Learning Forum

[Patrick Kirby]

I'm working with a mutliphase simulation of a bioreactor. The simulation involves an impeller that does not appear to lend itself well to the inclusion of inflation/refinement near the surface. The mesh statistics are overall much worse when I include inflation on this surface. I instead decrease the mesh size in this region and select the capture proximity and curvature options. I'm curious if this seems to be appropriate when modeling turbulent flow, in general. I'm asking from a best practices stand point.

 

Thanks

 

 

[Rob]

Inflation and multiphase can be an interesting discussion. Whilst I can't go into too much detail consider the effect of particles/bubble that are bigger than the near wall cell height. How does that compare with your y+ theory? 

Check the Learning AIS content and also the newer Knowledge Materials, https://ansyskm.ansys.com/   The latter are "public" versions of the Customer Portal content, and I'm fairly sure I saw some mention of bioreactors when checking the materials. 

[Patrick Kirby]

Hi Rob, thanks for directing me. I believe you are hinting at the need to have a grid size large enough to properly model the bubble dynamics and yet at the same time capture the flow features. I agree that this is an interesting problem.

Thanks and have a great one

 

Pat

[Rob]

Pretty much. It's more that if your cell is much smaller than the bubble to meet the y+ criterion how do you then justify the multiphase assumptions: ie why aren't you using VOF. VOF isn't the right choice here!