Learning Forum

[tumulpurwar]

I would like to know, whether in new Ansys versions, adapt y+ criteria is valid for all types of cells.

As with old versiosn of Ansys Fluent, adapt y+ only valid for Tetrahedral cells.

Also if there is any better substitute  in place of  adapt y+ Adaption in new versions?

[Rob]

I don't remember that ever being a limitation, https://ansyshelp.ansys.com/account/Secured?returnurl=/Views/Secured/corp/v222/en/flu_ug/flu_ug_chp_adapt.html   The only issue I remember was trying anisotropic adaption on anything that wasn't a hex or prism cell. 

You can still create registers on y+, whether there is a better option will depend on why you think you shouldn't be looking solely at y+ 

[tumulpurwar]

I have gone through guide, but they didn't talk about Y+ adaption, other adaption they talked about. 

 

I already tried with register, 

But as I click adapt, it tells that polyhedral meshes can't be adapted. 

 

Why?? 

 

Tetrahedral it's accepting. 

[tumulpurwar]

Hi ROB

Is Adapted Tetrahedral can be converted in to polyhedral cells in Fluent?

[tumulpurwar]

[tumulpurwar]

[tumulpurwar]

[tumulpurwar]

For gradient based adaption for old versions, dynamic adaption is possible but only for tet.

Polyhedral dynamic adaption only coming from 2021 update

[Rob]

Which version of Fluent are you using? Poly adaption requires PUMA, which has been available for a while and is the default in the last few releases. 

[tumulpurwar]

Hi ROB

Just found my Enterprsie license contains both

But than, why its not adapting boundary layer cells )prism layer) for y+ adaption?

[tumulpurwar]

Some where i found this asnwer..is it true ROB?

 

I think i got the answer , i cant adapt boundary layer, even my mesh is created as polyhedral in Ansys Flent meshing

[tumulpurwar]

 

its 2019 R3..but i want know, i ran for a while my simulation, than tried to adapt y*, but why it gives answer that it cant adapt..?

PUMA I dont need for now, for my application, solution adaptive mesh refinement can work too

 

[Rob]

You can adapt the boundary layer, but there are rules for anisotropic adaption.  Please post exactly what panels you're using. 

[tumulpurwar]

Hi Rob

This is hull around which i have fine prism layer and currenntly after running some itaeration , my y is shown in photo,i want make it betwenn 30 and 300 , and due to this reason i want use y adaption, i am not sure, why you dont recommend y* adaption?

Anisotropic probably is expensive in terms of memory.and even if i use anoisotropic my y* will never come in range what i want

[tumulpurwar]

 

One doubt i have ROB,if 2019 R3 doesnot have “DYNAMIC y* adaption”

Why than it shows this box in adaption, i have even tick it.

FYI: My mesh is polyhedral, it should not show,if it doesnot have this feature, i think its incorrect

 

 

[Rob]

Dynamic adaption means Fluent will refine and coarse automatically as the crition are met. It's commonly used for free surface modelling where the free surface moves and we don't want to refine regions that we don't have to. 

If you need to refine near the wall you'll need to use anisotropic on the boundary layer. This has limitations (you must adapt the whole cell layer, and that layer must terminate at an external boundary or be "looped"). 

I have not recommended either y+ or y*, I am trying to explain adaption. 

[tumulpurwar]

Thnaks Rob, i understood what you said, but this is what i am asking, why i cant use y* adaption, if it is in fluent?

and why its not adapting prism layer in my case, when my mesh is only polyhedral ,made in fluent meshing?

Thanks

[tumulpurwar]

It means PUMA and Dynamic adaption in polyhedral is different things?

[tumulpurwar]

In my case i need to refine either prism layer or refine velocity gradient near hull wall, as well as need to refine with help of criteria pressure waves(sea waves hitting or resisting hull movement forward)

[tumulpurwar]

Also i dont want to refine further boundary layer, as you can see from my earlier photo of y* plot, i want to coarse and fine boundary layer so that in can come under 30 to 300, as i am using wall function, thats why i think anisotropic will not help here, what you say?

[tumulpurwar]

[tumulpurwar]

So i guess conclusion is

y* adaption for 2019 R3 license with polyhedral cells doesnot work for PRISM LAYERS

Anisotropic can work

For tetrahedral ,y* adapt can work

[tumulpurwar]

[tumulpurwar]

Hi Rob

I went through guide book, understood all issues now, thanks