[Mesa-users] Difference between "setting" and "relaxing" rotation

[Pablo Marchant]

Maybe the relaxation routines are capturing the artificial changes as a
lack of energy conservation. The artificial changes being made are of
course not meant to preserve energy, so that is not indicative of an issue.
Perhaps you can try setting the error to zero at the start

set_initial_cumulative_energy_error = .true.
new_cumulative_energy_error = 0d0

If you check the order in which the operations are done
in job/run_star_support.f90, you can see that the relaxation is done first,
and then the energy error would be set to zero.

One extra thing you should be careful about, can be different depending on
settings such as the number of steps. Since changes are made in a fixed
amount of steps, you cannot ensure for instance that the resulting model is
in thermal equilibrium. So it could be the case that using different number
of steps the actual state of the star is slightly different in each case
after relaxation and that is the cause of the discrepancy.

Cheers

On Wed, Jul 17, 2019 at 7:25 PM Nicholas J Nelson via Mesa-users <
mesa-users at lists.mesastar.org> wrote:

> Okay, this has led to a follow-up question -- we're now using
> relax_surface_rotation_v to get a ZAMS model with a specified surface
> rotational velocity. If we use the default number of relaxation steps (100)
> we get a "rel_E_err" error after the relaxation has finished and evolution
> begins. If we decrease the number of relaxation steps to 25 that goes away.
> Is there something else we should be changing along with
> num_steps_to_relax_rotation?
>
> The inlist we're using is attached.
>
> Thanks,
> Nick
>
> Nicholas J. Nelson
>
> Assistant Professor of Physics
>
> Faculty Adviser, Society of Physics Students
>
> California State University, Chico
>
> njnelson at csuchico.edu
> ------------------------------
> *From:* Nicholas J Nelson <njnelson at csuchico.edu>
> *Sent:* Wednesday, July 17, 2019 2:26 PM
> *To:* Francis Timmes <fxt44 at mac.com>
> *Cc:* mesa-users at lists.mesastar.org <mesa-users at lists.mesastar.org>
> *Subject:* Re: [Mesa-users] Difference between "setting" and "relaxing"
> rotation
>
> Thanks, Frank and Pablo. I figured someone (or some-two) would have a
> quick answer for that.
>
> Nicholas J. Nelson
>
> Assistant Professor of Physics
>
> Faculty Adviser, Society of Physics Students
>
> California State University, Chico
>
> njnelson at csuchico.edu
> ------------------------------
> *From:* Francis Timmes <fxt44 at mac.com>
> *Sent:* Wednesday, July 17, 2019 1:45 PM
> *To:* Nicholas J Nelson <njnelson at csuchico.edu>
> *Cc:* Frank Timmes <fxt44 at mac.com>; mesa-users at lists.mesastar.org <
> mesa-users at lists.mesastar.org>
> *Subject:* Re: [Mesa-users] Difference between "setting" and "relaxing"
> rotation
>
> some modifications are best done gradually over several steps in a pseudo
> evolution mode.
> these operations have “relax” in their names.
>
> many of these operations have an alternative, with “set” in name, that
> make the desired change all at once.
> the “set” version is fine if a model can converge the modifications, but
> for larger changes where
> that’s not possible, one will want to use “relax” instead.
>
> here is an example. say i have a white dwarf and i want to modify the
> interior composition.
> if i try to "set" a large composition change, the model will not converge.
> if instead i "relax" the composition, the model will usually proceed just
> fine.
>
> clearer?
>
> fxt
>
>
>
>
>
> > On Jul 17, 2019, at 1:32 PM, Nicholas J Nelson via Mesa-users <
> mesa-users at lists.mesastar.org> wrote:
> >
> > I'm working on a project that is looking at the effects of rotation on
> main sequence evolution and I've run into a question I can't seem to answer
> from the documentation. There appear to be two ways to activate rotation in
> Mesa -- namely one can use the group of inputs that begin with 'set' (e.g.,
> set_omega, set_omega_step_limit) or one can use another group of inputs
> that begin with 'relax' (e.g., relax_omega, num_steps_to_relax_rotation).
> What is the difference between setting and relaxing?
> >
> > Thanks,
> > Nick
> >
> > Nicholas J. Nelson
> > Assistant Professor of Physics
> > Faculty Adviser, Society of Physics Students
> > California State University, Chico
> > njnelson at csuchico.edu
> > _______________________________________________
> > 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/20190717/9f117831/attachment-0001.html>