Advanced Turbulence Parameters

[frederic.gaillard.7]

Hello, 

I'm looking to change my turbulence advanced setting. I just figure it out that the default  parameter of the turbulence intensity was 15% too low. And now i have the same question with CMu CE1 eand CE2. 

For CMu  can i use the eddy viscosity formula (here) to cumpute the CMu with these assumptions : 

• eddy viscosity (ut) correspond to ''effective viscosity''  in the result quantities 
• K correspond to TKE values 
• epsilon correspond to the TED values 

These assumption are correct ?  if it's not how i can calcul the CMu parameter ?

I, basically, have the same question for CE1 & CE2.  In your documentation it says to increase the CE2 for more mixing, mores shear ... but increasing by what amount, it's a little bit confusing ... 
Thank for your help. 

FRED

[David.Short. · Accepted solution]

Hi Fred,

Are you sure you need to increase the default turbulent intensity at the inlet boundary? 20% would be used if the incoming flow is highly turbulent. What is your sim?

Your assumptions are all correct however CMu is an empirical constant and not something that you calculate. The value is based on empirical evidence e.g. experimental data. If you find literature stating a different value that corresponds to a flow regime the same as in your simulation then feel free to use that.

The same goes for the CE1 and CE2 empirical constants. If you come across more suitable values that are based on experiments that match your simulation closely then change them.

So i would recommend either sticking with the defaults or diving into the wonderful world of CFD literature.

All the best,

David

[frederic.gaillard.7]

Hello David,

• Are you sure you need to increase the default turbulent intensity at the inlet boundary? 20% would be used if the incoming flow is highly turbulent. What is your sim?
  • I didn't know that this specific parameter only apply on a boundary condition. I though is was more a ''fluid caracteristic'' related to is angular momentum. His capacity/facility to produce eddy or vortex. 
  • My conclusion came from when I displayed Turb Intensity on my region of interest  (upper part of the duct, where it's highly meshed) The half of my values lied between 10% and 20%. So i though it was appropriate to ajust this parameter
  • I also use the bulk-weighted calculator to find velocities over a plane  and by knowing my TKE i use the following formula to figure my turbulence intensity (I) :     
  • It gave me approximatively the same percentage. 
•  CMu is an empirical constant and not something that you calculate. 
  • 
  • Knowing all my parameter in the previous formulas i exepected to find CMu. But it's seems that the eddy viscosity is computed regard to the values of CMu.
• So i would recommend either sticking with the defaults or diving into the wonderful world of CFD literature. 
  
  • I will clearly review some litterature about this topics . I consider to buy this book (CFD Book) if you have other specific publication in mind, feel free to send me some. I'll keep you posted from what I learn. 

Thank You David. 

[David.Short. · Accepted solution]

Hi Fred,

The turbulent intensity value is just to get things started at the inlet it is then simply dependent on the fluid velocities (specifically the fluctuating turbulent velocity divided by the mean flow velocities). It needs a value at the inlet and then settles out and responds to the flow. An accurate value is only needed if your region of interest is very close to the inlet. For most cases the entering flow is not highly turbulent thus the default of 5% will suffice.

But it's seems that the eddy viscosity is computed regard to the values of CMu.

Exactly!

With regards to literature I would just search online for the specific equations that you are interested in. You may buy a book and find that it doesn't even go into that sort of detail. Saying that it wouldn't hurt to read some chapters in a text book (I have tried a few times but don't have the patience since leaving university).

If you think that I  have answered your question / solved your problem please can you mark the relevant post as an accepted solution, alternatively you can allow me to mark it as such. This will help others to benefit from the discussion and easily find the solution if they have a similar issue.

All the best,

David

[frederic.gaillard.7]

Understood ! 
I will check on some specific litterature if i can add relevant information on this topic. I will close this discussion right after. 
Thank for your help !