I was just wondering after watching Houdini tutorial how can we get fire along with Volcano preset. the tutorial image shows some fire burning but the tutorial doesn't explain anything.
Do I need to use two presets one for fire and another for dark smoke.
I tried turning on the combustion model in the volcano preset but nothing comes up.

I also tried making it from scratch without using any preset but unfortunately the smoke was coming in wavy pattern which I am not looking for.

I have uploaded my version (please ignore the shader and color in my version),here are few references

http://www.youtube.com/watch?v=F6F4nj9hbw4&feature=related [youtube.com]
http://www.youtube.com/watch?v=xsKaD169y90&feature=related [youtube.com]
http://www.youtube.com/watch?v=tiPXKrWeLPI&feature=related [youtube.com]

Reduce the Gas Buoyancy and reduce the Temperature Diffusion (temperature field blurring) to get sharper and slower smoke sim.

Also try enabling the advection of the fuel as one way to carry the fire way up in to the smoke as your references show but you will have to cut down the gas expansion and amount of fuel and the other fuel consumption rate parameters.

Thanks Jeff,

If you look, reference it seems the smoke cluster or cloud are moving faster but they are bigger in form.
As my version was lacking details I tried to increase disturbance values a bit though I get the details but I still miss that huge smoke cloud effect and sim also appears bit slow.

Initial big cloud is also missing in my version,which usually appears in all OIL fires

Here, I also did make my source object a bit bigger to get bigger clouds didn't know whether it is smart practice or not (Please check the file enclosed)

I would also like to know does this Big cloud has to do something with the type of shader that I am applying ,because I have tried to render using the new pyro shader as well as the Billowy shader and my look get drastic changes in the rendering.

As I am very much new and I would like to apologize for my silly question but unfortunately I still didn't get which values in shader makes render look huge cloudy .
:shock:

Several things:

Temperature Output at 10 is fine and all but this is used by the Gas Buoyancy DOP inside the Pyro Solver to directly affect the vel field. Simply cutting down the Buoyancy Lift to say 1-2 from the default of 10 (so I know you didn't touch it ) will slow down the velocities (which were honking at 40+ meters per second using the velocity visualization range settings). That's a blast furnace for the scale you want and if you want that, then you need to increase timesteps. Just think of this. You are blowing the velocity probably over 100 voxels at that speed per timestep when set to 1fps. Gas Project Non Divergent will whack that way down and the advection of the various fields with huge velocities across the timestep will cause shredding and tearing and a tremendous amount of fluid loss (as that amount of smoke is smeared across many voxels) as you are well beyond anything that is reasonable to simulate again without increasing time steps.

So with Buoyancy Lift now set to 1, the plume that evolves is much nicer.


Next thing. I see you want to blow the simulation off in the -X direction. By manipulating the Buoyancy Direction is not how to go about this. Well only if you want to have a “Tower if Pizza” leaning smoke sim which will just look plain weird.
You can add a Gas Wind DOP to do this. The Gas Wind DOP is a force that directly affects the vel field and yes it expects a vector vel field to be present on the simulation object which the Smoke Object definitely creates for you.
Removing the -1 in the X field for the Buoyancy Direction and setting the Gas Wind DOP in a -1 X direction and adjusting the wind force now gives you a nice wind effect blowing across -X.

Now that the smoke plume is evolving nicely, time to actually work on the effect itself: The interplay between the combustion of the fuel, the gas released, the temperature released, the height of the flames, secondary combustion of the soot (smoke with a lot of unburned hydrocarbons typical of petroleum fires that can re-combust causing massive amounts of expansion mid-plume) and the biggest question of all, whether to advect the fuel or not.

That last one, whether to advect the fuel or not is one you have to make right up front as it will change all the settings so you have to ask yourself the question: Does this fire carry fuel up in to the plume or not? Compare the burning of say wood or Oil/Gasoline? Wood does release gas and yes the soot generated can re-combust if it is a slower fire but no fuel is advected. Petrol fires burn from fumes/vapours as the liquid can't burn (no oxygen). So it all depends how quickly the fumes burn off or how hot the fire is. Diesel and Oil fires are more tricky to ignite and again require a lot of temperature and off gassing of fumes to combust and again can carry fuel up in to the fire. By how much?

If you choose not to advect the fuel, then you can supply some initial velocity on your source to give some initial kick to the combustion.

If you choose to advect the fuel, then no need to supply initial velocity but adding a bit is always a good thing and this brings us to the emitter geometry.

