[Anna Tian]

Hi,

When I use 'Workbench -> Table of Design Points' to run the simulation, I always have error messages like 'Update of the Solution component in simulation1 for design point3 failed: The solver failed with a non-zero exit code of : 2.'

The simulation is definitely able to converge. And if I use the button 'Update the selected Design points' to run the simulation again, there is 50% of the chance to obtain the correct solution on the table of design points (another 50% of the chance: just have the same error message as before).

And if i copy a design point to 'current', I will have 10% of chance to have this problem. The error message in solver is:

'

End of solution stage.

+--------------------------------------------------------------------+
| The results from this run of the ANSYS CFX Solver have been |
| written to F:/long_slit_porous_model/workbench calculation/heat |
| transfer |
| 1.5_pending_tasks/dp0_inlet_velocity_150_Solution/inlet_209.res |
+--------------------------------------------------------------------+


+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| Error removing file F:\long_slit_porous_model\workbench |
| calculation\heat transfer |
| 1.5_pending_tasks\dp0_inlet_velocity_150_Solution\ inlet_209.dir\m- |
| on.old: Permission denied |
+--------------------------------------------------------------------+


+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| Unable to remove working directory: Directory not empty |
+--------------------------------------------------------------------+


This run of the ANSYS CFX Solver has finished.


'

How to solve this problem?

[Anna Tian]

Originally Posted by Anna Tian

Hi,

When I use 'Workbench -> Table of Design Points' to run the simulation, I always have error messages like 'Update of the Solution component in simulation1 for design point3 failed: The solver failed with a non-zero exit code of : 2.'

The simulation is definitely able to converge. And if I use the button 'Update the selected Design points' to run the simulation again, there is 50% of the chance to obtain the correct solution on the table of design points (another 50% of the chance: just have the same error message as before).

And if i copy a design point to 'current', I will have 10% of chance to have this problem. The error message in solver is:

'

End of solution stage.

+--------------------------------------------------------------------+
| The results from this run of the ANSYS CFX Solver have been |
| written to F:/long_slit_porous_model/workbench calculation/heat |
| transfer |
| 1.5_pending_tasks/dp0_inlet_velocity_150_Solution/inlet_209.res |
+--------------------------------------------------------------------+


+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| Error removing file F:\long_slit_porous_model\workbench |
| calculation\heat transfer |
| 1.5_pending_tasks\dp0_inlet_velocity_150_Solution\ inlet_209.dir\m- |
| on.old: Permission denied |
+--------------------------------------------------------------------+


+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| Unable to remove working directory: Directory not empty |
+--------------------------------------------------------------------+


This run of the ANSYS CFX Solver has finished.


'

How to solve this problem?

Could anyone answer me this question?

[Halle]

Hi, have you solved this problem? I get the same error message.

[mvoss]

Whats your "initialization option" ? If the dp table fails in the very first iteration there is an error in the cfx-setup.

[Halle]

Thanks for your reply. I used "Update from Initial Conditions". And it's not the first but number 3 and 4 for instance.

I now used a bigger sample size for the Screening and something happened that might be useful to know: Most errors were within the first 40 DPs (which was the original sample size). From there on errors occured very rarely.

[mvoss]

It´s pretty hard to guess where the error comes from. Especially since the error warning is referring to the *.mon file (which is the monitor output for the cfx solver ) - and this file isn´t related to any kind of dp study.
Have you been "watching" the solver via the solver monitor or was the run completely in the wb-background ?

[bzoolim]

In "solution" menu, Just Mouse right click and push the "reset", restart!!