[90nash]

Hello Foamers,

I am trying to compare particle trajectories with and without a temperature dependent viscosity model. For this i modified the simpleFoam solver to include temperature equation and also generated a new library file to include a temperature dependent viscosity model.
Now i need to modify the icoUncoupledKinematicParcelFoam solver so that it reads the nu values from the 0 directory, generated by the above solver, instead of reading it from transportProperties directory (just like icoUncoupledKinematicParcelFoam reads U file from 0 directory of case)

Can anyone please tell me how this can be done? It will be a tremendous help.

Thanks

[ybapat]

Hello,

You can read nu as a volScalarField from 0 file very similar to U field.

-Yogesh

[90nash]

Hello Yogesh,
Thank you for your reply!! I played around with createFields.H file and was able to successfully compile a solver which takes nu file as input from 0 folder.


I would really appreciate if you could help me with another question. I was using power law model for viscosity and used k value as 0.035 [Pa.s^(n)] and n 0.6 (got these values from a reference paper on blood flow). However i am getting huge time step continuity errors (simpleFoam solver, OF 2.1.1) for this k value.
I searched the forum and found that some people have used very large values for k (2500). If i use such high value, the solution does converge. Can you tell me what i am doing wrong. Is k value supposed to be this high?
The same issue is also reported in this post:
http://www.cfd-online.com/Forums/ope...-openfoam.html

[ybapat]

Hello,

As you have taken k value from some paper it should be correct. What you can try is to do a few iterations with constant viscosity value so you get some flow field and then switch to power law for viscosity. This should help you in converging your case.

Regards,
-Yogesh

[90nash]

Excellent Advice Yogesh. Thank you!!
I ran two time steps with constant viscosity model and then switched to power law. The solution converged.
Although i don't understand clearly why this happens.

Best,
Nadish

[ybapat]

Hello ,

Reason is that power law uses velocity gradients which are not stable when you start your simulation. Doing a few iterations with constant mu helps to stabilize velocity gradients.

-Yogesh