yade-users team mailing list archive

[Xu <question703169@xxxxxxxxxxxxxxxxxxxxx>]

Question #703169 on Yade changed:
https://answers.launchpad.net/yade/+question/703169

Xu gave more information on the question:
Hi Karol and Jan,

Thank you for your MWE. It looks like 'utils.getStress()' may give
slightly different results for clumps and spheres. However, I have used
the clump which is made up of more than 2 spheres (3 or 4 or dozens).
The more the number of spheres members in clump, the larger the
difference between 'inwork' and 'Etot', when the volum of clump remains
the same. And i think the slightly different results of
'utils.getStress()' may not cause such obvious error between input work
and microscale energy. Because the value of total microscale energy is
more than twice the input work value when the number of clump members
reaches 40-50.

So that's why I said the cause of the difference may lie in line 20,21
in [1]. As the particle volume remains the same,  the radius of sphere
member in clump becomes smaller when the number of memeber increases.
And the kn and ks calculated in [2] is smaller, hence, the microscale
energy computed by [3-4] is larger than the real value. So i guess this
is the reason why 'Etot' is larger than 'inwork' .

I think my opion may be wrong, but I can't find any other reason. So if
you can give me some more reasons for this phenomenan, i would really
appreciate it.

Best wishes,
Xu

[1] https://gitlab.com/yade-dev/trunk/-/blob/master/pkg/dem/FrictPhys.cpp#L20-L21
[2] https://gitlab.com/yade-dev/trunk/-/blob/master/pkg/dem/FrictPhys.cpp#L35-L37
[3] https://gitlab.com/yade-dev/trunk/-/blob/master/pkg/dem/ElasticContactLaw.cpp#L91-L94
[4] https://gitlab.com/yade-dev/trunk/-/blob/master/pkg/dem/ElasticContactLaw.cpp#L97-L100

-- 
You received this question notification because your team yade-users is
an answer contact for Yade.