TP 11-1
TP 11-1
TP 11-1
Derive the unsteady state heat conduction equation for three dimensional flow:
Heat can only pass through the faces of the element by conduction.
T
K y. z
Rate of heat input through ABCD at position x = x
T
K (T x) y. z
Rate of heat output through EFGH at position (x + dx) = x x
T
K x. z
Rate of heat input through ADHE at position y = y
T
K (T y ) x. z
Rate of heat output through BCGE at position (y + dy ) = y y
T
K x. y
Rate of heat input through ABEF from position ‘z’ = z
T
K (T z ) x. y
Rate of heat output through ‘DCGF’ at position (z + dz) = z z
T T
K y z K x. z K T x. y
x y - z
T T T
K x (T x x) y. z K y (T y y ) x z K z (T z z ) x. y
=
T
cp x. y. z
t
2T 2 T 2T T
2 2 2
α x y z t
K
cp
Where α, thermal diffusivity = m2/s
This is the three dimensional unsteady state heat conduction equation.
Types of boundary conditions:
The values of the dependent variable are given at all points on a particular boundary
then the condition is said to be of the first type.
Consider the unsteady state Heat condition in one dimension equation.
2T T
x 2 t where T f x, t
The boundaries will be defined as either fixed values of x or fixed values of ‘t’.
Two boundary conditions of the first type for this problem are:
T f x
(i) at t = 0 , ---(1)
T g t
(ii) at x =0, ---(2)
Equation (1) gives temperature at all values of x for one value of ‘t’. Equation (2) gives the
behavior of the temperature at all times at one value of x.
The boundary will be defined by keeping one independent variable constant, but
instead of giving the direct behavior of the dependent variable, its derivatives is restricted by
a boundary condition of the second type.
This situation occurs with radiant heating or cooling of a surface, electric heating of a
surface, or thermally insulated surface.
In these cases the heat flow rate in known but not the surface temperature.
But the heat flow rate is related to the temperature gradient.
This is the second type, boundary conditions
T
At x=0, if it is insulated, x =0.
In this type of boundary condition, the derivative of the dependent variable is related
to the boundary value of the dependent variable by a linear equation.
Ex: The surface rate of heat loss is government by a heat transfer coefficient.
Fig. 11.2 Unsteady - state Boundary element
h T T0
Rate of heat removed from the surface per unit Area =
T
K
The rate of heat is conducted to the surface internally per unit are =- x
These two rates must be equal.
T
h T T0 K
x ---(3)
Equation (3) in true for any surface which faces in the negative direction of the x-axis, but the
sign of the right hand side must be reversed if the surface focus is in the position direction.
The sign must not be changed according to whether the surface is being heated or
cooled.
The only negative signs arise from the following two conditions:
(i) The heat condition equation contains a negative sign by definition.
(ii) The outward normal from the surface may be in the negative coordinative
direction.
Equation (3) contains both of these negative signs and is correct as written.
For ordinary differential equation the number of boundary conditions is equal to the
order of the differential equation.
But for partial differential equation no such rule has yet been discovered.
But the unsteady state heat conduction equation in one dimension flow is:
2T T
x 2 t
Two boundary conditions are needed at fixed values of x and one at a fixed value of ‘t’.
When only one condition is needed in a particular variable it is specified at one fixed
value of that variable.
When two are more conditions are needed,
Note:
1. Laplace transform can only be applied to initial value problems because the variable
removed by the transformation must have an open range.
2. the numerical methods are seen that the calculation techniques are quite different
depending on whether the problem is initial value or boundary value and that the
methods are not inter changeable.
But q x.t
n
q n
t
x
T T q T
. tn
x q x q ---(3)
2T n T q n T q
2
t
t
x 2 q q x q 2 x
q n
t
But x
2T 2n T
2
t
x 2 q 2 ---(4)
Now partially differentiate the equation (2) WRT ‘t’
T T q T
. .nx.t n 1
t q t q
T dT
.nx.t n1
t dq
q
q x.t n x
But tn
T dT t n1
.nq. n
t dq t ---(5)
From (1), (4) & (5)
d 2T q dT
t 2n 2
n n t n1 .
dq t dq
d 2T dT
2
nq t n 1 n 2 n .
dq dq
d 2T dT
2 nq t 1 2 n .
dq dq ---(6)
The above steps are valid for any constant value of n , and in particular it is true for
1 1
n
2 (n can be eliminated by choosing the value as 2 ) ---(7)
Form (6) and (7)
d 2T 1 dT
2
q .
dq 2 dq ---(8)
d 2T 1 dT
2 q .
dq 2 dq
After the Integration
dT q 2
q
A. e 4 T B erf C
dq 2 ---(9)
Where A is an integral constant
1/2
But q x.t x.t
n
x
T B erf C
Or 2 t ---(10)
Where B and C are arbitrary constants. These can be found by using the boundary
conditions.
2T T
x 2 t ---(1)
The boundary condition are
T T1 At t 0, x 0
T T0 At x 0, t 0
T T1 1 T
Take non-dimensional term , T0 T1 x T0 T1 x
2T 1 2 1 T
2 and
x T0 T1 x 2 T0 T1 t t ---
(2)
from (1) and (2)
2
x 2 t ---(3)
With the boundary conditions
T1 T1
T T1 0, At t 0, x 0
For T0 T1 ---(4a)
T0 T1
T T0 1, At x 0, t 0
T0 T1 ---(4b)
The equation is
2 1
f q where q x.t n and n
x 2 t , 2
x
B er f c
2 t ---(5)
T T1 x
f B er f c
T0 T1 2 t ---(6)
T T1 T1 T1
0
T0 T1 T0 T1
x
B er f c
0 2 .0 B C 0 , since erf (∞) =1
B C B 1 ---(8)
T T1 x
1 er f 1
T0 T1 2 .t
T T1 x x
1 er f er fc
T0 T1 2 .t 2 .t
OR
T T1 T0 T1 x
1 er f
T0 T1 2 .t
T T0 x
1 er f
T0 T1 2 t
T T0 x
er f
T1 T0 2 t
Solution:
For constant properties, the differential equation for the temperature distribution in y-
direction with time ‘t’
T1
2T 1 2 1 T
and
x 2 T1 T0 x 2 T1 T0 t t (3)
from (2) and (3)
2
x 2 t (4)
With the boundary conditions
T0 T0
T T0 0, At t 0, y 0
For T1 T0 (5a)
T1 T0
T T1 1, At y 0, t 0
T1 T0 (5b)
2
y 2 t where = F(x,t) (6)
Now assume that the solution of equation (1) can be written as:
f q
(7)
But q y.t
n
q
tn
y
q
. tn
y q y q
2 n q n q
2
t t
y 2 q q y q 2 y
q
tn
But y
2 2n
2
t
y 2 q 2 (8)
Now partially differentiate the equation (2) WRT ‘t’
q
. .nyt n1
t q t q
d
.ny.t n 1
t dq
q
q y.t n y
But tn
d t n 1
.nq. n
t dq t (9)
From (1), (8) & (9)
d 2 q d
t 2n 2
n n t n 1 .
dq t dq
d 2 d
2
nq t n1n2 n .
dq dq
d 2 d
2
nq t 12 n .
dq dq
The above steps are valid for any constant value of n , and in particular it is true for
1 1
n
2 (n can be eliminated by choosing the value as 2 )
d 2 1 d
2
q .
dq 2 dq (10)
d 2 1 d
q .
dq 2
2 dq
After the Integration
d q 2
q
A. e 4 B erf C
dq 2 (11)
Where A is an integral constant
1/2
But q y.t y.t
n
y
B erf C
Or 2 t (12)
Where B and C are arbitrary constants. These can be found by using the following boundary
conditions.
T0 T0
T T0 0, At t 0, y 0
For T1 T0 (5a)
T1 T0
T T1 1, At y 0, t 0
T1 T0 (5b)
From equation (12) and (5a)
y
0 B erf C
2 0
0= B + C, B = - C (13)
From equation (12) and (5b)
0
1 B erf C
2 t
, erf (0) = 0
1=0+C
Therefore c = 1 and B = -C = -1, then the solution is
y
1 erf 1
2 t
y y
1 erf erfc
2 t 2 t
----------------
Consider a slab of finite thick ness in x direction and extending indefinitely in the y and z
– directions.
Initially the slab is at uniform temperature. If raise the temperature on both sides of the slab,
then to solve the problem we can use separation of variables method.
T F x, t
Let ---(1)
Where T is a dependent variable
x & t are the independent variables.
