Temperature floating point exception
 

I receive the following error while simulating an ANSYS Icepak model:

Version: 15.0
The simulation is using double precision, and the temperature discretization schemes are standard for pressure and first for momentum/temperature.
The error occurs reliably ~20 minutes into a simulation that should last 8 hours (running on a HPC 2 core allocation).

Previous very similar models didn't have this problem. Any ideas? Thanks!

hi
 

Most of the times this error could be due to mesh / boundary conditions.
Keep an eye on the solver (fluent window). this will help you with identifying the reason. Whether the error message is reported during mesh check, or mesh split or solution initialization.

This can also happen after start of first iteration and unfortunately we won't be having any way to check the location of divergence. In such a case i suggest using a Top down appraoch.

Deactivate all assemblies leaving behind only one assembly. Then start solving. If it goes well, activate next and start again.

I most of the times adhere to this approach. This is a bit hectic and time taking but would surely give you a solution to the problem.. :)

give it a try and let me know.. :)