Jump to content

Flip problems


Recommended Posts

I am trying to recreate a desktop fountain that I bought recently because it is really cool and uses a lot of complex fluid dynamics in it. I have the modeling nearly finished, but the fluid dynamics are getting a little sticky. Below is an image of what I am making.

wfl-gls_inset.jpg

So far I figured out that I need to turn the fountain steps into a rigid body to have the water fall over it. Now I need to figure out how to make it continually emit water. Then there is the matter of the surface tension driven reaction that makes the water flow across the surface of the steps and flow around to the bottom and drip off. I am currently using Flip fluids, but I saw in the help that SPH has the surface tension built into it.

I tried to add an emission node into the POP network, but that just sent particles flying everywhere. Is there an option somewhere in the default network that will cause it to emit continuously?

Link to comment
Share on other sites

Ok, the particlefluidemitter node got them moving in a stream, which looks great, but now comes the hard part. How do I get them to stick to the geometry? I understand the hydrostatic forces that cause it to happen in the real world, but now it's time to recreate it in Houdini. I am hoping that I can do this without getting too deep into VOPS as I haven't really gotten there yet. Is there any way to do this using strictly DOPS? I was assuming that either viscosity or surface tension would work, since the real world forces stem from those two properties of a fluid, but they don't seem to do what I am looking for. Here is what I have so far.

Fountain1.png

How it should work is that the water falls on the top surface and fills it up, the water tension holding up the edges until it weighs too much, then the water falls and it's surface tension clings to the underside of the shelf until enough collects that it falls under it's own weight. I have a feeling that, with the number of times I said weight, I am looking at a rather complex VOPS network to do this, and I will need some help to figure it out, though I don't want the answers handed to me too much. General pointing would be nice though.

Link to comment
Share on other sites

I am currently using Flip fluids, but I saw in the help that SPH has the surface tension built into it.

Tbh, FLIP is meant to be 'splashy', very inviscid & for larger scale sims from what I know, so SPH might be better. One tip is to use the 'Sticky Fluid' option in the solver node tho, as this might get a better result. Not sure if it's on by default but might be worth a look

Link to comment
Share on other sites

Slow is fine, so long as it is working. I have been playing around with SPH and have found very limited success... which is now gone. What did I do to my poor simulation? It was coliding just fine, but now the fluid just hits my fountain and then falls through it. What is going on? can anyone help with how to get it to cling to the underside of the shelves?

Edited by Adam Ferestad
Link to comment
Share on other sites

Getting custom velocity vectors to direct flip particles:

Even though less related to this effect in particular I still highly recommend you to take a look at this file where flip particles are being held in place only for a certain amount of time -shows you how to play with the velocities:

about custom velocity fields -- In houdini 11 there are plenty "attribute to volume" or "volume to attribute" and other volume manipulation tools added, the rest of the thesis is still quite practical - even if it is more directed towards smoke/fire, flip uses fields as well underneath, so you could either manipulated the fields or you could manipulate the velocity on the particles:

http://www.peterclaes.be/tutorials/PeterClaesThesis/index.html

As an exercise I would recommend you to do it withouth fluids, just see if you can make it work with just particles -- A lot of the same knowledge will feed into your flip sims. With particles the key will be to play with the states (collided, sliding,...) and probably creating an sdf out of your collision geometry that you can use to lookup the normals and closest point on surface. That way you can decide when a particle is allowed to detach from a surface.

Have a look at this file in regards to the states:

http://www.sidefx.com/index.php?option=com_forum&Itemid=172&page=viewtopic&p=97495#97495

Have a look at this file in regards to pop collisions with sdf:

Flip to me is the fluidity SPH offers with the control particles and vopsops offer through sopsolver, it is fast, stable and you can cheat/control a lot of things. SPH is much more of an unstable black box in comparison. But some things you can not easily achieve with flip yet and sph with a lot of substeps might give you more accurate results.

  • Like 1
Link to comment
Share on other sites

Hmmm... I looked over those links and I can honestly say that I can't make heads or tails of them. I guess I have much more to learn. Looking over the networks rarely helps me at this point since I am still trying to figure out what the nodes do at all, let alone how to work tricks and magic with them. I really do need to do some more thorough tutorials I think. Perhaps I will do the bridge and elevator tutorials from 3dbuzz.com next week. It will give me something to do at least and should teach me some stuff about the whole process of creating things in Houdini. I have no idea if it will help with any of the 3 projects I am currently trying to get working, but one can hope.

Link to comment
Share on other sites

There are a lot of tutorials out there, especially the ones Peter Quint makes should take you more step by step.

It is true that some of the above links are more intermediate to advanced and each one of those files can take some time to understand, but do come back to them and spend the time looking through them. They are no quick solutions, but instead a series of techniques that you will be able to use elsewhere too. They are cool techniques though. As soon as you don't understand what a node does, go to the help file and read up on it and check the examples that come with it. Or perform a search on the forums for more example files... try searching by author for Macha, pclaes, old school - or others that you have found to provide good example files. Sometimes I'm just curious to what other type of subjects members are interested in and what kind of files they have shared, particularly if they have just shared some bright file :) - you can check what other posts they have made from their profile.

hmm... the bridge tutorial :) it is a long time ago since I was part of that one. It won't help you all that much with particles as it is 90% sops and 10% vopsops, but there is a bit of vops in there that goes over linear interpolation which you'll probably enjoy - remember when I said in that other thread how I used to implement math in Houdini when I learned some of it in uni... that linear interpolation is a bit of that. It is a fun tutorial that shows a good dose of proceduralism.

Link to comment
Share on other sites

Yeah, the proceduralism is what I am trying to focus on right now, but for some reason I keep trying to do some really advanced things with it apparently. I am loving all of it though. I really wish I had known about Houdini 12 years ago when I started my path down 3d, because it really does work in my head the way I need it to. Much more so than 3ds Max ever has. Unfortunately, I am only like 3 or 4 months into the whole program, so I should probably focus on walking before I take off at a dead sprint. It hurts more when you sprint into a wall.

Link to comment
Share on other sites

I did a quick test with the gas surface tension solver node and this is what I came out with. For some reason the particles seem to be getting accelerated and thrown off of the model I am using as a test collision. I was already rendering this (god I need a new computer, this took something like 35 hours to render) when I posted about taking a break from this, but I still found it odd that the particles are being accelerated. I am curious why that is going on, so here is the file that I used and I hope someone can tell me what I did wrong in the surface tension solver.

foutain_experiment.hipnc

Link to comment
Share on other sites

one thing you want to understand about gas microsolvers is they generally need several data objects attached what ever's being processed. eg. the gas surface tension microsolver requires a Velocity Field, a Surface field, and a Collision field, none of which exist out of the box in an SPH object. If you wanted to use that node you would need to generate all those fields beforehand, same thing with the gas surface snap node you have in your scene

Link to comment
Share on other sites

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.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...