[Sargam05]

Hello everybody,

I am solving a thermal stress analysis problem in OpenFOAM. In my case, there is an initially undeformed cube subjected to a uniform temperature rise ΔT=T-Tr
.
As per my understanding, for this problem I do not need to solve energy equation. Does it means that I need to give some fix scalar value of the temperature in momentum equation for solving stresses due to temperature rise only?

I appreciate if anyone could help me to understand the case.


Thanks in advance.

Best regards,
Sangeeta

[bigphil]

Hi Sangeeta,

Yes you are correct that since you know the temperature distribution you do not need to solve the energy equation. And then you would subtract the thermal stress term grad(3*K*alpha*deltaT) from the momentum equation.

However, to save you changing the code you could just set fixedValue T boundary conditions and solve the energy equation using solidDisplacementFoam or elasticThermalSolidFoam, it probably wouldn't add much computation time and will give you the same temperature and displacement distributions.

Philip

[Sargam05]

Originally Posted by bigphil

Hi Sangeeta,

Yes you are correct that since you know the temperature distribution you do not need to solve the energy equation. And then you would subtract the thermal stress term grad(3*K*alpha*deltaT) from the momentum equation.

However, to save you changing the code you could just set fixedValue T boundary conditions and solve the energy equation using solidDisplacementFoam or elasticThermalSolidFoam, it probably wouldn't add much computation time and will give you the same temperature and displacement distributions.

Philip

Hi Philip,

Thank you for the reply. Yes, I am using elasticThermalSolidFoam for this case. Do you think that only fixing temperature in T boundary condition is fine to get good results. Because I can only fix temperature in boundary patches for e.g.

TOP
{
type fixedValue;
value uniform 100;
}
BOTTOM
{
type fixedValue;
value uniform 100;
}
LEFT
{
type fixedValue;
value uniform 100;
}
RIGHT
{
type fixedValue;
value uniform 100;
}
FRONTANDBACK
{
type fixedValue;
value uniform 100;
}

Does it means it rising uniform temperature in whole cube geometry?

Kind regards,
Sangeeta

[bigphil]

Hi Sangeeta,

If you fix the boundary patch temperatures and then set the ddtScheme for T to be steadyState then the internal temperature field will be uniform 100.

Philip

[Sargam05]

Hi Philip,

Thanks a lot for the information. I will try this.

Kind regards,
Sangeeta