Pyro Expand + Pump + Sink examples?

[catchyid]

Hi guys, I have been testing SOP Fluid Source that is configured as sink,pump and expand. Sometimes I get "good" results and sometimes I don't. I am not sure exactly what I am doing wrong?! Does anyone have examples on how to configure use these fluid sources in pyro simulation? The kind of problems I am having are similar to this:

-although I am using sink, smoke still passes through the volume! I managed to increase the absorption amount by scaling Source Volume DOP (scalar volume velocity)

-in a small example, expands pushes away smoke, in another example it does not. Does expand use surface normals to determine the direction in which to push smoke (but it's actually a volume, so normals exist!)

Thanks, merry Christmas and happy new year everyone

[Atom]

Quote

although I am using sink, smoke still passes through the volume!

You may need to enforce the boundary for the heat field.

Check out this thread.

Edited December 28, 2016 by Atom

[catchyid]

Thanks for the link Atom on the collision problem 

if someone knows the internals/examples of pump/expand, pls let me know  to be more specific, what direction do pump/expand use to add velocity, is it object normal for example?

[bunker]

not sure how the source volume work for this, but it's an OTL so you can check the inside...
for sink/pump/expand operations I normally use a "Gas Field Wrangle" microsolvers.

if you set your "input 1" to a SOP path with a volume then:

// load volume
float vs = volumesample(0,0,@P);
// expand
f@divergence += vs * chf("expand_amount");
// sink
f@density *= clamp(1-vs,0,1);

of course you'd need to adjust the volume values if you're using an SDF/surface field

for pump, you'd need to load 3 volumes since it's velocity:
float vx = volumesample(0,0,@P);
float vy = volumesample(0,1,@P);
float vz = volumesample(0,2,@P);
vector v = set(vx,vy,vz);
v@vel += v * chf("velocity_scale);

Edited January 2, 2017 by bunker