Chapter 6

Finned surfaces

6.1 Introduction
Convection heat transfer is governed by the equation
Q = hA9
With given values of h and 9 the only way of increasing the heat
transfer rate is to enlarge the surface area A. This may be achieved
by extending the surface with projecting fins of various shapes. The
most common types are strip fins on plate surfaces and annular fins
on tubular surfaces; both types may have rectangular or tapered
profiles. In many heat exchangers, particularly liquid to gas applica-
tions, the h values on the gas side are much smaller than those on the
liquid side and fins would be added to the gas side until the thermal
resistance of each are approximately equal. An example of this
technique can be seen in the car radiator. The addition of fins alters
flow patterns and h values and it is unlikely that h will be constant
over the whole fin, but this effect is ignored in simple analysis.
Account is, however, taken of the change of temperature along the
fin length by a fin efficiency.

6.2 Strip fins of rectangular profile

Fins are considered long and thin enough for one-dimensional con-
duction laws to apply. Figure 6.1 shows a fin and by considering the
balance between conduction and convection heat transfer to and
from a small element distant x from the wall (root of the fin) a
differential equation to the temperature distribution is established:

M-nte
z = o
dx
where 9 = T — 7} is the temperature difference between fin and
12
fluid at any point
m = (ph/kA) '

123
124 Finned surfaces

rF i n0t
°i Q (convection)

Fin t i p

Distance f r o m r o o t

Figure 6.1

p is the fin perimeter

/ is the fin length
k is the thermal conductivity of the fin material
A is thefincross-sectional area
h is the average convection heat transfer coefficient
and T wis the wall temperature.

The solution to this equation for long, thin fins with negligible heat
loss from the tip is obtained by using the boundary conditions:
(a) at the wall where x — 0 and 9 = 90 = T w— 7};
d.9
(b) at the tip where x = 1 and = 0.

It is found that the temperature distribution is given by (Figure 6.1):

9 _ cosh m(l — x)
90 ~ cosh ml

and hence the heat transfer rate from thefinis given by

Q = mkA90 tanh ml (6.2)
 Annular fins and tapered fins 125

Since this heat transfer is less than that which would occur if the
whole fin was at the wall temperature a fin efficiency is defined by

fni _ actual heat transfer rate

^ ~~ heat transfer rate if whole fin were at wall temperature

In this case
tanh ml ,r _)
- - u s -

This analysis also applies to pin fins with various cross-sections.

6.3 Annular fins and tapered fins

If a similar analysis is made for annular fins of rectangular profile
(Figure 6.2) the resulting differential equation is
2
d0 1 d0 (2h\Q

Similarly for a strip fin (Figure 6.2) with tapered triangular profile,
analysis yields

dx
2 + I
x dx
_
\kb J x
I =o
Both these equations can be solved using tables of Bessel functions
[1] or by numerical integration by fourth-order Runge-Kutta
scheme [2] or by finite difference techniques. However, because of the
widespread use of fins the concept of fin efficiency is normally used
and graphs of fin efficiency are plotted from solutions to the
equations [3]. Figure 6.3 shows such a chart. The fin heat transfer

A n n u l a r fin Tapered fin F i n array

Figure 6.2
 126 Finned surfaces

f
Fin efficiency

A n n u l a r fin efficiency chart f o r

fin o f constant thickness

Figure 6.3

rate is then obtained from the definition of fin efficiency:

n actual heat transfer rate

M
fi = . —

heat transfer rate if whole fin were at wall temperature

6.4 Arrays of fins

Although the use of fins enhances the heat transfer area the extra
area is inefficient due to the change in temperature along the surface
which is expressed for the single fin by the fin efficiency. For arrays
of fins (Figure 6.2) the effect can be calculated as shown below and
for complex surfaces with fins (Figure 7.5) data is included on the
chart for use in equations (6.5) and (6.6).
Let Afa be the total fin area per unit length.
Let A be the total area including fin area per unit length:

The total heat transfer rate from a finned surface is

where 0Ois the temperature difference at the wall

 Program 6.1 Heat transfer, rectangular profile 127

Q' is the heat transfer rate per unit length

and h is the convection heat transfer coefficient

fafin^fin + (A - Afin
)]hA6'0(
A
0! = [rizJ + (1 -
fofin/M P)]hA00 (6.4)
Equation (6.4) may be written
Q! = ri'hA00 where >/' = [f/fin J? + (1 - j8)] (6.5)
In equation (6.5), y\' is the area weighted fin efficiency which may be
incorporated into the calculation of U value for heat exchangers. If
UA is the overall heat transfer coefficient based on the enhanced
surface area A where the heat transfer coefficient is hl9 then

(6.6)

where A2 is the area of the unfinned side where the heat transfer
coefficient is h2 (Program 7.4).

6.5 Use of fins

Although fins increase the surface area the extra material increases
the thermal resistance at the fin position and there may be situations
where the heat transfer rate is not improved. As a guide, analysis
shows that if pk < hA fins will not enhance the heat transfer rate.

References
1. Eckert, E.R.G. and Drake, R.M. Analysis of Heat and Mass Transfer, McGraw-
Hill (1972).
2. Adams, J.A. and Rogers, D . F . Computer Aided Heat Transfer Analysis, McGraw-
Hill (1973).
3. Gardner, K.A. 'Efficiency of extended surfaces', Transactions of the American
Society of Mechanical Engineers, 67, 621-31 (1945).

WORKED EXAMPLES

Program 6.1 Heat transfer in a strip fin of rectangular profile

In the programs given in Chapter 3 the finite difference method was
used to solve two-dimensional conduction problems using differen-
128 Finned surfaces

Wall

Fluid r
f
Figure 6.4

tial equations obtained by analysis. In the programs given in this

chapter a similar method is used to solve one-dimensional conduc-
tion problems with convective boundaries. Clearly, there are alter-
native methods for this particular program because the fin we are
considering can be solved by direct analysis and the resulting
equations (6.1) and (6.2) could be programmed to give the tem-
perature distribution in the fin and the heat transfer in the fin.
However, this alternative is not always available and here we make
a simple model of the fin in the computer and obtain an approximate
solution without mathematical analysis.
Figure 6.4 shows a fin divided into four sections (five nodes). For
this fin the wall temperature T(0) is 100 °C and the surrounding fluid
temperature is at 10 °C. The fin is 0.06 m long, 0.003 m thick and is
2
made of material of thermal conductivity 50 W/m K. The surface
heat transfer coefficient is 40 W/m K. Consider unit width and write
a program to determine the temperature distribution in the fin and
the heat transfer rate from the fin.
Consider P nodes numbered from 0 to (P — 1). At the wall where
p = 0, T(0) = 100 and at the far end, where the node number
N = (P — 1), the conduction to the end must equal the convection
from the side surfaces since in simple long fin theory there is no
convection from the end. Thus

k(w\) h(Tr - TN) 0

TN =

The main body of the fin (nodes = 1 to P - 2) will be governed by

the balance between conduction into the section, conduction out of
the section and convection from the side surfaces:
 Program 6.1 Heat transfer, rectangular profile 129

X
+ 2h(x-\){Ts - TN) = 0

r„_, + r , +1 + 2 (g]r r

7"- \ V

<••(£
The heat transfer rate from the fin must be equal to the heat transfer
into the fin at the wall element:
To-TA , Jx_1\J(T0+Tl

0 = k ( w \ ) ( ^ ^ ) + feW + r, - 27})

The necessary reading for this program is Section 6.2.

