Hi Giovanni,

Can I also have the two mesh files you are using to test it?

Before I test, I can see the particles you are using are small (0.5mm), and 
the mesh is stiff and has a high Coefficient of Restitution. This makes 
contacts with this mesh challenging to resolve and I doubt if 5e-6 as the 
step size is enough. About the message you got, that should be a warning, 
but it is a warning when the physics is getting bad so I guess it's close 
to an error (you can still turn it off by setting the *Verbosity *from 
STEP_METRIC to INFO), and this could tie back to my suspicion that the time 
step size is far from enough.

But if I can get the mesh, I can say for sure.

Ruochun

On Wednesday, July 19, 2023 at 4:37:13 PM UTC-5 Giovanni Bianchi wrote:

> [image: Immagine 19-07-23 - 23.34.jpeg][image: Immagine 19-07-23 - 
> 23.34.jpeg][image: Immagine 19-07-23 - 23.33.jpeg]Thanks, Rouchun,
>
> I link here the code that I am using and the graphic output I get from the 
> simulation.
>
> I don't know the exact Young modulus of the material I am using, and I am 
> testing different values. I get this problem of particles passing through 
> the mesh only for low values of E; instead, for high values, the simulation 
> crashes, giving the error: " N contacts were active at time xxx on dT, but 
> they are not detected on kT, therefore being removed unexpectedly!".
>
>
>
> Giovanni
>
> Il giorno mercoledì 19 luglio 2023 alle 23:14:13 UTC+2 Ruochun Zhang ha 
> scritto:
>
>> Hi Giovanni,
>>
>> It's hard to say without seeing the script and the rendering. Since you 
>> are doing a compression test and DEM is penalty/penetration based, it could 
>> be just that there is so much compression that the penetration is too big, 
>> so the particles appear to be phasing through the mesh (note that if *E* 
>> is low and the particles are relatively large, then having visible 
>> phasing-through can be normal). It could also be that the particles are 
>> somehow under a large load (say, being compressed by an invisible boundary 
>> without the user's knowledge) and it forces the particles through the mesh. 
>> Or maybe the mesh is not oriented correctly (triangle facet normal 
>> direction) so the particles approach it from the inner side, and then phase 
>> through.
>>
>> If you put here a rendering of the problem, and if you can, show the 
>> script you are using, it should be easy to diagnose.
>>
>> Thank you,
>> Ruochun
>>
>> On Wednesday, July 19, 2023 at 3:53:15 PM UTC-5 Giovanni Bianchi wrote:
>>
>>> Hello everyone!
>>>
>>> I am simulating the compression of a granular material with DEM-Engine, 
>>> and in some cases it seems that some particles pass through the mesh 
>>> boundary.
>>> I adapted the mixer demo to my case without changing anything in the 
>>> contact detection parameters.
>>>
>>> What am I doing wrong?
>>>
>>> Giovanni
>>>
>>

-- 
You received this message because you are subscribed to the Google Groups 
"ProjectChrono" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/projectchrono/4c7fb82e-4052-43d6-b218-c25525c5dbcfn%40googlegroups.com.

Reply via email to