sparkChan Posted June 19, 2011 Share Posted June 19, 2011 (edited) hi. Assuming that I have three waterdrop in the ceiling and I want those waterdrops mix together and become a big waterdrop, then fall down from the ceiling. I used Magnet Force DOP to mix those waterdrop into a waterdrop, but for the waterdrop falling down from the ceiling, I have no idea. so what will I do next? In this process, I used flip solver to do that! thanks. Edited June 19, 2011 by sparkChan Quote Link to comment Share on other sites More sharing options...
Adam Ferestad Posted June 20, 2011 Share Posted June 20, 2011 I am trying to picture exactly what you are asking. You are wanting to have the three drops join into one as they fall through the scene, then have the large drop continue to fall? How physically accurate are you trying to make this? How are you handling the camera work? Is the camera moving with the drops or is the camera stationary and the drops are falling past it? I am somewhat confused on how the scene is setup, because I would probably approach it from different angles depending on what the end goal is. Also, are you trying to do this with just the 3 drops, or a continuous system, 3 drops at a time? Quote Link to comment Share on other sites More sharing options...
pclaes Posted June 20, 2011 Share Posted June 20, 2011 Flip should be fine for this. As you can override/adjust the velocity of the particles. You will basically have three initial blobs of fluid particles, each blob should have a different group to them, but they need to be part of the same particle fluid object. So you can treat each blob separately inside a sop/popsolver. You will initially need to give them an upwards force to counter gravity and also a directional force to direct them towards each other. Since it is such a specific amount of drops, you will want to time the release of the surface manually. This is basically when you manually turn the upwards force to zero, therefore letting gravity take over. In order to fake surface tension you should not turn the upwards force to zero on all particles at once, instead reduce the force gradually from a center point outwards. That way the particles will probably appear to peel off the surface. If you want an example of how to split off and manipulate different flip particles and mess with their velocities, have a look here: Quote Link to comment Share on other sites More sharing options...
sparkChan Posted June 22, 2011 Author Share Posted June 22, 2011 thanks. this is the test I've done earlier. but when they got together and fall down, the whole water goes away, but I want a little water to fall down form the big water. hi, pclaes, thanks for your example. I'm a newer to flip solver, so can you give me some more information about it? thanks. Quote Link to comment Share on other sites More sharing options...
pclaes Posted June 22, 2011 Share Posted June 22, 2011 Flip will give you overall fluid motion. It will not give you any density control. But you have the option to override the velocities. If you have a look in that example you will see that the "bubbles" particles of the flip fluid object are branched off and treated differently. A magnet sop is not bad, but is very rough in terms of control. By using pointclouds and/or assigning individual velocity vectors to your particles you can guide each particle to where it needs to go. You can link up your solvers with a multisolver. First input: flip ojbect, second input flip solver, third input sopsolver. If you are looking for more in depth information on how flip fluids works I would recommend having a look at Robert Bridson's book on fluids. (Or have a look at some of the siggraph papers he published). Quote Link to comment Share on other sites More sharing options...
sparkChan Posted July 1, 2011 Author Share Posted July 1, 2011 Flip will give you overall fluid motion. It will not give you any density control. But you have the option to override the velocities. If you have a look in that example you will see that the "bubbles" particles of the flip fluid object are branched off and treated differently. A magnet sop is not bad, but is very rough in terms of control. By using pointclouds and/or assigning individual velocity vectors to your particles you can guide each particle to where it needs to go. You can link up your solvers with a multisolver. First input: flip ojbect, second input flip solver, third input sopsolver. If you are looking for more in depth information on how flip fluids works I would recommend having a look at Robert Bridson's book on fluids. (Or have a look at some of the siggraph papers he published). hi, pclaes, I'v seen your file, but it's difficult for me to control the velocity of particles in sopsolver. just like this picture showed. I really want to get this result, so I created a velocity decreasing from center to side,but next I don't know how to go. hope your advise! thanks! this is the file. waterdrop_setup.hip Quote Link to comment Share on other sites More sharing options...
sparkChan Posted July 1, 2011 Author Share Posted July 1, 2011 this is the test I've done today. waterdrop_02.hip Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.