[zeoffle]

Hello.

I'm going to use the properties calculated through the User Defined Real Gas Model (UDRGM) in Fluent.

I made the UDRGM by referring to the examples presented in the Fluent manual.

Compile proceeded without any problems. However, when use it to perform calculations, the following message appears and interrupts:

'Divergence detected in AMG solver : temperature'
'Divergence detected in AMG solver : pressure coupled'

In some solution to this problem, it said this may be caused by an inappropriate mesh.
However, the example UDRGM in the Fluent manual is calculated with the same mesh, it proceeded without interruption.
Also, someone said that the boundary condition can also be affected, but it was also calculated without interruption by using the example UDRGM with the same boundary condition.
Besides this, I checked if there was a problem writing the code, and there was nothing wrong with the code I wrote.

So, if anyone has done Fluent's calculations using UDRGM, can you tell me how to solve these problems or some tips related UDRGM?

Additionally, in the Fluent manual, it said the complexity of UDRGM can also affect, is this related to the number of variables used by UDRGM?

Thank you.

[AlexanderZ]

beliving that you've checked mesh and boundary conditions, we may assume, that problem is in UDRGM model
as UDRGM is complicated
it's hard to find the exact reson, which lead to divergence
so I recommned you to simplify equation inside the model as much as possible (you may use just constant values in the beggining)
once you confirm, that simplified model works -> move on and add equatios, so you may find the exact reason of divergence

[zeoffle]

Hi

Firstly, thank you for reply to my question.

And I also agree that complexity of UDRGM is one of the reasons of divergence.

After reading your advice, I re-checked my UDRGM code.
It is quite complicated then example in Fluent manual. Especially, it has so many variables and equations.

So, I will try your method to simplify the equation and code.

Thank you.

[zeoffle]

Originally Posted by AlexanderZ

beliving that you've checked mesh and boundary conditions, we may assume, that problem is in UDRGM model
as UDRGM is complicated
it's hard to find the exact reson, which lead to divergence
so I recommned you to simplify equation inside the model as much as possible (you may use just constant values in the beggining)
once you confirm, that simplified model works -> move on and add equatios, so you may find the exact reason of divergence

Hi

After reading your advice, I modified some equations simpler than I was first made.

And It works! There are no divergence! I'm really appreciating to you and your advice.

But there are another issue about properties' phase.

Calculating vapor phase properties using UDRGM are good. But I think that it doesn't calculate liquid phase properties.

So, I used some solutions in Fluent manual to solve this issue.

First, I changed the Boolean variable from 'vapor_phase' to 'Liquid_phase'. But it didn't work and results are diverged.

I also set phase in the text-interface command, that in Fluent manual.

'define/user-defined/real-gas-models/set-state'
'define/boundary-conditions/modify-zones/change-zone-state'

Also, it didn't work and results were diverged.

I re-checked equation and code, but there were no special error.

And I searched about this problem, there were not enough information.

So, If you or someone had experience about this problem and know how to solve this, Can tell some hint or solutions about how to set the UDRGM to calculate liquid phase properties?

Thank you.