10 PRINT " LONGITUDINAL F I N "

20 PRINT
30 DIM T ( 1 0 )
35 C = 10000000
40 P R I N T "HOW MANY N O D E S DO Y O U C H O O S E ? "
50 I N P U T P : P R I N T "THERE ARE " ? P ; " NODES"
55 REM G U E S S THE I N I T I A L TEMPERATURES
58 T(0) = 100
60 FOR N = 1 TO I P - 1)
70 T(N) = (T(0) + TF) / 2
80 NEXT N
90 P R I N T " WHAT I S T H E F I N L E N G T H ? "
100 I N P U T L : P R I N T " F I N L E N G T H I S " ; L ; " M"
110 P R I N T "WHAT I S T H E F I N T H I C K N E S S ? "
120 I N P U T W: P R I N T " F I N T H I C K N E S S I S " ; W } " M"
130 P R I N T "WHAT I S T H E T H E R M A L C O N D U C T I V I T Y ? "
140 I N P U T K : P R I N T " T H E R M A L C O N D U C T I V I T Y I S " ; K ; " W/M K"
150 P R I N T "WHAT I S T H E S U R F A C E H E A T T R A N S F E R COEFFICIENT
?"
160 I N P U T H : P R I N T " H E A T T R A N S F E R C O E F F I C I E N T I S " ? H ? " W/
M ' 2 K"
170 P R I N T "WHAT I S T H E F L U I D T E M P E R A T U R E ? "
180 I N P U T T F : P R I N T " F L U I D TEMPERATURE I S ' ; T F ; " C"
190 X = L / (P - 1)
2 0 0 F = ( H * <X " 2 ) ) / (K * W>
210 REM I T E R A T I O N R O U T I N E TO F I N D T E M P E R A T U R E D I S T R I B U T
ION
220 A = 0
230 F O R N = 1 TO ( P - 2)
240 T(N) = ( T ( N - 1) + T < N + t ) + 2 * F * TF> / ( 2 * ( 1 +
F) )
250 NEXT N
2 6 0 N = (P - 1)
270 T(N) = (T(P - 2) + F * TF) / (1 + F )
280 A = A + 1
290 P R I N T "NUMBER O F ITERATIONS";A
130 Finned surfaces

3 1 0 Q = <K * W * 1 * < T ( 0 ) - T U ) ) ) / X + H * (X / 2) * 1
* (T(0) + T(l) - 2 * TF)
320 I F (C - Q) < 0 . 1 THEN GOTO 3 3 5
325 C = Q
330 GOTO 2 3 0
335 PRINT
340 P R I N T "NODE NUMBER T E M P E R A T U R E I N C"
350 PRINT
360 F O R N = 0 TO ( P - 1)
370 T(N) > ( INT ( T ( N ) * 100 + 0 . 5 ) ) / 100
380 PRINT TAB( 6 ) N , TAB( 17)T(N>
390 NEXT N
400 PRINT
410 P R I N T "THE NUMBER OF I T E R A T I O N S I S " ; A
420 PRINT
430 P R I N T " T H E H E A T T R A N S F E R FROM T H E F I N I S "5Q;" W/M"

RUN
LONGITUDINAL FIN

HOW MANY N O D E S DO Y O U C H O O S E ?
?5
THERE ARE 5 NODES
WHAT I S T H E F I N L E N G T H ?
?.06
F I N LENGTH I S . 0 6 M
WHAT I S T H E F I N THICKNESS?
?.003
F I N THICKNESS I S 3 E - 0 3 M
WHAT I S T H E T H E R M A L C O N D U C T I V I T Y ?
?50
THERMAL C O N D U C T I V I T Y I S 5 0 W/M K
WHAT I S T H E S U R F A C E H E A T T R A N S F E R A
COEFFICIENT?
?40
HEAT TRANSFER C O E F F I C I E N T I S 4 0 W / M 2 K
WHAT I S T H E F L U I D T E M P E R A T U R E ?
?10
F L U I D TEMPERATURE I S 1 0 C
NUMBER OF ITERATIONS1
NUMBER OF ITERATI0NS2
NUMBER OF ITERATI0NS3
NUMBER OF ITERATI0NS4
NUMBER OF ITERATI0NS5
NUMBER OF ITERATI0NS6
NUMBER OF ITERATI0NS7
NUMBER OF ITERATI0NS8
NUMBER OF ITERATI0NS9
NUMBER OF ITERATIONS10
NUMBER OF ITERATIONS11
NUMBER OF ITERATIONS12
NUMBER OF ITERATIONS13
NUMBER OF ITERATIONS14
NUMBER OF ITERATIONS15
NUMBER OF ITERATIONS16
NUMBER OF ITERATIONS17

