[luzikato]

Hello, I'm simulating a thermoelectric generator with a mesoscale combustor as the main heat source. As we know, the thermoelectric generator must have a temperature difference to generate electricity. To ensure this condition, I use a cooling jacket on the cold side to absorb transferred heat from the heat side so the cold side keeps cooled.

My problem is when I try to simulate the whole system with the solid and fluid domain, I can't reach convergence. To see the problem, what I did was to simulate one system at a time and suppress the other system. (Only combustor then combustor plus the TEG).

I tried to simulate the combustor and TEG and it ran ok with good residuals. But when I try to include the cooling jacket my I can't reach convergence. So tried to run only the cooling jacket domain without energy equation, just the momentum. And as I predicted the cooling jacket domain is the one that makes my whole simulation didn't converge.

The cooling jacket consisted of one inlet and outlet. I use mass flow-inlet and mass flow-outlet as the boundary condition. I set the mass flow rate to 0.0665 kg/s as the experiment both at inlet and outlet. A pump is used to pump cooling water in and out of the cooling jacket. From the pump specification, I get the maximum flow rate is 240 L/h that's why I get 0.0655 kg/s as the mass flow rate to my cooling jacket simulation.

Any suggestion to solve my problem?

Thanks in advance. Best regards.