Laminar Flame Speed Overprediction in CHEMKIN

[shezanghamat]

Hi

I am using premixed laminar flame speed simulator in CHEMKIN to calculate flame speed at equivalence ratio (ER) range 0.5-2. Mixture is,

Fuel = H2 (1)

Oxidizer = O2 (0.21008), N2 (0.78992)

numbers under parenthesis are mole fractions. Operating condition are 1 atm, 300 K. I am using complete combustion products as H2O and N2. Only ER of 1.1 is excessively overpredicting the flame speed in order of 10^11 cm/s. Rest of the range of ER are valued as expected. I have tried refining the grid, increasing number of time steps, changind the adaptive grid values, etc. Still the output remains same. Also, plot of number of grid points vs ER shows very less grid points assigned to ER 1.1.

[Didier Bessette]

Clearly there is something wrong in your calculation particularly on Equivalence Ratio 1.1 case. Maybe remove that point from your post-processing.

And then you can review what was wrong on this particular point. (Try another initial mesh or an initial inlet velocity)

If you use Model Fuel Library file, here is what you should get.

MFL file is located in:

But you could use other 

You can use also Multicomponent transport as well for H2-air. But the difference in result is no huge. (But don't use Unity Lewis number)

On some very lean point or very point, I also change initial velocity

Note that I'm using SI units in preferences.

In my case the only failed point was ER=0.2 (lean limit)

(In your case, unselect ER=1.1 before launching post-processing tool. I guess the rest maybe ok but grid point looks very very small.)

Hopefully this will help.

The recommendations are general but tuning is mechanism dependent. Usually it is harder to converge close to flammability limit (lean limit or rich limit).

 

[shezanghamat]

Dear Didier Bessette

I tried with reduced inlet velocity as per your suggestion. It worked. Thank you so much for your time and quick response.