CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   turbulent kinetic energy is not zero at the wall (http://www.cfd-online.com/Forums/fluent/112440-turbulent-kinetic-energy-not-zero-wall.html)

 yogi06_sati January 29, 2013 00:25

turbulent kinetic energy is not zero at the wall

hi all
i am solving multiphase turbulent flow problem. velocity profile is correct but turbulent kinetic energy and intensity is not zero at the wall, although it should be zero at the wall. turbulent kinetic energy and intensity is showing maximum on the wall. i do not why is it so?

i need help

thank you

 yogi06_sati January 29, 2013 03:14

i am waiting for guidance from experts. it will really help me a lot.

 flotus1 January 29, 2013 05:07

What kind of wall modeling are you using? If you are using a wall function with Y+ around 60 it is normal that the TKE has a maximum value in the first cell away from the wall. If I recall correctly, the maximum for TKE is occurs at around Y+=30 for a flat plate.

 yogi06_sati January 29, 2013 08:04

In my case fluid is flowing in a circular pipe. i am using k-omega viscous model. As you told wall model and Y+ value, i am not clear how and where to use wall model and Y+ value. i mean where is the input for Y+ value in gui.

probably you are getting my point

thank you very much

 flotus1 January 29, 2013 12:26

Y+ is not an input parameter.
It is the non-dimensional wall distance of the first cell.
You can evaluate it as a post-processing variable at solid walls.

The only way to influence it (besides the flow conditions) is with the mesh size.
If you want to resolve the boundary layer and thus capture the "correct" behavior of the flow variables near the wall, Y+ should be around 1 as a rule of thumb. Additionally, you will have to use the "enhanced wall treatment" in the turbulence modeling section.

 yogi06_sati January 30, 2013 07:00

thank you very much Alex S

 mattyg101 January 30, 2013 10:47

It may help to use the scalable wall treatment if it is available with the turbulence model you are using. This applies the different wall treatments based on the value of Y+ or Y* as far as i am aware.

I had this problem as well a while ago. Good luck.

 micro11sl April 28, 2014 11:03

Quote:
 Originally Posted by yogi06_sati (Post 404688) hi all i am solving multiphase turbulent flow problem. velocity profile is correct but turbulent kinetic energy and intensity is not zero at the wall, although it should be zero at the wall. turbulent kinetic energy and intensity is showing maximum on the wall. i do not why is it so? i need help thank you
Hi guys,
I think the question is to ask why turbulent kinetic energy and turbulence intensity are not zero at the wall, but the replies are all related to Y-plus!

The answer is, FLUENT doesn't implement zero boundary condition for kinetic energy k (k omega based model)! It uses zero gradient boundary condition (the change rate across the wall boundary is zero, instead of fixing a zero value for k), so you will see a finite value of k on the wall. I don't know why FLUENT does this. Among literature, I see some use zero, while some use zero gradient. You probably need a reference discussing boundary condition. The k change very near wall surface may be quite not trustful. One guy from my university shows me the difference from FLUENT and in-house DNS code. He then implements a UDF turbulence model using zero value boundary condition and also finds difference from FLUENT default turbulence models.

Sheng

 All times are GMT -4. The time now is 03:03.