[mesa-users] Convection and Ledoux Criterion

[Josh Shiode]

All,

I'm trying to evolve some massive stellar models with MESA 4631 on a macbook pro (with the mesasdk).

I have use_Ledoux_criterion = .true. and alpha_semiconvection = 0.1 (see attached inlist for more details), and I'm finding some strange behavior in models post-He burning. 

1) There are zones marked as convective (mixing_type = 1) where brunt_N2 > 0, brunt_B > 0, brunt_nonB < 0. 
With Ledoux criterion and semiconvection, I would have expected these zones to return as semiconvective, since, as far as I can tell, brunt_B is used to calculate Ledoux stability.

2) There are also zones marked as non-mixed (mixing_type = 0) where brunt_N2 > 0, brunt_B > 0, brunt_nonB < 0. I would have expected semiconvection here, as mentioned above.

3) In some cases there are also zones marked as mixing_type = 1, where both components of the brunt are positive... so I would have thought it should be stable (mixing_type = 0).

Note: these are not isolated zones, they can span many zones in the grid...

I have tried to follow the chain of commands through star/private/brunt.f, mlt_info.f, and mlt/private/mlt.f, but cannot seem to track down what could be causing this behavior. 

Any help would be much appreciated!

Thanks,
Josh

P.S. You can find the inlist I used attached. I've also tried with the brunt_smoothing turned off, to no avail. A good example would be the 19th profile that this inlist spits out (tried including it before, but I think that's why my message didn't get delivered the first time).

-------------- next part --------------
A non-text attachment was scrubbed...
Name: inlist_massive
Type: application/octet-stream
Size: 8589 bytes
Desc: not available
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20121213/3f79595a/attachment.obj>
-------------- next part --------------