Laminar flame speed: CONVERGE vs CHEMKIN

man1 · April 24, 2020, 10:25

Hi all,

I am comparing laminar flame speed of Hydrogen calculated by CHEMKIN and CONVERGE using the same chemical kinetic mechanism from Burke et al (2012).

The figure attached shows that there is a significant difference between the flame speed calculated by two solvers under same unburned temperature and pressure conditions. The domain length and tolerances are also same both solver. In CONVERGE Newton solver is used for calculation.

Can someone explain what causes the difference in flame speed of the two solvers?

Thank you

tmalewic · May 4, 2020, 13:18

The differences in the results between CONVERGE and CHEMKIN could be that as a default above 600 K for flamespeed calculations CONVERGE CFD uses TCUT (cutoff_temp) feature.

TCUT will lower the flamespeed. TCUT is per case so it can be changed in one_d_cases.in input file. Using TCUT makes it easier for the simulation to converge especially at these higher temperatures.

From the v3.0 manual'
"Combustion temperature cutoff (K). Source term is not evaluated for
cell temperature lower than the given value."

Please try for each case have the cutoff_temp to be 0.0.

man1 · May 8, 2020, 11:37

Quote:

Originally Posted by tmalewic

The differences in the results between CONVERGE and CHEMKIN could be that as a default above 600 K for flamespeed calculations CONVERGE CFD uses TCUT (cutoff_temp) feature.

TCUT will lower the flamespeed. TCUT is per case so it can be changed in one_d_cases.in input file. Using TCUT makes it easier for the simulation to converge especially at these higher temperatures.

From the v3.0 manual'
"Combustion temperature cutoff (K). Source term is not evaluated for
cell temperature lower than the given value."

Please try for each case have the cutoff_temp to be 0.0.

Hi Tom,

Thank you very much for your explanation. I will try this method.

EDE16 · May 10, 2020, 11:47

not to hijack the original post, but is flame speed a possible output in converge? or only with certain combustion model, i.e. not SAGE?

MFGT · May 11, 2020, 01:48

You can do 1D or 0D simulations of flames speed and auto-ignition times in the chemistry tab.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Difficulties with premixed laminar methane flame in openFoam	DustExplosion	OpenFOAM	16	August 9, 2019 05:31
Laminar premixed flame with the same species speed	Gruesome Marc	Main CFD Forum	0	October 16, 2008 07:30
Looking for Laminar Flame Codes	John	Main CFD Forum	0	September 17, 2005 19:13
About the laminar jet diffusion flame length	Sheng Chen	Main CFD Forum	1	July 10, 2004 01:07
correlation for laminar flame speed	cfd-user	Main CFD Forum	0	November 28, 2003 07:27

May 4, 2020, 13:18		#2
tmalewic New Member Tom Malewicki Join Date: May 2020 Posts: 1 Rep Power: 0	The differences in the results between CONVERGE and CHEMKIN could be that as a default above 600 K for flamespeed calculations CONVERGE CFD uses TCUT (cutoff_temp) feature. TCUT will lower the flamespeed. TCUT is per case so it can be changed in one_d_cases.in input file. Using TCUT makes it easier for the simulation to converge especially at these higher temperatures. From the v3.0 manual' "Combustion temperature cutoff (K). Source term is not evaluated for cell temperature lower than the given value." Please try for each case have the cutoff_temp to be 0.0.

May 10, 2020, 11:47		#4
EDE16 Member Join Date: Jun 2016 Posts: 52 Rep Power: 9	not to hijack the original post, but is flame speed a possible output in converge? or only with certain combustion model, i.e. not SAGE?

May 11, 2020, 01:48		#5
MFGT Senior Member Tobias Join Date: May 2016 Location: Germany Posts: 264 Rep Power: 10	You can do 1D or 0D simulations of flames speed and auto-ignition times in the chemistry tab.