freeflame (adiabatic) & Soret & Cantera 2.1

[dirk.g...@googlemail.com]

Hi,

I am run the adiabatic flame example in version 2.1 of Cantera for CH4.

In contrast to the previous version the Soret effect is not included in the example.

I tried to implement it, but got the message:  " AttributeError: 'FreeFlame' object has no attribute 'enableSoret' "

Am I doing anything wrong or does the free flame not have this attribute anymore in the current Cantera version ?


We see quite some differences in CH4 flames for the H2 and CO profiles over the temperature using the Soret effect, and this agrees well with our experimental species measurements.
We can not compare simulation and experiment in meaningful way if we neglect the Soret effect.

Greetings,
Dirk

[Ray Speth]

Dirk,

Including the Soret effect is still an option, there's just a syntax change in the new Python module. The new syntax is:

    f.soret_enabled = True

where 'f' is the 'FreeFlame' object. I've updated the migration guide for the new Python module to reflect this change.

Regards,

Ray

[dirk.g...@googlemail.com]

Hi Ray,

thanks, I guessed the call, but did not use the True/False statemenet.


I hope I am not a niusance, but there is another question which came up:

With windows 7/64 I do not get any response from Cantera  to the screen using any of the following two commands:

loglevel_mc = 5
f.solve(loglevel=loglevel_mc, refine_grid=refine_grid)
f.show_solution()

I also tried "f.show_solution" without the brackets, which shows also no response on the screen.

Greetings,
Dirk

[dirk.g...@googlemail.com]

Hi Ray,

I did run a freely propagating flame for CH4, with multicomponent transport and w as well es w/o the Soret effect.

The comparsion to results from Cantera 2.0 shows, that the soret effect is not switched on after the call:

f.transport_model = 'Multi'
f.enable_soret = True

Turning on the Soret effect does not affect the results in 2.1, while the Soret effect worked in 2.0 . The results with and without the Soret effect are identical.

Attached is pdf file that shows the results for Cantera 2.0 and 2.1.  The curve from Cantera 2.0, multicomponent transp., is identical with both results from Cantera 2.1.

Greetings,
Dirk



On Saturday, November 2, 2013 11:02:47 AM UTC+1, dirk.g...@googlemail.com wrote:

[Nick Curtis]

Dirk,
It looks to me that the problem may be that you used: 
f.enable_soret = True
instead of
f.soret_enabled = True

Nick

[dirk.g...@googlemail.com]

Nick,

stupid me, you are right. Gosh, that took me hours..

Thanks,

Dirk

On Saturday, November 2, 2013 11:02:47 AM UTC+1, dirk.g...@googlemail.com wrote: