Hi, Ruochun
Thank you for your reply.
1. I have a question. Can the geometry boundary in DEME be a
surface? Or must it be a solid?
2. In addition, I observed the speed of the two particles during
the calculation, and they did not exceed 5 m/s. My time step is 1e-6, the
Young's modulus of the particle material is 1e8, and the particle radius is
0.04. By the way, how to set the time step during the calculation?
According to Rayleigh waves? Is there a relevant calculation formula?
Best regards,
Wenxuan
在2024年7月31日星期三 UTC+8 13:49:49<Ruochun Zhang> 写道：
> Hi Wenxuan,
>
> Since you have 2 spheres but each sphere had more than 200 contacts, it
> could be one of the two causes...
>
> 1. Please check your overall box domain (*InstructBoxDomainDimension*) is
> larger than the geometry you have. If it is smaller, internally the contact
> detection may behave incorrectly.
>
> 2. The particles may be traveling at unusually large velocity so too many
> potential contacts with the mesh facets are detected, a.k.a. *"the
> simulation may have diverged and relaxing the physics may help"*. It
> might be related to the way how you are rendering the mesh, but to me, the
> mesh appears to have a suspicious texture. Can you show us if the inner
> wall of the mesh has a correct inward normal direction? This is important
> as the mesh--sphere contact in DEME is directional and if the normal is
> flipped, upon contact with the mesh, the spheres will be ejected with high
> velocity.
>
> Thank you,
> Ruochun
> On Tuesday, July 30, 2024 at 11:05:46 PM UTC+8 [email protected] wrote:
>
>> Hi, dear
>> When I simulate two sphere particles drop from the hopper, it
>> crashed during the simulation, what causes this? The crash information show
>> as follow:
>> terminate called after throwing an instance of 'std::runtime_error'
>> what(): On average a sphere has 223 contacts, more than the max
>> allowance (100).
>> If you believe this is not abnormal, set the allowance high using
>> SetErrorOutAvgContacts before initialization.
>> If you think this is because dT drifting too much ahead of kT so the
>> contact margin added is too big, use SetCDMaxUpdateFreq to limit the max dT
>> future drift.
>> Otherwise, the simulation may have diverged and relaxing the physics may
>> help, such as decreasing the step size and modifying material properties.
>> If this happens at the start of simulation, check if there are initial
>> penetrations, a.k.a. elements initialized inside walls.
>> This happened in contactDetection.[image: Screenshot from 2024-07-30
>> 23-01-58.png]
>
>
--
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/3e95137e-f3c6-4bf1-a0dc-4d417f29c25an%40googlegroups.com.
