Nu D
μ
Laminar flow μ
( Re<2100 )
(22-42)
(23-32)
(23-50) (23-51)
Heat transfer coefficients
Dimensional Analysis
- can be used for both type flow (laminar/turbulent)
) ,
, Pr,
, (
) ,
, Pr,
(Re,
L L
s s
L Ra D
f Nu
L f D
Nu
μ μ μ
μ
=
= Turbulent flow:
Laminar flow:
Example 24
Example 24 - - 1 Dimensional Analysis 1 Dimensional Analysis
u b L
D k, h m , C p , μ , g c , J
ρ , β , g, Δ t
Heat-transfer coefficient in natural convection in a heated pipe
6 Fundamental dimensions: Length [L], mass [M], time [θ]
temperature [T], heat [H], force [F]
- Heat & force can be expressed in terms of the other four dimensions
- Include gravitational constant, g & mechanical equivalent of heat, J
u
bL
D k, h
m, C
p, μ , g
c, J
ρ , b, g, Δ t
Variable Symbol Dimension
Length of heated section L [L]
Fluid density ρ [M/L
3]
Fluid viscosity μ [M/L θ ]
Fluid thermal conductivity k [H/L θ T]
Dimensional constant g
c[ML/F θ
2]
J [FL/H]
Mean heat-transfer coefficient h
m[H/ θ TL
2] Temperature difference, t
s-t
0Δ t [T]
Coefficient of thermal expansion β [T
-1]
Specific heat of fluid C
p[H/MT]
Gravitational acceleration g [L/ θ
2]
Bulk velocity of fluid u
b[L/ θ ]
Diameter of pipe D [L]
No.
1 2 3 4 5 6 7 8 9 10 11 12 13
Example 24
Example 24 - - 1 Dimensional Analysis 1 Dimensional Analysis
6 Fundamental dimensions:
Length [L], mass [M], time [θ], temperature [T], heat [H], force [F]
u
bL
D
k, h
m, C
p, μ , g
c, J ρ , b, g, Δ t
Example 24
Example 24 -1 Dimensional Analysis - 1 Dimensional Analysis
The total number of variables is 13, and we have chosen to express these in terms of six dimensions. As is often the case, the maximum number of variables which will not form a dimensionless group is equal to the number of fundamental dimensions.
According to Buckingham’s theorem, 13 variables, 6 dimensions
Number of dimensionless groups: (13 – 6) = 7 7
7 dimensionleess 7 dimensionleess group group =π =π 1 1 , , π π 2 2 , , π π 3 3 , , π π 4 4 , , π π 5 5 , , π π 6 6 , , π π 7 7
Example 24
Example 24 -1 Dimensional Analysis - 1 Dimensional Analysis
According to Buckingham’s theorem,
13 variables, 6 dimensions → Number of dimensionless groups: (13 – 6) = 7 7 7 7 dimensionleess dimensionleess group = π group = π
11, π , π
22, π , π
33, π , π
44, π , π
55, π , π
66, π , π
77( )
g f e c d c b a
g b f e c d c b a
g f e c d c b a
g p f e c d c b a
g f e c d c b a
f g e c d c b a
g m f e c d c b a
D J g k L
u J g k L
g J g k L
C J g k L
J g k L
t J
g k L
h J g k L
μ ρ π
μ ρ π
μ ρ π
μ ρ π
β μ
ρ π
μ ρ π
μ ρ π
=
=
=
=
=
Δ
=
=
7 6 5 4 3 2 1
The variables which we choose to be common to all groups are first six in the table above.
Each of the remaining seven variables
will, in turn, be added to the first six, to
give seven groups.
