[vinz]

Hi everibody,

I'm pleased to anounce the release of the first version of the small converter gridpro2FOAM. This tool is made for the conversion of meshes createdwith GridPro ( for more informations) into files usable with OpenFOAM.
If you don't have this software, I included a 2D mesh of a cylinder in a tunnel in order that you can try the program and also try the mesh under OpenFOAM.
I also included some screenshots of the cases that I've run, and others will be available for people who would be interested.
This grid generator is the one that we use in my company. It's why we've dcided to make a small converter, when we begun to work with OpenFOAM. I f it can be useful for someone else, why not!
I'll hope you'll like it, and if there is any question I'll be pleased to try to answer.



Vincent

[vinz]

Ok I made a mistake. It seems impossible to upload files of that size (more than 50ko) so, I will put it on the website of my company and it will be available tomorow.
I tried to edit the post, but it was impossible since the post was "more than 0minute old". This was funny!!!
Anyway, I'm sorry for the inconvenience, gridpro2Foam will be available soon on the website www.rtech.fr.

Vincent

[vinz]

Hi everybody,

Due to some ftp problems I haven't been able to upload the file so far. But it's now done. The converter comming with an exemple of GridPro files can be downloaded following this link:
GridPro2FOAM

Hope you'll enjoy. Don't hesitate if you have questions either on GridPro or on the converter.

Vincent

[lr103476]

Hi Vincent,

Thanks for the converter, good job! Since a week I am running gridPro (trial version). I am really impressed about its capabilities to create smooth hexahedral meshes and all very quick. The main part which I find rather difficult is where to put extra boxes for cell clustering. Especially around wingtips and stuff like that.

I was wondering, could you maybe post some more GridPro examples (*.fra files)? (or the ones from your screenshots as a start).

GridPro is only used by a few bunch of people. Have you got a clue about that? In my oppinion it is much better than gridgen / gambit in generating hexahedral meshes.

Regards, Frank

[vinz]

Hi,

I've modified the code GridPro2FOAM. It's now able to create coraser grid automaticaly.
An example:
You want to create a mesh four times as coarse as you input file. If it's not possible (if the mesh is not divisible by 4 in one direction) it will try to make it three times as coarse, and then twice as coarse or finally leave it like this if the mesh cannot be coarsen, and this, independently for each direction.
The code has been quickly tested. So, if some people have the possibility to test it on some meshes, it would be great.
If thre is a problem, don't hesitate to report it.
The code can be downloaded there: GridPro2FOAM

Regards,

Vincent

[maddalena]

Hello everybody!
I have some problems to export a 3D mesh from GridPro, using both the gridpro2FOAM utility and the fluentMeshToFoam utility.
First I created my mesh using GridPro, setting the boundary conditions with the automatically boundary labeling. I chose Fluent as solver. Than I copied my .tmp file in ./GridPro2FOAM directory and run the converter. But when I looked to the boundary file in OpenFoamFiles/Temp, all the boundary where defined as "Unknown". More or less the same happened when I saved my grid as .msh and than used the fluentMeshToFoam utility.
It seems to me that there is some incompatibilities in the boundary definitions... Does anyone have the same problem? and, more important... Does anyone solved it?
thanks in advance!
maddalena.

[vinz]

Hi maddalena,

Don't worry, your problem is totally welknown. I don't convert the boundary conditions with gridpro2FOAM.
This comes from the fact that, in gridpro, each user can define is own bc names. This would be very difficult to convert each specific bc.
I was thinking about another tool to convert the bc but I don't have so much time to do it right now.
However, you just have to edit the bc file in openfoam, and you will see that everything else works fine.
What type of case are you studying?
DOnét hesitate to ask if you need some help.

Good luck

Vincent

[maddalena]

Hi Vincent,
thanks for your fast answer! This morning I succeeded in editing the boundary file and my simulation is running now. I am dealing with a 3D flapping wing, the mesh itself is not difficult to realize, but I have to assure that cell are fine enough especially on the leading and trailing edge of my wing, that is a 6% thick ellipse.
That's all! Thank you one more time!
maddalena.

[vinz]

Hi Maddalena,

I assume you're working with Frank Bos on the flapping wing in TU Delft?
I've seen the presentation of Frank during the last workshop in zagreb and the work done so far is very interesting.
Waiting forward to see your results too.
Good luck

Vincent

[maddalena]

yes, you are supposing right! I have almost finished with 2D wing validation, and I hope in some weeks I will have some nice results with the 3D wing also... I will post something when everything is finished.
regards,
maddalena.

[mrangitschdowcom]

Vincent,
Is the gridproToFoam converter still available? It's not on the link from 12-2006...

Thanks!


Mike Rangitsch

[vinz]

Hi Michael,

My mistake, we changed the ftp server and I didn't think about updating this link.
The new link is then:
http://www.rtech-engineering.com/GridPro2FOAM_3.0.zip

GridPro2Foam is now in version 3.0 with the possibility to scale a model before using it in OpenFOAM.
There is still the cylinder mesh as example going with GridPro2FOAM.

Hope it will be usefull, and if you have any problem, don't hesitate to ask.

Regards,

Vincent

[vinz]

Hello everybody,

A new version of GridPro2FOAM is now available. No new features, but bug correction which was causing problem on few grids.
New version can be found at:
http://www.rtech-engineering.com/Gri...OAM_4.0.tar.gz

Don't hesitate to contact us if you need any information about the converter or GridPro itself.

