Hi there,

Can any one tell me briefly about Y Plus and its Significance ?

I am new in this field,

Thanks in advance

rakesh.

Dear Rakesh,

The concept of y+ is important in turbulent flows, precisely because it controls the resolution of what is referred to as the viscous sublayer. Depending on various models used, the grid resolution must be such as to have y+ below a specified value, generally 3-5. You can think of y+ simply as a Reynold's number, based on the distance from the wall and friction velocity. As a result, y+ is a measure of how close the first layer is to the wall, the grid resolution and hence the ability of the grid to represent the turbulent b.l. properly. For more details, look into Tennekes and Lumley ( for the basic ideas) and Wilcox (for Turbulence modelling)

Regards,

Ganesh

There is some information available in CFD-Wiki here.

The page is not that complete yet though, but it does contain a definition of y plus.

hi,

after reading the basic explanation by ganesh, i've a question...

i'm doing laminar flow with possibility of flow seperation. Can i use the y+ value to estimate my grid spacing near the wall for a given Re no.?

Or is there other better, more accurate way to estimate my grid near wall for a given Re no.?

Dear Zonexo,

The concept of y+ is used more in turbulent flows, than in laminar flows. The defintion of y+, as I have mentioned earlier arises from the need to resolve the viscous sublayer very close to the wall. In a laminar boundary layer, though a defintion of y+ is possible, it is probably not employed becasue of difference in structure between the b.l. in laminar and turbulent flows. In any case, at the point of separation, the use of frcition velocity(through the wall shear stress) is not justifiable, and a different scaling has to be looked out for.(You can see this in Tenekkes and Lumleys's book). A rough estimate of the grid size near the wall can be estimated from the knowledge of the b.l. thickness. The b.l. thickness is related to the Re and the knowledge of Re can give the approximate b.l. thickness. Roughly, for a lmainar flow (1/20) th of this thickness, should do good enough.(This means there are 20 layers indie the b.l.). This is a rough thumb rule that is good enough for attached laminar b.l., even with separation. Note that an issue of heavily separated flows is important in turbulence and existing turbulence models find it difficult to make accurate predictions in such situations.

Hope this helps

Regards,

Ganesh