Example 24
Example 24 - - 1 Dimensional Analysis: the 1 1 Dimensional Analysis: the 1 st st group group
( ) (
b) (
c) (
d) (
e) (
f)
ga
g m f e c d c b a
L T H
FLH MLF
T HL
ML ML
L
h J g k L
2 1 1 1
2 1 1
1 1 1
1 3
1
−
−
−
−
−
−
−
−
−
−
−
=
−=
θ θ
θ θ
μ ρ π
L : a-3b-c-d+e+f-2g=0 M : b+c+e=0
θ : -c-d-2e-g=0 T : -d-g=0
H : d-f+g=0 F : -e+f=0
∴ a=g, b=c=e=f=0, d=-g
k L h k
L
h m g m
⎟ =
⎠
⎜ ⎞
⎝
= ⎛ π 1
π π 1 1
Nusselt number
Example 24
Example 24 - - 1 Dimensional Analysis: the 2 1 Dimensional Analysis: the 2 nd nd group group
( ) (
b) (
c) (
d) (
e)
f( )
ga
g f
e c d c b a
T FLH
MLF T
HL ML
ML L
t J
g k L
1 2
1 1
1 1 1
1 3
2
( )
−
−
−
−
−
−
−
−
=
−Δ
=
θ θ
θ μ
ρ π
L : a-3b-c-d+e+f=0 M : b+c+e=0
θ : -c-d-2e=0 T : -d+g=0 H : d-f=0 F : -e+f=0
∴ a=b=2g, c=-3g, d=e=f=g=1
μ
π
2= L
2ρ
2kg
cJ Δ t
π π 2 2
Example 24
Example 24 - - 1 Dimensional Analysis 1 Dimensional Analysis
D L D
J g k L
u u L
J g k L
g g L
J g k L
k C C
J g k L
J kg J L
g k L
t J kg t L
J g k L
f e c d c b a
b b
f e c d c b a
f e c d c b a
p p
f e c d c b a
c f
e c d c b a
f c e c d c b a
=
=
=
=
=
=
=
=
=
=
= Δ Δ
=
7 7
6 6
2 2 3 5
5
4 4
2 2
3 3
3
3 2 2 2
2
π μ
ρ π
μ π ρ
μ ρ π
μ π ρ
μ ρ π
π μ μ
ρ π
ρ β π μ
β μ
ρ π
μ π ρ
μ ρ π
(24-1)
How to make seven groups
Grashof number
Prandtl number
Reynolds number
Example 24
Example 24 - - 1 Dimensional Analysis 1 Dimensional Analysis
η ζ
ε δ γ
β π π π π π
απ π
π π
π π
π π
π
7 6 5 4 3 2 1
7 6
5 4
3 2
1 ( , , , , , )
=
→
= function
(24-1)
The seven groups obtained in Example 24-1 can be used to correlate the results of experiments in natural convection. One common way of doing this is to write an equation of the form
determined be
to are η
and ζ
ε, δ, γ, β, al,
experiment
By
Example 24
Example 24 - - 1 Dimensional Analysis 1 Dimensional Analysis
The constants represented by the Greek letters can be determined by a statistical analysis of the experimental results, often the least-squares method.
Groups which have no effect on h
mcan easily be located because the exponents on these groups will be very small.
Two groups may have the same exponent, indicating that the variables they contain may be combined in a single group. For example, in Ex 24-1, it is found that the combined groups, π
2, π
3, π
5, give a single dimensional group, L
3ρ
2gβΔt/μ
2, known as the Grashof number, which together with π
1, the Nusselt number, and π
4, the Prandtl number.
The group π
6and π
7are meaningless in this analysis because u
b=0
and D=∞.
Dimensional Analysis Dimensional Analysis
The correlating equation is
The Prandtl number is approximately constant at unity for most gases, so the absence of that group would be expected.
The Grashof number is in Eq. (22-25) differs from the product π
2π
3π
5only in that for an ideal gas β equals 1/T, where T is the absolute temperature.
25 . 0 2
2 3
4 5
3 2 1
478 .
0
) (
) (
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎛ Δ
=
=
T t g L
k L h m
μ ρ
π π
π π α
π β δ
(22-25)
(24-2)
Dimensional Analysis Dimensional Analysis
( )
γ β
η ζ β δ
α α
π π π π
π π α μ π ρ
π
μ β π ρ
π π
π μ π
Pr Re
, convection forced
In
478 .
0 ,
convection natural
In
Re Pr
Gr
Number Reynolds
:
Number Granshof
: number Prandtl
:
number Nusselt
:
25 . 0 7
6 4 5
3 2 7
6
2 2 3 5
3 2
4 1
=
=
=
=
=
×
= Δ
×
×
=
=
m c
b a
m b p
m
Nu
Gr Nu
Nu
u D
t g L
k C
k
Lh
Dimensional Analysis Dimensional Analysis
pipe circular
in metal liquid
k for μ C μ
ρ 0.025 Du
7.0 Nu
sphere k for
μ C μ
ρ 0.6 Du
2.0 Nu
: ts coefficien -
fer heat trans In
cylinder horizontal
from convection
natural for
Pr 0.53Gr
Nu
pipe circular
in flow nt
for tubule Pr
0.023Re Nu
: equations Most
0.8 p
0.8 b
1/3 p
1/2 0
3 1 2
1 1/4
1/4
1/3 0.8
3 2 1
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝ + ⎛
=
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝ + ⎛
=
+
=
=
=
=
δ γ γ
β
π βπ α
π π απ
π L
Predicting convective heat
Predicting convective heat - - transfer coefficients transfer coefficients Summary of Equations used for
Summary of Equations used for Predicting convective heat
Predicting convective heat - - transfer coefficients transfer coefficients
Equations Equations
Fundamental Fundamental
derivations derivations
Empirical correlations
Empirical correlations
of dimensionless groups
of dimensionless groups
Predicting Convective Heat-transfer Coefficients: Nu
xx h
xHeat Exchanger Design Heat Exchanger Design
- Fundamental derivations - Empirical correlations
of dimensionless groups
Calculation of overall heat
Calculation of overall heat- -transfer Coefficients transfer Coefficients
Calculation of heat transfer area Calculation of heat transfer area
γ β
π απ π
1=
2 325 .
