[kerstin]

Hi OpenFOAM-Users,

In my thesis I have programmed a solver for laminar fiber suspension flow in OpenFOAM and before everything gets lost, I would like to contribute it to the community.
Unfortunately I have never written an english paper and didn't clean the code up. And I have not written a documentation. So everything is very messed up and done with a lot of workarounds and clumsy programming. It is possible, that there are some files already dispensable in the archive...But I give the short facts here:

-Basis: icoFoam (version OF-1.3)
-last version I compiled was with OF-1.3. I get some errors with OF-1.4...didn't try to resolve it so far

- name of the solver: nonNewton
-Fiber suspension Model:
- equations are documented in my (german) thesis
http://hdl.handle.net/2003/24559 (chapter 6)
- closed model, closure for 4th order configuration tensor, differential equation for second order configuration tensor
- hybrid closure from Advani and Tucker
- Invariant based closure IBOF-5 from Chung and Kwon
- Fiber-Fiber interactions with approach from Folgar and Tucker, interaction coefficient with approach from Bay and Phan-Thien

- models are selected with switches in constant/fiberProperties and constant/RheologyControl and constant/MatrixProperties

-Phan-Thien/Tanner (PTT) model for viscoelastic matrix. (I never really tested the implementation and didn't use it!!!)

I don't read here regularly, so if someone is really interested, I will do my best to help. Unfortunately the solver is not as ready to use, as I would prefer to have it...
So you may write an email to
thesis at kerstinheinen.de

- solver packed in nonNewton.tar.gz
download at www.kerstinheinen.de/nonNewton.tar.gz
- example 2D channel with a cylinder www.kerstinheinen.de/nonNewton_example.tar.gz

(sry, actually I am to stupid, how to upload the files in this post)

Thanks to the OpenFOAM developers. Keep it running! It's great work!

Kerstin

[kerstin]

Hi all,

I updated and renewed the implementation of the viscoelastic Phan-Thien/Tanner Model for laminar flows in OF-1.5.

This is a multi-modular implementation for use of superposition of different parameter sets.

Attached a description of equations and implementation and also related shortcomings. This is more or less an unsophisticated hack, but still better than nothing for all interested in PTT actually.

[kerstin]

laminarPTTMultimod.tar.gz
case_laminarPTTMultiMod.tar.gz
Description_of_PTT_Model_in_OpenFOAM-1.5.zip

[kerstin]

Hi all,
I cleaned up my old implementation of a laminar fiber suspension code and updated to OF-1.6.

Now its hopefully more straight forward to understand and include in other codes..
This code did not have extensive testing so far. It still needs more validation...So handle with care and don't blame me, if you get strange results

Actually I just copied the fiber files into icoFoam and nonNewtonianIcoFoam. Long term plan would be to have it in the same general way like the viscoelastic solver...

The code writes out also the eigenvalues of A2. lambdamax is the largest eigenvalue... Still missing is a post-processing utility to view the fiber orientation as vectors or orientation ellipsoids...

- The test case attached is a channel with cylinder.
- just run the script runAll.
- general advice: the fiber suspension solver should always have as a first step running for a newtonian velocity field with icoFoam...otherwise it can be quite unstable...

- the example case is not perfect. You will get some warnings for negative eigenvalues, which dissapear after some time...its maybe due to bad initialisation...

One nice extension would be a model for turbulent flows...where orientation of the fibers is coupled to turbulence properties... Is there actually anyone out there working on fiber suspension models (based on the orientation tensor approach) for turbulent flows?

Kerstin

[kerstin]

Next step done: attached a pdf doc with the equations used in the solver uploaded above...



Kerstin

[pawanhv]

Hello, I am a beginner here.. I am interested to know the difference in results in the example here if I change the duration from 2mins to 4mins? , It crashes after 4.1mins I guess

[krikre]

Thank you, Kerstin!

I've added your code to openInjMoldSim and credited you in the readme https://github.com/krebeljk/openInjMoldSim