[Mesa-users] Writing to history.data during relax_mass method

[Pablo Marchant]

On Thu, Oct 4, 2018 at 12:21 PM Bill Paxton <paxton at kitp.ucsb.edu> wrote:

>
> > On Oct 4, 2018, at 9:51 AM, Eoin Farrell <efarrel4 at tcd.ie> wrote:
> >
>
> > (I plan to use the relax_mass method in stars with a more complex
> internal structure than ZAMS)
>
> do not do that if you care about the results.   do not even do ZAMS models
> that way.
>
> do the adjust mass during the fully convective early stage of pre-ms long
> before the ZAMS structure is determined.
>
> (that is almost always the only safe time to do other major changes such
> as adjusting composition)
>
> (if you want an elephant, build an elephant.  don’t try to relax a giraffe
> into an elephant)
>
> if you are only changing the mass over a fairly narrow range, use create
> pre ms to build a prototype of the “typical” mass and save it while it is
> still fully convective.   then use that as your starting model for the runs
> that will do adjust mass at the start of the run when still fully
> convective.  that just saves the overhead of doing create prems for each of
> the later runs.  as a check, compare the case with the largest change in
> mass done is this manner to what you get if you start with that mass at the
> beginning with create pre-ms.  decide if there are signficant differences.
>

+1 to this. Almost all relax routines will produce final models which could
be unphysical. There's no guarantee on the physical time the relaxation
will take, so you cannot say the final model will be in thermal
equilibrium. In many situations I perform a relax and the let the model
evolve while removing changes in composition (mix_factor = 0d0 and
dxdt_nuc_factor = 0d0) so the model can reach a configuration in thermal
equilibrium (though in certain situations this might also be unphysical).
If you want to fix the state of the star after relaxation, you can use the
relax_entropy option, but this requires that you have an idea of what the
actual final configuration is.

If you know the timescale in which the mass changes then you should model
the change in mass as an actual accretion process. This will ensure that if
the properties inside the star change too much its actually a realistic
physical change and not a fluke from relaxation.

Cheers


> cheers
> bill
>
>
>
>
>
>
> >
> > Cheers,
> > Eoin
> >
> >
> >> On 4 Oct 2018, at 17:36, Francis Timmes <fxt44 at mac.com> wrote:
> >>
> >> hi eoin,
> >>
> >> what is your goal that appears to require a relaxation to write history
> and profile files?
> >> in several respects a relaxation is usually a quasi, not a proper,
> evolution ...
> >>
> >>
> >> fxt
> >>
> >>
> >>
> >>
> >>
> >>
> >>> On Oct 4, 2018, at 1:35 AM, Eoin Farrell <efarrel4 at tcd.ie> wrote:
> >>>
> >>> Dear Mesa-users,
> >>>
> >>> I am trying to use the relax mass method to add mass to the envelope
> of a 15 Msun star. This works fine, however during the relax_mass method it
> does not write to the history.data file or the profile data files. Once the
> relax_mass method ends and the normal evolution begins it begins to write
> to the history and profile files.
> >>>
> >>> Among other things, I have tried looking through the relax.f90 routine
> and modifying
> >>>
> >>> s% do_history_file = .false.
> >>> s% write_profiles_flag = .false.
> >>>
> >>> to true (line 3722) but this doesn't seem to work. I am using r10398
> MESA. I have attached my inlist files.
> >>>
> >>> Is there a way to get MESA to record what is happening to the star
> during the relax_mass method in the history.data and profile data files?
> >>>
> >>> Cheers,
> >>> Eoin
> >>>
> >>>
> >>> <inlist><inlist_project>_______________________________________________
> >>> mesa-users at lists.mesastar.org
> >>> https://lists.mesastar.org/mailman/listinfo/mesa-users
> >>>
> >>
> >
> > _______________________________________________
> > mesa-users at lists.mesastar.org
> > https://lists.mesastar.org/mailman/listinfo/mesa-users
> >
>
> _______________________________________________
> mesa-users at lists.mesastar.org
> https://lists.mesastar.org/mailman/listinfo/mesa-users
>
>

-- 
Pablo Marchant Campos
M.Sc on Astrophysics, Universidad Católica de Chile
PhD on Astrophysics, Argelander-Institut für Astronomie, Universität Bonn
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20181004/59be47ab/attachment.html>