[mesa-users] Low mixing length alpha values -> convergence problems

[Ireland, Lewis]

Not sure if my previous email sent (having some problems)!

I am not sure where the run has convergence problems in particular. For small alpha values, model numbers can reach a few thousand before stopping and complaining about convergence.

I will give what you said a go in the meantime. Thank you.

Lewis

________________________________
From: Aaron Dotter <aaron.dotter at gmail.com>
Sent: 19 August 2015 10:37
To: Ireland, Lewis
Cc: mesa-users
Subject: Re: [mesa-users] Low mixing length alpha values -> convergence problems

Hi Lewis,

Setting aside the issue of why you want to use alpha_MLT=0.01...

Where in the run does it have convergence problems?  Near the beginning or sometime later?

Because you can create a saved model that has evolved down the Hayashi line for a while using a "normal" alpha_MLT value and then start from that instead of using create_pre_main_sequence_model.  See mesa/star/defaults/star_job.defaults:

      save_model_number = -111

         !### save_model_when_terminate
         ! save final model when a run terminates

      save_model_when_terminate = .false.

         !### save_model_filename
         ! saved model root filename

      save_model_filename = 'undefined'

and

      load_saved_model = .false.
      saved_model_name = 'undefined'

No guarantee that this will solve your problem but it is worth a try.


Aaron


On Wed, Aug 19, 2015 at 8:11 PM, Lewis Ireland <lgi201 at exeter.ac.uk<mailto:lgi201 at exeter.ac.uk>> wrote:
Hello,

I am using the work directory (tutorial folder) to create 0.5 solar mass models, all at different mixing length parameters (to see the change in R when it nears ZAMS). I have these set:

! begin with a pre-main sequence model
    create_pre_main_sequence_model = .true.

  ! starting specifications
    initial_mass = 0.5d0 ! in Msun units

mixing_length_alpha = 0.55d0

I have the following set to end the run:

  ! stop when the star nears ZAMS (Lnuc/L > 0.99)
    Lnuc_div_L_zams_limit = 0.99d0
    stop_near_zams = .true.

  ! stop when the center mass fraction of h1 drops below this limit
    xa_central_lower_limit_species(1) = 'h1'
    xa_central_lower_limit(1) = 1d-3

 I need to go to real low values of mixing length parameter, e.g. 0.1, 0.01 etc., but when I go to or below alpha = 0.5, I get:

hydro_newton_step failed to converge
       retry cnt, step, log10(dt/yr), retry_factor     1   543      -14.045150        0.500000
                                                     dt    5.6884209599392411D-07
                                     min_timestep_limit    9.9999999999999995D-07

 stopping because of convergence problems dt < min_timestep_limit


terminated evolution: convergence problems

termination code: min_timestep_limit

How can I go about using lower values of alpha without this problem? Is there another test_suite you recommend me to use that would benefit my models or get past this convergence issue?

Cheers,

Lewis

------------------------------------------------------------------------------

_______________________________________________
mesa-users mailing list
mesa-users at lists.sourceforge.net<mailto: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/20150819/5ce0bbfe/attachment.html>