478
0.
0 Gr
Nu
m= Nu = 0 . 023 Re
0.8Pr
1/3( ) Re
1/2Pr
1/3332
.
0
xNu
x=
( )
1
0 0
2 /
1
( )
2 exp Pr
Re
∞ −
⎟ ⎟
⎠
⎞
⎜ ⎜
⎝
⎛
⎟ ⎟
⎠
⎞
⎜ ⎜
⎝
⎛ −
= ∫
η∫ f η d η d η
Nu
x xo lm c
o c
c lm
b o b
b i
di o i
i o o
h A
A k
r A
A k
r A
h A A
h U A
1 1
, ,
Δ + Δ +
+ +
=
DL A A
U A q t
A U
q= Δ lm = =
π
0 0
Shell and Tube Heat Exchanger
1/3 0.8
Pr D Re
0.023 k
h ⎟
⎠
⎜ ⎞
⎝
= ⎛
i
556 . 0
482
max. 0
h ⎟⎟
⎠
⎜⎜ ⎞
⎝
⎟ ⎛
⎠
⎜ ⎞
⎝
= ⎛
μ ρ u D D
k
o
o lm
c o c
c lm
b o b
b i
di o i
i o o
h A
A k
r A
A k
r A
h A A
h U A
1 1
, ,
Δ + Δ +
+ +
=
DL A A
U A q
t A U
q = 0 Δ lm = = π
Summary of Equations used for Summary of Equations used for Predicting convective heat
Predicting convective heat - - transfer coefficients transfer coefficients
☺ Convection in circular pipe (Laminar/turbulent)
☺ Convection in non-circular conduits
☺ Convection normal to a cylinder
☺ Convection normal to a bank of circular tubes
☺ Convection from spheres
☺ Convective heat transfer between a fluid and a packed bed
☺ Convection from a plane surface
☺ Heat transfer to liquid metals
☺ Heat transfer to non-newtonian fluids
☺ Effect of surface roughness on heat transfer coefficient
Convection in circular pipe (Laminar flow) Convection in circular pipe (Laminar flow)
Fig. 22-3: convection in circular pipe Fig. 22-4: convection in circular pipe with uniform wall temperature
Parabolic flow, uniform heat flux: overall temperature difference is constant
3
4 1
62 1 10
/ p
m πkL
. wC Nu
,
Gz ⎟⎟⎠
⎜⎜ ⎞
⎝
= ⎛
>
6 3 5 , Nu ~ .
Gz <
m Low flow rateLong pipe
Convection in circular pipe (Laminar flow) Convection in circular pipe (Laminar flow)
For the case of constant wall temperature: condenser
L Pe D L
Gz D
vection fer by con
heat trans
duction fer by con
heat trans kx
Gz wC Number
Graetz x p
⋅
=
⋅
⋅
=
=
=
= Pr
Re
3
4 1
62 1
10
6 3 5
/ p
m m
πkL . wC
Nu ,
Gz
.
~ Nu
, Gz
⎟⎟ ⎠
⎜⎜ ⎞
⎝
= ⎛
>
< (a low flow rate or a long pipe)
t
s=constant
Convection in circular pipe (Laminar flow) Convection in circular pipe (Laminar flow)
For the case of constant wall temperature : condenser
14 . 3 0
/ 1 3
/ 1 3
/ 1
3 1
Pr Re
86 . 1
62 4 1
10
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎟ ⎛
⎠
⎜ ⎞
⎝
= ⎛
⎟⎟ ⎠
⎜⎜ ⎞
⎝
= ⎛
>
s m
/ p
m
L Nu D
πkL . wC
Nu ,
Gz
μ μ
Account for the effect of temperature
on viscosity Sieder and Tate
(22-40)
(22-42) 4
D
2u w = ρ π
Wall temperature
Arithmetic average temperature
Gz>10
Convection in vertical tube (Laminar flow) Convection in vertical tube (Laminar flow)
Natural-convection flow effects are usually significant only when the forced convection flow is laminar and the temperature-difference driving force is large.
