Collision problem

# 1 27-09-2005 , 04:30 PM

Registered User
Join Date: May 2004

Location: Utah, USA

Posts: 369

I put a 3D fluid emitter in a model of a bottle that was positioned so that the fluid could pour out of it. I created a very thick fluid That required a lot of negative bouyancy (>-80) to get it to flow out and onto the floor that I created. I then established a collision between the fluid and the bottle and also the fluid and the floor.

The simulation worked very well except that some of the fluid penetrated the bottle and also the floor. I increased the tesselation on the bottle to 40,000 but I couldn't stop the penetration. Same with the floor. If you looked at the bottom side of the floor you could see fluid.

I discovered that if I made the bouyancy more positive (~-20) the penetration decreased and if I also lowered the viscosity then I could completeley eliminate the penetration. I could also play with the opacity graph but it severly decreased the amount of visible fluid coming out of the bottle.

This little project I was working on was just an exercise to better understand fluids but I guess I failed since I couldn't get a super thick fluid to flow out of the bottle without having it penetrate the bottle's surface. I tried making a less viscous fluid with less bouyancy then I turned up the dampening to slow down the simulation but it didn't look good. It just looked like a thinner fluid flowing slowly. I was trying to simulate lava right as it begins to cool down.

One last thing. I made sure that the emitter was not close to the bottle's inside surface to make sure the penetration wasn't happening while the fluid was being emitted. Basically, the fluid was created in the bottle then it dripped down until it hit the inside of the bottle after which time it began flowing out.

Any ideas how I can keep the penetration of the bottle from happening??

Last edited by Velusion; 27-09-2005 at 04:33 PM.

Quote

# 2 27-09-2005 , 07:26 PM

Subscriber
Join Date: Feb 2005

Location: Winter Park, FL

Posts: 62

if you're using surface render - hard/soft surface fiddle around with the surface threshold and surface tolerance, that should get you started to fixing your penetration problem...

if you're using volume render, just keep fiddling with the opacity graph and also make use of texturing in the fluid container...

the only bad thing about messing with the surface threshold and tolerance is if you have a transparent container like a glass or a coke bottle you lose the effect in which case particles would help out the situation a little maybe even more to give it a splashy feel when it hits the surface...

another difference is you're not using a volume emitter also, though volume emitters in this kind of simulation are funky and hard to get going most of the time... but this sounds like you've got it set to omni in which case will cause a penetration...

for true liquid collisions use realflow you'll get much better simulations for this type of scene

Neil
www.delta3d.com
https://twitter.com/frescalus

Last edited by Razor Blade; 27-09-2005 at 07:40 PM.

Quote

# 3 28-09-2005 , 01:59 PM

Registered User
Join Date: May 2004

Location: Utah, USA

Posts: 369

Thanks, Razor.

The threshold and surface tolerance are not active in the attribute editor for some reason. I can't change them.

I am using a volume emitter and surface render (soft).

I've been meaning to give RealFlow a try!

Quote

# 4 28-09-2005 , 02:07 PM

Subscriber
Join Date: Feb 2005

Location: Winter Park, FL

Posts: 62

Thanks, Razor.

The threshold and surface tolerance are not active in the attribute editor for some reason. I can't change them.

I am using a volume emitter and surface render (soft).

I've been meaning to give RealFlow a try!

Originally Posted by

wierd, try making a new fluid container with something cheesy and see if they show up when you switch to surface render....

what version of maya are you using for this again?

Neil
www.delta3d.com
https://twitter.com/frescalus

Quote

# 5 28-09-2005 , 03:03 PM

Registered User
Join Date: May 2004

Location: Utah, USA

Posts: 369

6.5

One other thing; I actually made a bottle within a bottle for my test. The idea was that even if the fluid penetrated the inner bottle that maybe it wouldn't go any further but it did. Later on I'll pull the file up again and take a look but I'm rendering something right now (see voxel problem thread)....

Last edited by Velusion; 28-09-2005 at 03:16 PM.

Quote

Collision problem

Posting Rules Forum Rules

Similar Threads

Topics

Free Courses

Full Courses

VFX News