CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Y+ value (http://www.cfd-online.com/Forums/fluent/62817-y-value.html)

ashishmgad March 20, 2009 12:18

Y+ value
 
hi,

am working on turbulent convection in cavities.
in RSM model i have kept standard wall function option
i got y+ value 0.3
which matches with literature alongwith convection coefficient

but fluent manual says y+ should be approximately 30
if i think of maintaining it to be 30 there will be hardly any cells in the domain !!
my nearwall cell size is 0.1

am confused whether i have set problem correct or not
pls give suggestion thanks

Micael March 20, 2009 22:22

Use standard wall function when y+ > 30
Use enhanced wall treatment when y+ < 5

y+ < 5 mean yours cells at the wall are in the viscous layer and standard wall function (log law) are not valid there.

If the convection coefficient matches with literature it may just be a
coincidence. Be carefull with it.

y+ = 0.3 is a very fine grid. If you evaluate your grid as coarse, then this appear strange to me. What is you geometry and set up? Anyway, try first the enhanced wall treatment, may be y+ was not correctly calculated.

Have a nice day.

Micael

Micael March 23, 2009 09:23

I just received your email.

May be the viscous (laminar) layer near the walls is thick. It can explain why y+ is so low and why the mesh is too coarse when y+ > 30.

Normally, we use y+ > 30 when the viscous layer is thin. Indeed, in those cases, resolving the flow down to the viscous layer is very computationally expensive.

Anyway, the rules of thumb are:
- Use standard wall function when y+ > 30
- Use Enhanced wall treatment when y+ < 5
- Avoid to have a mesh so that 5 < y+ < 30
Check the user guide (12.10 and 12.11) for information about those functions.

ashishmgad March 23, 2009 11:22

Quote:

Originally Posted by Micael (Post 210430)
I just received your email.

May be the viscous (laminar) layer near the walls is thick. It can explain why y+ is so low and why the mesh is too coarse when y+ > 30.

Normally, we use y+ > 30 when the viscous layer is thin. Indeed, in those cases, resolving the flow down to the viscous layer is very computationally expensive.

Anyway, the rules of thumb are:
- Use standard wall function when y+ > 30
- Use Enhanced wall treatment when y+ < 5
- Avoid to have a mesh so that 5 < y+ < 30
Check the user guide (12.10 and 12.11) for information about those functions.

Thanks a lot Michael


All times are GMT -4. The time now is 13:28.