NODE NUMBER TEMPERATURE IN C

0 100
1 77.51
2 63.13
3 55.14
4 52.59

THE NUMBER OF ITERATIONS IS 17

THE HEAT TRANSFER FROM THE FIN IS 272.186489 W/M

 Program 6.2 Array of fins of the type used in Program 6.1 131

Program nomenclature
C parameter to end program when iteration completed
P number of nodes
T(N) temperature at a node
TF fluid temperature
L fin length
W fin thickness
K fin thermal conductivity
H surface heat transfer coefficient
X length of a section
A iteration counter
Q heat transfer rate per unit width

Program notes
(1) The analytical solution to this problem gives a heat transfer rate
of 275.4 W per unit width.
(2) The program structure is as follows:
Line 30 allows for up to 10 nodes.
Line 35 is used to set a value for C far in excess of the heat
transfer rate to allow the first iteration (line 320).
Lines 40-200 set up the problem with lines 60-80 estimating the
initial temperatures.
Lines 210-330 are the iteration routine used until the change in
heat transfer rate is considered negligible, C being reset by line
325 to the latest value.
Lines 340-430 display the results.
(3) The program can be run with different numbers of nodes and
different levels of accuracy: (C-Q) of line 325.

Program 6.2 An array of fins of the type used in Program 6.1

Fins are often used in arrays for which the fin pitch is known. Fin
pitch is the distance between centre lines of adjacent fins. The total
heat transfer rate for a finned surface can then be determined by one
of two methods:
(a) by adding the heat transfer rate from the fins to the heat
transfer rate from the unfinned area;
(b) by using the analysis of Section 6.4 to determine the area
weighted fin efficiency (rj') and hence equation (6.5) to deter-
mine the heat transfer rate per unit length of finned surface
(unit width fins)
 132 Finned surfaces

= ri'hA00 Q'
The necessary reading for the program is Sections 6.2 and 6.4.
This program uses method (a) and determines the heat transfer rate
from a single fin determined in Program 6.1 which is multiplied by
the number of fins. The heat transfer rate from the unfinned area is
then added:
Q = y & n + (i - yw)he0
where y is the number of fins per unit length of finned surface and
w is the fin thickness. (The surface is of unit width.)
500 PRINT " F I N A R R A Y OVER 1 M L E N G T H "
510 PRINT " U S I N G THE F I N D E S I G N ABOVE"
520 PRINT
530 PRINT "WHAT IS THE FIN PITCH?"
540 INPUT Z: PRINT "FIN PITCH IS ";Z;" M"
550 Y = 1 / Z
560 PRINT "FINS PER METRE= ";Y
570
580
G
Q
=
=
U Q * Y )
( INT (Q
+
*
((1
100
-
+
Y * W)
0.5) /
* 1 *
100)
H * (T(0) M
- T F ) ) )

590 PRINT "HEAT TRANSFER FROM FINNED SURFACE= ";Q; W/M*

RUN
LONGITUDINAL FIN

HOW MANY N O D E S DO Y O U CHOOSE?

?5
THERE ARE 5 NODES
WHAT I S T H E F I N L E N G T H ?
?.06
FIN LENGTH IS.06 M
WHAT I S THE FIN THICKNESS?
?.003
FIN THICKNESS IS3E-03 M
WHAT IS THE THERMAL CONDUCTIVITY?
?50
THERMAL CONDUCTIVITY IS50 W/M K
WHAT
?40
IS THE SURFACE HEAT TRANSFER A
COEFFICIENT?

