0
$\begingroup$

I watched this video about CFD (

) where flow was assumed to be incompressible and inviscid, but how can vorticies occur behind the ball at the start of the video, shouldn't air in case of inviscid flow smoothly leave the ball

$\endgroup$

2 Answers 2

2
$\begingroup$

Euler CFD solutions are subject to something called artificial (or numerical) viscosity. How much you have depends on whether your method is odd or even order -- dictating whether a diffusive or dispersive error term dominates.

(sometimes more artificial viscosity is added to an Euler code, but even if you do nothing, there will be some).

This is what allows an Euler code to separate flow from a sharp corner -- instead of achieving infinite acceleration and staying attached like a potential flow code will do.

I believe the artificial viscosity plays a role in setting up a Von Karman Vortex Street in an otherwise inviscid simulation.

$\endgroup$
2
  • $\begingroup$ Does numerical viscosity have the potential to be used to estimate pressure drag? In other words, does numerical viscosity have some physical background? $\endgroup$ Mar 30 at 6:16
  • $\begingroup$ In certain cases, where the separation point is easy to predict (such as a flat aft face, or reasonably sharp corners), then yes, it will get the base drag reasonably correct. However, for something like a cylinder or a streamlined body (with separation), it will likely get the separation point wrong and will therefore mispredict the separation drag. $\endgroup$ Mar 30 at 17:57
1
$\begingroup$

This is a well known phenomenon, a Kármán vortex street. Wikipedia explains it adequately. I won't quote that entire article, though, daring for now to post a link-only answer.

$\endgroup$
2
  • $\begingroup$ Wikipeida says that such vortex can take place only with a specific range of Re numbers, but in the case of inviscide flow, the Re number approaches infinity. I am not sure about this because I was not able to find any mentions of the invisvid case in this article. $\endgroup$ Mar 29 at 21:07
  • $\begingroup$ Vortices can form in Re=oo superfluids such as 3He, but the descriptions of that are way beyond me. The Physics SE may be a better place to ask. $\endgroup$ Mar 29 at 21:39

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .