Heat Transfer Simulation
Hello experts,
In a steady-state forced convection heat transfer simulation, is there a distinction between concurrently solving the flow and temperature equations versus solving the temperature equation subsequent to resolving the flow equations?
Find more posts tagged with
If nothing (material properties, etc) in the solution depends on temperature - there should be negligible difference between solving both in the same run versus solving flow first, then using restart to solve the thermal on that solved/frozen flowfield. With larger models that take a long time to run, two separate runs like this may reduce overall run time.
In certain scenarios, particularly in cases involving completely enclosed rotating flows, the energy equation may fail to converge when solved simultaneously with the flow equation. However, in such instances, convergence is often achieved when the energy equations are solved separately after the flow field has been solved or 'frozen'.
Could you please explain the reason behind this?
Altair EmployeeIn certain scenarios, particularly in cases involving completely enclosed rotating flows, the energy equation may fail to converge when solved simultaneously with the flow equation. However, in such instances, convergence is often achieved when the energy equations are solved separately after the flow field has been solved or 'frozen'.
Could you please explain the reason behind this?
Do you have an example set of input/Log files that shows this? It likely depends on the level of change the flowfield is still experiencing - if the flow is still changing enough that causes changes in the thermal field, we would see that behavior.
Altair EmployeeIf nothing (material properties, etc) in the solution depends on temperature - there should be negligible difference between solving both in the same run versus solving flow first, then using restart to solve the thermal on that solved/frozen flowfield. With larger models that take a long time to run, two separate runs like this may reduce overall run time.
If nothing (material properties, etc) in the solution depends on temperature - there should be negligible difference between solving both in the same run versus solving flow first, then using restart to solve the thermal on that solved/frozen flowfield. With larger models that take a long time to run, two separate runs like this may reduce overall run time.