Spalart-Allmaras model, boundary and initial conditions

vinz · December 15, 2009, 08:48

Dear OpenFoamers,

I have few questions about the Spalart-Allmaras model in OpenFOAM.
I found this web page dealing with the different versions of SA model: http://turbmodels.larc.nasa.gov/spalart.html
I'd like to know which one is implemented in OpenFOAM.

I also have questions about the initial values for nut and nutilda. I have been surprised to see that values of 0.14 are imposed for nut and nutilda in the airfoild2D tutorial. Those values look pretty high to me but I may be missing something here.

I would like to know how should I choose my initial far field values of nut and nutilda for a flow around an airfoil at 50m/s for example. Can someone help me with that please?

Any information on the model and/or the boundary conditions and initial conditions would be very welcome.

Regards,

Vincent

Simon Lapointe · December 15, 2009, 09:41

Hi Vincent,

Unless there has been a recent change, the version of the Spalart-Allmaras RANS model in OpenFOAM is the one with fv3 Term (SA-fv3). This version is not recommended but is said to be more stable, especially in complex geometries. If you want to use the "standard" version, it is easy to implement.

Concerning the farfield values, it depends of your problem. In the case of an external flow around an airfoil, you might want to use rather small values of nut (maybe a viscosity ratio of 0.1 or smaller). For nuTilda, Spalart recommends using a farfield value of about 3*nut.

Hope this helps.

vinz · December 15, 2009, 10:01

Hi Simon,

and thank you for your fast reply. You advise a viscosity ratio of 0.1, that means I should use nut around 1e-6 if I use a nu value of 1e-5, right?
And then a nuTilda of 3e-6? does it look right for you?

Do you happen to know why the values of nut and nuTilda have been set so high (0.14) in the airfoil2D tutorial? Unless we should impose the viscosity ratio instead of the variable value itself inside nut and nuTilda initial dict?

Thanks again.

Vincent

Simon Lapointe · December 15, 2009, 10:32

Yes those values seem right. You might also want to try different values to clarify the influence of the farfield values on your solution. In the case of an external flow around an airfoil, I did not see major differences when the upstream boundary is far from the airfoil.

I'm sorry but I can't really help you with the airfoil2D tutorial. I'm pretty sure these are nut and nuTilda values (especially since the dimensions are specified) and not viscosity ratios.

Sofia · July 7, 2010, 08:36

Hi,

Do you know the difference between nu and nut in the SA turbulence model,

thank you

Sofia

Simon Lapointe · July 7, 2010, 08:56

Hi Sofia,

nu is the kinematic viscosity, a property of the fluid. nut is the turbulent kinematic viscosity, calculated by the turbulence model.

In the Spalart-Allmaras model, nut is calculated by the means of a transport equation for nuTilda (modified turbulent kinematic viscosity). nut is obtained by multiplying the transported variable nuTilda by a function (fv1).

Sofia · July 9, 2010, 08:11

Thank you Simon,

I choose nut=nu and it works well...

AshwaniAssam · July 20, 2015, 04:57

The SA model in openFOAM has been updated with the standard Spalart-Allmaras Model.

link1 link2

December 15, 2009, 08:48	Spalart-Allmaras model, boundary and initial conditions	#1
vinz Senior Member Vincent RIVOLA Join Date: Mar 2009 Location: France Posts: 283 Rep Power: 18	Dear OpenFoamers, I have few questions about the Spalart-Allmaras model in OpenFOAM. I found this web page dealing with the different versions of SA model: http://turbmodels.larc.nasa.gov/spalart.html I'd like to know which one is implemented in OpenFOAM. I also have questions about the initial values for nut and nutilda. I have been surprised to see that values of 0.14 are imposed for nut and nutilda in the airfoild2D tutorial. Those values look pretty high to me but I may be missing something here. I would like to know how should I choose my initial far field values of nut and nutilda for a flow around an airfoil at 50m/s for example. Can someone help me with that please? Any information on the model and/or the boundary conditions and initial conditions would be very welcome. Regards, Vincent