OMC(Opposing Mixed Convection) Lower heat-transfer coefficients AMC(Assisting Mixed Convection)
Higher heat-transfer coefficients
Forced convection
Natural convection
Forced convection
Natural convection
3 / 1 3
/ 1 3 / 1
Pr Re
62 .
1 ⎟
⎠
⎜ ⎞
⎝
= ⎛
L Nu
xD
Nu Nu >Nu > Nu x x Nu Nu <Nu < Nu x x
(22-40)
Convection in circular pipe (Turbulent flow) Convection in circular pipe (Turbulent flow)
Pr > 0.7; L/D > 60
4 . 8 0
. 0
3 . 8 0
. 0
023 . 0
023 . 0
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝
= ⎛
=
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝
= ⎛
=
k Du C
k Nu hD
k Du C
k Nu hD
b p b p
μ μ
ρ
μ μ
ρ : cooling of the fluid
: heating of the fluid t
bt
sDittus and Boelter (t b ) t
bt
12 t
2= +
14 . 3 0
. 8 0
. 0
3 . 8 0
. 0
023 . 0
023 . 0
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝
= ⎛
=
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝
= ⎛
=
b p b p
k Du C
k Nu hD
k Du C
k Nu hD
μ μ μ
μ ρ
μ μ
ρ
Colburn (t f )
Sieder & Tate (t b )
Temp effect on viscosity in the laminar sub-layer.
t
2t
1For Pr>10
42 2 2
2
1 s s
f
t t t t t
+ + +
=
2
2 1
t t
bt +
=
Convection in circular pipe (Turbulent) Convection in circular pipe (Turbulent)
Pr > 0.7; L/D > 60
0.4 or 8 0.3
. 0
023 .
0 ⎟⎟
⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝
= ⎛
= k
Du C k
Nu hD
b pμ
μ ρ
Dittus and Boelter (t b ) (predicting within <13% for 651 data points) t
bt
s2 2 2
2
2 1
2 1
s s
f b
t t t t t
t t t
+ + +
=
= +
t
2t
1[ 3 . 796 0 . 205 ln Re 0 . 505 ln Pr 0 . 0225 (ln Pr)
2]
exp − − − −
=
p
b
C
u h
ρ (24-7)
Least Square Correlation (t b ) (predicting within 10.2% for 651 data points)
(24-4)
Convection in circular pipe (Turbulent flow) Convection in circular pipe (Turbulent flow)
4 . 8 0
. 0
3 . 8 0
. 0
023 . 0
023 . 0
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝
= ⎛
=
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝
= ⎛
=
k Du C
k Nu hD
k Du C
k Nu hD
b p b p
μ μ
ρ
μ μ
ρ : cooling of the fluid
: heating of the fluid
Dittus and Boelter (t b )
Temp effect on viscosity in the laminar sub-layer.
If two turbulent systems of the same fluid at the same bulk temperature are compared in heating and cooling experiment, the system being heated will have a higher temperature in the laminar sublayer than the system being cooled.
Consequently, if the fluid is a liquid, the sublayer of the heated system will be thinner, and the heat-transfer coefficient will therefore be higher than for the system being cooled. Since the Prandtl number is greater than 1 for most liquids, raising the exponent the above equation has the effect of raising h.
For most gases, the Prandtl number is near unity and approximately independent of
temperature, so that the value of the exponent on Pr is of minor consequence. An
understanding of this lack of effect for gases may be gained from the fact that viscosity and
thermal conductivity increase at almost the same rate with temperature, so that the thermal
U
0U
0U
0Re=5x10
5Heat Transfer with turbulent flow
c s
i
g L u
τ
= μ
Laminar sub-layer thickness
0.4 or 0.3 8
.
0
Pr
Re 023 .
