Open Foam problem

Replies to This Discussion

Permalink Reply by Mostapha Sadeghipour Roudsari on April 2, 2017 at 3:28pm

Everything looks fine in this screenshot. Did you get any error when you tried to run the analysis?

Permalink Reply by Pietro Scarpa on April 3, 2017 at 1:17pm

Thanks for quick answer, was a problem of the same naming I managed to fix that...

But now I got a strange final result, some value are completely out of scale and that obviously affects the visualization, any idea why?

Permalink Reply by Aris V on April 4, 2017 at 5:55am

Try to view results in paraview and see if they are ok. I had the same issue, which I think has to do with the location of probes. To fix the visualisation in grasshopper just put a reasonable upper limit to the gradient.

Permalink Reply by Mostapha Sadeghipour Roudsari on April 4, 2017 at 9:08am

This can be the reason as you have a number of probes inside the building geometries itself. We should give a more explicit warning for cases like that.

On the other hand, just seeing the residual values you need to let the analysis run for more iterations. The values are far from convergence (10^-5).

Permalink Reply by Theodoros Galanos on April 6, 2017 at 10:04pm

I think we do have to include a warning on this, seems to be happening a lot.

Is it possible you think to automatically split the sampling surface with the mesh boundaries, and then scale it down a bit? Or is it easier to just filter out those huge values?

Kind regards,

Theodore.

Permalink Reply by Mostapha Sadeghipour Roudsari on April 7, 2017 at 6:10am

That's a little bit of blackbox. I think this is something that should be set by users, as it can make major differences. We can have a separate component to help the users automate this process but it's quite straight forward to do it with GH native components.