Heat Loss in Fins
Heat Loss in Fins
Heat Loss in Fins
\
|
Q
dx
dT
k
dx
d
where Q represents the internal
heat generation, in W/m
3
.
Note:
In general, thermal conductivity, k
varies along the x-direction.
For isotropic, homogeneous
material, k is uniform and has the
same value in all directions.
(7-1)
The Governing Equation
( ) ( )
( )
=
T T
A
ph
dx A
T T h dx P
Q
c
c
SME 3033 FINITE ELEMENT METHOD
Finite Element Modeling
Since heat transfer through thin fin is assumed one-dimensional, we will
model the fin using one-dimensional heat transfer elements.
The fin is discretized into three elements, as shown.
A single element will be in a local coordinate system.
Global coordinate
Local coordinate
SME 3033 FINITE ELEMENT METHOD
The temperature varies from T
1
at node 1 to T
2
at node 2 in an element.
To predict temperature at any point between the two nodes we need to
establish a temperature function.
For simplicity, it is assumed that the temperature varies linearly within the
element. Therefore, we establish a linear temperature function in the
form,
( ) | |{ }
e
T N T N T N T = + =
2 2 1 1
1
2
1
;
2
1
2
N
N
| |
=
|
\ .
+
| |
=
|
\ .
Temperature Function
where N
1
and N
2
are linear shape functions given by
(7-2)
(7-3)
SME 3033 FINITE ELEMENT METHOD
Element Conductivity Matrix
A one-dimensional heat transfer element is used to model thin fins, which is
similar to the element used to model the plane wall (Chapter 6).
Therefore the element conductivity matrix for the plane wall can be used
for the thin fin, so that
| | | |
2
1 1
W
1 1 m K
e e
e
T T
fin wall
e
k
k k
l
(
= =
(
where k is thermal conductivity of the fin material (W/m K) and l
e
is the element
length.
(7-4)
SME 3033 FINITE ELEMENT METHOD
Element Heat Transfer Matrix
In thin fins, heat is dissipated primarily by convection heat transfer.
The effect of conduction through the fin is represented by the conductivity
matrix [k
T
] just established.
To include the effect of convection heat transfer, we need to establish a
heat transfer matrix, [h
T
].
It can be shown that the element heat transfer matrix is given by,
| |
2
2 1
W
1 2 3 m K
e
e
T
hl
h
t
(
~
(
where,
h = heat transfer coefficient (W/m
2
K)
l
e
= element length (m)
t = thickness of the fin (m)
(7-5)
SME 3033 FINITE ELEMENT METHOD
Exercise 1
A circular rod of diameter D = 5 mm and length L = 190 mm has its base
maintained at T
b
= 100C. It is made of copper with thermal conductivity k =
398 W/m K. The surface of the rod is exposed to ambient air at temperature
T
~
`
)
where,
T
+ =
where [k
T
]
e
= element conductivity matrix (due to conduction);
[h
T
]
e
= element heat transfer matrix (due to convection);
{T}
e
= nodal temperature vector (unknown);
{r
}
e
= element heat rate vector (due to convection).
(7-7)
Note:
If the fin is insulated, then the magnitude of both [h
T
]
e
and {r
}
e
will be zero.
SME 3033 FINITE ELEMENT METHOD
Exercise 2
Reconsider Exercise 7-1. a) Assemble the global heat rate vector for the
rod; b) Write the global system of linear equation for the problem.
A protruding rod, which can be treated as a pin fin.
SME 3033 FINITE ELEMENT METHOD
Thermal Boundary Conditions
Temperature at the base of the fin, T
b
is usually assumed to be the same
as the temperature of the structure.
The tip of the fin can be assumed as adiabatic or insulated. Therefore, the heat
flux q = 0. Hence, the boundary conditions for the fin are,
at 0, i.e. specified temperature
0 at , i.e. specified heat flux
b o
T T x
q x L
= =
= =
Global coordinate
0 b
T T =
SME 3033 FINITE ELEMENT METHOD
Exercise 3
Reconsider Exercise 7-2. a) Apply the boundary conditions; b) Solve the
modified global system of linear equations for the unknown temperature
distribution in the rod.
A protruding rod, which can be treated as a pin fin.
