ParaView: stress on facet =/= force * unit area

Hi Christopher,

>
> Some of my co-workers have been running experiments in which a group of
> spherical particles is exerting forces on a facet. Ideally, the stress on a
> body should equal the force on the body per unit area. However, in our
> experiments, we have found a large discrepancy between stress and
> (force*area). The product of force and area in one of our facet tests was
> 18 kN, while the stress recorded for the same facet was 66 kN.
>

please try to be more consistent in your stress, force and physical unit
relations (e.g. "discrepancy between stress and (force*area)" - should be
stress and force/area? or stress = .. kN - should be kPa?)

You can have a look on the implementation in the source code, personally I
am a bit confused by it so I would rely more on force/area :-)

>
> What could be the cause of the discrepancy? How exactly does the
> VTKRecorder (I presume this is the engine used) calculate the force values
> it sends to ParaView?
>

forces are exported correctly, I am not sure with stresses..

cheers
Jan

Ah yes, I did mean force/area. Had a late night last night :) Thanks, will inspect the source code.

Where exactly in the source code can the stress calculation and exportation code be found?

Look for REC_STRESS here [1].

[1] https://github.com/yade/trunk/blob/master/pkg/dem/VTKRecorder.cpp

Anton