Issue with assigning mesh as the domain

Thanks for reporting this @anurags!

This should usually be no problem, already reported that to our engineers to fix this issue. We will keep you up-to-date about the progress!

Jousef

I will run it and let you guys know. Thanks again to @anirudh2821998 for the effort.

1 Like

Hello,
Unfortunately, I have run exactly into the same problem. my case is here:

I have also tried to modify the case prepared by anirudh2821998, but it has a different problem: the blades have a rough surface * case: https://www.simscale.com/workbench/?pid=8801242825754213671&mi=spec%3A6b497a44-009d-4025-b3cf-00d8d3a5e0b3%2Cservice%3AMESHING%2Cstrategy%3A19&sh=2)

Do you know another way around that issue?

Thank you,
Stan

Hi @ephiopia93: Please take a look at small demo project I’ve just created. Internal CFD with MRF. This should allow you to sort out those problems.

Cheers,

Retsam

3 Likes

Retsem,
Thank you very much for your answer. I have tried to rerun your case, but the same issue occurs: the mesh is not assigned to the domain. In addition, using the method presented by [anirudh2821998] gives a poor mesh quality and crashes. I use all boundary conditions as in one of the examples I found in the public domain, so I suspect the problem is in the mesh. Could you help me with this issue?

Thank you,
Stan

Hi Stan,
I do not see, in the posts of this topic, where Retsam provided you a case to run, how were you able to re-run his case?

I opened your case in your post #7 and the first thing I see is that there are still 3 solids, in the only geometry, in that project.

In Retsam’s Post #8, he shows you a project that he created to help people currently having a problem with internal MRF simulations. Currently you must follow his recommendations and if you did, you would have only 2 solids in your geometry.

Please modify you geometry accordingly and try everything he suggests in his sample project…

Regards,
Dale

2 Likes

Hey guys!

The issue has been resolved and the error message “Mapping of assignments from geometry to mesh failed” should no longer appear. Let me know if you have any further questions.

1 Like

Hello @anirudh2821998,

During the assembly of body 1 and body 2, wouldn’t the surfaces of of blade cavity in body 1 overlap with the surfaces of blade cavity in body 2? So, though the surfaces will be joined to each other , they still act as two different surfaces. How to provide boundary conditions for both the surfaces, to make them act as one?

Thanking you.

Regards,
Yash Agiwal.

Hi.

I believe the project is already public:

image

Thanks,
Goku

Hi Goku!

Was a problem with the project name that it converted the link incorrectly. So I do not see any problems (let alone errors) inside your project - can you tell us what the issue was?

Best,

Jousef

Hi Jousef,

I was following the tutorial that I linked above, and set all of the values as it said to. But, when i tried to begin a simulation i got these two errors:

I am not sure on how to resolve the second error.

Thanks,
Goku

Thanks for reporting this Goku!

Will forward that to our engineers - we will tackle this as soon as possible. Getting back to you when I know more.

Have a nice weekend!

Jousef

I am still getting similar errors:

  1. Mesh cannot be selected for boundary conditions
  2. Prop inside the refinement boxes not being meshed - I tried both making it a solid and negative geometry inside the fluid domain.
  3. After meshing (and spending core hours) I get this message image

Here is the link to the project: https://www.simscale.com/workbench/?pid=3724086337920935062&mi=spec%3A49cdd096-10f0-4d5b-b222-ce2f8d931fd7%2Cservice%3ASIMULATION%2Cstrategy%3A24&ps=analysis%2FIncompressible--oneOf%2FboundaryConditions%2F0

Hope you guys have some insight

Hi @joshuaxdmb!

You can just click on the button to use the mesh as the domain, that’s fine. Also regarding your setup before you jump back to it.

You want the cylinder to rotate using the MRF approach, right? I would personally increase the domain to make sure the boundaries do not affect the results. CFD Squad, what is your take on that?

Best,

Jousef

You’re right domain will probably need to be larger, but I want to make it work first.

Using the mesh as domain still does not solve the problem tho. The prop is still not being meshed - neither as a part or as a cavity inside a part.

I am trying hex-dominant parametrix. It’s taken 12 core hours and only 50% done.

That’s not that much tbh. Could also be that your refinements are too coarse. CFD Squad, can you make some suggestions here?

All the best!

Jousef

Hi Retsam, I am trying to setup a simulation similar to your example, but for some reason it is not working. The propeller inside the rotating region (which is negative geometry) is not being meshed. Do you have any insights?
Here is the link to one of my setups - I have been trying different setups with small variations to see what happens: https://www.simscale.com/workbench/?pid=2591225129477023134&mi=spec%3A29848356-7db5-4d95-8729-f14339973ac8%2Cservice%3AMESHING%2Cstrategy%3A47&sh=24

Appreciate any help!

Hi @joshuaxdmb: it looks like you solved already the meshing problem in simulation ‘Incompressible 2’. Please, before running simulation, set ‘Potential foam initialization’ on in Simulation control.

Mesh could be better, as for the moment BL are not crated. But start by running a simulation is a simple setup and tune it later on.

Take care,

Retsam

It creates the mesh when I choose multiregion mesh, which is different from your setup. I think it is creating overlapping meshes, and the boundary layer does not seem to be created. Any insights?

I am also curious, does the name of the parts matter? Does the fact that I am linking directly to Onshape instead of uploading an STL matter, in your experience?

Thanks for the help!
-Joshua

Hi @joshuaxdmb,

Naming convention does matter: you need to use names which make sense. Now you have 3 geometries and two of them have the same name. This is not ‘tidy’ and will not sort it for you.

Please use the geometry where you have only two volumes: Geometry with fluid volume to simulate and MRF zone (a cylinder).

Current simulation fails because you have three volumes, propeller being a separate part. Please read again explanations in my project.

OnShape is a perfect source for directly importing geometries into SimScale. If I did it in STL in the past, it is not relevant at all.

Also, if you want to create a BL, you should calculate y+ value being > 40 for a given flow (use y+ calculators) and use Wall function.

I suggest you start all over again, using automatic meshing with BL. This should help you top run first simulations.

Take care,

Retsam

1 Like