Dear All,

I am modeling a 2D square cross-section geometry 2x2 copper. The copper has an energy source term of 1000 W/m3. The walls of the solid are exposed to ambient conditions: heat transfer coefficient, h, is 100 W/m2K, and free-stream temperature is 300 K.

The solver is pressure based, 2D, implicit, steady-state; energy equation is set. The energy residuals have an absolute criterion of 1e-09.

When the case is run, the solution converges to 1e-09. The temperature distribution for the convergence is uniform! When the heat transfer coefficient value is increased by a factor of 1000%, the residuals flatten out at around 1e-16 but the temperature distribution is the same!

This solution is physically impossible in not having a temperature gradient in a heat generation and convection boundary conditions. Any help in solving this matter would be greatly appreciated.

Thank you in advance.

Regards,

Damian

it seems that you set heat generation wrong. how do you set heat generation in your case? is the heat generation in your whole domain or it is set on the boundary layer?

But the heat generation is supposed to be in the whole domain, not the wall.

hi what is your initial temperature of the domain? i think there is no temperature difference between your ambient and domain.... try to create some temperature difference so that u can see the variation of heat transfer coefficient.

truly sudhir

Any idea of time being taken to reach steady state in the two cases (HTC =100 and HTC = 1000)? You may have seen Biot number of the model and seen the effect of thermal conductivity.

Yes but the issue I am having is that the domain's temperature is unknown. Only the heat source is applied. Is there any way to just apply the heat source without specifying temperature inside the domain?

you can try heat flux condition... i hope u know the heat flux of ur domain