Emitter Geometry was at way too low a resolution.
The rule of thumb with any smoke/pyro sims these days is that the emitter shape and emitter velocity determines the look of the sim for the first meter or two implicitly so this has a dramatic effect over the look of the simulation.
At a division size of 0.3 and the scale of the sim, the volume container had a res of something like 13^3. Way too low a resolution to create a detailed simulation. Decreasing the division size to say 0.1 for this sim gives you a lot more detail in both the source noise and the velocities.
The sim after these changes has a lot more power and life at the beginning now. Nice.
Have a look at the first attached non_advected_fuel example file with documented changes from above.

For that previous file, you can now tweak away at the secondary combustion of the soot and the gas expansion.

Remember to turn on the heat, temperature, fuel, burn, heat, and divergence fields when dialing in the various combustion parameters.

Temperature drives various noise masks and the Gas Buoyancy.
Fuel is what you are burning. Simple.
burn shows you that there is ignition of fuel and where.
heat is the field that shows you where temperature is being added due to combustion and is an indicator as to where expansion/divergence can happen.
divergence shows you the result of the Gas Expansion parameter or how much the simulation is blowing outward.

Tweak away!

No two explosions are alike and give it your own personality.

Now if you want to advect the fuel, expect to completely rework every parameter and forget about anything regarding direct-ability or predictability as you try to tame the combustion.

To advect the fuel, you simply enable this feature on the Pyro Solver and you better have the fuel field visualization on for display because with the previous settings, you will have one massively expanding fireball. Too massive as you are carrying fuel in to the fire causing combustion further up driving way more temperature causing way more combustion and heat and expansion.

After enabling Advection of fuel at it's default of 0.1 with the previous file, you get a massive expanding fireball.
The first thing to do is cut back the amount of fuel in the Source Volume DOP then reduce the combustion temperature, the amount of heat generated and the gas expansion until the simulation is not running out of control (unless you want a massively expanding fireball that is…).

Too much tweaking… Argh. I also forgot to mention that this technique is very sensitive to simulation division size as well. Decreasing the division scale (increasing resolution) will cause the simulation to behave quite differently so you have to settle in on a higher resolution as well. I warned you that this is not suited to direct-ability.

I just noticed that your emitter noise scale creates elements that are too large so greatly reduced both the scale of the noise and tied to vel curl noise scale too.

Stopping the endless tweaking and see the attached file.

Btw I modified the emitter geometry noise values that would work equally well in the other non-advecting fuel example.

I also adjusted the shredding upwards a bit.

Thanks Jeff,I have studied your files and it helps a lot to understand how advection works with temperature.

Now after that I have tried applying “Flame” shader to my sim and again I find myself bit confused with options available.
Most confusing I find was in “smoke Tab” which are Float and Color
to my dismay when I select Float option I found my smoke color changes from dark to slightly light.
but on the contrary when I use color ramp nothing will change.I don't know how I could use these options and how they would help me in the final rendered look of my render.
I also tried to find out any description which is available on the help menu but applying changes to these options nothing significant appears in the render specially in the case of color

Hi,
I would like to apologize for inserting my issue in this old thread but I find that it is the most appropriate thread for what I am trying to achieve.
I was trying to make the Oil Fire as the reference video uploaded but some how I am unable to achieve the desired result .
I am using Advect as mentioned in the discussion above.
#I am missing the Rolling/twirling/swirling effect in the fire as well as in the smoke.
#Also I am unable to control the smoke dissipation and losing the overall shape of my fire which I would like to be something like my reference mov.
I would request to share your expert suggestions what should I be doing or missing in order to achieve the desired result.

what is correct way to create blending fire and smoke in houdini?
i see many training and get a lot of example for blend fire with smoke in houdini but all of that have big problem:
smoke and fire both generate from one source therefore color of fire influenced by smoke and we get bad blending fire and smoke also if you change parameter of smoke shader color of fire change too!!!
thanks.

NimaNolan
what is correct way to create blending fire and smoke in houdini?
i see many training and get a lot of example for blend fire with smoke in houdini but all of that have big problem:
smoke and fire both generate from one source therefore color of fire influenced by smoke and we get bad blending fire and smoke also if you change parameter of smoke shader color of fire change too!!!
thanks.

Dont know about “the correct way”, but out of the box the solver got controlls for were you want your smoke to emit.

Set to burn it will emit from the burn field. That is were you got fuel and a temperature higher then ignition temperature. Basicly the source for the flames.

Set to heat, it will emit from the heat field that is advected from the burn field and often used as flames. With this option you got controll over were on the flames to emit and a blending region.

However there is nothing that stops you from creating your own smoke from fire emitter.

set to burn or set to heat give you same result!
we need more control about smoke and fire like maya
i try to approach to this fire in houdini:
but The difference is i want very heave smoke on top of the flame.