Thermal stress

DaveStupple

We have an assembly that will be operated with two surfaces at fixed temperature. I did a thermal steady state model with these two temperatures imposed to get a temperature distribution, then Loads & constraints>Transfer temperatures from solution. I set a thermal stress reference temperature at 25°C then solved as 3D static. It solved OK, but I think I have modelled cooling from the temperature distribution to 25°C. I want to be modelling the effect of operating at the set temperatures compared to room temperature. Would the effect be identical but with the signs on the stresses reversed? Is there a simple way to do it the right way round? Thanks for reading, Dave.

Victor

It's the other way round (reference temperature is the initial temperature) so it will already be doing what you want.

Yes, you can reverse the effect by negating the 6 stress components in linear analysis because stress is proportional to load and thermal load is proportional to temperature delta. Or negate all the temperatures (reference temperature as well as node temperatures).

DaveStupple

OK, that's good. Thanks Victor.