Regards,

Vincent

[vinz]

Dear openfoam users,

We are pleased to announce the last realease of GridPro2FOAM which can be downloaded at the following address:
www.rtech-engineering.com/GridPro2FOAM_5.0.tgz

Thanks to the feedbacks from different users, we managed to improve the converter in order to handle both periodic boundary conditions, and specific conditions per block.

This will be useful to create meshes with static and rotating parts for instance, or meshes on periodic geometries.

This should permit to convert almost every mesh created with GridPro.

If you are a user of GridPro, or that you want to try it, but that something specific to your case is missing into the converter, do not hesitate to contact us. We will do our best to implement it.

Waiting forward some more feedback.

Regards,

Vincent

[hjasak]

Hello Vincent,

I would like to include this into the dev line. However, and after some basic cleaning I get tons of warnings:

Examples


RTechTimer.C: In member function 'void RTechTimer::Start()':
RTechTimer.C:43: warning: use of old-style cast
RTechTimer.C:45: warning: use of old-style cast
etc

Block.C: In member function 'void Block::PrintEquivalent()':
Block.C:132: warning: comparison between signed and unsigned integer expressions

exampleVector.C:14: warning: ISO C++ forbids declaration of 'main' with no type
exampleVector.C: In function 'int main()':

exampleVector.C:20: error: 'printf' was not declared in this scope
exampleVector.C:22: warning: comparison between signed and unsigned integer expressions

pair.C:30: warning: format '%d' expects type 'int', but argument 2 has type 'size_t'
etc


There is lots, this is just a selection. Do you want to do a cleaning job on it or shall I?

Also, I take it this is a stand-alone application and you did not use any of the OpenFOAM container types etc. Any particular reason?

Hrv

[vinz]

Hello Hrvoje,

Actually the warnings you get are quite strange.
I do not get any of them when compiling on my machine (fedora 9)
I do not get any warning neither compiling on two other machines with different distributions.
Furthermore, why do you get warning on C files while we wrote the code in C++ and the files have a .cpp extension?

Here is the output of my compilation:
g++ -c -g Patch.cpp
g++ -c -g Node.cpp
g++ -c -g Block.cpp
g++ -c -g gridpro2FOAM.cpp
cc -c -g util.c
g++ -c -g Triplet.cpp
g++ -c -g Face.cpp
g++ -c -g RTechTimer.cpp
rm -f gridpro2FOAM
g++ -o gridpro2FOAM Patch.o Node.o Block.o gridpro2FOAM.o util.o Triplet.o Face.o RTechTimer.o \

So actually, I don't really know what i should clean! :D
Tell me a bit more so I can try to help you.
By the way, what are the cases you try to use GridPro on? If you have some questions about topology building, do not hesitate to contact us directly, we can talk about that.

On the second hand, this is indeed a stand alone code since the beginning.
To be honest we did not want to be dependant on changes made in OpenFOAM from one version to the other which might have impacted the converter.
And the last but not the least, we wanted to do something general to able to adapt it quite rapidly to other CFD softwares if needed.

However a direct export to OpenFOAM should be present in the next release of GridPro, which will make the use of GridPro with OpenFOAM even more easy!

The last thing I would like to add is that if someone has questions or improvement requirements, one can contact us at the following adress:
support@rtech-engineering.com

Regards,

Vincent

[vinz]

To everybody,

Small changes have been done like adding scaling feature and handling of axisym cases.
New version can be downloaded here:

www.rtech-engineering.com/GridPro2FOAM_5.1.zip

Note also that starting from version 5.0 and later the new command to convert a GridPro case is:
gridpro2FOAM <gridfile.tmp> <openfoam> [--option]

options are now:
--medium: creates a mesh removing half of the points of the GridPro mesh
--coarse: creates a mesh keeping one fourth of the points from the GridPro mesh
--axisym: creates an axi-symmetric grid from a 2D GridPro file
--scale x: scales the GridPro mesh with a factor x


Regards,

Vincent

[vinz]

Dear OpenFoamers,

Here comes a new version of GridPro2FOAM.
The latest version 5.6, now handles the internal surfaces boundary conditions as used in OpenFOAM. The cyclic patch needs to be defined as an internal surface in GridPro. By default the internal surfaces are usualy treated as standard block connectivities. However, by specifying the option "--cyclic surface_number" at the end of the command line, the internal surface will be treated as a cyclic patch and converted into a separate boundary condition inside the OpenFOAM boundary file.
It is also possible to specify more than one cyclic patch unlike in the previous versions.
This update has been possible thanks to the feedback and the help from Mikko Auvinen. I encourage people using the converter to not hesitate to send us there needs or comments.

The software can be downloaded at the following page:
http://www.rtech-engineering.com/news.html

[egp]

Hi Vincent,

I just used gridpro2Foam, and wanted to say that it worked just fine, even the --coarse and --medium options (although a few blocks gave me warnings that certain directions were not divisible by 2).

Thanks!

[vinz]

Hi Eric,

Thanks for the feedback. Sorry for the late reply, just coming back from holidays. There is a new version 5.11. I don't have it with me right now but will upload it as soon as possible. There are some new features to handle actuator disk and some other things which could be interesting.
If it's not confidential, which kind of mesh did you try to convert with gridpro2FOAM? I'm always interesting in the complexness of cases it can convert.

Regards,

Vincent