CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM Programming & Development (https://www.cfd-online.com/Forums/openfoam-programming-development/)
-   -   any plans for Chimera grids?? (https://www.cfd-online.com/Forums/openfoam-programming-development/63790-any-plans-chimera-grids.html)

abhishek_basak April 20, 2009 05:27

any plans for Chimera grids??
 
Hello all,

I wonder if there are any plans for Chimera Grids in OpenFOAM. As per this link (http://www.outreach.psu.edu/programs...erson-Eric.pdf).. a version of openFoam called overFOAM is being developed with this feature, it is going to be made open source??

Abhishek Basak

thomas78 May 17, 2009 07:04

Hello Abhishek,
I was hoping that somebody would have answered your question, because I have similar interests.

I have started my own experiments writing a 2D chimera grid code, but I'm afraid the donor search is far too slow for any practical purpose.

If any libraries like dirtlib and suggar (mentioned in Abhisheks link) were available to Openfoam user, I'll be happy to work on the integration with the Openfoam libraries. Does anybody know available codes or would you like to join efforts to make a chimera library for Openfoam?

Cheers
Thomas

boger January 30, 2010 00:46

overset grid implementation in OpenFOAM
 
3 Attachment(s)
Following his proposal at the Overset Symposium (which Abhishek referenced above), Eric tasked me with this work. After a few months of ramping up on OpenFOAM, I was able to implement a dynamic overset grid capability.

I worked with Ralph Noack and used his SUGGAR and Suggar++ libraries to perform the overset assembly and his DiRT library to handle the intergrid communication and interpolation. I also wrote an lduSolver based on PETSc to facilitate implicit treatment of the intergrid boundary conditions.

The result is a library that inserts easily into solvers in both 1.5-dev and 1.6.x. The overset assembly may be static or dynamic, with grid motions that can be prescribed or governed by 6DOF equations of motion.

The attached pictures are simple teasers from potentialFoam and rasInterFoam. As I am cleaning up the final details, we're just now shifting more into applications, so expect to see the flashy stuff at the various upcoming workshops (no doubt beginning with the First http://www.cfd-online.com/Forums/ope...enn-state.html, which Eric is organizing for March 2010).

David

kris2010 February 1, 2010 17:53

Hi David! This is fantastic news. Especially with moving geometries I can imagine that your work will really improve Openfoam performance. Unfortunately the mentioned overset libraries seem to be difficult to find. May I ask if you plan to share parts of your work? The pictures look great, but I'm sure I am not the only one who would like to understand how they are produced.

Kris

egp February 2, 2010 07:42

http://www.cfd-online.com/Forums/images/editor/menupop.gif
 
David has done an awesome job developing the foamedOver library. However, public release of this library is uncertain and at the discretion of the sponsor. If it becomes available, David will announce it here on the forum.

Overset meshing is a major enabler for dynamic bodies, and complex geometries. An example for a moving-body/water-entry problem is shown in the following animation:

http://dl.dropbox.com/u/2182201/waterEntry2Dalpha.mp4

kris2010 February 2, 2010 16:15

I am sure that everyone agrees that he has done a good job.

Abhishek started this thread with the question whether such a thing could be made open source. If Davids very beautiful results are presented to answer the question about open source features, he should have written 'Look, what interesting demonstrations can be done with Openfoam. But hey, there are only few people who will be able to get access to the code.' Can you possibly help with information whether it is possible to use those overset libraries for an absolutely non-commercial bachelor thesis? Who to contact?

Of course nothing is wrong with extending Openfoam privately with closed source. But what is the purpose of publishing the pictures in this place?

Kris

egp February 2, 2010 16:24

Hi Kristina,

The purpose for posting on the forum is simply to show that it can be done, in response to the OP. That's it. Nothing more.

One of the "problems" with overset methods is that there are a number of other pieces of software that all have to work in concert with the flow solver. Obtaining, installing, and learning how to use is still difficult, and keeps it in the realm of specialists (e.g., see the overset symposium proceedings).

Eric

kris2010 February 2, 2010 17:07

I find this argumentation weird. You say it is difficult to obtain and use the necessary pieces, that's why only specialists get access to it. I say if you shared the pieces, they would become well documented and easier to use.

So I will have to face the facts and choose a less interesting project. Hopefully future generations of students will be in a better situation.

Kris

boger February 3, 2010 08:22

Hi Kristina,

I'm sorry if the pictures are making you angry. Why post them? As Eric said, partly to show that it can be done. Partly in response to Abhishek's original post to say we'd followed through with our original plan. Partly to advertise that we'll have more to say at the upcoming workshop to help you understand how they are produced. Partly to generate interest from, or maybe even motivate, people like you, which can ultimately help to put us into a position where we are able to hand you the pieces. Unfortunately, someone has to pay us to do our work, and we don't always get the final say in how the result gets distributed.

Having said that, we, and more to the point, I personally, have written and released open source software to the overset community before. If you are interested in a free copy of USURP (http://pdf.aiaa.org/preview/CDReadyM...V2006_1148.pdf), for instance, please drop me a line.

David

kris2010 February 5, 2010 14:35

Thank you. But it seems that USURP depends on Suggar and Dirtlib. This does not solve the problem. I have given up to study overset grid methods.

Kris

openfoam_user February 20, 2012 07:33

Dear David,

Any new about a possible release of the foamedOver library ?

Best regards,

Stephane.

openfoam_user November 21, 2012 09:18

Chimera grids in OpenFOAM
 
Hello all,

I wonder if there are any plans for Chimera grids in OpenFOAM.

Stephane.

elvis January 6, 2013 08:22

Hi there is a commercial version available to those who qualify for a EAR-99 export licence.
See http://www.celeritassimtech.com

hxaxtma October 14, 2015 11:11

Just in addition to the threads above, couple of years ago, there is still no open source library for chimera capability in OF or foam-extended, right?

And to all the "commercial versions" of the lib, I think this was not the "spririt" of OpenFOAM (acutally foam-extended). This is really sad.

So I am having an idea here, what about creating a group of OF users which are willing to code a chimera lib which is really open source? Of course it is seen as a working project in your free time, not paid, but it is a great opportunity to work and learn together. I would be in!

Flowkersma August 16, 2016 10:29

Hi,

I read from http://www.openfoam.com/services/community-projects.php that there is an ongoing community project OverFOAM which purpose is to develop overset meshes to OpenFOAM. Does anyone know more about this project? I could not find any mention about it from their repository even though they mention on the website that it is "actively progressing".

Regards,
Mikko

akidess August 16, 2016 10:54

I don't think that repository includes any code that isn't ready for user deployment. If you are interested in OverFOAM, most straightforward is to directly contact OpenCFD/ESi.

elvis October 13, 2016 13:30

If I Unterstand Professor E. Patterson correctly Wiki implemented a commerial overset Version for a Client.

elvis December 2, 2016 13:17

Application of an Overset Grid Method for Aerodynamic Problems
Andreas Groß (Uni Rostock)
just to post another hero


All times are GMT -4. The time now is 09:23.