= 0 Nu
U
iU
bMost liquid (Pr>1) : T , μ , L and k h x
Most gas (Pr~1) : T , μ , L but k , h x
Influence of heating on velocity profile in laminar tube flow
Parabolic velocity profile
If the fluid is heated or cooled, the velocity profile can be greatly altered because of the effect of temperature on viscosity.
heating
cooling Liquid heating
Gas cooling
Gas heating Liquid cooling
μ ↓
Poiseulli’s flow
μ ↑
Example 24
Example 24 - - 2: find 2: find h h for the water for the water
water t
1=60
oF u
b=7ft/s
t
2=140
oF
t
s=200
oF (assume) Steam condensation
t
∞=300
oF
D
i=0.870 in
F t
t t
t t
t F t t
s s
f b
o o
2 150 2
200 140
2
200 60
2
2 2
2 100 140 60
2
2 1
2 1
= + +
+
= + +
+
=
+ = + =
Arithmetic =
mean bulk temperature :
Arithmetic mean of the wall and bulk-fluid
temperatures
1” 16-gaguge copper tube
Example 24
Example 24 - - 2: find 2: find h h for the water for the water
water t
1=60
oF u
b=7ft/s
t
2=140
oF
t
s=200
oF (assume) t
∞=300
oF
D
i=0.870 in
Example 24
Example 24 - - 2: find 2: find h h for the water for the water
53 . 4
364 . 0
) 3600 )(
000458 .
0 )(
998 . 0 ( Pr
000 , 68
000458 .
0
) 0 . 62 )(
7 )(
12 / 87 . 0 ( Re
=
=
=
=
=
=
k C
D u
p i b
μ μ
ρ
73 . 2
384 . 0
) 3600 )(
000292 .
0 )(
00 . 1 ( Pr
000 , 107
000292 .
0
) 2 . 61 )(
7 )(
12 / 87 . 0 ( Re
=
=
=
=
=
=
k C
D u
p i b
μ μ
ρ
100 o F 150 o F
Example 24
Example 24 - - 2: find 2: find h h for the water for the water
) )(
)(
/(
1560
) 53 . 4 ( ) 600 ,
68 ) (
12 / 870 .
0 (
) 364 .
0 )(
023 .
0 (
023 .
0
2
4 . 0 8
. 0
4 8 0
. 0
F ft
h Btu
h
k Du C
k Nu hD
. b p
= o
=
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝
= ⎛
= μ
μ
ρ (24-4)
(a) Using the Dittus and Boelter equation (t b )
heating
Example 24
Example 24 - - 2: find 2: find h h for the water for the water
) )(
)(
/(
1800
) 73 . 2 ( ) 000 ,
107 (
) 0 . 1 )(
2 . 61 )(
3600 )(
7 )(
023 .
0 (
023 .
0
2
3 / 2 2
. 0
2 . 3 0
/ 2
F ft
h Btu
h
Du k
C C
u
h p b
p b
= o
=
⎟⎟ ⎠
⎜⎜ ⎞
⎝
= ⎛
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎛ −
μ μ ρ
ρ
(b) Using the Colburn equation (t f )
(24-5)
Example 24
Example 24 - - 2: find 2: find h h for the water for the water
) )(
)(
/(
1580
000205 .
0
000458 .
) 0 53 . 4 ( )
600 ,
68 ) (
12 / 870 .
0 (
) 364 .
0 )(
023 .
0 (
Pr Re
023 .
0
023 .
0
2
14 . 0 33
. 0 80
. 0
14 . 0 33
. 0 8
. 0
2 . 0 14
. 3 0
/ 2
F ft
h Btu
h k hD
Du k
C C
u h
s
b p s
p b
= o
⎟ ⎠
⎜ ⎞
⎝
= ⎛
⎟⎟ ⎠
⎜⎜ ⎞
⎝
= ⎛
⎟⎟ ⎠
⎜⎜ ⎞
⎝
= ⎛
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝
⎛ −
μ μ
μ ρ μ
μ μ
ρ (24-6)
(c) Using the Sieder and Tate equation (t b )
Example 24
Example 24 - - 2: find 2: find h h for the water for the water
(d) Using the Empirical Method (@100 o F) (t b )
[ ]
) )(
)(
/(
1640
) 998 .
0 )(
3600 )(
7 )(
0 . 62 )(
00105 .
0 (
] 89 . 6 exp[
] ) 53 . 4 (ln 0225 .
0 53 . 4 ln 505 .
0 600 ,
68 ln 205 .
0 796 .
3 exp[
Pr) (ln 0225 .
0 Pr ln 505 .
0 Re ln 205 .
0 796 .
