[mesa-users] controlling timestep using diffusion timescale

[Alina Istrate]

Dear all,

I've been studying convergence for 1.2 solar mass models including the
effect of gravitational settling (as you might recall from a previous
email). In order to limit timesteps, I ended up using
max_years_for_timestep low enough such that the main sequence evolution
does not display large jumps in Teff, caused by the coupling of Ledoux
criterion and settling. However, I would like to control the timestep
directly using a timescale for diffusion, so the code adjusts properly
at different masses, instead of needing to tune max_years_for_timestep
each time.

I saw that there is an option to reduce the timestep size for the
diffusion substeps (diffusion_dt_div_timescale). This parameter is not
used when calculating diffusion with isolve, is there a reason for this?
Also, would it be possible to have the value computed by
get_timescale(...) in star/private/mod_diffusion.f90 accesible through a
public interface, and provide a timestep limit associated to diffusion
using this timescale (something like dt_div_diffusion_timescale_limit)?

Cheers!

Alina


-- 
Alina Istrate 
PhD student


Argelander Institute fur Astronomie
Auf dem Hugel 71
D-53121 Bonn
Germany
Room 2.024