
[Sponsors] 
November 14, 2003, 13:00 
y+=1 or y+>30 in SpalartAllmaras model

#1 
Guest
Posts: n/a

Hi.
The tutorial "Airfoil" in step 7 says that when using the SpalartAllmaras model, the y+ is to be in the order of 1 or larger than 30, then the grid is fine. If not, the grid needs modification. I understand that the y+ need to be in the order of 1. But why larger than 30? Wont the resolution be too coarse? And why is BOTH y+ values ok to choose from and not only one of them? Is there something in the model that states that y+ larger than 1 and smaller than 30 is unacceptable, or is there another thing with this range? Hope someone can help me. Cheers Christian 

November 14, 2003, 14:07 
Re: y+=1 or y+>30 in SpalartAllmaras model

#2 
Guest
Posts: n/a

When y+ is larger than 30 fluent uses a wallfunction approach (which is valid for y+ in the range between 30 and 200 or so). When y+ is below 30 the wallfunction approach is not valid. Instead the equations have to be integrated all the way to the wall. However, this requires that the grid is sufficiently fine to resolve the inner parts of the boundary layers (y+ around 1). If the grid is not this fine you run into problems.
Hence, the problems is that there are two types of wall treatment  wall functions valid for y+ between 30 and 200 and lowRe integration downto the wall which requires y+ around or below 1. There are hybrid methods which are able to "mix" these two wall approaches and allow you to use any y+ you want. However, these hybrid methods aren't always that robust and they are certainly not as mature as the two original approaches (lowRe or wallfunctions). Btw, the nonequlibrium wallfunctions in fluent are surprisingly good if you use a good y+ (between 30 and 200). 

November 18, 2003, 12:47 
Re: y+=1 or y+>30 in SpalartAllmaras model

#3 
Guest
Posts: n/a

But..... The manual says that the wall law is initiated at y+ smaller than 11.something. Or am I reading it uncorrect?
And another question: The enhanced wall function can combine the enhanced features and the 2 zone model. I read that as no matter the y+, my calc will be correct. And the normal wall function starts at 11.something and if my y+ is above that but smaller than 30, my calc will be uncorrect. Is that correctly understood? 

November 18, 2003, 21:29 
Re: y+=1 or y+>30 in SpalartAllmaras model

#4 
Guest
Posts: n/a

11 is still of the order of 1, 10 to the power 1


November 19, 2003, 04:17 
Re: y+=1 or y+>30 in SpalartAllmaras model

#5 
Guest
Posts: n/a

< 11 is still of the order of 1, 10 to the power 1
???? I dont understand your point. According to Jonas the Fluent wall function starts at y+ above 30, not above 10. 

December 7, 2003, 23:27 
Re: y+=1 or y+>30 in SpalartAllmaras model

#6 
Guest
Posts: n/a

hi,christian,i agree with jonas's opinion, and i also read the words about y+ euqals to 11, but i can't remember that clearly. would you please tell me where is the words? i am now also confused about that and the wall function just like you. and another words, did you do the calc test about y+ you mentioned above or you encountered that problem in practice or it is just what you deduced? thanks in advance.


Thread Tools  
Display Modes  


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
LES and combustion model  Margherita Cadorin  CFX  0  October 29, 2008 06:24 
Spalart Allmaras Wall Function  vandraren  OpenFOAM Running, Solving & CFD  3  February 8, 2008 09:42 
Yplus for Spalart Allmaras  andimb  OpenFOAM PostProcessing  1  April 25, 2006 05:04 
Question to the new Version of Spalart Allmaras in 13  andimb  OpenFOAM Running, Solving & CFD  0  April 7, 2006 11:31 
High Re Spalart Allmaras  jj  CDadapco  0  October 3, 2005 15:31 