Vellum balloon sim behaving weird when not at the origin

   549   1   2
User Avatar
Member
2 posts
Joined: April 2021
Offline
Hi everybody. Noob here

I'm facing an issue with a Vellum balloon sim.

Basically, if the geometry is not centered to y=0, it behaves in a rather unexpected way. Do you have any insights on why this could be happening?

Correct behavior


Weird behavior



All my values are at the defaults set by Vellum Configure Balloon, except for pressure rest length (set to 30) and gravity which I deactivated in the Vellum solver node.

The situation came up while following this tutorial. I tried to lift a little bit my geometry and everything got screwed.
https://www.youtube.com/watch?v=8ge2W7KKwf8 [www.youtube.com]

Any ideas? I uploaded my .hip file in case anyone wants to check.

In any case, thanks for reading!

Cheers

Manuel
Edited by ManuN3 - Oct. 17, 2023 00:43:55

Attachments:
Y = 0 _Expected behaviour.jpg (626.2 KB)
Y != 0 _Weird behaviour.jpg (665.2 KB)
Upload.hiplc (523.7 KB)

User Avatar
Member
334 posts
Joined: April 2018
Online
I'm experiencing the same thing now. The geometry will inflate correctly when it is not at the origin if it is not fractured, but if it is fractured, it will behave as if the inflation pressure is being directed at it from the origin, and not from within the geometry.

Edit: found this explanation on a Reddit post:
The problem here is your weld setup. You probably cut the geometry with edge fracture, which means it's not closed anymore. This means the pressure constraint will calculate the volume not only inside, but also the outside part and gets all confused, because there is no closed geometry where the volume ends at.

You can make sure the pressure constraint will see the geo as one by changing "Define Pieces" from "From connectivity" to " By Attribute" and provide an attribute that's the same on all pieces. Problem is: Once the pieces are actually breaking, the same problem will again arise, so here you should probably delete the pressure constraint once the object is broken. So still not ideal.

An idea to solve this easier is to switch the sim at some point by taking the result of one sim (pumping up the geometry) and feed the frame before it's supposed to burst into another sim where the weld is prepared, probably add a start velocity to the pieces and let it burst in that new sim. Pretty straight forward and actually even allows for more control then an all-in-one approach. It's all about the end result and having control over the behaviour, so this is actually the approach I would prefer for better results.

(If someone knows a better approach, I'm all ears)

https://www.reddit.com/r/Houdini/comments/10f4y9b/vellum_pressure/ [www.reddit.com]
Edited by eikonoklastes - May 15, 2024 07:38:16
  • Quick Links