CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Siemens (
-   -   fsi (

ALFIN July 3, 2007 13:45


i heard about star CD recently and i would like to know whether can i model FLUID structure interation and to find the stress and deflection of the structure using this software along with the fluid properties



Mike July 3, 2007 23:14

Re: fsi should be able to do this with ES-FSI.

Peter July 4, 2007 04:45

Re: fsi

Star has an interface with Abaqus for example. For high end / difficult applications like this start by calling CD-adapco.

On the subject of es-tools. Only is worth paying a license fee for. This last I knew about es.fsi was about six months ago. es.fsi is a collection of subroutines, no GUI, that you have to hack and play with before compiling for each job. This position may have changed, but adapco wants to see a return on investment before it spends more money on the this or any other tool.

Also es.fsi has limited functionality. It reads the super element format of the FEA stiffness matrix. It works for simple small displacement linear problems, but not much else. This is very clear when you see the es.fsi presentation that all CD and / or adapco salesmen have.

Another issue with es.fsi is that because of its unfinished nature, it is difficult to use, therefore it is sold only after a consultancy project first. so you give your first job to adapco at $5K per week minimum, then you pay for training, then you pay for the license, then you ... With es.fsi you are going to need deep pockets and long arms for this project.

Yes, Star can handle FSI problems, and handle them in various ways. It actually gives good results as well. Just call both the FEA and CFD vendors to get a balanced picture first.


Dave July 4, 2007 06:20

Re: fsi
Hi Alfin,

The good news is that yes, you can do FSI with STAR-CD. The bad news is that nobody (including all other CAE vendors) has come up with a "one-size-fits-all" solution for FSI; depending on your case you may be able to do it all in STAR, or need to couple to an FEA code (like abaqus), couple with a 6DOF solver, etc. There's a good article that outlines the various methods, giving examples, here:

Luca July 7, 2007 12:58

Re: fsi
Can you be more precise and give a little more detail of what you're going to do? I had no problem in setting up a FSI problem for aeroelasticity analysis of aircrafts inside FLUENT. Maybe I can help you a little and give you some suggestions. Luca

hayate July 8, 2007 06:05

Re: fsi

the company i work for (defence sector) is trying since last summer to find the best solution for FSI , we try everything... star cd with abaqus , fluent with abaqus , star with nastran (neinastran) ... about 5 or 6 codes cupling. the only thing that i understand is 1) for real FSI on tanks and pipes only dytran can give you the best results 2) the compination of abaqus with fluent is a full disaster... and of course now abaqus is a part of dassault and fluent ansys so if someone buy this solution in the near future it will not works..... 3) the only code that really is worth is ansys cfx -ansys mechanical every other solution... just too risk to put money on it , if someone buy star with fluent is also too risk due to the seperate companies policy, star with abaqus...too expensive nearly 200 000 euros .... now we are going to test a software named smartcoupling from to see the star cd with neinastran... we will see about that...

best regards

Peter July 8, 2007 07:07

Re: fsi
Dear Hayate,

In the days when Fluent had an exclusive arrangement with Fluent, MSC was getting cosy with CD-adapco. For obvious reasons. Some projects were talked about and development work to make the codes talk to each other on a better level involved CD London. This was back in 2005.

Once Ansys bought out Fluent, the Fluent / Abaqus arrangement quickly went to out of fashion. Abaqus then chased CD-adapco for a working agreement. CD-adapco was happy to chased by all relevant and potential partners. It is clearly in the interest of CD-adapco to make non binding partnerships with as many FEA vendors as possible. I would expect Star-CD to talk to MSC products, Abaqus and Permas as a minimum. Permas because adapco is (or at least was) the US reseller of Permas.

Therefore linking Star to various FEA codes for FSI should be pretty much deliverable and therefore supportable in 2007.

What I cannot understand without further information is the €200K price. This is very high.


hayate July 8, 2007 09:58

Re: fsi
hello again

well i also believe that cd adapco should stay along and make agreements with as many fea vendos as they can because now they have the opportunity to take eveyone that do not want to use a standar company like ansys (mechanical and cfx/fluent) or abaqus with something else....

i know the people in italy that create the smartcoupling software for FSI and as they told me their software is working with several codes .

unfortunatelly i never wor with pernas so i have no idea about that software , is it realy good? worth to see that?

about the price of the fluent abaqus .... my friend ... 100 000 euros for abaqus and 85 000 eyros for fluent...

best regards hayate

All times are GMT -4. The time now is 03:27.