Pancho Posted January 25, 2018 Share Posted January 25, 2018 (edited) Hi there, I just wonder where these lines/ripples come from. They look ugly and not realistic. My velocity volume is about 200 voxels in width and covers the flip sim pretty close. So I guess that's not the problem. Has it got to do with substeps or the grid size? The sim runs very slowly so changing settings and get "instant" feedback is a "small" problem.... : ) Roughly 36m particles and 6m per frame. Cheers Tom Edited January 25, 2018 by Pancho Quote Link to comment Share on other sites More sharing options...
Pancho Posted January 25, 2018 Author Share Posted January 25, 2018 In this sim it seems to be present, too. But not as visible. Quote Link to comment Share on other sites More sharing options...
coltonmil Posted January 25, 2018 Share Posted January 25, 2018 Stepping like this in faster sims is usually resultant of too few substeps. Perhaps dropping point count to test more substeps might help you get quicker feedback, before cranking it back up for the final sim. A 200x200 velocity volume might be contributing a bit, as for the number of points you have that's a little coarse, but that'd probably be smoothed out as opposed to stepping. Is the velocity volume animating? And if so, is it cached? It could be reading in and not substepping properly if that's the case. Quote Link to comment Share on other sites More sharing options...
Pancho Posted January 25, 2018 Author Share Posted January 25, 2018 The velocity volume is static, though the vector within the volume are changing. I guess I'll try both. More substeps and higher voxel count in the vel volume independently to see what helps most. Cheers! Quote Link to comment Share on other sites More sharing options...
chrisdunham95 Posted January 26, 2018 Share Posted January 26, 2018 Does it still hold these artefacts when meshed? - meshing may help smooth out and eradicate some of those not natural ripples and lines - (Otherwise I agree as you guys have discussed above!) Quote Link to comment Share on other sites More sharing options...
Pancho Posted January 26, 2018 Author Share Posted January 26, 2018 I did encounter some major problems today. One of them was that the bounding box with padding spawned particles all over the bottom and the side of my fluid tank, way beneath the narrow band. Even if all the forces were disabled, same effect. Lowered the grid scale from 2 to 1 and it was gone. Now I got a sim in which I started to change the substeps. Well quite a difference! What's going on here? Why is the surface crashing to the bottom of the bounding box? How do I set the surface volume correctly (Flip solver -> volume limits)? CAn I just use the ground plane in here or how do I create a correct SDF volume for this purpose? Cheers Tom Quote Link to comment Share on other sites More sharing options...
coltonmil Posted January 26, 2018 Share Posted January 26, 2018 So you're trying narrowband/surface style now? What could be happening is the volume vel is pushing the particles outside of the threshold of the surface... I don't have too much experience with using narrowband, but I'll do some experiments! Quote Link to comment Share on other sites More sharing options...
Pancho Posted January 26, 2018 Author Share Posted January 26, 2018 Yes. In order to get the details I guess there is no way around it, BUT I'm too unexperienced with flip fluid workflows in general. Is there a way to upres the medium resolution sized sim , based on the velocity field and the particle positions? My idea of simulating flips is to get the high res detail in one go, which might be the wrong approach. I guess the most important step now would be to get the volume surface right, so that the waterlevel stays the same. This is annoying to have a nice narrow band simulation which crashes down to the bottom of the sea (sim tank). Afterwards I can check the other problems. I'll try to post a scene file tomorrow. Cheers! Quote Link to comment Share on other sites More sharing options...
Pancho Posted January 28, 2018 Author Share Posted January 28, 2018 Guess I discovered why the fluid drops down to the bottom of the fluid tank! The reason seems to be the whirlpool itself. As soon as the forces start to tear a center hole into the flip fluid, the fluid drops from the center towards the border to the bottom. There are still particles generated at the former borderline which crawl down the outer limits of the sim tank. So the question seems to be, how to generate a swirl in the center of the sim without tearing a hole, but a tornado like tube going down as far as possible. The thicker the narrowband, the later the waterline drops down. But the point is to keep it shallow in order to get more details at the same particle count. Wonder whether one needs to sim this in a flat tank and then deform the surface or particles afterwards to get the swirl/tornado depression in the surface. Quote Link to comment Share on other sites More sharing options...
johnLIC Posted January 28, 2018 Share Posted January 28, 2018 Or model the deforming tornado whirlpool shape ( maybe in NURBS so the tangents can drive the vel?), and sim on top of it? Quote Link to comment Share on other sites More sharing options...
Atom Posted January 29, 2018 Share Posted January 29, 2018 Here is a way to make a whirlpool mesh surface. Maybe you can use it to feed the fluid source? ap_whirlpool_pattern.hipnc Quote Link to comment Share on other sites More sharing options...
Pancho Posted January 29, 2018 Author Share Posted January 29, 2018 Thanks so much for the imput. I wonder how to use a whirlpool shape in order to shape the particles. I guess there are to possibilities. a) Have a sim with a collisin/attraction surface or straighten out a flat sim and reshapie it to the desired form. in case anyone successfully managed to use a narrow band fluid for a whirlpool sim, please step forward and report about your endeavour. Wonder whether it is possible. All my undertakings were unsuccessful as the surface always dropped below the waterline. Quote Link to comment Share on other sites More sharing options...
Dixtra Posted July 15, 2020 Share Posted July 15, 2020 On 1/30/2018 at 1:36 AM, Pancho said: Thanks so much for the imput. I wonder how to use a whirlpool shape in order to shape the particles. I guess there are to possibilities. a) Have a sim with a collisin/attraction surface or straighten out a flat sim and reshapie it to the desired form. in case anyone successfully managed to use a narrow band fluid for a whirlpool sim, please step forward and report about your endeavour. Wonder whether it is possible. All my undertakings were unsuccessful as the surface always dropped below the waterline. Hi Pancho, i was trying to replicate the same thing that you wanted, driving flip fluid with mesh. Have you had any luck and could you share some tips? thanks 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.