[mesa-users] Tricks to create CO white dwarfs

[Aaron Dotter]

Hi Pablo,

This is not a solver adjustment, but I've had good results from limiting
the Fe-bump opacity.  This works nicely if you plot the opacity profile in
the slow part of your evolution and then adjust these factors to smooth
that profile for logT < 6.  This is definitely stellar engineering, not
stellar physics!

Aaron

         !### opacity_max

         ! limit opacities to this value (ignore this is value is < 0)

      opacity_max = -1


         !### opacity_factor

         ! opacities are multiplied by this value

      opacity_factor = 1


         !### min_logT_for_opacity_factor_off
         !### min_logT_for_opacity_factor_on  and
         !### max_logT_for_opacity_factor_on
         !### max_logT_for_opacity_factor_off

         ! temperature controls for where the `opacity_factor` is applied
         ! if, for example, you only want the opacity factor to apply
in the iron bump region
         ! you can give a logT range such as

         !     min_logT_for_opacity_factor_off = 5.2
         !     min_logT_for_opacity_factor_on = 5.3
         !     max_logT_for_opacity_factor_on = 5.7
         !     max_logT_for_opacity_factor_off = 5.8

         ! ignore these if < 0.

      min_logT_for_opacity_factor_off = -1
      min_logT_for_opacity_factor_on = -1
      max_logT_for_opacity_factor_on = -1
      max_logT_for_opacity_factor_off = -1


         ! if you need cell-by-cell control of opacity factor,
         ! set the vector "`extra_opacity_factor`" using the routine
"`other_opacity_factor`"



On Sun, May 17, 2015 at 1:22 AM, Pablo Marchant <pamarca at gmail.com> wrote:

> Pgas/P ~ 0.2 at its minimum. I am precisely looking for these "engineering
> tricks", and whether or not others have found successful ways to do this in
> other, less tricky ways. Something like playing with the solver tolerances
> in just the right way to boost a bit performance.
>
> On Sat, May 16, 2015 at 1:03 PM, Aaron Dotter <aaron.dotter at gmail.com>
> wrote:
>
>> Hi Pablo,
>>
>> What are the initial and current masses of the stars when you have
>> convergence issues?
>>
>> Do you find regions in these stars where Pgas/P << 1?
>>
>> If you're experiencing this sort of problem, there are a few
>> "engineering" tricks, like limiting max opacity or MLT++, that will get you
>> through faster.  Otherwise you need patience.
>>
>> Aaron
>>
>>
>> On Saturday, May 16, 2015, Pablo Marchant <pamarca at gmail.com> wrote:
>>
>>> Hi all. For the last days I've been playing with some systems that will
>>> end their lives as CO white dwarfs, but I always run into convergence
>>> problems close to stripping the CO core. At that point maybe it manages to
>>> go through after several hundred thousand steps, but that makes it a bit
>>> unpractical if I want to model several. So I was wondering if others with
>>> experience on the topic could share their tricks to get through this phase.
>>>
>>> I know the test_suite case make_co_wd makes use of MLT++, so in
>>> particular I'm looking to find if others have managed to get things to work
>>> with plain old MLT.
>>>
>>> Cheers!
>>>
>>> --
>>> Pablo Marchant Campos
>>> M.Sc on Astrophysics, Universidad Católica de Chile
>>> PhD student, Argelander-Institut für Astronomie
>>>
>>
>
>
> --
> Pablo Marchant Campos
> M.Sc on Astrophysics, Universidad Católica de Chile
> PhD student, Argelander-Institut für Astronomie
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20150517/1087f682/attachment.html>