On Monday, August 4, 2014 2:21:20 PM UTC-4, Robert Cimrman wrote:
On 08/04/2014 05:21 PM, Geoff Wright wrote:
> Hi Robert,
>
> When I try the extend_cell_data on the surface of a cube, I get an error:
> "ValueError: The weights and list don't have the same length." I attached
> the file. It uses a mesh that is provided with sfepy.. can you take a look?
>
> This is another example where the flux doesn't make sense. If I use only
> vertices on one face as the source then the flux values are good. But if
> you run the file as-is, the source extends a bit into the adjacent faces,
> and the flux results are incorrect. The electric field values look okay,
> and the voltage field looks fine. My results are as follows:
>
> Source flux: 1.09348063065
> Sink flux: -0.90821822135
> Block flux: -0.040501111711
> Entire surf: 0.298397270469 0.0181174920822 -0.0124582155367
> Source surf area: 3.05555373528
> Sink surf area: 1.0
>
>
> I'm pretty sure there is an issue with the surface_flux function. Is this
> something you will be able to look at? I'm trying to decide whether to
> continue investing in this or switch to another tool/software.
I am actually up to something, stay tuned. (And the surface_flux function is
innocent - it comes down to region definitions and gradient computation accuracy).
r.