[eng.abdul]

How the NUMECA take care of turbulent intensity at the inlet for different turbulent models in Fine/Turbo. I need to know its default value for SA turbulent model. A reply is appreciated please

[DarylMusashi]

Dear eng.abdul,

the turbulent intensity at the inlet can be defined for the following turbulence models:
-ke
-kw
-SST
-v²f

The SA model works with the kinematic turbulent viscosity [µt] as input parameter.

I think it is possible to calculate the Turbulent Intensity based on the turbulent viscosity with equations mentioned in the theoretical manual (at the very bottom of the page, equation 3 (equation 1 as attachment))

FINE™/Turbo > User Guide > Flow Model > Turbulent Navier-Stokes > Best Practice for Turbulence Modelling > Defining Initial & Boundary Conditions

With this equation you can estimate k at the inlet on basis of µt/µ, Ny und Cµ (which is 0.09 in SA model). With k, you can rearrange equation 2, which yields the streamwise fluctuating velocity u'². Finally, you can use the attached equation 3 to compute the related turbulent intensity Tu.

I hope this works and yields reasonable values for Tu.

Kind regards,
Holger

[eng.abdul]

Thank you very much Mr Holger for prompt response.

1-Are these three equations valid for SA model? because in the manual these are expressed for k-ε, k-ω, EARSM and v²-f Models.?

2-Does Numeca considers 5% turbulent intensity for one equation SA Model?

[DarylMusashi]

Dear abdul,
indeed you are right, these equations are only mentioned in context of the other turbulence models and not for SA model, I am not sure if they can be used for the SA model. Secondly, the constant Cµ doesn't exist in the SA model, as I claimed in my last post.

Unfortunately, I did not find any clue in the manuals which value of turbulent intensity is assumed at the inlet for SA model. It seems to me only the kinematic turbulent viscosity is necessary as turbulent quantity at the inlet when using the SA model.

Sorry that I can't provide a more helpful answer at the moment.

[eng.abdul]

thank you very much for your replies ... much appreciated..... One last question please .... can you tell me what is the suggested value of kinematic turbulent viscosity by the user manual . what is your comment

[DarylMusashi]

Hi,
for internal flows (like turbomachinery) a ratio of µt / µ = 1 to 5 is reasonable. From my experience (mainly turbomachinery applications) the default value of µt gives good results in comparison with experimental data.
For external flows a ratio of 1 is recommended.

You can also use the expert parameter NUTFRE to define define the initial value of the turbulent viscosity in the entire flow field.

Kind regards
Holger

[eng.abdul]

Thank you Holger!!

The default value at the inlet in boundary condition is νt=0.0001 m2/s, and the initial value of turbulent viscosity by the expert parameter NUTFRE is 0.0003... How these two values correspond to the ratio of νt / ν = 1 to 5 for internal flow. I mean what will be the ratio of νt / ν for turbulent viscosity of 0.0001 m2/s at the inlet and NUTFRE of 0.0003. I think the kinematic viscosity ν, needs to be calculated for a particular case. what is your take

[DarylMusashi]

Dear abdul,

µt is defined at the inlet. The resulting ratio µt/µ (which should be between 1 and 5, as discussed) depends on the fluid properties. To be more precise it depends on the viscosity of the fluid.
In FINE/Turbo you can model different fluids (for example Air Perfect Gas, Real Gas) with different laws for computing the viscosity. For Air Perfect Gas the Sutherland-Law is used to compute the viscosity on basis of the temperature. If you know the temperature at the inlet (or at a certain location in general) you can calculate the viscosity of the fluid at this location and then compute the ratio µt/µ.

I hope this answers your question? If not please don't hesitate to ask.

[eng.abdul]

Thank you

But if I know the dynamic viscosity and density of a fluid then i think it is easy because v is ratio of dynamic viscosity and density. But by the sutherland law (rather than constant viscosity) with the temperature at the inlet gives more accurate value of vt/v.... thank you

Regards

[eng.abdul]

Holger

I know the temperature at the inlet for a gas . how to determine the constants used in Sutherland's viscosity equation for a gas . the literature mention these constants only for air.


https://www.cfd-online.com/Wiki/Sutherland's_law

[DarylMusashi]

Hi there,
May I ask which fluid you are going to model? I found some definitions following the links in this old thread:

Sutherland's law coefficient for viscosity

D.Cook's link shows some constants for different gases.

Kind regards,
Holger

[eng.abdul]

well i am talking about the real gas made of combustion products...gamma=Cp/Cv=1.33 .... i think the constants for air can be used for this....but dynamic viscosity is certainly different than the air at elevated temperatures