|
[Sponsors] |
November 30, 2005, 05:00 |
StarCD or Fluent
|
#1 |
Guest
Posts: n/a
|
Hello CFD giants,
I am a fluent user and have not so far used star CD.But i have a plan to. Please tell me frankly which one is better(User friendly and stable) |
|
November 30, 2005, 05:18 |
Re: StarCD or Fluent
|
#2 |
Guest
Posts: n/a
|
It depends what you want to do and how you want to do it, if you a fluent user you would like STAR-CCM+ as it is very user friendly and really nice to use, although it currently doesnt have all the models that fluent or STAR-CD has, but they are on the way from what I understand. I know a couple of ex-fluent users who say that CCM+ is a lot nicer than fluent.
I think it would be fair to say that STAR-CD and pro-STAR are harder to learn than flunet but if you are doing complex modelling is ultimately a better code, ceratinly the moving mesh capabilities and complex chemistry is better and the user coding in STAR-CD is very impressive and allows you to customise pretty much everything you can think off. But as I say it all depends what you are doing.... |
|
November 30, 2005, 05:45 |
Re: StarCD or Fluent
|
#3 |
Guest
Posts: n/a
|
Thanks Ben for ur valuable comments.
|
|
November 30, 2005, 21:45 |
Re: StarCD or Fluent
|
#4 |
Guest
Posts: n/a
|
Ben, it really boils down to what one intend to do. there always been debates about which solver is better. For instance take starCCM+ and fluent6, when it comes to segregated solver (i did not read about coupled details so no comments about them), they both are exactly same as far as descretisation is concerned. There are two very minute differences, one the treatment of boundary conditions is bit different, and then the application of URF, starCCM+ uses implicit urf implementation that increases the diagonal dominance and is thought to be better, but can screw up if it is too low 0.00001 (sometimes its debated) fluent on the ther hand has the form phi_new = phi_old + urf (phi_new - phi_old), they are minute difference in the end.
So the questions when we compare them really boils down to what model you are going to use, for example, if one has to do lot of user written input, taking fluent will be pain in arse howsoever good that might be. But if someone wants to do combustion anaylsis, he then has to do a reasearch whether the model he intend to use is present in the solver of not, for example fluents Eddy dissipation Concept (as i understand is not present in other solvers but the model si sbest as i understand .. i might be wrong wincei am not much aware of cfx etc upgrades). Internally all this solvers are same (almost) and the reason is the best way of doing the things is known to everybody and they all have tried to implement the best, the example for this is , take gradient limiters , fluent uses venkatkrishanans limiter and so does starCCM+ , because this is best limiter known. So if we can peel off the skin we see everything the same or similar. just we need to dig up the docs. |
|
December 1, 2005, 05:07 |
Re: StarCD or Fluent
|
#5 |
Guest
Posts: n/a
|
I think that was the point I was trying to make although I guess maybe it doesnt come across quite as intended. Certainly there is little variation between the codes and the "which code is best" argument is fairly irrelevent for general flow situations where complex physics are not concerned.
As you say if you want to do something specific then it is a different matter, Star is widely regarded as best for moving meshes and probably considered best for automotive and in-cylinder apps of the "big three", Fluent is noted for its ease of use and it popular for aero and CFX is pretty much untouchable when it comes to turbo machinery. Outside of that it is just what you like. A lot of users are fairly imbeded in the code they use and are generally reluctant to swap due to the time and cost of retrainning on what is, no matter how user friendly, still one of if not the most compicated type of software you can buy (although I hear that structural dynamics packages are fairly nasty too). |
|
December 1, 2005, 06:27 |
Re: StarCD or Fluent
|
#6 |
Guest
Posts: n/a
|
i think i was just enforcing what you said.
|
|
December 5, 2005, 07:37 |
Re: StarCD or Fluent
|
#7 |
Guest
Posts: n/a
|
I cant say anything from a solver perspetcive, as I am a new learner of Star. But from user point of view Fluent is better and easier to learn too compared to Star, which I feel is difficult. One more aspect I learn today only. In Fluent one can apply a velocity /mass_flux inlet "normal to boundary". No need to worry about the plane and and the axis of the region. In star, either one should know exact u,v,w or need to create a local plane aligned to B.region and then apply the BC. No way to apply a "normal to boundary".
|
|
December 5, 2005, 10:13 |
Re: StarCD or Fluent
|
#8 |
Guest
Posts: n/a
|
I generally agree with the previous posts, but I think one of Star-CDs biggest strong-points compared to Fluent (and CFX) is the mesh-handling capabilities. For complicated and/or awkward geometries this can be a great advantage. Apart from that, Star-CD is tricky to get into and sometimes not very user-friendly , but very versatile. Fluent is easier to use for 'standard' problems. The new Star-CCM+ is much easier to use than Star-CD (I've heard that CD-Adapco have lured quite a few developers from Fluent...), but is not complete yet with regards to physics.
|
|
January 22, 2006, 23:29 |
Re: StarCD or Fluent
|
#9 |
Guest
Posts: n/a
|
hi guys,
I am a new user of star, along with i am using Comet also for ship hydrodynamics application. Comet is developed by ICCM and is taken over and marketed by Star. I validated Fluent for the ship hydrodynamic applications. Although i concluded RSM and SST k-omega performs well there is always a confusion about cd cl convergence. it is taking a very long time for cd cl convergence, about 4 to 5 days. But using comet one can accelerate the body, and thus we can get the drag values over a range of Fn in a single run. But we have to go for ICEM for the creation of mesh. with regards Sulfi |
|
February 8, 2006, 10:00 |
Re: StarCD or Fluent
|
#10 |
Guest
Posts: n/a
|
Interesting comments about Star's Mesh generation. I friend of mine used to use Star for mesh generation as it offered great scope for moving individual nodes etc. However he had to do this a lot and whilst it was a great facility to have he would have prefered to not to have to do it at all. Hence he moved to ICEM. Now he no longer needs to use that great Star feature. Question: was it actually a feature or a necessity ?
|
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Fluent & StarCD | İbrahim Utku | FLUENT | 0 | March 28, 2006 08:14 |
Fluent - StarCD | chusmann | Main CFD Forum | 11 | September 3, 2005 04:03 |
FLUENT/CFX/STARCD as CFD modelers | carlos | Main CFD Forum | 2 | December 9, 2004 01:17 |
[:::o:::] Iges ? Fluent vs StarCD | Natacha | Main CFD Forum | 5 | April 3, 2002 05:33 |
Why STARCD slower than Fluent and CFX | hsu | Main CFD Forum | 6 | June 5, 2000 08:59 |