3 exp
2
2 2
F ft
h Btu h
C u
h
p b
= o
=
−
=
−
−
−
−
=
−
−
−
−
ρ = (24-7)
Example 24
Example 24 - - 2: find 2: find h h for the water for the water
Equation h (Btu/(h)(ft 2 )( o F)
(a) the Dittus and Boelter equation (t b ) (24-4) 1560 (b) the Colburn equation (t f ) (24-5) 1800 (c) the Sieder and Tate equation (t b ) (24-6) 1580 (d) the Empirical Method (t b ) (24-7) 1640
water t
1=60
oF u
b=7ft/s
t
2=140
oF
t
s=200
oF (assume) t
∞=300
oF
D
i=0.870 in
Convection in Non
Convection in Non - - circular Conduits circular Conduits
45 . 0
1 2 4
. 0 8
. 0
023 .
0 ⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝
= ⎛
= D
D k
C u
D k
Nu hD eq eq b p μ μ
ρ
Wiegand (heat transfer coefficient on the outer wall of inner pipe)
( )
1 2
1 2
1 2
12 22
4 4
4 4
D D
D
D D
D D
D D
l r A
r D
eq p H
H eq
−
=
= − +
π
⎟ ⎟
⎠
⎞
⎜ ⎜
⎝
⎛ −
π
=
=
=
Concentric pipes
D 1 D 2
(24-8)
h
(14-4)
(14-5)
Concentric pipes
D 1 D 2
Convection in Non
Convection in Non - - circular Conduits circular Conduits
2 . 14 0
. 3 0
/ 2
023 .
0
−
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎛
μ
= ρ
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎛ μ
⎟⎟ μ
⎠
⎜⎜ ⎞
⎝
⎛ μ
ρ
b s eq
p p
b
u D k
C C
u h
heat transfer coefficient on the inner wall of the outer pipe
( )
1 2
1 2
1 2
12 22
4 4
4 4
D D
D
D D
D D
D D
l r A
r D
eq p H
H eq
−
=
= − +
π
⎟ ⎟
⎠
⎞
⎜ ⎜
⎝
⎛ −
π
=
=
=
(24-6)
h
Convection in Non
Convection in Non - - circular Conduits (turbulent) circular Conduits (turbulent)
( ) 4
4 4
4
1 2
1 2
2 1 2
2
D D
D D
D D
l r A
r D
p H
H eq
= − +
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎛ −
=
=
=
π π
Concentric pipes
D 1 D 2
a
b a b
ab b
a ab l
D A
p
eq
= +
= +
= 2
) (
2 4 4
b<<a
b b a
ab b
a ab l
D A
p
eq
2 2
) (
2 4
4 =
= +
= +
b =
a
실린더 주위에서의 열전달계수 분포
왼편 그림은 실린더 표면에서의 열전달계수 를 원주방향에 따라 나타낸 것으로 θ=0°가 실린더의 맨 왼쪽을, θ=180°가 실린더의 맨 오 른 쪽 을 나 타 낸 다 . 경 계 층 의 박 리 (separation)가 θ∼80°에서 일어나는 층류 경계층의 경우 박리 전에 열전달계수가 지속 적으로 감소하다가 박리 이후 난동성분의 증 가로 다소 증가하게 된다. 그러나 난류로의 천이가 일어나는 오른쪽 그림의 경우 천이로 인한 급속한 열전달계수 증가가 나타난다. 이 는 평판 위를 지나는 유동의 경우에서도 나타 났던 것이다. 그 이후 경계층의 박리에 의해 열전달계수는 다시 급속히 감소하게 된다.
Convection Normal to a Cylinder Convection Normal to a Cylinder
Average coefficients for the entire surface
(
1/3)
0 0
Pr 1 . 1
n n
b Du k
hD
b Du k
hD
⎟⎟ ⎠
⎜⎜ ⎞
⎝
= ⎛
⎟⎟ ⎠
⎜⎜ ⎞
⎝
= ⎛
μ ρ μ Gases: ρ
Liquids:
Du 0 ρ/μ n b
1-4 0.330 0.891
4-40 0.385 0.821
40-4000 0.466 0.615
4000-40,000 0.618 0.174
40,000-250,000 0.805 0.0239
(24-9)
Local number for cross-flow about a circular cylinder
At low Reynolds numbers At high Reynolds numbers
Nu Nu versus Re for flow normal to a single cylinder versus Re for flow normal to a single cylinder
Natural convection from horizontal cylinder Natural convection from horizontal cylinder
A single horizontal cylinder A single horizontal cylinder
10 10 3 3 < < Gr Gr · · Pr Pr < 10 < 10 9 9
4 / 1 4
/ 1 2
2 3
53 .
