[Muhammad Ibrahim]

Hey Everyone!

I am a PhD candidate, working on Thirdparty extension of OpenFoam like rheoTool. Solution of Navier Stocks equations are quit tricky with stress tensor of viscoelastic models. I read out many forums to collect knowledge about it but i didn't find any discussion.

Numerical instability is very common i observed with any kind of polymer model provided in rheoTool version7. Is any professional is working on it? i want to learn "system/fvSchemes,fvSolution" with PISO, SIMPLE, PIMPLE coupling methods.

If I am not wrong with writing in correct forum, please confirms if there are some rheoTool professional.

Moreover, please let me know if i am not on correct forum, thanks for your valuable times.

[dlahaye]

Could you pls. elaborate on what kind of instabilities you are referring to?

Do you have examples that illustrate your point?

Is the problem caused by the reduction of the viscosity is shear-thinning situations?

Is reducing the relaxation factor or reverting to a coupled velocity - pressure solver a solution here?

[Muhammad Ibrahim]

Mr. dlahaye i am very thankful! for your reply, which highlight that i am on the right forum. I will first describe setting of FENE-PLog model in (1) then problem in (2).

(1) Geometry contains microcylinders with no-Slip boundary condition for velocity on Up/Lower walls and cylinder. outlet pressure fixed value 0 for out flow. tau and theta linear extrapolation on walls and cylinders.

i apply SIMPLE coupling method but i am also confused why i need to apply other methods like PISO or PIMPLE. Moreover, i am also suffering to select solvers with preconditioner and smoother for fields (U, p, tau, theta).

(2) I can introduce functions in solver for validity but currently i comments all of them. i need to compare contours first with experimental work. Attached diagram has velocity bubble at inlet contradict experimental results.

please go through diagrams attached below and advise me, please don't mind if i write something wrong. experts can guess my level of knowledge.

thanks! again and waiting for your reply

[dlahaye]

1/ Use SIMPLE for time-independent problem (picture at final destination). Use PIMPLE for time-dependent problems (video of voyage towards final destination). You probably want to stick to SIMPLE for time being. Please consult the documentation.

2/ The use of linear solvers is likely to be an issue of secondary importance for time being.

3/ I am guessing (without any sound basis except for own experience) that it is more beneficial for you at this point to monitor the convergence of the SIMPLE iteration. Please share plots of the convergence of the residual of the SIMPLE iteration. One can produce these plots by generating log file, running the utility foamMonitor on these log-files and plotting the results. The plotting part is discussed elsewhere in this forum. Ask here in case that ambiguities persist.

4/ Please use small enough relaxation factors for *all* fields (U, p, tau, theta) until obtaining reasonable convergence. Increase relaxation factors after obtaining reasonable convergence.

[Muhammad Ibrahim]

thanks once again!

Please find 'gnu' residuals graphs and guide me with your precious advice,

i want to share my setting but first i will try with your advice and after is better to share with you. i am try now with PIMPLE.

thanks!

[dlahaye]

Congrats for the residual plots! Well done.

Here is what I understand:

1/ Your residual graph shows that residual for the pressure drops from 1.0 (one) to 1e-5 (0.00001). This seems to be a home-run. You can tell your crew to pack and go home.

Here is what I am happily confused about:

1/ The x-as of the plot fails to show iteration number (as it now goes from 0 (zero) to 1 (one). How many SIMPLE iterations did the simulation require?

2/ How did you generate the graphs? Can you please use

simpleFoam >& log.simpleFoam 
foamMonitor log.simpleFoam 
ls logs/
more logs/p_0

and share what you learn?

3/ What is the residual plot for tau and theta?

4/ How does the computed field look like? Did the visuals change with respect to earlier post? Do we learn something new?

Keep up your efforts. Good luck, D.