Talk:2d thermal expansion model: Smith

A few issues (Lurbano 13:19, 10 April 2008 (CDT))

in the second line of the 2D diffusion equation temperature you cancel out dx, however since dx is in the numerator on one side of the equation, and the denominator on the other side you should not do this. Propogate the dx from the dT/dx term and you should end up with a dx² in the denominator of the S term in the final equation.
Double check your C value. Based on [Wikipedia's specific heat page] the value for granite is 0.79 J/(gK), which needs to be converted to J/(kg K) by multiplying by 1000.
The thermal conductivity should be returned to 3.0 W/(mK)
Reduce the timestep to 100 yrs to become stable
Finally, you'll need to multiply the input heat flux "Q[20,20] = K * ((NT-T[21,21])/dz_crust[21,21]) * dx * dx" by a factor somewhere around 5000.0 in order to see a significant effect. I'm not sure why yet but I'll look into it tonight and get back to you.

A Few Issues Addressed (User:Ajsmith1 23:31, 10 April 2008 (CDT))

I made the above changes and also attemped to attach Q to the ball. The error produced is: rastermap has no property get.