0 ⎟⎟
⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝
⎛ Δ
= k
t C g
D k
hD p μ
μ β
ρ (24-10)
2 + ∞
= t t t f s
t ∞
t s
Example 24
Example 24 -3: find - 3: find h h between the surface and air between the surface and air
D=0.1m
C t
s= 56
oC t
∞= 20
ot C
t
ft
s38
o2 20 56
2 + =
+ =
=
∞K kg J C
s Pa K m kg
K m W k
p
= ⋅
⋅
×
=
=
=
⋅
=
−
−
/ 1000
10 92 . 1
00322 .
0
/ 14 . 1
/ 0266 .
0
5 1 3
μ β
ρ ( )
722 . 0
0266 .
0
) 10 92 . 1 )(
1000 Pr (
10 00 . 4
10 92 . 1
) 36 )(
00322 .
0 )(
8 . 9 ( ) 14 . 1 ( ) 1 . 0 (
5 6
5 2 2
3
=
= ×
×
=
= ×
−
Gr
−K m W
k h t C
g D
k
hD 1 / 4 p 1 / 4 2
2 2 3
/ 81
. 5 53
.
0 ⎟⎟ ⇒ =
⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝
⎛ Δ
= μ
μ β ρ
steam Heat loss air
(24-10) 10
3< PrGr < 10
9Convection Normal to a Bank of Circular Tubes Convection Normal to a Bank of Circular Tubes
Grimison Equation (> 10 rows)
3 / max 1
max
Pr )
1 . 1 (
n n
b DG k
hD
b DG k
hD
⎟⎟ ⎠
⎜⎜ ⎞
⎝
= ⎛
⎟⎟ ⎠
⎜⎜ ⎞
⎝
= ⎛
μ Gases: μ
Liquids:
G
max= ρU
max(velocity at the minimum cross section)
2 +
∞= t t t
f s(24-11)
b=0.482 b=0.229
n=0.632
Convection Normal to a Bank of Circular Tubes Convection Normal to a Bank of Circular Tubes
Grimison Equation (< 10 rows deep)
3 / max 1
max
Pr )
1 . 1 (
n n
b DG k
hD
b DG k
hD
⎟⎟ ⎠
⎜⎜ ⎞
⎝
= ⎛
⎟⎟ ⎠
⎜⎜ ⎞
⎝
= ⎛
μ
Gases: μ
Liquids:
2 +
∞= t t t
f s(24-11)
The ratios given in Table 24-4
1 2 3 4 5 6 7 8 9 10 1 1.10 1.22 1.31 1.35 1.40 1.42 1.44 1.46 1.47 1 1.25 1.36 1.41 1.44 1.47 1.50 1.53 1.55 1.56
triangle square
N
Convective heat transfer exchange Convective heat transfer exchange
between liquids in laminar flow and tube bundles
between liquids in laminar flow and tube bundles
Energy transfer and frictional loss for liquids
Energy transfer and frictional loss for liquids
in laminar and transition flow past tube bundles
in laminar and transition flow past tube bundles
Convection from Spheres Convection from Spheres
Froessling Equation
2 1 0
3 1
6 0 0
2
/ /
p
μ ρ Du k
μ . C
k .
Nu hD ⎟⎟
⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝ + ⎛
=
=
2
) )(
( )
( 2
2 / 1 1
4 4
) 4
(
0 2
0
0 2
/ 2
2
0
=
−
=
−
=
−
⎟ =
⎠
⎜ ⎞
⎝
⎛ +
− ∞
−
=
−
=
−
=
∫
∫
∞k hD
t t D h t
t Dk q
t D t
k q
r dt dr k
q
const dr
r dt k
q
s s
s t
t
D s
π π
π π
π
Steady state conduction:
t
0t
s 2 / D Heat transfer by conduction in an infinite stagnant medium(24-12)
(24-18)
(24-13) (24-14)
(24-15)
(24-16) & (24-17)
Nu Nu versus Re for air versus Re for air - - flow past single sphere flow past single sphere
2
Convective Heat Transfer between a Fluid and a Packed Bed Convective Heat Transfer between a Fluid and a Packed Bed
fraction void
: ,
) 1
50 ( 2 Pr
2 / 1 3
/ 2
ε ε
ε ρ
b bs
bs p
bs H
u u
where
μ
ρ . Du
C u
j h
=
⎥ ⎦
⎢ ⎤
⎣
⎡
= −
=
−
(24-19)
Bradshaw (1963) 400<Re p <10,000
) 1
Re (
ε μ
ρ
= −
bsp
Du
b
bs
u
u
(14-20)
ε
=
Ub= Average interstitial velocity at any cross section in the bed
Convection from a Plane Surface Convection from a Plane Surface
Natural convection
9 3
/ 1 3 / 1
9 4
4 / 1 4
/ 1 4 / 1
4 / 1
10 Pr
13 . 0
10 10
59 . 0 Pr
59 . 0
478 .
