Heat flux balance in the solution is greater than the maximum tolerance

Noeline Tharshan

unread,

Sep 27, 2019, 3:42:06 PM9/27/19

to THERM

"the heat flux balance in the solution is greater than the maximum tolerance. Simulation results may not be valid"

What does this mean? How do I fix this?

Simon Vidanovic

unread,

Sep 27, 2019, 3:46:29 PM9/27/19

to Noeline Tharshan, THERM

Noeline,

simulation engine sometimes have problem to converge and it will try to achieve best possible tolerance. By default, tolerance is set to 1e-5 and if that has been reached, program will just report results without any message. However, if achieved tolerance is less than what is set then program should report tolerance achieved.

Regards,

Simon

On Fri, Sep 27, 2019 at 12:42 PM Noeline Tharshan <noeline...@kmai.com> wrote:

"the heat flux balance in the solution is greater than the maximum tolerance. Simulation results may not be valid"

What does this mean? How do I fix this?

--
You can view the THERM forum online at: http://windows.lbl.gov/software/THERM/forum
---
You received this message because you are subscribed to the Google Groups "THERM" group.
To unsubscribe from this group and stop receiving emails from it, send an email to LBNL-THERM+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/LBNL-THERM/7210ba2b-9e37-4b07-96ed-41d200f1759e%40googlegroups.com.

Noeline Tharshan

unread,

Sep 27, 2019, 4:14:47 PM9/27/19

to THERM

Hi Simon,

What is the best way to move forward with this? Simplify the model? Simplify the boundary conditions?

Kind regards,

Noeline

On Friday, September 27, 2019 at 3:46:29 PM UTC-4, Dragan Vidanovic wrote:

Noeline,

simulation engine sometimes have problem to converge and it will try to achieve best possible tolerance. By default, tolerance is set to 1e-5 and if that has been reached, program will just report results without any message. However, if achieved tolerance is less than what is set then program should report tolerance achieved.

Regards,
Simon

On Fri, Sep 27, 2019 at 12:42 PM Noeline Tharshan <noeline...@kmai.com> wrote:

"the heat flux balance in the solution is greater than the maximum tolerance. Simulation results may not be valid"

What does this mean? How do I fix this?

--
You can view the THERM forum online at: http://windows.lbl.gov/software/THERM/forum
---
You received this message because you are subscribed to the Google Groups "THERM" group.

To unsubscribe from this group and stop receiving emails from it, send an email to LBNL-...@googlegroups.com.

Simon Vidanovic

unread,

Sep 27, 2019, 4:37:42 PM9/27/19

to Noeline Tharshan, THERM

In Option->Preferences->Simulation you can check "Save Conrad Results File". Once that is checked, you can run the simulation again and this time program will create .O file. In that file you can find the following section:

********************************************************************************

*** s o l u t i o n ***

relax = 1.0000E+00
number of tries = 1
achieved convergence = 2.8837E-09
solution converged

This will tell you what tolerance that engine has achieved.

Alternatively, you can try to reduce convergence tolerance in Option->Preferences->Simulation and then change Convergence Tolerance to something bigger.

Best,

Simon

To unsubscribe from this group and stop receiving emails from it, send an email to LBNL-THERM+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/LBNL-THERM/1c89218b-67ff-40c3-80d8-171fefeb95fb%40googlegroups.com.

Reply all

Reply to author

Forward