H15 cloth, plastic deformation?

[Skybar]

Are there any good workflows for accomplishing plastic deformation with cloth and/or FEM in H15? Or do I still have to go back to H12.5?

[Guest tar]

Fwiw. Haven't investigated an answer yet but did notice that the 'plastic' node is unconnected in the Solid/Cloth Dop.

[carlo_c]

I was trying to achieve something similar and with some help got a simple setup working switching an rbd object to cloth and back. Not sure if it's the effect you're looking for exactly but it might help spark some ideas https://www.sidefx.com/index.php?option=com_forum&Itemid=172&page=viewtopic&t=42111

[Jason]

Plastic deformation was a casualty of unifying FEM with cloth a couple of years back and it is sorely missed.

[anim]

it is commonly faked by updating the rest attribute over time, but that's not the same as it prevents from doing independent rest updates for selective parts of the geo based on other attributes, unless maybe if you are doing it on disconnected lines as a vertex attrib 

 

For simple cases you can maybe use Wire Solver

Edited October 30, 2015 by anim

[Skybar]

Ah right I see. Thanks guys

[glomm]

its really easy to implement in 15, just add a sop solver to update rest if tet deforms too much

each tet has primitive intrinsic attribs measuredarea, perimeter, volume

you can use them to mix your restP and your newrestP

the only catch is you need to transform your deformed tets to initial position in order to get good blends

since restP can be vertex you can update each tet separately from its neighbors

it works really good, i did a few tests w crashing car juts few days ago

and since in 15 fem is so much faster its starting to become usable

[wateryfield]

its really easy to implement in 15, just add a sop solver to update rest if tet deforms too much

each tet has primitive intrinsic attribs measuredarea, perimeter, volume

you can use them to mix your restP and your newrestP

the only catch is you need to transform your deformed tets to initial position in order to get good blends

since restP can be vertex you can update each tet separately from its neighbors

it works really good, i did a few tests w crashing car juts few days ago

and since in 15 fem is so much faster its starting to become usable

Hi, glomm.

I have not deep in new fem solver yet, but looks like your way works great.

Do you have any video or simple hip show the result.

Thanks.

[Follyx]

Yep, a hip would be great.

[Skybar]

Oh that sounds interesting! I have to look into that, thanks Sergey

[Juraj]

Hi,

thanks for sharing ideas. I tried mixing current P with restP (on vertices) based on volume change. I did some tests and it looks promising.

 

 

But I am not sure I understood this sentence - what do you mean?

...

the only catch is you need to transform your deformed tets to initial position in order to get good blends

...

 

Also I had no succes with fracturing and plastic deformation. When tets become too deformed they kind of exploded and simulation went funky.

 

Juraj

[ChargingImage]

its really easy to implement in 15, just add a sop solver to update rest if tet deforms too much

each tet has primitive intrinsic attribs measuredarea, perimeter, volume

you can use them to mix your restP and your newrestP

the only catch is you need to transform your deformed tets to initial position in order to get good blends

since restP can be vertex you can update each tet separately from its neighbors

it works really good, i did a few tests w crashing car juts few days ago

and since in 15 fem is so much faster its starting to become usable

 

It would be awsome to have some simple hip file. I don't know how to do that stuff . Yet.

[Follyx]

yep a hip would be great from someone.

Thanks in advance.