Hello Benoît,
For your information, the habit is now rather to ask questions via
Launchpad interface (https://answers.launchpad.net/yade/+addquestion,
have also a look at https://answers.launchpad.net/yade)
My personnal feeling is that your problem is calibration of the
parameters of your numerical discrete model towards the experiments you
want to simulate.
I'd recommand you first to have in mind that what is called Young
Modulus in Yade has NOTHING to do with a Young Modulus in continuum
mechanics... I dared to upper case "NOTHING" since the equations, and
the roles in these equations of these two "Young Modulus", have nothing
in common, even if correlations can be deduced.. I wouldn't be able to
answer you why it is called "Young modulus", then...
Maybe some " " around Young's modulus in
https://www.yade-dem.org/doc/yade.wrapper.html#yade.wrapper.ElastMat.young
would be helpful to not mislead anyone ?
An other thing is also that 200 particles were surely too few to have a
behaviour that depends only on the mechanical parameters of your
numerical model. 2000 is already better, but 10000 is maybe the most
common, I would say.
Le 10/02/2012 11:36, CHARLAS Benoit 224833 a écrit :
Hi everybody
I have some problems on YADE at the moment. I’ll try to make it clear.
I did an experimental oedometric test on a metal powder with a very
low punch stress (~5MPa). Metal powder for gears manufacturing are
usually pressed up to ~900MPa.
So I expect there is no plastic deformation of the grains in my case.
I wanted then to reproduce this with an elastic Contact Law on YADE
(with “Law2_ScGeom_FrictPhys_CundallStrack()”)
The *numerical results* show a higher volumetric density than
experimental measurements (that was expected …) but also *forces
around 2000 times too important.*
**
I checked the following:
* *Units*of my parameters
* *Unbalanced force*(to remain quasi-static)
* *Number of particle*(from 200 to 2000 and soon to 10000)
* Change initial *volumetric density* with friction or clumps
* Reduce the *velocity* of the upper punch
* Reduce by 1000 to 3000 times the *Young Modulus*
From all the things I checked, the sole way to obtain numerical forces
near the experimental ones was to reduce the Young Modulus by 2000 or
3000 times.
My current problem is that I can’t justify such a result.
Do you have an *explanation *on how to justify this*or *some*tips* on
what could be checked?
Thanks
Benoit
_______________________________________________
Mailing list: https://launchpad.net/~yade-users
Post to : [email protected]
Unsubscribe : https://launchpad.net/~yade-users
More help : https://help.launchpad.net/ListHelp
--
Jerome Duriez
Post-Doctorant LTDS (Ecole Centrale de Lyon)
04.72.18.62.17
_______________________________________________
Mailing list: https://launchpad.net/~yade-users
Post to : [email protected]
Unsubscribe : https://launchpad.net/~yade-users
More help : https://help.launchpad.net/ListHelp
