How to get a flame with a length of 3 m and a flame temperature of 1700 °C

[brad_fds6]

Hello every body,

I tried to model a flame in FDS with a length of 3 m and a temperature flame of 1700 °C and a heat flux of 5 kW/m2 at 15 m from the fire source.

But when i put a mass flow rate in FDS, the length is bigger than 3 m and the heat flux is of 1.8 kW/m2 at 15 m from the source.

I don't know the problem, and i would have help about the way to simulate it in FDS correctly.

Thanks for your help

[fde]

Is the question about modelling the fire source or measuring correctly?

[dr_jfloyd]

For the flame height:

-You could have an input error for your fire

-You grid resolution may be inadequate

--Your assumption that a given fire size will give you a 3 m flame height may be incorrect.  How do you know the flame length is 3 m? If you are using something like Heskestad's flame height correlation, recall that it is just a correlation, and its flame height predictions are only an approximation.

Why do you need to resolve the actual flame temperature?  

If you want to resolve the temperature of the flame sheet, you need to have sufficient grid resolution to resolve the temperature gradient through the flame which will have a thickness on the order of 1 mm.

[brad_fds6]

Hello for your firts answers

In my problem, I want to reproduce the conditions of an experiment in an open environment. During this experiment, the flame has a height of 3 m and fluxes have been measured at 15.
I have flux values and flame height.

Do you have an example of modeling to reproduce this case.

A mesh resolution of 3mm would not be good for me, with a range of 20 m, my computing time will be huge despite the use of a Cluster.

Waiting for your help

[dr_jfloyd]

I don't know the details of what you are modeling, and I don't know the details of how you have tried to model it. All I can do is repeat my early comment:

For the flame height:

-You could have an input error for your fire

-You grid resolution may be inadequate

[brad_fds6]

Dear Jason Floyd,

I have tried diffrent possibilties, but i have always numerical instability.

I start in open configuration, but i have not good result.

I put in attachement my fds file.

Could you help me to slove my problem in order to have a flame with 3m.

Thank for your help

Best regards

Le jeudi 11 avril 2019 09:31:10 UTC+2, brad_fds6 a écrit :

[dr_jfloyd]

You have specified a mass flux of 1 kg/s of propane from a 2 m x 2 m burner. 4 kg/s of propane is a 180 MW fire. This will have a flame height that is close to 30 m.

[brad_fds6]

What is the law that you used, may be i have made a mistake in the law that i use.

So, the good value is 18 MW.

Thanks for your help

Le jeudi 11 avril 2019 09:31:10 UTC+2, brad_fds6 a écrit :

[dr_jfloyd]

This is basic fire dynamics:

&SURF ID='FIRE SOURCE', MASS_FLUX=1.0,SPEC_ID='PROPANE',TMP_FRONT=730.,COLOR='RED'/

1 kg/s/m^2 of burner area

&VENT XB=-1.0,1.0,-1.0,1.0,0.0,0.0, SURF_ID='FIRE SOURCE'/

(1 m - -1 m) = burner is 2 m on a side

The heat of combustion of propane is ~ 45 MJ/kg

1 kg/s/m^2 x 2 m x 2 m x 45 MJ/kg = 180 MW

Heskestad's Flame Length Correlation = -1.02 D in m + 0.235 Q in kW^(2/5) = -1.02 x 2 m + 0.235 x (180,000 kW)^(2/5) = 28 m

[brad_fds6]

Thank you dear jason floyd,

I made a mistake on my calculation.

My question now is on the way to get this flame with a temperature of 1700 °C.

Thanks for your help

Best regard

Le jeudi 11 avril 2019 09:31:10 UTC+2, brad_fds6 a écrit :

[dr_jfloyd]

Same answer as in April. If you want to resolve the temperature of the flame sheet, you need to resolve the flame sheet on your grid. However, it is unlikely that your specific application of FDS requires you to do this.