HEAT TRANSFER C O E F F I C I E N T I S 4 0 W / M 2 K
WHAT I S T H E F L U I D T E M P E R A T U R E ?
?10
F L U I D TEMPERATURE I S 1 0 C
NUMBER OF ITERATIONS1
NUMBER OF ITERATI0NS2
NUMBER OF ITERATI0NS3
NUMBER OF ITERATI0NS4
NUMBER OF ITERATI0NS5
NUMBER OF ITERATIONS6
NUMBER OF ITERATI0NS7
NUMBER OF ITERATI0NS8
NUMBER OF ITERATI0NS9
NUMBER OF ITERATIONS10
NUMBER OF ITERATIONS11
NUMBER OF ITERATIONS12
NUMBER OF ITERATIONS13
NUMBER OF ITERATIONS14
N U M B E R OF ITERATIONS15
 Program 6.3 Circumferential fin 133

NUMBER OF ITERATIONS16
NUMBER OF ITERATIONS17

NODE NUMBER TEMPERATURE IN C

0 100
77.51
2 6 3 . 13
3 5 5 . 14
4 52.59

THE NUMBER OF ITERATIONS IS 17

THE HEAT TRANSFER FROM THE FIN IS 272.186439 W/M

FIN ARRAY OVER 1 M LENGTH
USING THE FIN DESIGN ABOVE

WHAT I S T H E F I N PITCH?
?. 01
FIN PITCH IS . 0 1 M
F I N S PER METRE* 100
H E A T T R A N S F E R FROM F I N N E D SURFACE* 29738.65 W/M'2

Program nomenclature
Z fin pitch
Y fins/mtransfer rate
Q heat

Program notes
(1) The program is very simple but it requires to be added to
Program 6.1 to produce answers. (Program 6.1 determines the heat
transfer from a single fin.)
(2) The program could be rewritten to use the second method (area
weighted fin efficiency) to be of more general use for any fin shape
(Problem 6.2).
(3) Is this program valid if Y is not an integer?

Program 6.3 Circumferential fin

This program considers an annular fin on a circular pipe. In such a
fin the heat transfer area increases with increase of radius and an
analysis yields a differential equation which can be solved with Bessel
functions or by numerical integration. In this program, as in
Program 6.1, we model the circumferential fin and use the simple
one-dimensional heat transfer equationInk AT(2.4):
(2.4)
134 Finned surfaces

Figure 6.5

To use this we consider the fin as a series of annular rings (Figure 6.5)
and by a balance between conduction into a ring, conduction out of
a ring and convection from the ring surfaces we write

2nkw(TN_{ - TN) 2nkw(TN+l - TN)

+ 2(2nRNx)h(T( - TN) = 0
from which

ta
fe) + ta
fe) + 2 J W
- (^ > "fe)
h ! ,

where F = hx/kw.
 Program 6.3 Circumferential fin 135

At the outside half-width ring by energy balance, ignoring con-

vection from the outer ring edge surface, we obtain
2nkW
(TN.t - TN) J+ / _ A x \ _ ^ _ Q

from which
n R
TN-t + F(RN-j)ln' T,
4 \R»

1 +
-l

The heat transfer rate at the wall is given by

. _ 2nkw(T0 -

The necessary reading for this program is in Sections 2.4 and 6.3.
Write a program to determine the heat transfer rate from an
annular fin and test it with the following data:
Fin outer diameter 0.2 m
Fin root diameter 0.15m
Fin thickness 2.5 mm
Thermal conductivity 48 W/m K2
Surface heat transfer coefficient 280 W/m K
Fin root temperature 170°C
Ambient temperature 25 °C
10 PRINT "CIRCUMFERENTIAL FIN"
20 PRINT
30 C = 10000000
40 DIM T(20): DIM R(20)
50 PRINT "HOW MANY N O D E S DO YOU CHOOSE?"
60 INPUT Ps P R I N T "THERE ARE";P?' NODES"
70 REM CUESS THE INITIAL TEMPERATURES
80 PRINT "WHAT IS THE WALL TEMPERATURE?"
85
C"
INPUT T:T(0) = A T: PRINT "WALL TEMPERATURE IS ";T(0);"

