What influences ignition temperature and oxygen consumption

[Ivorne]

Hello,
This is for my independent game development project. I am aiming to simulate simple environmental properties in grid-based world. I have world made of cells (or blocks) and I consider each physical property to be constant and uniform in given cell. Some properties spread from a cell to neighbouring cells in update steps. Processes I want to simulate: oxygen partial pressure, inert partial pressure (co2, vapor, others), temperature, fire. I also consider adding toxig gases, vapor (as separate item), radiation and radioactive contamination (radioactive elements).

I want to ask you about some things about combustion. My current model (not implemented yet) goes like this: A block has given mass (kgs) of flammable material and an ignition temperature. Upon reaching this temperature, it will ignite and start consuming flammable material and oxygen and start producing heat (energy). This heat is generated into the same block where the combution happens. This does not only sustain the fire in this block but beause the heat propagates to neighbouring cells, it can possibly causing them to ignite too. There are three ways in which the fire can stop: 1. oxygen depletion, 2. flammable material depletion, 3. heat propagates to neighbouring cells too fast (faster than what is created by combustion) and the temperature of this cell goes under ignition temperature.

So now my questions:
1. What influences ignition (autoignition) temperature? Is it good low resolution aproximation to say that it depends on a) material (property of the material, one number) and b) partial pressure of oxygen (linear dependency, the higher oxygen level, the lower ignition temperature)? This would mean that with extremely high oxygen pressure, anything will ignite in any temperature. I read on wikipedia that ignition temperature depends on oxygen level and pressure, is it ok to conflate them together and say that it depends (instead of on these two) only on partial pressure of oxygen?

2. How to easiest compute / aproximate oxygen consumption rate? Could I use just [oxygen consumed] = [energy created] * [flamable material related constant]? Or does it even depend on the material?

3. Can be energy created per time constant for given material (aproximately)? Or should it depend on oxygen partial pressure? And how should it depend?

Thank you for your advices. Formulas would be apperciated. I just was not ready to spend whole day aggregating all formulas about combustion to get the dependencies I want. I thought it would be nice to once ask someone with more knowledge about this to tell me something about the dependencies between those values.

Thx again,
Ivorne

 

[Roark]

Flame ignition/propagation doesn't only depend on temperature. For example, flowing grain can spontaneously explode at room temperature (electrostatic discharge). I can light a match and drop it into a pool of ethanol and it will light on fire.

There are people who study how fires can propagate through rooms and buildings. Their models are complex and if you want that level of description you need to do your own research. Since this is not an academic project, I recommend just starting with the simplest models first. There are just too many things that could be talked about if you want accurate models. Make many assumptions and have a game that runs.

 

[Ivorne]

:D
Ok, I forgot to mention that the simulation cells are going to have size around 1m^3. Of course I want to work with simple approximative model. I made lots of approximative and precision models for various physical processes. I know very well how to do them. I am not asking you how to program it. I just want to map all the dependencies between given variables - pressure, volume, weight, temperature-induced autoignition, material constants, heat generated and oxygen partial pressure in air.

 

[Roark]

Maybe I can help add a little complexity. You can try adding in the fact that only a certain portion of that mass can be considered flammable (x). I think it would be good to give it a property that describes how fast it can burn. This property can depend on porosity and maybe other factors. For the things that do burn it won't be complete combustion either. Not all will be turned into CO₂. So consider putting in another factor for the amount of heat generated. Some engineered flame-resistant or flame-retardant materials can leave a coat of carbon on the material surface, inhibiting any further combustion. If you are doing this simulation in a closed system (no gas escapes) then the pressure will increase over time and also increase the heat capacity of the gas. How are you simulating the heat of the gas escaping by the way? I think a good approximation might be the flame temperature, but even that depends on material and the environment.

Let's talk about oxygen. I think using the partial pressure of oxygen (P_O2) is a good idea for ignition temperature and combustion rate. I think T^Ignition = C*1/P_O2 + D might be better than the linear relationship you are talking about. With this model, infinite P_O2 will lead to ignition at D temperature. Zero P_O2 will lead to infinite ignition temperature.

I suggest looking into Dr. Ya-Ting Tseng Liao (https://scholar.google.com/citations?user=UfKp-foAAAAJ&hl=en&oi=sra) and her work. She is a simulation person, doing what you are doing at an academic level. I think this article might be a good place to start(http://www.tandfonline.com/doi/abs/10.1080/13647830.2013.831486).

 

[Ivorne]

Thanks, these are good ideas, I will try to incorporate them. About the heat simulation: I use discrete approximative model of thermal conduction. That means that the fire will generate heat to one cell (the burning cell) and the heat will be transferred to surrounding cells - the bigger the difference in temperatures of cells, the faster. So I expect that a specific flame temperature will arise from the model without explicitly specifying it (considering somehow stable state of surrounding cells). You made a good point that the combustion it self should generate some gas and therefore possibly increasing the pressure. I am definitely going to implement that.