CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   y plus (http://www.cfd-online.com/Forums/cfx/22823-y-plus.html)

lcm July 10, 2006 07:08

y plus
 
hello, using a k-e model for a basin with roughly 100x40x3m, and a very low inlet velocity (<1m/s), the solver gives me a reynolds number of 10^5!! the y plus also goes up to 1000 or greater for 50% of the cells. Is there an explanation for this? Is this ok? or should i refine the mesh even more? i have already about 85000 elements. Any comments? Thanks!

V. Kumar July 10, 2006 07:51

Re: y plus
 
you may calculate Re with hand calculation as well. The absolute velocity is not an indicator of yplus. Yplus depends on your Reynolnds number. For your case, a grid cell with 0.1% of the domain size, perpendicular to flow direction, should resolve your boundary layer up to yplus of 1-10. 50% cells above yplus 1000 seems to be ok but it depends on your grid and Re. it is difficult to comment on this number.

VK

Glenn Horrocks July 10, 2006 18:00

Re: y plus
 
Hi Icm,

Do a mesh size sensitivity study and find out. For some flows that y+ is OK and for some it is not.

Regards, Glenn

Adam July 11, 2006 05:39

Re: y plus
 
If you're using the k-e model, then there's no point in having a yplus < 11 as the model uses wall functions to model the sublayer. Just make sure you have 10 elements in the boundary layer for good resolution.

Adam



All times are GMT -4. The time now is 04:29.