2T
f 11 x, .g t
x 2 ---(4)
T
f x , .g 1 t
And x ---(5)
f x
is a function of ‘x’ only, the differential can be performed with respect to x.
Now substitute equation (4) and (5) in equation (2)
f 11 x, .g t f x .g 1 t
Then ---(6)
f x .g t , thus
The variable dependence can be separated by dividing by
f 11 x, f x .g 1 t
f x g (t ) f x .g t
f 11 x g1 t
f x g t
x ---(7)
f 11 x
f x
Here the left hand side of equation (7) is independent of ‘t’ and the ratio of must
there fore be constant for all time ‘t’
g1 t
g t
is independent of x and the ratio must be constant for all ‘x’.
f 11 x g1 t
2
f x g t
d2 f 2 dt
f 11
x f x 0
2
. or dx 2 dx 0
---(8)
dg
g 1 t 2 g t 0 . or 2 g 0
And dt ---(9)
Equations (8) & (9) are both linear ordinary differential equations with constant coefficients.
T A cos x B sin x e t
2
---(15)
Case (ii): if 0
2
f 11 x
0
f x f 11 x 0 f 1 cons tan t
c4
f c4 x c5 ---(16)
g1 t
0 1
g t g t 0 g c6
And ---(17)
The solution of equation (2) is
T c4 x c5 c6
=
A0 x B0
Where A0 c6 c4 , B0 c5c6
Solution:
Here heat is lost at the two faces and since the solid extends indefinite in the y and z –
direction. It is apparent that the heat flows only in the x –direction.
2T T
x 2 t ---(1)
The boundary conduction are:
BC-1:
T T0 at x 0, t 0 .........(2 A)
T T0 at x 2 R, t 0 ......(2 B )
T T1 at t 0, x 0 .........(2C )
T T0 at t , x 0 ........(2 D)
And
T T0
Let take , T1 T0
T 1
.
x x T1 T0
2 2T 1
2.
x 2
x T1 T0 ---(3)
1 T
And t T1 T0 t ---(4)
from equation (1), (3) & (4)
1 2
.
T1 T0 x 2 t ---(5)
Then the B.C can be redefined as:
For T T0 0 at x 0, t 0 ---(6A)
T T0 0 at x 2 R, t 0
---(6B)
T T1 1 at t 0, x 0 ---(6C)
And T T0 0 at t , x 0 ---(6D)
Now the equation is:
2
.
x 2 t ---(6)
F x, t
Where .
Then the solution by the method of separation of variable can be expressed as:
f x .g t
---(7)
f 1 x .g t
Then x
2
f 11 x .g t
x 2
---(8)
f x .g 1
And t ---(9)
From equation (6), (8) & (9).
. f x g t f x .g t
---(10)
. f x .g t
The variable dependence can be separated by dividing by a , thus
. f 11 x g t . f x .g 1 t
. f x .g t . f x .g t
f 11 x g t . f x .g 1 t
f x g t . f x .g t
=
f 11 x g1 t
f x g t
(11)
f 11 x
f x
Here the left hand side of equation (11) in independent of ‘t’ and must therefore be
constant for all time ‘t’.
g1 t g1 t
g t g t
- is independent of ‘x’ and the ratio of must therefore be constant for all ‘x’.
f 11 x g1 t
2
f x g t
f 11 x 2 f x 0
---(12)
g 1 t 2 g t 0
And ---(13)
Then the solutions are:
If 0 :
2
A cos x B sin x e t
2
---(16)
If 0
2
f 11 x
0 f 11 x 0 f 1 cons tan t
f x
f c4 x c5 ---(17)
g1 x
0
g x g 1 x 0 g c6
---(18)
the solution of the equation (5) is
c4 x c5 c6
A0 x B0
---(19)
A cos x B sin x e t
2
0 A .0 B .0 e t
2
2
A .e t 0 A 0
2
e t 0
BC-II:
0, at x 2 R, t 0
0 0 B sin .2 R e t
2
2
B e t .sin .2 R 0 sin .2 R 0
.2R n
n
2R ---(20)
Where n- is a non zero integer.
The solution of equation (5) is
n 2 2
n . 4 R 2 t
Bn sin .x .e
2R ---(21)
Where n- is an integer and subscript has been attended for connivance
The general solution of this type can be written as:
n 2 2
n . 4 R 2 t
Bn sin .x .e
n 1 2R ---(22)
Bn can be determined by using the initial boundary conditions.
