[Mesa-users] 2.0 M_sun star did not reach the AGB phase

[Francis Timmes]

log_L_lower_limit always refers to a stopping condition in any inlist:

https://docs.mesastar.org/en/r15140/reference/controls.html#log-l-lower-limit

evolving a model completely through the agb phase can be challenging
and usually requires, as mentioned below, much smaller timesteps than
the default set of timestep controllers that are apparently being used.
non-convergence of the solver can take many forms, including the form
indicated in your terminal output.

fxt

p.s. - in the future, please use plain text copy & paste.
its often clearer and always more bandwidth friendly than 
sending everyone on mesa-users a ~0.6 mb screenshot image.






> On Sep 6, 2021, at 6:58 PM, Ebrahim Hassani <ee.haji at gmail.com> wrote:
> 
> Hi dear Francis,
> 
> Thank you very much for your attention and also for your answers.
> 
> After reviewing the code several times, I could find the problem with my code. In fact, I had changed the value of the log_L_lower_limit variable in the inlist_start file from 1 to -4. After returning the value of log_L_lower_limit to 1, the code worked correctly and I  could evolve the 1_M_Sun star until the white dwarf phase. Honestly speaking, I don't know what the value of log_L_lower_limit determines in the inlist_start file (but in the inlist_to_wd file, the value of log_L_lower_limit determines the stopping condition for the star).
> 
> After correcting the above-mentioned mistake, I could run the code for stars with 1.0-2.0 solar masses successfully (and until the white dwarf phase). But when I increased the mass of the star to 2.5 solar mass (or higher masses e.g. 3.0 M_Sun), I encountered a new problem. The attached file is the screenshot of the final error message that I had got for the 2.5 solar-mass star. The code produced the same problem for other higher mass stars (e.g. 3.0 M_Sun) too. This time stars are evolving until the AGB phase, but they don't evolve further, and then they don't reach the white dwarf phase.
> 
> May I ask again what might be the reason for the new error message? It is worth stressing that, this problem happens just for high-mass stars (2.5-3.0 solar masses) and not for low-mass ones (1.0-2.0 solar masses). My guess is, this problem happens because of the new source of energy that we defined in the src/run_star_extras.f90 file and through dark matter annihilation, but I am not sure about this.
> 
> I would really appreciate any of your feedback and any comments
> 
> 
> All the best
> Ebrahim
> 
> 
> 
> On Mon, Sep 6, 2021 at 5:34 AM Francis Timmes <fxt44 at mac.com> wrote:
> hi ebrahim,
> 
> your other_mesh_delta_coeff_factor routine is sufficiently complicated
> that you may need to do additional debugging. in addition, there is an 
> apparent reliance on the default timestep controls; tightening those
> should be considered. 
> 
> if the model works to your satisfaction for a mass of 1.0, then i suggest 
> making some plots of what the model is doing and also noting what 
> physical variable (central temperature, central density, etc) controls the timesteps. 
> then try a model with a smaller mass change, say, 1.1 msun. what changes and why? 
> repeat this analysis for a mass of 1.2 and so on. perhaps by the time you get 
> to 1.5 msun or so it will be clearer why your model is stalling. also, note that
> helium ignition is rather different for 1.0 msun that it is for 2 msun ...
> 
> happy hunting!
> 
> fxt
> 
> 
> 
> 
> 
> 
> > On Aug 29, 2021, at 9:46 PM, Ebrahim Hassani via Mesa-users <mesa-users at lists.mesastar.org> wrote:
> > 
> > Dear mesa-users
> > 
> > I used 1M_pre_ms_to_wd star in mesa-r15140/star/test_suite to consider dark matter effects in stars. I could edit and run the code successfully. But when I changed the mass of the star from 1.0 M_sun to 2.0 M_sun the code again worked without any problem. But for the 2.0 M_sun star, the star did not finish the run. It took about 10^5 steps until now and it did not reach the AGB phase or later phases. In addition, until 10^5 step, the mass of the star did not reduce due to e.g. stellar winds, etc.
> > 
> > The code of the 2.0 M_sun star is available at (see the attached files too):
> > https://drive.google.com/file/d/1FpFxApMOeZ6tK2Pzr3XbkAk1ZKqhsQhm/view?usp=sharing
> > 
> > I used this code (in jupyter notebook) to plot the H-R diagram (see the attached files too):
> > https://drive.google.com/file/d/1iDOTf6L1sJXUkeO3cupHQOas-RIaRnwc/view?usp=sharing
> > 
> > It is worth mentioning that I mostly edited the run_star_extras.f90 file in the src folder to consider dark matter effects. I also uncommented "max_model_number = ..." command in inlist files. I also changed the amount of "log_L_lower_limit" values in inlist files.
> > 
> > May I ask what might be the possible problem of the code that it did not reach to AGB phases? And how can I solve this problem?
> > 
> > I would greatly appreciate any of your comments.
> > 
> > 
> > Best regards,
> > Ebrahim Hassani
> > <Cap_Rate.ipynb><2M_pre_ms_to_wd_Mwimp_1D2_Rho_0.zip>_______________________________________________
> > mesa-users at lists.mesastar.org
> > https://lists.mesastar.org/mailman/listinfo/mesa-users
> > 
> 
> <ScreenShoot_Ubuntut.png>