particle advection problem

[wodH]

Hi there,

First time trying to advect particle to my fluid source, however the particle seems to be pushed to certain direction. I have checked my velocity on my particle, dont seem to have any velocity pointing anywhere towards to where its going. 

Im pretty sure there are many other ways to advect  it, this is the only way I come up with since I just started to use houdini, I have attached the file down below please sim a few frame of this pyro and to be able to check the particles. I have tried to name everything as clear as possible at the same time.

Thank you guys in advance!


 

particleAdvectionProblem.hip

[dchow1992]

It's working as intended. Fluid simulations in computer graphics tend to be "incompressible", meaning during the solve they undergo a pressure solve that essentially makes the velocity field continuous on the grid, also called divergent free. As a result, there will be velocity present everywhere on the grid. To visualize this, you can plug in a vector field visualization into your sop vector field and check on streamers.

If anyone notices any incorrect information feel free to correct me, I am still learning as well. 

These are great videos that I have watched several times

https://vimeo.com/119694897   (it has english subtitles)

https://vimeo.com/42988999

 

anyways, after your pyro simulation you can multiply your velocity field by density to mask it off before advecting your points with it.

Edited July 6, 2016 by dchow1992

[Solitude]

Can't open the file at the moment, but if you set the advection to set velocity and set the method to trace (sometimes lower the cfl to 0.9 or something)  you should get it pretty accurately close to your density field.  If you are using it as a force then it will give it a push in the right direction, but won't follow the sim exactly. Depends on what you're after to what you would use. 

[wodH]

15 hours ago, dchow1992 said:

It's working as intended. Fluid simulations in computer graphics tend to be "incompressible", meaning during the solve they undergo a pressure solve that essentially makes the velocity field continuous on the grid, also called divergent free. As a result, there will be velocity present everywhere on the grid. To visualize this, you can plug in a vector field visualization into your sop vector field and check on streamers.

If anyone notices any incorrect information feel free to correct me, I am still learning as well. 

These are great videos that I have watched several times

https://vimeo.com/119694897   (it has english subtitles)

https://vimeo.com/42988999

 

anyways, after your pyro simulation you can multiply your velocity field by density to mask it off before advecting your points with it.

thank you for the reply, I will check those videos out right the way. Main concern was why those particles got pushed to one direction when my fluids wasnt going to that axis.

If I figure out later on I will post the result with explanation.

Thanks

[wodH]

57 minutes ago, Solitude said:

Can't open the file at the moment, but if you set the advection to set velocity and set the method to trace (sometimes lower the cfl to 0.9 or something)  you should get it pretty accurately close to your density field.  If you are using it as a force then it will give it a push in the right direction, but won't follow the sim exactly. Depends on what you're after to what you would use. 

Thanks for the reply, I will dig into that and play with some settings. Once again if I figured something out later on, I will post the result with explanation of it.



thank you