Register to reply

D^2T/dx^2 + d^2T/dy^2 + d^2T/dz^2 = C

by timsea81
Tags: d2t or dx2, d2t or dy2, d2t or dz2
Share this thread:
Jan30-13, 05:08 AM
P: 90
I'm trying to solve the heat conduction formula in 3 dimensions when there is constant generation from electrical resistance q'''. This creates a constant C on the right hand side that is equal to q'''/k.

d^2T/dx^2 + d^2T/dy^2 + d^2T/dz^2 = C

I found a solution using separation of variables for when the right hand side equals 0, but it doesn't work with a non-zero constant on the right, because you end up with:

X'''/x + Y'''/y + Z'''/z = C/XYZ
Phys.Org News Partner Science news on
Experts defend operational earthquake forecasting, counter critiques
EU urged to convert TV frequencies to mobile broadband
Sierra Nevada freshwater runoff could drop 26 percent by 2100
Jan30-13, 06:59 AM
P: 90
I think I got it, maybe. I can solve the homogeneous equation:

d^2τ/dx^2 + d^2τ/dy^2 + d^2τ/dz^2 = 0

and then assume the particular solution to have the form:

T = τ + Ax^2 + Bx^2 + Dx^2

That makes

d^2T/dx^2 + d^2T/dy^2 + d^2T/dz^2 = 0 + 2A + 2B + 2D,

So 2A + 2B + 2D = -C

and I can use boundary conditions to find A, B, and D

Feb1-13, 01:03 PM
HW Helper
P: 1,008
You need to start with a particular solution ([itex]Cx^2/2[/itex] will suffice, but if you expect your solution to have certain symmetry properties then it might be worth looking for a particular solution which shares those properties) and then add complementary functions to satisfy the boundary conditions.

Register to reply