Pop Grains Friction Problem

[breadbox]

I have a grain setup with clusters using attraction weight. The particles with no cluster attraction weight come to rest nicely but the clumps of clustered particles look like they are on ice and skid along as if they had no friction. I tired increasing the scale kinetic value and although that does hold the big chunks with more added friction it really makes all the small individual particles come to rest way too fast and feel sticky.

 

Any idea what the parameter is that I might use to get those bigger clusters to come to rest with friction a bit better while maintaining just a little bit of skid overall. 

  

 

popGrainIceSkate.hipnc

[toadstorm]

I think that extra energy is coming from your clumping strength being much too high, and your total weights adding up to >1. Lower your collision and clumping weights down a bit so that they add up to 1 and try a clumping strength of maybe 20, and probably enable a drift threshold to minimize slow sliding. If the big chunks melt too much you could try explicit constraints instead.

[breadbox]

Thanks Henry,

Clumping stiffness lowered to 20 seems a bit more stable. It's a little more slumpy than I'd like but it might work. 

In regards to the weights all adding up and == 1, its interesting... I can have a weight of 0.9 internal collision + 1.0 clumping = 1.9 and it seems to resolve more "stable". I could lower each one to 0.5 and 0.5 which would be technically more "correct mathematically" however then the clusters then really slump a lot.

I guess its a balance of cheating physics until it becomes unstable, and we are allowed to do that with these controls but you never know when you might pass over that threshold and get a weird result!! 

I might look into explicit constraint for the grains to hold them together in clusters... I've never used them much but I imagine it works kind of like a glue constraint in an RBD simulation? but then explicit constraints also have a weight and stiffness so I'm kind of curious what's the benefit of using "clumping" over "constraints" they seem to be able to do the same thing in just a slightly different way?

[toadstorm]

In the Grains masterclass I remember Jeff saying that the weights should add up to 1.0, but I suppose this is just a guideline and it's still possible to have stable sims with weights all over the place... I wish the documentation were a little more clear about best practices.

Yes, explicit constraints are a lot like glue networks. They're better at keeping distance between individual grains, rather than a sort of generalized clumping force. There's also an option to treat explicitly constrained particle clumps as rigid bodies, though it's expensive.

[vahuar6]

On 10.5.2018 at 9:02 PM, toadstorm said:

There's also an option to treat explicitly constrained particle clumps as rigid bodies, though it's expensive.

Hi,

Could you care to explain a little bit about this? I need some very stiff chunks that break off few grain particles when impacting ground, but still retain the overall rigid shape. Right now all i get is melting type clumps.

[toadstorm]

The option I was referring to is the "Enable Rigid Shape Matching" parameter on the POP Grains node. 

If you need mostly completely rigid chunks that only break off at a few points, you might have better luck simulating your large chunks using Bullet and have some rbd chunks or grains splinter off from that.