yade-dev team mailing list archive

[Anton Gladky <gladky.anton@xxxxxxxxx>]

Hi, guys!

Can this fix have an influence on this one:

http://www.mail-archive.com/yade-users@xxxxxxxxxxxxxxxxxxx/msg01656.html

......

I have recently discovered, is that it is
better to use as less possible friction coefficient between acting part of
the press (facets) and specimen (better 0). So, we imitate greased by oil
the surface of the press.

In this case we get the most stable results. If the friction coefficient is
large (for example, 0.4-0.5), results are changing randomly, depending on
size specimen, number of grains etc...

..........


Thank you


2010/5/11 Bruno Chareyre <bruno.chareyre@xxxxxxxxxxx>

> Yeah, I got a banana too, and compilation went to plugin24 in the
> meantime... (matter of days before I get a decent pc).
>
>  the answer while staring at a banana: since the stress _is_ right after
>> the slip (as it is computed independently), the force is right as well.
>> COOL!
>>
>>
> Yes, I wouldn't have notice anything without changing friction.
> Still wondering if it couldn't result in different force orientation in
> some cases (after many iterations), since cp1rel was based on wrong values.
> Not sure.
>
>
>  That also explains (_big_ thanks)
>>
>
> Glad to help. :)
>
>   why I had bogus values for dissipated
>> energy on interactions; typically it was something around 1e10, though
>> it didn't look like garbage values (which puzzled me really) at all
>>
>>
> Indeed, the return value 2*diff.norm() was nonsense but finite. I guess you
> used that one for plastic dissipation?
> By the way, there are a few useless operations in the whole dem3DOF thing
> (like sqrt computed twice here and there, or diff.norm() computed but not
> used - and equivalent to (currDist-displacementTMax)). Ok, to fix this at
> the same time?
>
> Bruno
>
>
>
> _______________________________________________
> Mailing list: https://launchpad.net/~yade-dev<https://launchpad.net/%7Eyade-dev>
> Post to     : yade-dev@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~yade-dev<https://launchpad.net/%7Eyade-dev>
> More help   : https://help.launchpad.net/ListHelp
>