Announcement

Collapse
No announcement yet.

Particle Tuner not affecting object

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Particle Tuner not affecting object

    I have a popsicle object that I'm trying to melt from the top. I have a animated sphere with a particle tuner linked to it, that passes through the top of the popsicle, but for some reason I cant get any reaction from the sphere. I had to increase my scene scale to 500 just to keep the gravity from affecting the popsicle shape. If there is another way to do this, I would love to hear it.
    I attached the scene file in hopes that someone can help me with this. Thank you so much for taking the time.
    -B
    Attached Files

  • #2
    Ah, dang, looks like the rar file is not uploaded fully - I get an error trying to open it. Could you try uploading it once more? Or alternatively - sharing it with google drive or wetransfer.

    In general, the approach should be like in this video:



    But if you are doing what I am doing in the video and getting different results, then the first thing I'd check would be the geometry - maybe run an STL check on it, or temporarily replace it with simple primitives like spheres and boxes to make sure there are no issues like overlapping polys, inverted normals, etc, which could be breaking the Tuner.

    Cheers!
    Svetlin Nikolov, Ex Phoenix team lead

    Comment


    • #3
      Thanks for replying Svetlin. I went ahead and uploaded the max file to my dropbox. I'll check what you said also.

      https://www.dropbox.com/s/d10b6d4cga...sicle.max?dl=0

      -B

      Comment


      • #4
        So I checked for overlapping polys, inverted normals and ran an STL check. Everything came back clear as far as I could tell. Next thing I did, was replaced my geometry for a standard cylinder. There is a very very small amount of reaction from the tuner. What I mean is it looks as if one vert jumps.

        Comment


        • #5
          Ah, just increase the distance in the Tuner - currently it's 1 and that is in voxels.

          Another thing that might be better would be to use Volume Brush mode in the Source instead, though in this case Surface Force seems to work well enough.

          Cheers!
          Svetlin Nikolov, Ex Phoenix team lead

          Comment


          • #6
            That worked with increasing the distance. Thanks so much. I was intrigued by your suggestion of using the volume brush mode, so I gave it a try. I had to turn off motion Vel. It was creating a tidal wave effect when it passed through the popsicle. Even with it turned down to .01, it still created some stickiness to my tuner. Another thing I did was, make the tuner stationary and give it enough distance that the melt starts more subtle. I'm still playing with it, but it seems to be working. I do think I will need to greatly decrease the voxel size to get it to look more like a watery liquid. Right now even with viscosity set to 0.0 in the tuner, I'm not getting any separate trails from the melt.
            Another question I had was. I was thinking about adding some sweat to the popsicle itself to help with the realistic look. Is this something that can be achieved with Phoenix and if so can I use more than one Phoenix FD in a scene? What are your thoughts on that.

            One other thing. I had to increase my scene scale 5000 to get the popsicle to not be affected by gravity. Is this normal?

            I cant thank you enough for all your help.
            -B
            Last edited by brandon_hall; 14-05-2023, 09:42 AM.

            Comment


            • #7
              Oh yes, for a better look you would definitely need more resolution, but 15-20 million voxels should be enough.

              In order to get some trails, try enabling the Surface Tension option and try Droplet Formation 0 or non-zero.

              The sweat droplets could be simulated with Phoenix as well indeed, check the droplets on a windshield tutorial for the technique: https://docs.chaos.com/pages/viewpag...xitMobile=true

              And finally, in order for the viscosity to hold the shape more strongly, make sure to enable wetting and sticky liquids and increase the Steps Per Frame option.

              Cheers!
              Svetlin Nikolov, Ex Phoenix team lead

              Comment


              • #8
                Thanks so much Svetlin. Having a blast playing/learning this and is 10x's much easier to wrap your around than Houdini is.

                Will report back if I run into any more issues.

                Thanks again!

                Comment


                • #9
                  Well, new problem. My UV's don't seem to transfer over to my Liquid source correctly. I have tried everything, even rotating my map. Nothing works. The actual popsicle is unwrapped. You can see in the first screengrab I have "Inherit TexUVW from geometry" checked. Any ideas?
                  Attached Files

                  Comment


                  • #10
                    Could it be that you're using another map channel for the uvs? By default it's using 1 in the source, but you might have used something else for the emitter object?

                    You also need to switch the texture applied over the simulator to use Explicit mapping so that it can use the texuvw channel properly.
                    Georgi Zhekov
                    Phoenix Product Manager
                    Chaos

                    Comment


                    • #11
                      The map channel that I'm using for the UV's is set to 1(See attached).

                      Corona bitmaps do not give you the option of explicit mapping, but I did try to use a standard bitmap in a duplicated material and got the same issue.

                      I uploaded a file with the bitmap this time, incase I'm just being completely blind and stupid here and not looking in some obvious direction.

                      https://www.dropbox.com/s/2e1xay6t26...sicle.zip?dl=0

                      Attached Files

                      Comment


                      • #12
                        Any ideas by chance?

                        Comment


                        • #13
                          So I tried using a gradient ramp instead of a bitmap to see if it would work any differently (Image 1). No luck. I have tried rotating U,V, and W. I tried using vertex color channel and planar. None of them worked. I tried applying a standard UVW map with planar mapping (Image 2). That does work, but only to a point. When the object melts, instead of keeping the color of the melt that it should be, it inherits the color that its passing through.

                          Is this something that Phoenix is even capable of, or did I just waste a lot of time, that will not give me the result I need? Could really use some feedback on this.
                          Attached Files

                          Comment


                          • #14
                            Ahaa, my colleagues took a look at the scene:

                            The popsicle geometry has Initial Liquid Fill in tis right click Phoenix Properties - this means it gets filled with liquid at the start of the simulation, and that liquid has no UVWs. Furthermore, when the Source also emits, it adds to the liquid created by the Initial Liquid Fill. This is the reason why Brush mode behaved weirdly - it doubled the amount of liquid and created double the normal pressure.

                            So I suggest you turn the Initial Liquid Fill off and leave the Source do that instead. And now you can safely switch to Volume Brush mode at 100% strength.

                            Cheers!
                            Svetlin Nikolov, Ex Phoenix team lead

                            Comment


                            • #15
                              I did what you said. I turned off "Initial Liquid Fill" on the Geometry and turned on "Emit Liquid" in the LiquidSrc and changed the emit mode to Volume Brush at 100% and the uv's still did not correct themselves. Any ideas?
                              Attached Files

                              Comment

                              Working...
                              X