Is ANSYS capable of doing this ? 1

[sadiq3210]

i m working on a project with fiber glass that has to be saturated with resin by the application of vacuum.

Is it possible to find the vacuum pressure, fill time of mold, point(s) of application of vacuum and resin inlet point(s) through ANSYS ?

 

[rickfischer51]

Depends on what you mean by Ansys. The Fluent Polyflow module might be able to do this, but I cant imagine how this would be done in Multiphysics. This is a polymer flow problem which requires a volume of fluid solution with a shear thinning fluid. Not sure what you are doing, but this sounds like it is essentially a mold filling simulation, with a channel filled with class fibers that the resin must flow around and through. Flotran (now undocumented) cant do it, and I'm not familiar with CFX.

Rick Fischer
Principal Engineer
Argonne National Laboratory

 

[sadiq3210]

Any CFX user/expert here?
please reply to this thread

 

[Dave442]

Hi Sadiq3210,

Since you haven't gotten much of a response here I would try the CFX forum on www.cfd-online.com. There are plenty of CFX experts there and I've gotten great advice in the past.

Good luck,
Dave

 

[sadiq3210]

thanks for the info. can you please tell me what is CFX ?

 

[rickfischer51]

CFX is the CFD package Ansys purchased to replace Flotran. It is part of Multiphysics. Geeze, RTBM. Or google it.

Rick Fischer
Principal Engineer
Argonne National Laboratory

 

[loki3000]

what is the difference between fluent and cfx then?

 

[rickfischer51]

Fluent is arguably the best CFD program out there. It is marketed as its own product line independent of Ansys Multiphysics, and is very pricey. CFX is part of Multiphysics, although it can be purchased separately. It is less capable than Fluent. It replaces Flotran, which was undocumented at R15.

Rick Fischer
Principal Engineer
Argonne National Laboratory

 

[sadiq3210]

so, flotran, fluent and CFX are all software for simulating CFD ?

 

[loki3000]

so they have two products that do basically the same thing, if i understand correctly (one of them being inferior, but still under dev)?
got to love development via acquisitons strategy. simulia doesn't have those problems.

 

[loki3000]

meanwhile APDL, which is still used a lot in academic circles (and industry, but somewhat less), is getting exactly 0 time of development and is allowing abq to overtake it from the right.

 

[rickfischer51]

Flotran is an old finite element formulation that was easily integrated into Multiphysics, but the FE formulation has certain computational drawbacks. For instance, it does not calculate the divergence term, so it gives poor pressure drop predictions for a less than ideal mesh. It has been left in place to satisfy legacy issues, and has not been developed for some time. CFX was acquired as a replacement. I wouldn't call CFX inferior to Fluent, as what it does it does well. It just doesn't have all the capabilities of Fluent. Or the huge price tag.

Not sure what you mean by APDL gets zero development. Ansys Parametric Design Language (APDL) is the Ansys command language. The Workbench environment writes an APDL script everytime it runs a job. No APDL, no Workbench. APDL is still used extensively in industry. All the Ansys power users use it extensively. Newbie millenials raised on GUI's dont like it because it requires learning, and gravitate towards Workbench. MAPDL (Mechanical APDL) is the latest name the marketing twits have tackled on traditional non-Workbench Ansys, which was known as Ansys Classic. The primary difference is that MAPDL uses the XOX geometry kernal, which is obsolete and no longer developed, while workbench is Parasolids based. Any new capability in Workbench, except for meshing, is in MAPDL. So new elements, new solvers, new materials are in MAPDL. There are new scripting capabilites in Workbench, and ACT (Application Customization Toolkit) which looks like it's aimed at APDL. I havent looked at it yet, but I suspect it also writes APDL scripts. I wonder what the future holds

One thing that concerns me is that at R15 Ansys undocumented Design Optimization, Topology Optimization, and HF Emag. The excuse was that these were old technologies, and by the way, we have new products to replace them, at a price. They claim this is not a bait-and-switch tactic, but really how dumb do they think we are? At the rate they are going, they will have to change the name to Monophysics. Ansys, as a company, has been going down hill since Swanson left. The marketing twits seem to hate the original Ansys product, and are hell bent to change it for the sake of changing it. I wonder what John Swanson thinks of all this.

Rick Fischer
Principal Engineer
Argonne National Laboratory