CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   "Y plus" very important!!! (http://www.cfd-online.com/Forums/fluent/121958-y-plus-very-important.html)

 Qube August 8, 2013 15:28

"Y plus" very important!!!

Hi!

I need some information about the expected Y plus value for each turbulence model.

 behest August 8, 2013 15:38

hello,
Y+ is very effective in the calculation, and the value may be 5-24 or 24-500 for RANS turbulence model. But it can be changeable for your simulation and results

 Qube August 8, 2013 15:57

thank u for your response but i need a detailed information, such as how much acceptable y plus for realizable KE?

 behest August 8, 2013 18:31

1 Attachment(s)
Actually, every boundary layer is divided into three parts: 1-sublayer 2-buffer layer 3-log law layer. All RANS turbulent models estimate viscosity and this viscosity is also included of molecular and/or turbulent ones. The importance of y+ lies in the fact that which part the center point of a mesh situates. Therefore, turbulent model estimates viscosity due to that part. In the buffer layer, we cannot have a good anticipation, as a consequence, it makes a recommend to avoid a y+ value in buffer layer. I enclose a picture to give the points

Quote:
 Originally Posted by Qube (Post 444634) thank u for your response but i need a detailed information, such as how much acceptable y plus for realizable KE?

 flow_CH August 10, 2013 01:53

Quote:
 Originally Posted by behest (Post 444644) Actually, every boundary layer is divided into three parts: 1-sublayer 2-buffer layer 3-log law layer. All RANS turbulent models estimate viscosity and this viscosity is also included of molecular and/or turbulent ones. The importance of y+ lies in the fact that which part the center point of a mesh situates. Therefore, turbulent model estimates viscosity due to that part. In the buffer layer, we cannot have a good anticipation, as a consequence, it makes a recommend to avoid a y+ value in buffer layer. I enclose a picture to give the points

 Qube August 10, 2013 06:48

my case study is a simplified car body and i want to use RKE or SST KW turbulence models. for more accurate results, how much should be y plus values?!
which one is correct? 30< y+ <300 or y+ <5 ?

 behest August 10, 2013 08:15

Dear friend
The drag coefficient is depend to y+ value. I suggest to you to consider 30< y+ <300 initially; but if the results was not sufficient precision, then use another y+.
It has had an experience that the y+ <5 is achieved more accuracy

Quote:
 Originally Posted by Qube (Post 444837) Dear saeed! my case study is a simplified car body and i want to use RKE or SST KW turbulence models. for more accurate results, how much should be y plus values?! which one is correct? 30< y+ <300 or y+ <5 ?

 Qube August 10, 2013 10:33

thank you so much!

 All times are GMT -4. The time now is 08:54.