Fluid Emitter Offsets

Become a member of the CGSociety

Connect, Share, and Learn with our Large Growing CG Art Community. It's Free!

THREAD CLOSED
 
Thread Tools Search this Thread Display Modes
  07 July 2012
Fluid Emitter Offsets

I noticed in a lot of tutorials for fluid explosions how the keying for the fuel, heat, and density attributes are usually offset when keyed.

Whenever I try approaching this it gets tricky and I can never really get my head around why. Perhaps I am over thinking it.

Any suggestion for some sort of formula/ratio to offset these attributes for an explosion? For instance, I read somewhere in another tutorial that a good start up point for explosions or fire is to have twice the amount of fuel than there is in heat (or it's vice versa).
 
  07 July 2012
Dont lose sleep over it, try it out for yourself. A lot of fluid stuff is subjective and it depends on what kind of explosion you are after. By adding more fuel and less heat I believe just makes the fluid burn a lot slower and produce a consistant rolling explosion. Too much heat and you'll end up with a fireball and excessive velocity vectors.
Offsetting the attributes adds randomness rather than having them all finish the same time and therefore looking unrealistic.
__________________
Vimeo

Last edited by Aikiman : 07 July 2012 at 09:06 AM.
 
  07 July 2012
Hi, Vincent!

I'm building a dusty explosion right now so I'm just gonna roughly tell you what I'm doing.

I'm emitting a little blast of nParticles with a very low life span (they last about 9 frames) and I'm using them as fluid emitter to emit Density and Fuel (with some inherit velocity). Density is set to 3, Fuel is Set to 8.
Fuel emission starts at frame 4.
Density and fuel emission turns off automatically after the nParticles die off.

The heat (set to 5) is being emitted by another small volume emitter from frame 4 to frame 10. But since the emitter is very small there is not much temperature emission. Its just used to ignite the fuel.

So, for the first few frames I have lots of density and fuel and very little amount of temperature. That little temperature ignites the fuel. As the fuel burns quickly (I have high reaction speed) more temperature is released. The temperature has high buoyancy and some temperature turbulence which is then responsible for making it explode big.

Also I have two volume emitters at the bottom of the containers which are only used for speed emission to push the fluid side ways. There is one big volume emitter in between which I have animated moving upwards. This big volume emitter has inherit velocity set to 3 so as it animates up it also takes the fluid up creating a mushroom shape.

I'm also using Auto resize on the fluid container.

So you can see I'm not using the "emit Fuel twice as much as Temperature" method in here. Maybe because of the type of explosion I'm making. For a much firey explosion like a nuclear blast you may need to pump lots of temperature and fuel at the same time. I'm uploading some pics and a play blast here of a work in progress of the dust explosion I'm making which will give some sense to all I've written above.:

The Emitter Rig:





First few frames of explosion. Emitting just Density and Fuel:




Temperature released on 4th frame till 10th frame: You can see that temperature is burning the fuel.



Here you can see the big volume emitter which is animated upwards is creating the mushroom cloud formation.



A playblast: http://www.youtube.com/watch?v=7z6Z...eature=youtu.be
I'm learning how to make explosions so what I've posted here is just an experiment. Lots of tweaking needs to be done. But I hope it helps a little. : )
__________________

 
  07 July 2012
Nice one looks cool, heres my test...

https://vimeo.com/45175201

I havent really played with fluids since Maya 2011.
__________________
Vimeo
 
  07 July 2012
Nice! Looks cool the way heat travels up!
Also you said on vimeo you were using air/fuel ratio. Can you explain a bit on how you implemented it?
I probably use it wrong. Does it have something to do with how fast or slow fuel burns?
__________________

 
  07 July 2012
Thanks for the replies and time spent! I will give definitely give it another try with the info you guys gave and post up any results later in this thread.

Since we are on the topic of fluid explosions, I would like to ask what you both think of these playblasts of explosions? I am considering how I can improve these explosions and if fluids can be a solution. Download Link (in blue box)

As you can see the explosions are triggered by a collision event system when the rockets/bullets hits the ground. The plan for now is I am trying to create explosions using particles, which can then be shaded using David Schoneveld's Overburn method.

Right now I am concerned of a lack of rolling motion in the main Explosion Smoke. I can at least have the Hellfire explosion be a single fluid explosion instead of particles since it is one missile (and I probably will do just that). However, the bullet and Hydra explosions were trickier because of how they are happening multiple times after another (I may even consider holding off on the hydra explosion and focus on the bullet and Hellfire explosions instead).

I tried to put a fluid emission on the Ground Collision Particle resulting from the bullets/rockets hitting the ground (the debris and sparks are also emitted from this particle). But because of the very short lifespan of the Ground Collision Particle (around .1 seconds) and small emission area, it did not have enough time for the emitter to create enough fluid particles.

