[wshart]

Hello everybody,

this is my first post and unfortunately I've found a problem in FlowSimulation when using perforated plates.
There are big differences between the perforated plates feature with an opening ratio above 0.6 and a simulation model which uses the real 3d geometry for i.e. an opening. When I use the perforated plate feature the results of the velocity are complete nonsens, iteration time is absolutly out of order and convergence not given. For those who want see how you can run the simple simulations in the attachment.
I've reported this problem to our VAR and he redirected it to SolidWorks and it becomes a SPR. All strarted in August 2017. The first answer from SolidWorks was:

"Currently there is an inaccuracy when using perforated plates. The averaged velocity will be calculated correct, the maximum velocity might contain inaccurate value. Especially direct on the plate surface there are differences in the maximum values."

Because this answer wasn't acceptable (How could the averaged velocity be correct when there is a hundredfold incorrect velocity value? Do they eliminate or weight the cell values?) the problem becomes a SPR in November 2017.
With SW2018 SP2 it seems now, that they start to work on a solution, because the convergence is now much more better than in 2017 and and the results come closer to reality, but the calculation of the model with the perforated plates takes 10 times more as the real geometry. Nevertheless calculation is from my understanding still wrong. So I'm now nearly 6 month waiting to get a satisfying answer or solution for the problem.
Unfortunately it seems that there is no big interest in solving the problem. This might be caused in the fact, that we are only a small customer with only 3+1 licenses, or no one else is forced with this problem. In my opinion it is absolutely inacceptable to sell a simulation software for about € 14000 plus € 3500 for the annual subscription service and this software contains essential bugs and when they come up, it takes more than half a year to solve them.

So if anyone else has the same problem or knows any workaround it would be nice to get some help on this way, because in my daily work I simulate big electronics enclosures where the perforated plates should be a useful feature in simulation (if it works proper).

Kind regards

Wolfgang Schmidt

[Boris_M]

Hello Wolfgang,

I'm sorry that you face such problems.
I haven't used this feature as extensive as you apparently have. If you have a simplified model you can share, I can have a look at your problem and see if I can help you.
The perforated plates are based on handbooks where such losses were experimentally validated, similar to the VDI Wärmeatlas. The type of holes and their patterns were tested and relations derived. This might be a reason if such relations were not done with your dimensions where simply the relations will not match with the reality anymore. The perforated plate feature is a simplification and does not guarantee that it will work for all possible variations just like porous media are simply a pressure loss vs. flow rate on a block and not the real geometry. The simulation can only be as good as its boundary conditions and sometimes such simplification cannot be applied anymore.
I'm happy to look into your case but I cannot promise any high priority.

Regards,
Boris

[wshart]

Hello,

unfortunately I cannot upload a model because of the size limitation.
The model is about 5 MB and contains a simple model, which contains different studies to compare perforated plates with the corresponding 3d model.

Best regards Wolfgang