CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   STAR-CD (http://www.cfd-online.com/Forums/star-cd/)
-   -   StarCD 3.26 & 4 Importable File Types (http://www.cfd-online.com/Forums/star-cd/77179-starcd-3-26-4-importable-file-types.html)

trex930 June 16, 2010 01:43

StarCD 3.26 & 4 Importable File Types
 
Hi all,

I was curious as to whether anyone has a list of importable file types for each of the different versions.

Specifically I was hopping to import .ccm file into StarCD 3.26. But I have yet been able to find out how to do it. Any help would be greatly appreciated.

Regards
T

olesen June 16, 2010 02:15

Quote:

Originally Posted by trex930 (Post 263165)
Hi all,

I was curious as to whether anyone has a list of importable file types for each of the different versions.

Specifically I was hopping to import .ccm file into StarCD 3.26. But I have yet been able to find out how to do it. Any help would be greatly appreciated.

The only possibility that I can imagine is to save the ccm file as cel/vrt format (from star-cd v4). You then need to convert the cel/vrt files into a cel/vrt format that starcd v3 understands. I provided a small program (star-backport) sometime ago on this forum that would do that, with the caveat that it will not work for anything but primitive cell shapes (ie, hex, tet, prism but NO polyhedrals). If your ccm file has polyhedrals, you are stuck and will have to forget about importing it into starcd v3.

trex930 June 16, 2010 02:19

Cool

I just tried out StarCD v4.12 today and it worked. .ccm files can be imported by StarCd v4.12.

Now here is another challenge for those of you out there.

Is it possible to convert a StarCD v4.12 .mdl file to one thats usable in StarCD v3.26?

I initially thought that it was possible. But I keep getting errors. This could however be becaues of the fact that I imported the .ccm file in v4.12.
But im not too sure.

I really do appreciate the help!

Thanks in advance

T

RobertB June 16, 2010 07:24

Yes, generate a prismatic/tet mesh in CCM+.

Remember to set the option in the prism mesh model to generate tets/prism only.

Write out a ccm and read back into Star 4.

Write out ASCII cell and vertex files

CD/adapco have a short Fortran program that will convert the files to STAR 3 format (they should give it to you if you ask).

If you get errors you have Polys in the CCM+ (despite having asked for none). The usual error is that the boundary layer is too thick and the code collapses the prism layer, generating polys. Try a thinner boundary mesh layer.

You should then be able to suck the model into STAR 3.26.

You might want to use the command TETAlign after import to optimize the tes for STAR 3.26.

trex930 June 16, 2010 19:41

Hey Rob,

Thanks for the reply. I'll give it a shot today and keep you posted!

When you say prismatic/Tet. You're talking about prism layers, as well as a tetrahedral mesh right? Is it not possible to do it with a trimmer mesh? Because ideally that's what we would like to use.

Thanks in advance!

T

olesen June 17, 2010 02:36

Quote:

Originally Posted by trex930 (Post 263170)
Cool

I just tried out StarCD v4.12 today and it worked. .ccm files can be imported by StarCd v4.12.

One point you need to watch out for when reading the ccm files is the model scaling used. If you start from a fresh mdl file and then read the ccm file, the proSTAR will have all the vertices scaled per meter. If you start from an existing mdl and previously wrote a geometry with scaling (eg, geom,,0.001,ccm,nobackup), that scaling will be used when reading the ccm file.
You thus may or may not need to rescale your vertices accordingly.

Quote:

Originally Posted by trex930 (Post 263170)
Now here is another challenge for those of you out there.
Is it possible to convert a StarCD v4.12 .mdl file to one thats usable in StarCD v3.26?

You can only do this indirectly. Save the model as coded and read that it in the older version. With some luck it will understand most of the commands and ignore the unknown ones. Of course you need to make sure that the cread/vread within this .inp file use the correct v3-compatible format.

trex930 June 17, 2010 03:54

Hey again!

Thanks for your input! Just have a quick question for you!

Quote:

Originally Posted by olesen (Post 263346)
Of course you need to make sure that the cread/vread within this .inp file use the correct v3-compatible format.

Is there an easy way of going about this? Or is this the small little program that Im waiting for from CD Adapco??

Cheers,
T

olauer June 17, 2010 04:46

The real problem between 4.x and 3.26 is that in 3.26 only some very special polyhedral cells (called trim cells) are allowed. I do not think that it is possible to configure 4.x or CCM+ in a way to create trim meshes which can be read back to 3.26. The created polyhedral cells (even in trimmed meshes) are not backward compatible to 3.26 trim cell definitions.

The only thing I could imagine is - as already said - tets and prisms.

olesen June 17, 2010 08:35

1 Attachment(s)
Quote:

Originally Posted by trex930 (Post 263356)
Hey again!

Thanks for your input! Just have a quick question for you!

Is there an easy way of going about this? Or is this the small little program that Im waiting for from CD Adapco??

You can try using the attached star-backport and star-foreport utilities for converting star v3 <-> v4 cel/vrt format. They don't handle poly cells or trimmed cells, but should manage primitive cell shapes fine.
I believe they still work okay, but haven't used them for quite some time. See the '-h' option for the usage.

/mark

trex930 June 17, 2010 20:16

Quote:

Originally Posted by olesen (Post 263410)
You can try using the attached star-backport and star-foreport utilities for converting star v3 <-> v4 cel/vrt format. They don't handle poly cells or trimmed cells, but should manage primitive cell shapes fine.
I believe they still work okay, but haven't used them for quite some time. See the '-h' option for the usage.

/mark

Cheers man! Thanks a bunch! I wont be able to try it out for a couple of days. But I will let you know how it goes as soon as I can! Thanks again!:D

T

TMG June 18, 2010 11:01

Instead of jumping through so many hoops to force things that were never meant to work together, why don't you just use STAR4? What is the big deal about forcing everything back to STAR3? STAR4 was built to use general polyhedra and is fully compatible with anything you can generate in a .ccm file. You obviously have STAR4 so why are making so much work for yourself?

RobertB June 18, 2010 19:10

Have you actually used STAR 4 for real sized problems, 10 - 20 million cells? The plotting is so slow as to make it practically unusable.

If you like and use Prostar use 3.26. If you like CCM+ and Polys use it. Star 4, with the best will in the world, neither has the toolbox benefits of 3.26 nor the just 'leave it to me' benefits (?) of CCM+. I use both and have not, to date, found a compelling area where Star 4 is the answer.

There may be some area of physics in which it is compelling but I do not need that benefit.

trex930 June 19, 2010 05:55

Hey!

Thanks for you input guys!

@TMG: I understand what you mean. I have star 4, but the Engineer that needs the files that Im working on only has star3.26! Thats why im going through all this work! haha

may I also ask Mark: I have managed to extract the files from the link you posted. (Thank You very much by the way). But Im having trouble running it. Linux recognizes that its an executable, but when i type in the name (eg. star-backport) it says unknown command. Im fairly new to Linux, so any hints and tips would be super beneficial!

T

olesen June 21, 2010 02:19

Quote:

Originally Posted by trex930 (Post 263669)
Linux recognizes that its an executable, but when i type in the name (eg. star-backport) it says unknown command. Im fairly new to Linux, so any hints and tips would be super beneficial!

My guess would be that '.' is not in your PATH, thus you'll need to specify the complete path to the programs.
eg,
Code:

/some/path/to/program/star-backport
OR
 ./star-backport
(if it is located in the current working directory)

The other possibility is that perl is installed somewhere other than /usr/bin/perl on your machine, in which case you'll need to adjust the first line of the programs accordingly.


All times are GMT -4. The time now is 11:52.