1, at t 0, x 0 (BC-III) ---(23)
By using the B.C .III
n x
Bn sin . .1
1= n 1 2R
n x
Bn sin
n 1 2R ---(24)
m x
sin
Now equation (24) is multiplied both sides by 2 R and integrate the result from 0 to 2R
m x n x m x
2R 2R
sin
2R
.dx B n sin
2R
.Bn sin
2R
.dx
0 = n 1 0 ---(25)
0 if m n
m x
2R
1
Bm . sin .dx
R 0 2R
---(27)
The solution from (20) & (26)
n 2 2 2 R
1 n x . 4 R2 t n x
sin .e sin dx.
n 1 R 2R 2R
0 …(27)
n x
2R
n 2 2
1 2R n x . 4 R 2 t
1 1 .sin
n
.e
n 1 R n
2R
2 1 1
n
n 2 2
n x . 4 R2 t
.sin .e
n 1 n 2R is the required solution.
3.5.2 A slab occupying the space between y=-b and y=+b is initially at temperature
T0. At time t=0 the surface at y b are suddenly raised to T1 and maintained there. Find
T ( y, t ) .
Solution:
One dimensional unsteady state Heat conduction equation is:
T 2T
2
t x ---(1)
T T1
Take T0 T1
2
2
t y ---(2)
g1 t
g t
Lllly is independent of ‘y’ and the ratio must be constant for all ‘y’
f 11 x
2
f x f 11 y 2 f 1 y 0
---(11)
g1 t
1
g t g t 2 g t 0
Similarly ---(12)
Where is the separation constant. For satisfy the boundary condition, 0 equation
2 2
(11 and 12) are both linear ordinary differential equation with constant coefficient.
(i) For 0
2
d2 f dt
2
2 0
dx dx ---(13)
The above equation (13) solution is:
f c1 cos y c2 sin y ---(14)
Then the solution of equation (12)
dg
2 g 0
dt
dg
2 dt log g 2t log c3
dt
g 2 2
e t g c3e t
c3 ---(15)
If 0
2
A cos y B sin y e t
2
---(17)
g1 t
0
g t g t 0 g c6
1
And
f .g c4 y c5 c6
A0 B0 y ---(18)
For 0
2
A cos y B sin y e t
2
For 0
2
A cos y B sin y e t
2
0 A cos 0 B sin 0 e t
2
2
A e t
=
2
t
Here e 0
A 0
0 A cos y B sin y e t
2
2
B sin 2 B.e t 0 sin 2 B 0
2B n
n
2B
n
2B ---(19)
The solution of equation (2) is
n2 2
n . 2 t
Bn sin . ye 4 B
2B
The general solution of this type can be written as
n 2 2
n . 4 B 2 t
Bn sin . y .e
n 1 2B ---(20)
=
if mn
But 2
---(23)
sin .dy B . 2 B
2R m
2
0 = (: m=n)
m y
2B
1
Bm . sin .dy
B 0 2R
2B
1 m y 2B
Bm cos
B 2 B 0 m
2 2
cos n cos 0 cos m 1
m n
2 2
1 cos m 1 1 m cos 1
m m
2
1 1
m
Bm
m ---(24)
The solution of the equation from (20) and (24).
2 2
n
T T1
n . t
Bn sin
2
. y.e 4 B
T0 T1 = n 1 2B
n 2 2
2 n . t
. 1 1 sin
n 2
. ye 4 B
n 1 n
2B
T T1 2 1 1
n
n 2 2
n . 2 t
.sin . ye 4 B
T0 T1 n 1 n 2B
--------x------
Solution:
One dimensional unsteady state heat conduction in y- direction
2T T
y 2 t (1)
q0
2Q Q q
y 2
t where Q = q0 (3)
Equation (3) can be solved with the following initial and boundary conditions by using the
combining two independent variables in to one variable method.
Here Q = F(y,t)
At t = 0, q = 0 Q=0
T
At y = ∞ y =0 Q = 0
q y
erfc
Q= q0 = 2 t
T y
erfc
But q= -k y = q0 2 t
T q erfc y
- y = k 2 t
q0 y
ierfc
(T – T0) = k 4 t 4 t
-----------x----------