0
>
=
<
<
=
=
=
Ra Gr
Nu
Ra Ra
Gr Nu
Gr Nu
m m m Ideal gas heated by laminar
Natural convection from vertical plate
McAdams Eq.
Laminar natural convection
Combination of Laminar+Turbulent
The fluid properties are evaluated at the arithmetic average temperature (t
s+t
o)/2
(22-25)
(24-20)
(24-21)
to to
Determination of heat loss to the atmosphere Determination of heat loss to the atmosphere
by natural convection of air by natural convection of air
J.H. Perry (air, water, organic liq.)
7 5
4 / 1
7 5
4 / 1
9 3
4 / 1
9 3
4 / 1 0
9 3
/ 1
10 10
) ( 20 . 0
10 10
) ( 38 . 0
10 10
) / ( 28 . 0
10 10
) /
( 50 . 0
10 )
( 18 . 0
<
<
Δ
=
<
<
Δ
=
<
<
Δ
=
<
′ <
Δ
=
>
Δ
=
Ra t
h
Ra t
h
Ra L
t h
Ra D
t h
Ra t
h
All surface:
Horizontal cylinder:
Vertical plate:
Heated horizontal plates facing upward:
Heated horizontal plates facing downward:
[in]
[ft]
The simplified forms of Eqs.
(24-20) and (24-21) in which values of the physical
properties of air at typical
ambient conditions have
already been substituted.
Heat Transfer to Liquid Metals Heat Transfer to Liquid Metals
Liquid metal (same μ, high k) : Pr<<1, δ th >>δ
8 8 0
0
025 0
0 7
. p
. b
k μ C μ
ρ . Du
.
Nu ⎟⎟
⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝ + ⎛
=
For circular pipe, uniform heat flux, L/D>60, Pe>100 Lyon Equation
8 8 0
0
025 0
0 5
. p
. b
k μ C μ
ρ . Du
.
Nu ⎟⎟
⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝ + ⎛
=
For circular pipe, const wall temp, L/D>60, Pe>100 Seban and Shimazaki Equation
(24-27)
(removing heat from nuclear reactors)
Heat Transfer to Liquid Metals for Flat Plate Heat Transfer to Liquid Metals for Flat Plate
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎛ =
∂
= ∂
∂
∂
∂
= ∂
∂
∂
0 2
2 2
2
0 ;
u x y
t t
y t x
u t α θ
α θ
π
μ μ
ρ
πα α πα π
α
Pr Re 564 . 0
) /(
) (
/ 4 2
/ 4
2 / 2 1
/ 1 0
0 0
0 0
0 0
x x x p
s x s
s s
Nu
k xu C
k x h
x k u
dy
t t t t k d
h
x y u
u x y u
x erf y
t t
t t
=
⎟⎟ ⎠
⎜⎜ ⎞
⎝
⎟⎟ ⎛
⎠
⎜⎜ ⎞
⎝
= ⎛
⎭ =
⎬ ⎫
⎩ ⎨
⎧ − −
=
=
=
− =
−
small is
n 2 when
~
! 3 7
! 2 5
! 1 3
2
3 5 7π π n
n n
n n n
erf ⎟⎟
⎠
⎜⎜ ⎞
⎝
⎛ +
− ⋅ + ⋅
− ⋅
= L
Grosh and Cess
Effect of Surface Roughness on Heat-transfer Coefficient
10
310
410
510
610
7Re=Du b ρ/μ f
f=16/Re
f=0.046Re -1/5
surface roughness
U0 U0
0
Protuberance < laminar sub-layer No enhancement of h
If surfaces with transverse fins are used, the increases in the heat transfer area and in the heat- transfer coefficient are usually offset by an increase in the power required to pump the fluid past the heat-transfer surface
Effect of Surface Roughness on Heat-transfer Coefficient
Protuberance > laminar sub-layer Enhancement of h
Loss of energy due to friction
fin
Effect of Surface Roughness on Heat-transfer Coefficient
Copper tube
V-shaped groove Brouillette
Roughness ratio
= protuberance height/tubing diameter
=0.05
- Heat transfer coefficient become double - friction factor increase fourfold.
-Form drag is present !
Homework
24-1
24-4
24-6
24-14