CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   flame direction (

bamini October 18, 2012 09:17

flame direction
hello all,
I'm simulating the methane burning inside a big room,the burner is located on the floor at the middle of the room. the direction of flame is horizontal while that must be vertical. other thing seems correct, please help me to solve this problem.

michujo October 19, 2012 05:33

Hi, can you please explain further your case? Otherwise people might not be able to help you....

How is your burner? is it a jet flame? is it premixed? how is the geometry of the burner and the surroundings? boundary conditions you are using? are you considering buoyancy?... Post a (or some) pictures if possible.


bamini October 19, 2012 09:22

1 Attachment(s)
this is a room 6*5*4 m,full of air. the methane burner is 0.5*0.5 m right at the middle of the room,just methane comes out of that with 6 gr/sec mass flow rate. there is an opened door on one of the walls (1*2 m). I consider gravity and use transient solver.
burner is mass inlet and door is pressure outlet, walls roof and floor are just wall.

michujo October 19, 2012 17:16

Hi, I'm assuming that methane comes out of those four cells with blue border at the bottom at the room, right? if so, the mesh is extremely coarse, you cannot pretend to compute a flame with that mesh. Is it URANS, LES, or is it a laminar flame?

Do you mean that the injection hole is 0.5m*0.5m? if so it does not seem very realistic, that is a hell of a burner! How is the true configuration? is it a case you have experimental data or is it just an academic example? I have never seen such thing before...

Flames are very thin regions where large temperature and chemical species concentrations gradients are found, therefore you should refine a lot your mesh close to the flame if you want to get good results.

If there's no other inlet, where does air come from? after a while, the oxygen in the room will be completely depleted and your flame with extinguish, then the methane will just come out without burning and fill the room.

How do you initialise your calculation? how do you ignite it? do you model something like a spark or you just put some hot pocket at the beginning to ignite the mixture?

Did you have a look at the velocity profile? Maybe the velocity is blowing the flame towards the door and that's why you get something like a horizontal flame shape, anyway, with that resolution I don't think the results are meaningful.

Can you post a picture of the flame shape? (heat release field). Also, methane concentration field will be useful.


P.S: If you have access to the book y Versteeg & Malalasekera, there's an example of a room fire CFD model, maybe you can get some ideas from it.

bamini October 22, 2012 06:03

Hello and great thanks,
Honestly I want to simulate the pool fire ( liq fuel) so the geometry is for real situation but for liq fuel.
I think it is my mistake that I don't change the geometry, I try to do it, instead of 0.5m*0.5m hole use just small hole with finer mesh, I will write to you the result asap.
I use the tutorial guide:
Chapter 16: Modeling Species Transport and Gaseous Combustion
So according to that I don’t need any ignition although I try with ignition too.
There is a door, isn't it enough for coming fresh air in?

michujo October 22, 2012 06:51

Hi. The example I pointed to before (in the book by Versteeg&Malalasekera is precisely a pool fire so you might want to have a look at it).

What you want is to define a door that enables incoming air and outgoing burnt gases, right? I don't know if such thing can be modelled (maybe "opening" boundary condition? any Fluent user here that can help us?).

In the reference example they set up an air inlet and an outlet though.

Good luck.


bamini October 23, 2012 08:00

Hello again,
Thanks I have found the book it is useful.
Also I'm trying to take door in 2 parts and run the sample again, hopefully I get air coming inside.

All times are GMT -4. The time now is 05:35.