December 15, 2009, 09:41		#2
Simon Lapointe Member Simon Lapointe Join Date: May 2009 Location: Québec, Qc, Canada Posts: 33 Rep Power: 16	Hi Vincent, Unless there has been a recent change, the version of the Spalart-Allmaras RANS model in OpenFOAM is the one with fv3 Term (SA-fv3). This version is not recommended but is said to be more stable, especially in complex geometries. If you want to use the "standard" version, it is easy to implement. Concerning the farfield values, it depends of your problem. In the case of an external flow around an airfoil, you might want to use rather small values of nut (maybe a viscosity ratio of 0.1 or smaller). For nuTilda, Spalart recommends using a farfield value of about 3*nut. Hope this helps. alpemre and mechy like this.

December 15, 2009, 10:01		#3
vinz Senior Member Vincent RIVOLA Join Date: Mar 2009 Location: France Posts: 283 Rep Power: 18	Hi Simon, and thank you for your fast reply. You advise a viscosity ratio of 0.1, that means I should use nut around 1e-6 if I use a nu value of 1e-5, right? And then a nuTilda of 3e-6? does it look right for you? Do you happen to know why the values of nut and nuTilda have been set so high (0.14) in the airfoil2D tutorial? Unless we should impose the viscosity ratio instead of the variable value itself inside nut and nuTilda initial dict? Thanks again. Vincent tareqkh likes this.

July 7, 2010, 08:56		#6
Simon Lapointe Member Simon Lapointe Join Date: May 2009 Location: Québec, Qc, Canada Posts: 33 Rep Power: 16	Hi Sofia, nu is the kinematic viscosity, a property of the fluid. nut is the turbulent kinematic viscosity, calculated by the turbulence model. In the Spalart-Allmaras model, nut is calculated by the means of a transport equation for nuTilda (modified turbulent kinematic viscosity). nut is obtained by multiplying the transported variable nuTilda by a function (fv1). brucecumt, manojkv and hogsonik like this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Wind turbine simulation	Saturn	CFX	58	July 3, 2020 01:13
Superlinear speedup in OpenFOAM 13	msrinath80	OpenFOAM Running, Solving & CFD	18	March 3, 2015 05:36
Low Mach number Compressible jet flow using LES	ankgupta8um	OpenFOAM Running, Solving & CFD	7	January 15, 2011 13:38
RPM in Wind Turbine	Pankaj	CFX	9	November 23, 2009 04:05
Convective Heat Transfer - Heat Exchanger	Mark	CFX	6	November 15, 2004 15:55

December 15, 2009, 10:32		#4
Simon Lapointe Member Simon Lapointe Join Date: May 2009 Location: Québec, Qc, Canada Posts: 33 Rep Power: 16	Yes those values seem right. You might also want to try different values to clarify the influence of the farfield values on your solution. In the case of an external flow around an airfoil, I did not see major differences when the upstream boundary is far from the airfoil. I'm sorry but I can't really help you with the airfoil2D tutorial. I'm pretty sure these are nut and nuTilda values (especially since the dimensions are specified) and not viscosity ratios.

July 7, 2010, 08:36		#5
Sofia New Member Sofia Join Date: May 2010 Location: Toulouse, France Posts: 14 Rep Power: 15	Hi, Do you know the difference between nu and nut in the SA turbulence model, thank you Sofia

July 9, 2010, 08:11		#7
Sofia New Member Sofia Join Date: May 2010 Location: Toulouse, France Posts: 14 Rep Power: 15	Thank you Simon, I choose nut=nu and it works well...

July 20, 2015, 04:57		#8
AshwaniAssam Senior Member Ashwani Join Date: Sep 2013 Location: Hyderabad Posts: 154 Rep Power: 12	The SA model in openFOAM has been updated with the standard Spalart-Allmaras Model. link1 link2