90 FOR N 1 T O IP - 1)
100 T(N) = (T(0) + TF) / 2
110 NEXT N
120 PRINT "WHAT IS THE F I N ROOT DIAMETER?"
130 I N P U T DRJ PRINT "FIN ROOT DIAMETER IS ";DR;" M"
140 PRINT "WHAT IS T H E F I N T I P DIAMETER?"
150 INPUT DT: PRINT "FIN TIP DIAMETER IS ";DT;" M"
160 PRINT "WHAT IS THE FIN THICKNESS?"
170 INPUT W: PRINT "FIN THICKNESS IS ";W;" M"
180 PRINT "WHAT IS THE THERMAL CONDUCTIVITY?"
190 INPUT K: P R I N T "THERMAL CONDUCTIVITY I S ";K«, " W/M K"
136 Finned surfaces

200 PRINT "WHAT IS THE SURFACE HEAT TRANSFER COEFFICIENT

?" A
210 INPUT H: PRINT "HEAT TRANSFER COEFFICIENT IS ":H:" W
/ M 2 K"
220 PRINT "WHAT I S T H E F L U I D T E M P E R A T U R E ? "
230 INPUT T F : P R I N T " F L U I D TEMPERATURE I S " ; T F ; " C"
235 R(0) = A
DR / 2 : R ( P - 1 ) = DT / 2
240 X • (DT - DR) / ( 2 # (P - 1))
250 F = ( H # ( X 2 ) ) / ( K * W )
255 F = F / X
260 REM ITERATION ROUTINE TO FIND TEMPERATURE DISTRIBUT
ION
270 A * 0
280 FOR N = 1 TO ( P - 2)
290 R(N) - (DR / 2) + (N * X)
300 NEXT N
310 FOR N * 1 TO ( P - 2)
32 0 LL * R ( N + 1) / R(N)
33 0 MM « R ( N ) / R(N - 1)
340 E = (2 * R(N) * F * LOG ( L L ) * LOG (MM))
350 T(N) » ( T ( N - 1) * LOG ( L L ) + T(N • 1) * LOG (MM) +
E * TF) / s
( LOG ( L L ) + LOG (MM) + E )
360 NEXT N
370 N (P - 1)
380 I « LOG (MM) * F * ( R ( N ) - (X / 4))
390 T(N) * (T(N - 1) + I * TF) / (1 + I)
400 A = A + 1
41 0 P R I N T "NUMBER O F I T E R A T I O N S "?A
42 0 Q * <(2 * 3.142 • K * W * (T(0) - T ( l ) ) ) / ( LOG (R(l
) / R ( 0 ) ) ) ) + H * 3 . 1 4 2 * (DR + (X / 2 ) ) * (X / 2 ) * (T(0
) + T(l) - 2 * TF)
430 I F (C - Q) < 0 . 1 THEN GOTO 460
440 C * Q
450 GOTO 3 1 0
460 PRINT
470 P R I N T "NODE NUMBER RADIUS IN M TEMPERATURE IN C

480 PRINT
490 FOR N • 0 TO ( P - 1)
5 0 0 R ( N ) = ( INT <R(N) * 1000 + 0 . 5 ) ) / 1000:T(N) = ( INT
(T(N) * 100 * 0 . 5 ) ) / 100
510 PRINT TAB( 6 ) N . TAB( 17)R(N). TAB( 28)T(N)
520 NEXT N
530 P R I N T "THE NUMBER OF I T E R A T I O N S I S ";A
540 PRINT
550 P R I N T " T H E H E A T T R A N S F E R R A T E FROM T H E F I N I S ";Q;"
W"
560 Bl - (H * (DT - D R ) ) / ( 2 * K)
570 P R I N T "THE B I O T NUMBER I S "?BI

RUN
CIRCUMFERENTIAL FIN

HOW MANY NODES DO YOU CHOOSE?

?10
THERE ARE10 NODES
WHAT IS THE WALL TEMPERATURE?
?170
WALL TEMPERATURE IS 170 C
WHAT IS THE FIN ROOT DIAMETER?
?.15
FIN ROOT DIAMETER IS * 15 M
WHAT IS THE FIN TIP DIAMETER?
?.20
FIN TIP DIAMETER IS . 2 M
WHAT IS THE FIN THICKNESS?
?.0025
 Program 6.3 Circumferential fin 137

FIN THICKNESS IS 2 . 5 E - 0 3 M
WHAT I S T H E T H E R M A L C O N D U C T I V I T Y ?
?48
THERMAL C O N D U C T I V I T Y I S 4 8 W/M K
WHAT I S T H E S U R F A C E H E A T T R A N S F E R COEFFICIENT? A
?280
HEAT TRANSFER C O E F F I C I E N T I S 2 8 0 W / M 2 K
WHAT I S T H E F L U I D T E M P E R A T U R E ?
?25
FLUID TEMPERATURE IS 25 C
NUMBER OF ITERATIONS 1
NUMBER OF ITERATIONS 2
N U M B E R OF ITERATIONS 3
NUMBER OF ITERATIONS 4
NUMBER OF ITERATIONS 5
NUMBER OF ITERATIONS 6
NUMBER OF ITERATIONS 7
NUMBER OF ITERATIONS 8
NUMBER OF ITERATIONS 9
NUMBER OF ITERATIONS 10
NUMBER OF ITERATIONS 11
NUMBER OF ITERATIONS 12
NUMBER OF ITERATIONS 13
NUMBER OF ITERATIONS 14
NUMBER OF ITERATIONS 15
NUMBER OF ITERATIONS 16
NUMBER OF ITERATIONS 17
NUMBER OF ITERATIONS 18

NODE NUMBER RADIUS IN M TEMPERATURE IN C

0 .075 170
1 .078 145.13
2 .081 125.38
3 .083 109.83
4 .086 97.76
5 .089 8 8 . 5 9
6 .092 8 1 . 8 9
7 .094 7 7 . 3 3
8 .097 7 4 . 6 6
9 . 1 7 3 . 7 5
THE NUMBER OF ITERATIONS IS 18

THE HEAT T R A N S F E R R A T E FROM T H E FIN IS 564.748034 W

T H E B I O T NUMBER I S .145833333

Program nomenclature
c parameter to end program when iteration com-
pleted
p number of nodes
T(N) temperature at a node
R(N) radius of a node
DR fin root diameter
DT fin tip diameter
W fin thickness
K fin thermal conductivity
H surface heat transfer coefficient
TF fluid temperature
X width of radial slice
138 Finned surfaces

LL, MM, E, I convenient program variables

A iteration counter
Q heat transfer rate

Program notes
(1) The fin efficiency chart solution to this problem is 558 W.
However checking the program against other problems suggests that
the accuracy of the program may be influenced by the value of the
Biot number which is evaluated in lines 560 and 570. Bearing in mind
the work of Sections 2.13 and 3.8 this would not be surprising and
would be worth further investigation.
(2) The program structure is as follows:
Line 30 sets a value of C far in excess of the heat transfer rate
to allow the second iteration (line 430).
Line 40 allows for up to 20 nodes.
Lines 50-255 set up the problem with lines 90-110 estimating
the initial temperatures.
Lines 260-450 contain the iteration routine, lines 320, 330, 340
and 350 being used to obtain simple variables. The routine
continues till line 430 decides the problem is solved.
Lines 470-550 display the results.
(3) The number of nodes can be varied and different values of
(C-Q) tested.
(4) The program illustrates how situations may often be modelled in
a simple way but obviously clear thought is needed and assumptions
made must be valid. There are debatable points in the solution, for
example no convection from the outer edge surface is considered. Is it
worth allowing for the small increment in radius x/4 in the outside
equation? Such items can be examined by the user.
(5) If very narrow strips were chosen could the program be con-
sidered with the longitudinal fin program? This would avoid the need
for log ratios by using

6 - *A%

PROBLEMS
(6.1) Determine the effect of fin dimensions, number of nodes and
iteration cut off values on the accuracy of Program 6.1. The influence
of length to thickness ratio is important to establish since the
boundary condition at the end of the fin may need to be modified to
 Problems 139

allow for end convection when a short, fat fin is considered. For this
case analysis shows

cosh m(l — x) + [ ^ J sinh m(l — x)

cosh ml + ( i — sinh ml

and

tanh ml + -:—
\km
Q = mkA60
1 + t ha nl m
' km '

