[mesa-users] About defined time inteals

Robert Farmer rjfarmer at asu.edu
Thu Aug 4 12:13:04 EDT 2016


Hi
So there's no simple answer, MESA will take the largest timestep it can,
such that it is constrained by the choice of model sensitivity (things like
varcontrol_target etc). So to get bigger/smaller timesteps you
loosen/tighten the constraints on MESA, but to get timesteps at some exact
frequency is more difficult.

Two options i see:
Set max_timestep, but this will only work if its smaller than the step mesa
would otherwise take.

In your run_star_extras you could try manipulating the timestep (like how
its done in pablo's last email), but mesa may still take smaller timestep
than the one you choose. You could also use the, in extras_finish_step
http://mesa.sourceforge.net/run_star_extras.html#toc-1-4 :

s% need_to_update_history_now = .true.

condition to output information on some interval (with the need to also
manipulate the dt to get close to the chosen value, see pablo's last email)

Rob



On Thu, Aug 4, 2016 at 8:28 AM, Michal Simon <michal.simon at stonybrook.edu>
wrote:

> Greetings to all,
>
> This question is stimulated by Pablo Marchant's reply]
> to Kenny Van about setting time steps, and particularly
> his cautionary remark about units for the time steps.
>
> I have been using MESA to calculate evolutionary models for low mass
> (< ~2 Msun),  young (< 100 MY) stars, and that are not brown dwarfs.
> MESA calculates many, many models at extremely young ages, << 1 MY.
> That's fine and I can understand that MESA needs to do that to settle
> on models in quasi-static collapse.
>
> But, once a model gets to 1/10 MY age it starts getting
> interesting to me.  Is it possible to start to set a desired
> time interval then, at say 1/10 MY and to continue at steps of perhaps
> 1/20 MY until say 20 MY at which time the collapse proceeds very
> slowly and MESA can choose whatever time steps it prefers?   As you
> can imagine this would expedite making isochrones and side-step using
> MIST.
>
> Thanks in advance,
> Mike Simon
>
>
> ------------------------------------------------------------------------------
> _______________________________________________
> 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/20160804/73dad2f6/attachment.html>


More information about the Mesa-users mailing list