I could have an extra Ground particle created with a longer lifespan, but it will be tricky to do the keying of those three attributes when there are sequential bullet/rocket collisions happening over and over. But perhaps I am overthinking it again
 
  07 July 2012
Originally Posted by Parv: Nice! Looks cool the way heat travels up!
Also you said on vimeo you were using air/fuel ratio. Can you explain a bit on how you implemented it?
I probably use it wrong. Does it have something to do with how fast or slow fuel burns?


Its my first play with that setting and a value of around 2 worked with this effect, I didnt really know what I was playing with at first but I hit the sweet spot I think. If it was too low, all the fuel just became really heavy and burnt up too quickly but a high value meant there was not enough density for the fuel to burn so I didnt get any ignition. try using the extreme values and see what you come up with, its all about how the fuel mixes with density and how it burns.
__________________
Vimeo
 
  07 July 2012
Originally Posted by Venator11387: Right now I am concerned of a lack of rolling motion in the main Explosion Smoke.

But because of the very short lifespan of the Ground Collision Particle (around .1 seconds) and small emission area, it did not have enough time for the emitter to create enough fluid particles.

I could have an extra Ground particle created with a longer lifespan, but it will be tricky to do the keying of those three attributes when there are sequential bullet/rocket collisions happening over and over. But perhaps I am overthinking it again


I think they are looking great so far. Rolling motion can be achieved by plugging in the rotatePP into the textureRotate of the fluid, to be honest I havent tried it but that is the way to go if you are going to use (Peter Shipkov's) Overburn method.

For the ground emission just emit a ton of particles at collision and emit fluid from those particles with short lifespan - emit density, fuel and heat and velocity that will solve your problem. If you are familiar with SOuP you could use that too but I think the first method is best. GL looking cool so far
__________________
Vimeo
 
  07 July 2012
Thanks for the tip on the texture rotate, I will definitely use that!

However, what I meant by rolling motion is the curling of the smoke itself due to the forces of convection and air pressure. Like the upswelling shape of a mushroom cloud etc. Fluids does this easily. Particles I find are not as built in.
 
  07 July 2012
oh yeah, what you can do is affect particles using a fluid as a field. You can use a fairly low res fluid container say 100x100x100 and this will give you nice swirly effects on your particles. You still need use another field to affect just the fluid so you get velocity vectors inside it, dont use this other field to affect particles though, just let the fluid do the work. Turn particle conserve right down to zero.
__________________
Vimeo
 
  07 July 2012
Here is a rough of the explosion based on input from both you and Parv. I worked out the blast well enough. Some issues I am having some trouble figuring out are:

1. The way the particles shape are disspiating is pretty erratic especially at the end instead of being smoother.

2. The smoke shape is relatively solid and not as fluid after the blast. There is none of that upswelling rolling motion as it moves up despite there being that cylindrical volume field I put (as seen in the playblast).

The reference video in the zip file is the kind of overall motion I am trying to get in this explosion.

Link
 
  07 July 2012
Whats the method you are using, is it Overburn? Looks like you are having issues with transparency for a start. How are you using the field, is it affecting a fluid and then are you using the fluid to affect your particles?
__________________
Vimeo
 
  07 July 2012
I am using no methods outside of the given fluids system. I have not yet applied the fluid field method to my bullet explosion particles.

The fields are there to try to get that rolling motion. the flat cylinder on the ground was for the ground roll, while the tall one is for the motion going upwards.

You can take a look at my scene in the attached file.
Attached Files
File Type: zip Explode.zip (18.6 KB, 33 views)
 
  07 July 2012
Okay, I had a fiddle but its far from perfect but at least its on the way. I changed the opacity ramp - this was a big problem, also the timing of the bottom field happens slightly later than the explosion so density/fuel/temp has time to rise before the rest gets blown to the side. I think the timing here could be improved though.

You could have 2 emitters, one for the main explosion and the other just for the ground FX ie no fuel or temp (just a little at most). At the moment there may be too much fuel and temp being dragged along with the ground field.
Attached Files
File Type: zip Explode.ma.zip (19.1 KB, 38 views)
__________________
Vimeo
 
  07 July 2012
Hi, Vincent!

I was looking at that reference video you provided. I made an explosion similar to that. Here's the play blast: http://www.youtube.com/watch?v=kCgGN99F3AY
I thought you might wanna take a look.





I've attached the scene file. Maybe it can be of some help.

There are a few things hidden. You may wanna unhide them from the outliner.
Attached Files
File Type: zip House_Explosion.zip (16.4 KB, 57 views)
__________________


Last edited by Parv : 07 July 2012 at 02:08 PM.
 
Thread Closed share thread



Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
CGSociety
Society of Digital Artists
www.cgsociety.org

Powered by vBulletin
Copyright 2000 - 2006,
Jelsoft Enterprises Ltd.
Minimize Ads
Forum Jump
Miscellaneous

All times are GMT. The time now is 06:43 PM.


Powered by vBulletin
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.