Twin-screw extruder simulation - dynamic grids
 

Dear OpenFOAM-experts,

I have heard about a talk at the ProcessNet Meeting 2011 of Comitee for CFD and Mixing in Dortmund (in Feb.) with the title "Grid Movement in OpenFOAM at the Example of the Simulation of a Twin Screw Extruder".
Before this I did not consider CFD for extruder simulations, due to the high computational effort for the dynamic remeshing. Does anybody know about the capabilities of OpenFOAM for the simulation of a corotating twin-screw extruder, esp. for dynamic meshes?

Best regards,
Andreas

I'm also interested in that topic. I already thought of how to solve this problem, but none of these ideas are already implemented in OpenFOAM. I would be very much interested in how the TSEs are implemented in OpenFOAM in this presentation.

Hello,

if you are qualified to get access to the foamedOver =>see presentation: foamedOver -A library to Add a Dynamic overset Grid Capability to OpenFOAM” http://web.student.chalmers.se/group...SlidesOFW5.pdf . Well if you are allowed to use foamOver (see last slide), you will have no problems with the simulation of a corotating twin-screw extruder.

A second way is to use Radial Basis Function (RBF) and immersed boundary method.
Immersed boundary method was announced as in development, but if you look at http://www.sfbtrr75.de/website/index...mid=30&lang=de
Dr. Tuković
- Immersed boundary method with RBF (Radial Basis Function) interpolation in Open FOAM
=>One can think it has been implemented! but I do not know if it is publicly available=> contact Professor Jasak and Dr.Tukovic. But some funding might be nescessary before it goes public.

http://www.openfoamworkshop.org/2009...ynamicMesh.pdf =>slide 23 RBF slide 24 immersed boundary method

The work was done by Wikki for a commercial client and consists of automatic mesh generation, analytical mesh motion and lots and lots of GGIs and sliding interfaces. Simulations work well and it always impresses me how cool this is.

Hrv

Thank you Elvis and Hrvoje for your reply.

Hrjove - Would you be so kind as to give a more detailed information how this is done? I still don't have a clue how to achieve this with the already implemented methods, GGI and sliding interfaces, in OF. As I assume this will not go public, do you think that this is possible to achieve for an advanced OF-programmer (I'm actually far from being an expert...)?

Actually, no: this is expert work and if you've see the slides you will know what I have in mind. Henrik and I had to work very hard to get this to work fast, robustly and accurately and I learned something new in CFD as well! Bugs were fixed, algorithms improved and getting this to run in parallel is something I am particularly proud of!

Having in mind that I have personally written ALL the dynamic mesh and topological change machinery in OpenFOAM and that Henrik came up with a great idea for mesh motion, I doubt there's another team I know who is able to do this job: you need experts.

If you want to use something like this, we need to review the details, set up contracts etc.

Hrv