CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Boundary Layer - dimensionless wall distance y+ (http://www.cfd-online.com/Forums/main/93948-boundary-layer-dimensionless-wall-distance-y.html)

MartinRei October 31, 2011 16:21

Boundary Layer - dimensionless wall distance y+
 
Hello everyone,

first of all, if that is the wrong forum for my question, i kindly ask the admins to correct that, because i was not sure where to post it.

I am currently trying to generate a boundary layer mesh for a laminar flow (Re<10000) over a 2D airfoil.

The question is now, how "high" should the first row of boundary layer cells be in this laminar flow? y_{1}^+ = 1 or y_{1}^+ = 30?

Kind regards,
Martin Rei

Sonil November 2, 2011 02:10

Martin

Choice of what value of y+ to take depends on how accurately are you trying to capture the flow in laminar sub-layer in a turbulent flow problem. If you make a mesh with y+ 30 then you will have to perform your simulation using a turbulence modelling with wall scales (this will approximated the effects of laminar sub layer and buffer zone numerically) fora turbulent flow. y+=1 or less must be used if you have enough computing power and you need not use wall scale in this case.

In your case of a low reynolds number flow these things may not matter... Still i would suggest mesh as fine as possible, as a high resolution mesh is always preferable if we have enough computing resource.

MartinRei November 2, 2011 05:08

Quote:

Originally Posted by Sonil (Post 330385)
Martin

[...] in a turbulent flow problem. [...]

Is it different in a laminar problem? Because the problem at hand is laminar.

Sonil November 2, 2011 06:57

See it all boils down to resolving your boundary layer...
I would suggest you to calculate B.L. thickness with respect to chord using formula for calculating B.L. thickness for laminar flow over flat plate and try to have minimum of 20 layers inside that length... and keep the growth ratio of the grid inside 1.15 in the B.L.

Far November 3, 2011 02:14

Quote:

Originally Posted by MartinRei (Post 330214)
Hello everyone,


I am currently trying to generate a boundary layer mesh for a laminar flow (Re<10000) over a 2D airfoil.

The question is now, how "high" should the first row of boundary layer cells be in this laminar flow? y_{1}^+ = 1 or y_{1}^+ = 30?

Kind regards,
Martin Rei

Y+ is not a flow variable, it is by product of turbulence modeling and therefore it is meaning less in laminar flows. However, in laminar flows you should make the mesh independence study.

Quote:

Originally Posted by Sonil
In your case of a low reynolds number flow these things may not matter...

If the flow were turbulent, then in low reynolds number these things matter alot.

ps. If you refine mesh sufficienntly high, then you may reach the DNS with laminar flow solver !!!!


All times are GMT -4. The time now is 07:29.