It would also be interesting to consider two-dimensional effects in a

short, fat fin using the methods of Chapter 3.
(6.2) Write a program for a longitudinal fin array based on area
weighted fin efficiency. Compare the results with those of Program
6.2.
(6.3) Amend Program 6.3 to deal with an array of circumferential
fins.
(6.4) Consider the modelling of tapered strip fins. Figure 6.6
suggests one way of dealing with the varying cross-sectional area.

Figure 6.6
140 Finned surfaces

Compare the results with those obtained by using fin efficiency

graphs and by programming the analytical equation of Section 6.3.
(6.5) Model a tapered circumferential fin and compare the results
obtained with those obtained using fin efficiency graphs.
(6.6) Write a program for an array of tapered circumferential fins.
Compare the heat transfer rates obtained with this program with
those of Problem 6.3. Is there any advantage in the shape of the fin?
Is one lighter than the other to give the same heat transfer? Is there an
optimum fin shape to achieve a given heat transfer?
(6.7) A metal cylinder head on a small experimental engine is
required to dissipate lOkW by means of rectangular profile fins
across the total width of the head which is 200 mm square. The fins
may be up to 60 mm long. Write a program to determine a range of
choices of length, thickness and pitch. The pitch should not be less
than the fin thickness (see also Problem 6.10).

Data
Thermal conductivity of fin material 180 W/m 2K
Surface heat transfer coefficient 100 W/m K
Surface temperature at fin root 230 °C
Ambient temperature 28 °C
(6.8) To achieve a given heat transfer rate with an array of circum-
ferential fins with parallel faces is it better to use large or small outer
diameter fins if the total weight of the array is to be minimised?

Open-ended problems
(6.9) A thin steel tube of approximately d mm outer diameter is to
be manufactured with tapered strip fins along the length of the tube.
The fin pitch is to be such that there is an exact number of fins
around the outer circumference. The tube is to be used as a room
heater and is required to dissipate 1 kW from each metre run by free
convection when mounted horizontally as a 'skirting' heater. The
water in the tube is at a mean bulk temperature of 70 °C and flows
at a rate of 0.2 kg/s. The room temperature is 19 °C and the thermal
conductivity of steel is 48 W/m K.
Write a program to find if a suitable design with overall diameter
not exceeding 100 mm can be achieved. What other arrangement of
fins on a tube could be considered for skirting heating? Evaluate the
most likely alternative to find what size this would be.
(6.10) It has been assumed that the surface heat transfer coefficient
over a fin surface is constant. The determination of such a value
would require experiment in many cases. For example, annular fins
 Problems 141

around a tube in cross-flow would experience complex flow and

circumferential conduction would occur. The pitch of fins will also
be important because flow is basically over a flat surface and the
boundary layers developing on adjacent fins may interfere and cause
the surface heat transfer coefficient to decrease.
Consider rectangular longitudinal fins and write a program to
relate pitch, width, efficiency and heat transfer rate to boundary
layer thickness. Assume that such a fin can be treated as a plate and
that the hydrodynamic and thermal boundary layers both start at
the leading edge of the fin.
For laminar flow over
/ 2 a13flat
/ plate with constant wall
5 temperature:
Nu, = 0.332 Rei Pr for Re, < 5 x 10 , Pr > 0.1
and for turbulent flow:8 0 33 5
Nu, = 0.0288 Re? Pr for Re, > 5 x 10 , Pr > 0.5
where Nu, is the local Nusselt number hx/k and Re, is pVx/fi. All
properties are determined at the film temperature. The thickness of
the hydrodynamic boundary layer is
4.96*

and of the thermal boundary layer is

<5 =
t p l/3
r
where x is the distance from the leading edge.
If a set of fins is to be fitted to the 'sides' of a motor cycle engine
exposed to air flow, determine a suitable fin size and pitch to deal
with the worst expected running conditions.