Boundary layer in parametric mesh


#1

Hi, im trying to create a mesh with a viscous boundary layer for laminar regine through a pipe system and when i create it with Hex-dominat automatic the results are good but i need heat transfer for the proyect that i’m working on so when i try to create the mesh with hex-dominant parametric the results are really different.

It is posiible to create a boundary layer in parametric mesh like the one in the automatic?
I show you here both images.

the first one, with Hex dominant automatic: (the one wanted)
Captura2
And here is the one with parametric mesh:
Captura

Here i put the link of the proyect:
https://www.simscale.com/workbench/?pid=7228316951275979770#tab_0-0


#2

Hi @joaquinhor623!

So you want to perform a fluid flow analysis and a heat transfer analysis separately if I got that right, correct?

The creation of boundary layers in the parametric mesh should be straightforward and I can adapt your settings but unfortunately you deleted the mesh as far as I can see. On top of that let me say that the other meshes you created with one being over 30 million cells is definitely an overkill! I would perform a mesh convergence study and see if you really need that many volume cells which I doubt.

Create again the mesh as depicted in the second picture and I will have a look at your settings then.

All the best!

Jousef


#3

Hi @jousefm

Sorry that i didn’t explain myself well.

What i want is to perfom together both the flow analysis and the heat transfer so it is a kind of solar collector, where a plate (solid 1) is heated and transfer this heat to a pipe system (solid 2), where inside there is water flowing (solid 3) so i need this boundary layer on solid 3 while solid 0 is the insulation.

The mesh name is “Par 5mm mesh”, and also if you could help me to reduce the number of cells would be great as when i tried to reduce the refiniment settings the surfaces of the solids were not straight (it had lot of defects).

Thank you for your help!
Joaquin


#4

Hi @joaquinhor623!

So we are going to perform a CHT (Conjugate heat transfer) analysis. Let me check that out in the evening!

Best,

Jousef


#5

Hi @jousefm

Yes, CHT.

Thank you againg, i’ll wait for your answer,

Best,
Joaquin


#6

Hi @joaquinhor623 and sorry for the delay!

What we need to do first is to decrease the refinement levels and make sure to establish the boundary layer cells first without wasting core hours. On top of that when selecting the tube everything including the inlet/outlet gets selected which we want to avoid. We want the tube without the inlet/outlet to be refined for the boundary layer refinement. For that you need to split the tube inside your CAD program or directly inside the workbench.

Please feel free to contact me if you have any further questions.

Best,

Jousef


#7

Hi @jousefm

I tried with one simple cilinder in order to try to obtain the boundary layer before continue but the only results that i achieved was this: Captura

while with the hex-dominant automatic i got this:Captura2

It is possible to get the boundary layer with the parametric mesh?

Thank you again for your help!

Best,
Joaquin


#8

Hi Joaquin (@joaquinhor623)!

Yes of course! Please have a look at this validation case here: Turbulent Pipe Flow.

Cheers,

Jousef


#9

Hi @jousefm

Great! thats what i need. But i can not find the “y+” value to get the mesh that i need.
Where do i change this?

Best,
Joaquin


#10

Hi @joaquinhor623!

This is a good tutorial: Generate YPlus Wall Spacing.

Best,

Jousef


#11

Hi @jousefm

I still don’t achieve it, could you help me with the settings in the cilender examplo that a created?

Thank you in advance,
Best,
Joaquin


#12

Hi @joaquinhor623!

Let me Tag @Get_Barried and @vgon_alves to help out here. Will also try to get the layers and will post the results here.

Best,

Jousef


#13

Hi @joaquinhor623,

Applying a inflation layer mesh refinement at the circular wall of the pipe will allow you to generate the layers you desire for the inlet and outlet. What values to input to the mesh refinement is based off what Jousef has mentioned in the application of y+ post.

Try it out and see if you are able to get the results that you desire. Let me know if you still have troubles.

Oh and don’t forget to keep the material point within the pipe (and also ensure the pipe is closed/watertight).

Cheers.

Regards,
Barry