[Mesa-users] Relaxing angular momentum

Wed Oct 23 03:00:01 EDT 2019

Hi Pablo,

Many thanks for your explanations. It helps to understand what’s going on with the relaxation procedures.

Regarding my particular case. I’m simulating Sun solar-type stars from PMS till TAMS. I’ve implementing my routine of angular momentum loss using other_torque hook and I’m interesting in how the atmosphere abundances of different chemical elements vary depending on the parameters of my routine.

The simulations run ok but when I’m considering in details the results I see some computational artifacts (sawtooth profiles) that they don’t appear if my routine is not in place. To certain degree I’m able to eliminate them if both finer temporal and spatial resolutions are used. Even though I’m still able to detect “strange” behaviours that could have their root in the AML calculated for each cell for a particular moment in time. For that reason, I was asking if MESA offers a way for implementing gradually the AML in several steps and not in just one. In this way I could better trace better any relationship between my routine and the obtained results.

Many thanks in advance for your feedback.

Kind regards,

Roque

Obtener Outlook para iOS<https://aka.ms/o0ukef>
________________________________
De: Pablo Marchant <pamarca at gmail.com>
Enviado: Sunday, October 20, 2019 6:22:28 PM
Para: Roque Caballero Navarro <rcaballeron at hotmail.com>
Cc: Mesa-users ‎[mesa-users at lists.mesastar.org]‎ <mesa-users at lists.mesastar.org>
Asunto: Re: [Mesa-users] Relaxing angular momentum

Hi Roque,

a general thing about all the relaxation procedures is that there is no guarantee about what is the thermal state of the star after the relaxation. If you were to do this with a star in the main sequence, allowing for a relaxation timescale much longer than the thermal timescale of the star, then you would obtain a main sequence star in thermal equilibrium. This is because during the relaxation procedure mixing and composition changes from nuclear burning are turned off, but the energy generation from nuclear reactions is kept on. If you were to do the same thing with a white dwarf, it's much more uncertain since if you take a very long timescale the object would not settle into thermal equilibrium but simply go down its cooling track, making the final result sensitive to the timescale for the relaxation. In this case using a very short timescale would resemble an adiabatic process, happening much faster than energy can be transported through the interior, and should give a meaningful post-relaxation state.

So it all depends on your application, this is just stellar engineering, a method to get a specific outcome fro the code, so it depends entirely on what outcome yu're looking for. In applications where 3d models are imported for example, the entropy of the star can be relaxed as well to provide a very specific thermal state for the star.

Probably if you specify in more detail what you're trying to achieve, and what you've tried so far, I could provide additional help.

Cheers!

On Sun, Oct 20, 2019 at 3:53 AM Roque Caballero Navarro via Mesa-users <mesa-users at lists.mesastar.org<mailto:mesa-users at lists.mesastar.org>> wrote:

Dear all,

I'm aware of the existence of a number of "relax" options aimed to let some modifications be done gradually over several steps. In particular, I'm interesting in relaxing the angular momentum loss calculated in my other_torque routine.

According to MESA documentation, the following parameters are the relevant for my purpose:

relax_initial_angular_momentum = .false.
max_steps_to_relax_angular_momentum = 1000
timescale_for_relax_angular_momentum = 1d-10
max_dt_for_relax_angular_momentum = 1d-9
num_timescales_for_relax_angular_momentum = 1000
relax_angular_momentum_filename = ''

Additionally, it's mentioned that for relaxing angular momentum from current to specified over a certain amount of relaxation timescales. This is done by adding an extra torque term of the form and evolving the star without changing the composition for num_timescales_for_relax_angular_momentum times timescale_for_relax_angular_momentum.

s% extra_jdot(k) =  &
    (1d0 - exp_cr(-s% dt / (s% job% timescale_for_relax_angular_momentum*secyer))) * &
    (desired_angular_momentum(k) - s% j_rot(k)) / s% dt

I've some questions about how to modified the aforementioned parameters in a proper way, meaning that to make longer/shorter the period of time in which the angular momentum loss is applied:

  1.  If a want to extend the relaxing period, is it enough to assign to max_steps_to_relax_angular_momentum a bigger number, let's say 2000. What about num_timescales_for_relax_angular_momentum, should it be assigned the same value used in max_steps_to_relax_angular_momentum?
  2.  Which is the effect of modifying timescale_for_relax_angular_momentum and max_dt_for_relax_angular_momentum?

Basically I don't understand how to modify those parameters in the right way. I don't want just to throw there some values and try to figure out what's going on, so any hint about that will be very appreciated.

Kind regards,

Roque

_______________________________________________
mesa-users at lists.mesastar.org<mailto: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/20191023/7da73bfd/attachment.html>