Gauge pressure field started diverging

Hello all of you,

I saw this subject already in the forum a few times, but have not been able to resolve things. So I hope some one can help. It is about my project:

Simulation: Incompressible 6
Run 3 is using Realized k-epsilon and Run 4 is using k-omega SST.
Both have this ‘Gauge pressure field started diverging’; Run 3: at Gauge pressure = -1.04349e+15 at position: (-54.24 m, -76.35 m, 0.5804 m).
and Run 4 at: Gauge pressure = 1.41853e+15 at position: (-86.87 m, -46.04 m, 0.5278 m).
I get these errors also when changing the Mesh Fineness at 5, 7.5 or 9.

<the model worked (although I had a smaller height for ‘External Flow Volume’: in Simulation Imcompressible 6 it is 100m, while earlier [in Simulation: Realizable 225deg (7.09m/sec,terrain=10m)] it is 40m>

If I look at the Mesh I am not really able to see any strange things. But sometimes I get or a blue field or sometimes nothing, So it looks I don’t know how to debug this.
I wanted to set a point (like advised in: Divergence in Simulation: How to Tell When and Where? | SimScale ) but I can’t find this option: ‘Geometry primitives’.
The Mesh runs by the way gives this:
“The model contains faceted geometry. When Advanced concepts are used, the meshing pre-processing requires some boolean operations to be performed. The support for facet geometry is limited in this case.
This mesh meets our quality criteria.”

Any hint/help is welcome. THANKS.

All the best,

Victor

Hello @jjansman ! Thanks for posting your question on the Forum!

I’m currently investigating your project, and will get back to you as soon as possible.

Please all feel free to jump in if you reach a solution before my tests are concluded. :slight_smile:

Best wishes, Kaan

1 Like

Hello Kaan,

I have not yet been able to find a reason for this. I tried a few avenues, but to no avail. Hope you can help. Thanks.

All the best,

Victor

1 Like

Hello @jjansman , I hope you’re well and apologies for the delayed reply here.

I have noticed nothing unusual in terms of boundary conditions, however noticed there are some mesh cells that are beyond the acceptable quality range.

I would recommend having a mesh quality check first. Depending on the reasons, you can either apply mesh refinements to improve the quality, or you may need to perform cad clean-up operations. Please make sure that you’re referring to the following documentation page:

I hope this will be helpful.

I will have a check. And see if I can find these areas. Would you be able to show (in a grab) where “some mesh cells that are beyond the acceptable quality range.” perhaps I am overlooking this when checking the mesh quality? THANKS.

Hi @jjansman,

Applying the workflow in the link my teammate Kaan provided, you can see the faulty elements in your mesh. Using Volume Ratio as an example:

You just set the ideal max value (10) as the lower limit and the maximum volume ratio as the maximum limit!

I hope that helps :slight_smile:

Best,
Igor

This indeed helps!!! Now I see what to look for, it seems I had the min at a wrong value. I tried first 40 very thing becomes blue, then I used 50 and 60 that locates one locations. But changing it to 10 (as you did) I see more, Is ‘ideal min’ indeed the dieal? In the workflow I say 20 (which only shows one location).
it seems also I need to be patient;-) After change the min, it takes some time and then I get a blue space after some time, the blue is removed and I see one colored space:

Do I now need to change all locations where things are lighted in when using min=10 or could I start with the location that pops up at min-20 (or min-70) An

Anyway THANKS for being patient with me. At least I now can reproduce the things you see and thus I hope to be able to improve things. THANKS.
All the best,
Victor

1 Like

Hi @jjansman, you’re welcome :slight_smile:

The minimum value is taken from the Meshing Log, where you can inspect whether your mesh falls under the desired parameters:

I now checked Aspect ratio, Non-orthogonality, Volume ratio, Edge ratio, Cell volume and non of these show issues in the Mesh Quality at the locations mentioned in the Runs:
Run 3: at Gauge pressure = -1.04349e+15 at position: (-54.24 m, -76.35 m, 0.5804 m).
and Run 4 at: Gauge pressure = 1.41853e+15 at position: (-86.87 m, -46.04 m, 0.5278 m).

I was not able to find ‘Geometry primitives’ option to locate these points, so at the end I looked at an earlier successful run with the same model and with x, y, z (in Solution Fields) I found the locations. As said no issues were seen there in the Mesh quality view.

At the end I reduced the Roughness height of Custom (from 10m to 1m) and now I can Run it successfully. If I try Roughness heights 3, 5 or10m they all cause errors in run (I did not try 2 yet).

So I did not yet find a useful workflow to get it working.

I now got it working. I had not done Initial conidtions and enabling potential flow initialisation. And furthermore (perhaps the bigger error): I had no Inflate boundary layer (in the Mesh) for ground level (only Boundary conditions, but that is not enough).
When including these in the simulation, I get no Gauge pressure field started diverging anymore.

1 Like