Paul Curtis

Soothsayer
Just adding the gaswind will cause velocity differences within the smoke and make it expand as it moves along because of the turbulent nature of gas simulations. Same with confinement. So, the correct way would be to get rid of everything that causes velocity differences and turbulent behavior and start from there. You can of course try and massage the vel components a little. v@vel *= {0.9,1.0,0.9} or something along those lines but that will only bring you so far because you're forcing something unnatural but sometimes these kind of tricks work well enough. You could also build curves or a simple popsim and attract towards them.

Personally I'd get rid of anything that impedes a smooth upward flow sim, turbulence and forces and all and then add forces one by one.

Sometimes I do very low res and simple sims to get somewhat believable initial conditions of velocities and then I use that to help along the sim proper in its initial stages. That way you avoid simming right into empty 'air' (and that's when you're fighting those velocity differences that cause unwanted billowing, mushrooming, or expansions as you have here).

Stupid fairly newbie question - where would i be able to add that expression? Is there a node i can use? I have been using the Gas Limit node with the vel and limiting the speed of it, however i feel that i cannot smooth it off, in other words once it hits a speed i think the Limiter zero's the values.

I'm a lot closer but that expression, looks like it might be useful to not Limit the vel over all but to reduce it always.

Is there a micro solver node i can pop that expression in..?

thanks!
Paul

Soothsayer
It probably expands like this because your smoke has small and medium scale noise that builds up into more and more random velocity, aka tending to spread out equally in all directions. Remove all noise as much as you can, as well as any added sourcing vel, and see what the smoke does then and use that as a starting point.

Okay, that's interesting.

I've added a GasWind and of course i can push the smoke upwards but from what you say (which makes sense) how could i go about reducing the X and Z velocity vectors so i encourage a more upward motion?

I have vortex confinement which helps a bit and i tried a few other micro solvers but not really found what i am looking for. The initial velocity is good, the start of the smoke is how i want it. It just seems to reach a certain height and then spread.

thanks
Paul

This is probably pretty basic but i'm damned if i can work this one out

Nice smoke emitter, nice columns of smoke but after about 100m the smoke diffuses and expands out, bit like a funnel with a spout.

I'm trying to create vertically long smoke columns and i have tweaked and tested practically every parameter i can find.

So buoyancy is high, cooling is low and i have a Dissipate microsolver where evaporation is low. I can Vortex Confinement which helps keep detail.

I am increasing and decreasing the source as well to create randomised areas of density.

But what approach can i use to keep the upper smoke 'together' and not fan out so much, which aspect of the pyro causes this?

Imagine a fire, smoke floating straight up and expanding a *little*.

Really appreciate any pointers

(the image is a sample below, the buoyancy scale is 1 here but it's on 8 at the moment not that i makes too much difference...

cheers
Paul