Melting // FLIP

[akelian]

Hello Everyone,

 

Since a few months now I've been learning Houdini more and more.

A really good tutorial that introduced me to FLIP and DOP/SOP interaction recently was BW Design's Angel.

But since then I've been trying to recreate a real melting effect as in this video in houdini without luck.

 

 

What i'm looking for is a way to lock some particles but still have them interact with others (I managed to lock the velocity on some using gas but it looked strange), and also as you can see in the video, the sense of surface layers... I don't know if i'm really clear there, tell me if I'm not

 

Thanks

[Atom]

Didn't Ben cover that in his video? I think he used an attribute transfer to decide what melts when. Then he animated a sphere over the part he wanted to melt to target the attribute transfer.

Edited October 2, 2015 by Atom

[akelian]

Didn't Ben cover that in his video? I think he used an attribute transfer to decide what melts when. Then he animated a sphere over the part he wanted to melt to target the attribute transfer.

 

Sorry I was not very clear,  I want an object to be able to partially melt without the unmelted falling down

 

 

Here is a sketch of what i mean.

 

I found a thread on OdForce where someone had tweaked the FLIP Solver so that Velocity is upgraded before the Integrator. Quite old and I wasn't able to manage the same setup on FLIP H14

 

Hope it is clearer.

 

I'd love just point outs on how it could be done. DOP is the part of Houdini I feel the less confident for now, especially Gas, I don"t quite understand all of them.

[vijayakumar]

You can play with viscosity.. Make the bottom part high viscous and the area u want to melt as low.. Animate the viscosity value from where u want to melt.

[Skybar]

Set the un-melted particles velocity to zero at the second input of your flip solver.

[akelian]

Set the un-melted particles velocity to zero at the second input of your flip solver.

 

I cannot find how. The VEX condition on a Pop force as in Benn's Tutorial. Using Pop VOP & Using a Pop Speed Limit... The problem seems that once a part gets activated everything falls down without breaking into different parts per viscosity/velocity.... If any one has a clue! 

Edited October 3, 2015 by akelian

[vijayakumar]

Create a custom attribute. Say melt.. Then attribute transfer from top to bottom using sop solver inside dops.. It will evaluate every frame.sorry I cannot do a setup since I don't have my computer as of now.. Check with sop solver inside dops.

[akelian]

Thanks but I know that, and that's not what i'm looking for.

[IvanPulido]

I did a couple of year some tests using differents kind of melting, using noise, by temperature and controlling by null object, take a look to this, with any ways you can control the melting parts.

Using noise for example you can define differens areas to melt with no affect the other parts.
 

http://vimeo.com/141266292

Edited October 4, 2015 by IvanPulido

[alksndr]

Hey, try this for a simple example of how you could stop it by controlling velocity?

 

The idea is to use a group that will have your high temp particles and to ramp down their vel as they get placed into the group (threshold of x temp). So rather than me using  .y, use your temp value to define what is to be melted.

veloff.hip

[akelian]

I did a couple of year some tests using differents kind of melting, using noise, by temperature and controlling by null object, take a look to this, with any ways you can control the melting parts.

Using noise for example you can define differens areas to melt with no affect the other parts.

 

http://vimeo.com/141266292

 

Really cool stuff Ivan !

 

 

 

Hey, try this for a simple example of how you could stop it by controlling velocity?

 

The idea is to use a group that will have your high temp particles and to ramp down their vel as they get placed into the group (threshold of x temp). So rather than me using  .y, use your temp value to define what is to be melted.

 

Thanks a lot alksndr, I didn't know you could "separate" fluids using a sop solver, that's really what i was missing I think, going to retry my setup now.

 

I will post the result here if interesting