Hi,

I've got a "wood" structure that needs destroying and I'm getting stuck on the simulation part. The sim looks good for about 3-4 seconds and then the fractured pieces start to move of their own accord (i.e they freak out).
I've fiddled with substeps (both DOPNET and bulletrbdsolver), collision geo, constraint iterations, etc. I watched countless tutorials and searched the web looking for answers, but nothing seems to work.
I just don't know where the issue is coming from.





Thanks!

Hi, were you by any chance following an old tutorial? The way you have set it up is perfectly valid, but so much easier with more modern techniques. It was far easier to re-build the setup than debug yours.

I have attached an example file using the RBD MF node and other helper nodes. Let me know if this is what you were looking for.

You might want to try this tutorial I recently updated for the SESI docs: https://www.sidefx.com/docs/houdini/destruction/tutorials/intro_to_mbd_1.html [www.sidefx.com]

Sean

just had a quick look into Your file, and the rbd cluster node has "fuse cluster pieces" on. this results in bad collision geometry. there are no separate packed pieces anymore per cluster so the packed object creates a convex shape for all elements of a cluster resulting in intersecting geometry, this might cause the problems.
to see the problem use the exploded view node as source of the rbd packed to separate the clusters and turn on show guide geometry

for the wood constraints You also could try the soft constraints, as they allow the wood to bend before it breaks.

mrCatfish: Thanks so much for the file and the link to the tutorial. That's not really what I'm looking for. I'm trying to avoid the rbdmaterial fracture node at this point in time (although I have looked into it and your tutorial in the past ). The "blackbox" of that node is daunting and far more difficult to debug than my current set-up (which is, as you had guessed, composed of a few different methods I've picked up over time). Although, I'm having difficulty debugging my current "simpler" set-up, so...haha

sanostol: Thanks for the reply. Yes, I thought that might have been an issue as well (the "fuse cluster pieces"), but after turning it off, the problem persists.
I actually have another version with the conetwist constraint that allows for some nice bending before breaking, but I'm having the same issue with that file as well. I thought this example would be quicker and easier to debug when posting on the forum, so I removed the conetwist constraints and used glue and pin for now.

I thought it may be due to the constraints "fighting" with eachother over orientation, but I'm not sure how that would happen.
Also, it seems to be occurring on pieces that still have a hardconstraint still active (the "nails") after collision has taken place.

TheDude123
mrCatfish: Thanks so much for the file and the link to the tutorial. That's not really what I'm looking for. I'm trying to avoid the rbdmaterial fracture node at this point in time (although I have looked into it and your tutorial in the past ). The "blackbox" of that node is daunting and far more difficult to debug than my current set-up (which is, as you had guessed, composed of a few different methods I've picked up over time). Although, I'm having difficulty debugging my current "simpler" set-up, so...haha

Yup, fair enough, kudos to you for doing it the hard way and figuring out what's going on under the hood. Obviously, this will make you a better TD when you do have to dive into some of the black boxes and debug them.