[flytoceres]

Dear readers,

I am a new user of both Solidworks Flow Simulation (SFS) and ANSYS. I have been trying to understand more what types of problems SFS fail to provide reasonably good solutions. So far, after some extensive googling and reading, I see some people pointing out there are not enough turbulence models in SFS that could lead to insufficient results. I personally failed to reproduce some jet impingement heat transfer simulation with good results in SFS.

What types of relatively small-scale application/flow problems would you definitely not use SFS, and choose ANSYS or other more powerful software?

Thank you for your precious time.

[sbaffini]

While I surely am a Fluent advocate, this sort of comparison requires being proficient in both compared codes in order to produce unbiased results.

If you have access to both codes, try to replicate the same flow with settings as similar as possible, if not identical. Only in that case you can say something, but only for that specific setting you investigated.

However, in general, Fluent is a much mature, industry proven, solution that didn't had to adapt to any CAD to simplify the user job. So, it is probable that its lower user friendliness also comes with less compromises on accuracy or other implementation aspects.

[CFDfan]

Cradle's scStream and scFlow have a variety of turbulence models and are go0d choices too.

[CFDfan]

Originally Posted by sbaffini

While I surely am a Fluent advocate, this sort of comparison requires being proficient in both compared codes in order to produce unbiased results.

If you have access to both codes, try to replicate the same flow with settings as similar as possible, if not identical. Only in that case you can say something, but only for that specific setting you investigated.

However, in general, Fluent is a much mature, industry proven, solution that didn't had to adapt to any CAD to simplify the user job. So, it is probable that its lower user friendliness also comes with less compromises on accuracy or other implementation aspects.

difficult to comprehend that being more user friendly would compromise the accuracy

[sbaffini]

Originally Posted by CFDfan

difficult to comprehend that being more user friendly would compromise the accuracy

Probably a bad wording on my side, I didn't mention user friendliness as absolute (like a well designed GUI, etc.) but in terms of the need to adapt to a CAD and a CAD-based workflow.

If you have a CAD-based workflow, which means that, no matter what, it is going to be the way you do things (because that's the product you are selling), then chances are high that in some circumstances you have to sacrifice accuracy to follow that workflow. Automatic meshing without alternatives, for example, is always going to produce artifacts that, I guess, the solver is supposed to handle otherwise.

Now, I'm not into Solidworks, but I don't know of any available reference on the details of the solver and mesher. So, the truth is, we don't actually know what it does.

Besides this, when you are taken away the minute controls on the software, what actually happens is that you are taken away the possibility to properly verify the code.

Not a definitive proof, but enough for me to take conclusions based on experience. Of course, this is speculation due to lack of other information, in general it remains valid my first sentence on comparing two codes.

[hereford_jon]

Does anyone actually know who supplies the Solver in Flow Simulation for SolidWorks 2019? I found an article from Mentor Graphics in 2014 that stated the underlying solver was FloEFD. But a little googling seems to show that FloEFD is now in SolidEdge (a Siemens CAD package.) Is SolidWorks now using a different solver or do both SolidEdge and SolidWorks use FloEFD?

[hereford_jon]

Looking at the PDF document properties for the Technical Reference for Flow Simulation the author is Mentor Graphics and a lot of the diagrams in the manual are identical to those in the original article from 2014. However there is no reference to FloEFD by name.

[CFDfan]

Originally Posted by hereford_jon

Looking at the PDF document properties for the Technical Reference for Flow Simulation the author is Mentor Graphics and a lot of the diagrams in the manual are identical to those in the original article from 2014. However there is no reference to FloEFD by name.

Boris Marovic if he reads this could comment if the latest SWFS uses the same solver as the latest FloEFd (which is important), otherwise there is no wander that both Solidworks and Solidedge have embedded Flowefd module.