[lord_rayleigh]

Hello. First post.

I have a large 3-D conjugate heat transfer problem (3.6 million elements). Basically there is gas flow within several layers of insulation and metal. On the outside of the metal, I am trying to apply convection heat transfer (T~500 ºC, h ~ 20).

The problem converges fine without the energy equation. However, no matter what BCs or initial conditions (patch) I use, the energy equation diverges after 10 iterations.

Things I tried:

- energy relaxation of 0.05
- coupled solver
- second order energy...first order energy...MUSCL energy...QUICK energy
- set all BCs to 30 ºC.... Patched all zones to 30ºC.

This last one is puzzling. There are no gradients in temperature.... why is it diverging?

Thanks for your help,
Mike

[lord_rayleigh]

Well, if anyone is interested, I think I don't have enough fluid elements near the walls. I will try automatic inflation.

However, the trick is to NOT produce more than 5 million elements, which seems to be the max my machine can handle.