Particle Size Distribution Limits

[Jared Long-Fox]

Hello again all! My project involves simulating a granular material, specifically lunar regolith, which has a particle size distribution that spans 12+ orders of magnitude. Based on what I've seen from other studies, there are limitations on just how wide a DEM model's particle size distribution can (I assume in order to maintain numerical stability but do not actually know for sure).

Does anyone have information or experience in dealing with wide particle size distributions in Chrono? I am trying to understand and explain limitations and possible ways around them. Thanks in advance!

Best,
Jared Long-Fox

[Ruochun Zhang]

Hi,

To my understanding, the size distribution of lunar regolith particles (or at least a reasonable DEM representation of that) does not span 12+ orders of magnitude: That is like the difference between an atom and a mountain. Papers from NASA show about 2 orders of magnitude of difference. You have little hope of simulating 12+ orders of magnitude in Chrono either. Neither the force model nor the numerical resolution will be sufficient.

Chrono DEM-Engine is able to handle 2 orders of magnitude of particle size difference though, or even a bit more than that. The performance impact should be minimal as Chrono DEM-Engine tried to hide the cost of contact detection which is affected by particle size discrepancy more. We have a related paper here doing a similar task.

Thank you,

Ruochun