[mesa-users] White dwarf crystallization

[Dean Townsley]

Hi Reggie,

I believe the answer to your question is unfortunately yes this is 
normal behavior.  The entropy discontinuity at the phase transition is 
not treated in any special way, so the the integrator can "stumble" on 
it quite badly and do unphysical things.

There may be others out there who have tried to improve this.  If so I 
would invite them to contribute their work to the mainline code.

In terms of past references to this problem, I'm having trouble getting 
the mesa-users mailing list archives to behave as I expected.  It seems 
to be broken.

I found this message:
http://sourceforge.net/p/mesa/mailman/message/31696333/

But I can't seem to find my message that it refers to, which is the same 
issue you are talking about, nor the rest of the thread.

In the discusion at the end of 2013, this was all mixed up with platform 
dependencies.  Those got worked out by a sort of miracle worked by Bill, 
but not the crystallization issue.  Bill made some changes to allow the 
code to step over the problem, but I don't think that really fixes it 
since I don't believe it will converge.


Reggie

On 03/29/2015 09:54 PM, Reggie Caginalp wrote:
> Dear MESA-users,
>
> I am studying white dwarf cooling using MESA-r7184, MESASDK version 
> 4.10.0 20140710 on OS 10.9.4 Mavericks.
>
> Using the wd_cool_0.6M test suite case, with the default model, gives 
> oscillations in the value of the Coulomb parameter Gamma at the center 
> as in gamma_def.pdf. Furthermore, the gravitational energy oscillates 
> rapidly around the onset of crystallization, as in def_Lgrav.pdf (the 
> y-axis is Lgrav/Lsun). Attached are profiles of eps_grav 
> (epsgrav_prof_spike.pdf) and gamma (Gamma_prof_at_spike.pdf) from one 
> of the spikes. The part of the star that is just beginning to 
> crystallize experiences a huge spike in gravitational energy from 
> resettling, as compared with an eps_grav profile from well before the 
> onset of crystallization (epsgrav_prof_flat.pdf). (These last three 
> figures are from running the test suite case with a different model, 
> but using the default model gives similar results.)
>
> Is this normal behavior? If anyone knows a better implementation of 
> crystallization, I would appreciate it greatly!
>
> Thanks for your time!
>
> Sincerely,
> Reggie Caginalp
>
>
> ------------------------------------------------------------------------------
> Dive into the World of Parallel Programming The Go Parallel Website, sponsored
> by Intel and developed in partnership with Slashdot Media, is your hub for all
> things parallel software development, from weekly thought leadership blogs to
> news, videos, case studies, tutorials and more. Take a look and join the
> conversation now. http://goparallel.sourceforge.net/
>
>
> _______________________________________________
> mesa-users mailing list
> mesa-users at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/mesa-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20150330/54f671b4/attachment.html>