Turbulent Viscosity Ratio Limited not Vanished yet, but My Case Was Converged
 

Hello there, I have questions
I ran simulation about flow across inlet duct HRSG with Fluent 6.3.26, with the parameters were :
1. Turbulence model : Realizable k-epsilon
2. Standard wall function
3. k-epsilon model : Turbulence Intensity (10%) and hydraulic diameter
4. Residual monitor : 10-4
5. Inlet : Velocity inlet
6. Outlet : Outflow
7. Using Energy Equation

and i built the superheater tube geometri in my simulation.


http://i66.tinypic.com/214a5wp.jpg


while I was running the simulation, there was warning about "turbulent viscosity ratio limited to viscosity ratio of 100000 in ..... cells."

But in 1000 iterations my case was converged and the warning was not vanished yet. Can somebody explain what's wrong in my simulation?
Is my result valid ?

I'd appreciate for your help
Thanks

Regards

In general, the result is invalid and you need to fix the limited viscosity ratio.

Make a plot and look for regions where the turbulent viscosity is extremely high. That will tell you where you need to improve.

It could be your uniform velocity inlet.

Thanks LuckyTran,

So, do i just refine the grid in that area?

The statement about "limited viscosity ratio" appear, is it only because of my mesh, or something else like wrong boundary condition at the inlet?

It can and usually is caused by wrong inlet boundary conditions. i.e. a high uniform turbulence intensity applied to a uniform velocity inlet is non-physical.

It can be caused by the mesh if there are significantly skewed cells.

i have checked my worst skewness, and it is 0.554523 , is that bad ?

http://i67.tinypic.com/2qvwdg5.jpg

http://i68.tinypic.com/2hzk3tt.jpg

If it was my wrong boundary conditions at the inlet, can you suggest me the right one ?
I have my turbulence intensity set by 10 % (from other paper) and i used hydraulic diameter which is the hydraulic diameter of the inlet, that is 3.1409 m.

Thanks for your help LuckyTran.

The skewness seems okay so it's probably not the cause.

Check where the turbulent viscosity ratio is too high and determine where the problem is. Is it near the inlet? Is it near your walls or is it in the core?

Or maybe it is not near the inlet but clustered around your tubes.

The turbulence viscosity is highest at sudden expansion (exit of the duct). Is my mesh too rough at that area ?

http://i67.tinypic.com/2rf615v.jpg

And just curious, is there any case that turbulent viscosity ratio could really exceed 10^5 ?

There are certainly skewed cells in that area. Anyway, you know what locations need improvement.

Turbulent viscosity ratio > 10^5 is extremely rare and will certainly never occur for practical scenarios. Maybe for superfluids, critical point transitions, or inside a black hole or something... Since you doing a HRSG, I'll say that it's nearly impossible to have that high a viscosity ratio.

Black hole.. ? yeah right..

Ok then, thanks for your help. I really appreciate it.

Lucky Tran, sorry i forgot. In my simulation, i'm using standard wall function for my wall treatment, but i have tube (wall ) with heat transfer condition (conjugate HT). Can i just use standart wall function or i have to use the other else ?