hello there!
So i am trying to simulate my battery room in conjugate heat transfer analysis with fluid’s Viscosity model is sutherland. But it’s always running not completely. I tried to reduce the iteration, increase the maximum runtime, use the coarse meshing, and use boundary condition symmetry, but it’s still not running completely. So what should i do to make my simulation running completely with convergent results?
Thank you
Hi, make sure to post a link to your project so the community can help.
Cheers
sorry, the project link can not be shared. But i can give you the information that you need
It’s difficult to tell exclusively from the message but if the simulation is failing I would try to simplify the physics a little bit to get a stable solution and then add more physics as you go (if needed).
E.g. assuming you have a CFD use case that involves heat transfer start off with a simple “Incompressible” study (which is isothermal), then proceed to an incompressible CHTv2 study, then (if needed) add more elements to it.
Before i do this simulation, i have already simulated just batteries, rack and cooling block in conjugate heat transfer to know surface’s temperature of them and then i simulated my battery room with input that i get from previous simulation (surface temperature) in convective heat transfer to know ambient temperature of battery room. But i am not sure about the result and there is no relative humidity of battery room. So i tried to combine this simulations in one simulation, but when i am running the simulation, it’s always running not completely.
I was wondering if you could help me, I can share this project by adding your username or email to see my project and know what to do with my project. I would really appreciate it
¿What do you mean with the simulation is running “not completely”?
i mean the simulation is not running until finished. I have already used maximum runtime 30000s and 500 iterations
Could you also share your convergence plots?
The issue is that you ran out of runtime and are not achieving complete convergence, which causes the simulation to keep running until it converges or the runtime is consumed.
It is very difficult to give you further feedback without more details about the project. In my experience, for large meshes, using automatic sizing is not recommended, as it wastes resources by overrefining where it is not required and underrefining where a high-quality mesh is needed. This can result in divergence and stability problems, which can also cause your solution to fail to converge.
Jairo