[mesa-users] Is there any way to solve min_time_step convergence problem?

[Mathieu Renzo]

Hi Hyun,

You can try to play around with the resolution, i.e. change
mesh_delta_coeff and mesh_delta_coeff_for_highT for the spatial resolution
and varcontrol_target for the temporal resolution. Try increasing or
decresing these a few hundreds of steps before the code crashes. Sometimes
you can find a combination of values that works (but then would you trust
the result you get?) Also, I would not set the minimum timestep to 1d-99:
if the timesteps becomes very small there is a problem you want to solve.
Maybe you can pass through it with such extremely tiny timesteps, but
obtaining a wrong structure from MESA, which will cause problems later.

Bests,

Mathieu
On Oct 23, 2014 4:53 AM, "이현택" <diolet at unist.ac.kr> wrote:

>  Dear MESA users,
>
> I have faced on numerical problem during the simulation.
>
> When I try to make an iron core for very high mass star models(40M, 45M,
> 50M),
>
> it terminates with min time step convergence problem.
>
> For 50M case, it terminates with the output message "termination code:
> v_surf_div_v_kh_upper_limit".
>
> The inlist file is here(the only difference is the initial mass of the
> ZAMS) :
>
> &star_job
>    create_pre_main_sequence_model = .false.
>    change_initial_net = .true.
>    new_net_name = 'approx21.net'
>    pgstar_flag = .true.
>
> / !end of star_job namelist
>
> &controls
>   ! starting specifications
>   zams_filename = 'zams_z2m2_y28.data'
>       initial_mass = 40
>       initial_z = 0.02d0
>       Zbase = 0.02d0
>
>   okay_to_reduce_gradT_excess = .true.
>   varcontrol_target = 1.0d-3
>   mesh_delta_coeff = 3
>   min_timestep_limit = 1d-99
>   fe_core_infall_limit = 1d8 ! cm/sec
>   mixing_length_alpha = 2
>
>   RGB_wind_scheme = 'Dutch' !string means no RGB wind
>   AGB_wind_scheme = 'Dutch'! empty string means no AGB wind
>   RGB_to_AGB_wind_switch = 1d-4
>   Dutch_wind_eta = 1
>
> / ! end of controls namelist
>
> I'm using the mesa version 6794 and mesa SDK version 20140715.
> If you have any opinion, please notice me. It would be very helpful to me.
>
> -Sincerely yours, HYUN TAEK-
>
>
>
>
>
>
> ------------------------------------------------------------------------------
>
> _______________________________________________
> 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/20141023/5aaa3ec1/attachment.html>