SME 3033 FINITE ELEMENT METHOD
The heat dissipated, H
e
, from a single element can be estimated using the
Newtons law of cooling, given by
( )
e avg s
H h T T A
=
where
The total heat dissipated from the fin,
1
n
e
e
H H
=
=
=
1 1 0 0
1 2 1 0
0 1 2 1
0 0 1 1
10 33 . 3
360
2
T
K
1. Write the element conductivity matrices
Since the fin is made up of homogeneous material and that all elements
have the same length, the conductivity matrix for all elements will be the
same, i.e.
2. Assemble global conductivity matrix, we get
1 2 3 4
Connectivity with the
global nodes.
SME 3033 FINITE ELEMENT METHOD
| |
( )
| |
( )
| |
( )
(
= = =
2 1
1 2
001 . 0 3
10 33 . 3 9
2
3 2 1
T T T
h h h
| |
(
(
(
(
=
2 1 0 0
1 4 1 0
0 1 4 1
0 0 1 2
001 . 0 3
10 33 . 3 9
2
T
H
3. Write the element heat transfer matrices.
Since all elements have the same length and thickness, the heat transfer
matrix for all elements will be the same, i.e.
4. Assemble global heat transfer matrix
1 2 3 4 Connectivity with the
global nodes.
SME 3033 FINITE ELEMENT METHOD
{ }
( )
{ }
( )
{ }
( )
)
`
= = =
1
1
001 . 0
20 10 33 . 3 9
2
3 2 1
r r r
{ }
1
2
2
1
001 . 0
20 10 33 . 3 9
2
R
5. Write the element heat rate vector
Since all elements have the same length and thickness, the heat rate
vector for the elements will be the same, i.e.
6. Assemble global heat rate vector, we get
1
2
3
4
Connectivity with the
global nodes.
SME 3033 FINITE ELEMENT METHOD
7. Write the global system of linear equations, in the form
1
2
2
2
3
4
2
1 1 0 0 2 1 0 0
1 2 1 0 1 4 1 0
360 9 3.33 10
0 1 2 1 0 1 4 1 3.33 10 3 0.001
0 0 1 1 0 0 1 2
1
2
9 3.33 10 20
2 0.001
1
T
T
T
T
(
( (
(
( (
(
( (
+
`
(
( (
(
( (
(
)
=
`
)
| | | | ( ){ } { }
T T
K H T R
+ =
We have,
SME 3033 FINITE ELEMENT METHOD
8. Impose the thermal boundary conditions.
1
2
2
2
3
4
2
1 1 0 0 2 1 0 0
1 2 1 0 1 4 1 0
360 9 3.33 10
0 1 2 1 0 1 4 1 3.33 10 3 0.001
0 0 1 1 0 0 1 2
1
2
9 3.33 10 20
2 0.001
1
T
T
T
T
(
( (
(
( (
(
( (
+
`
(
( (
(
( (
(
)
=
`
)
Given, T
1
= 235 C. Using the elimination method, we delete the 1
st
row
and column of the global SLEs, and modify the right side term accordingly.
We have,
SME 3033 FINITE ELEMENT METHOD
2
2
3
2
4
2
2
2 1 0 4 1 0
360 9 3.33 10
1 2 1 1 4 1
3.33 10 3 0.001
0 1 1 0 1 2
2
9 3.33 10 20
2
0.001
1
1
360 235
0
3.33 10
0
T
T
T
(
( (
(
( (
+
`
(
( (
(
( (
)
=
`
)
2
1
9 3.33 10 235
0
3 0.001
0
) )
9. Write the modified system of linear equations
Imposing the boundary condition, the system of linear equations is
reduced to,
SME 3033 FINITE ELEMENT METHOD
2
o
3
4
211.7
197.0 C
192.2
T
T
T
=
` `
) )
2
1
2
2
2
3
235 211.7
9 20 2 1 3.33 10 121.9 W
2
211.7 197
9 20 2 1 3.33 10 110.5 W
2
197 192.2
9 20 2 1 3.33 10 104.7 W
2
H
H
H
+
(
= =
(
+
(
= =
(
+
(
= =
(
10. Solve the modified system of linear equations
11. Compute the heat loss from each element
Solving the modified system of linear equations yields the unknown
global nodal temperatures,
SME 3033 FINITE ELEMENT METHOD
12. Compute the total heat loss
The total heat loss from the fin is,
1 2 3
121.9 110.5 104.7
337.1 W
H H H H
H
H
= + +
= + +
=