Hybrido Particles Leaking Through

[Oldcode]

Hi Everybody,

Now that 2014 is out with the new Viscosity for Hybrido, I'm going to start using it for larger area that need much more than standard particles.

As a test, I tried some geometry that I used in a previous project for a small pond. It's the same geometry, from the same SD file, with the same scale (25), yet most of the hybrido particles leak right through it. Yes, I did make sure the relationship editor was set so the hybrido particles collide with the geometry. You can see some collisions going on, but most of the particles slip through and fall out the bottom.

Why is it that Standard particles have no problems with this geometry, but the Hybrido particles do?

Thanks in advance,

[tsn]

Collision detection in SPH and Hybrido is actually not comparable. With Hybrido, everything is based on fields and grids. The grid's resolution ("Cell size") determines the field's accuracy. This means that you have to adjust the field around the object to get proper collisions. Since you're working with a very large scale you could try to set the object's "Volume mode" to "Shell". This creates a waterproof "layer" around the object and avoids leaking. The thickness of this layer is: Cell size value * 3. So, you'll always see an offset or cache around the object, but for a pond this shouldn't be problematic.

You could also activate the "Display volume" feature to get an idea of how the solver "sees" the object:
http://support.nextlimit.com/display/rf ... lay+Volume

Instead of using the "Shell" mode it's also an option to use "Solid outside" and change the "Surface offset" parameter of the objects:
http://support.nextlimit.com/display/rf ... s+-+Volume

Finally, I recommend reading the page below to get a deeper understanding:
http://support.nextlimit.com/display/rf ... nce+Fields

Cheers,
Thomas

[Oldcode]

Thanks!

I had done some reading but didn't realize the object has a Cell size parameter too. I'll read up and see what happens.

[Oldcode]

Hi,

Thanks for your help. I reduced the Cell Size on both the emitter and the object and got it so the particles are not leaking through any more.

My next problem comes from me trying to fill up the pond. I set the emitter to Stream, put the emitter object at the bottom and hit simulate. I don't use the Time Line lock because if the program crashed, and it does more than it should, all the progress is lost. So, I started filling up the pond, let the simulation run, and when it got the end, I'd make an initial state and hit reset.

That worked for a while, but now, I hit reset, Frame 0 now has the particles from the initial state, but when I simulate, the emitter starts over and acts like I've hit reset without the initial state, and all the progress is lost. The particles from the initial state are gone and emitter has the particles as if I'm starting from scratch.

I'm sure I've got Reset to Initial State checked. What am I missing here?

Thanks,

[LuisMiguel]

Hi,

Make sure you are using the reset option in the command line options when simulating by command line again. If not, the simulation will always start form the frame zero.

luisM.

[Oldcode]

Hi Luis,

Thanks a lot, that did the trick.

Strange, I read about that in some of the documentation, but didn't think that was the solution for I never had to do that with Standard Particles.

I feel like I'm starting from scratch.

Thanks again,