[mesa-users] testing diffusion for 1.2 Msun with Ledoux criterion

Bill Paxton paxton at kitp.ucsb.edu
Wed Jan 28 14:18:01 EST 2015


Great news!   

Good luck with the next step in your project.

Bill




On Jan 28, 2015, at 11:13 AM, Alina Istrate wrote:

> Hi all, 
> 
> After trying a lot of parameters, I am really happy to let you know that the problem was solved by  constraining  max_years_for_timestep =1d6 (I tried everything else but this parameter so the  credit goes to Pablo! ). 
> Now I can move on to the next issue :) 
> 
>  Thanks a lot for your help and suggestions and especially to Bill for his patience :)
> 
> 
> Cheers, 
> 
> Alina 
> 
> 
> 
> 
> 
> 
> On 01/25/2015 08:05 PM, Bill Paxton wrote:
>> Hi Alina,
>> 
>> Try things!    Run the bad cases with various changes to see what effect they have.   Make plots of logdq to check the grid resolution; try using the various controls for mesh refinement to get more resolution around the convection boundary.   Doesn't look like diffusion is a part of this -- i.e., the problems happen with or without it.  So we can eliminate diffusion for now and focus on semiconvection.  Same for overshooting -- if it isn't a necessary prerequisite for the problem, turn it off for now.   Try different values for alpha semiconvection --- as it approaches 1 the results should approach Schw.    Do your problems go away for large alpha?    Try to get 
>> 
>> Good luck,
>> Bill
>> 
>> 
>> here are some controls that you might try --- no promises that they'll help.   just more options you need to explore.   check controls.defaults in star/defaults to get more info.    I notice that you have set 'xtra_coef_a_l_nb_czb' --- check the documentation again to see if that's really what you want for these tests.  recall what 'nb' means in the name.
>> 
>> 
>>       limit_mixing_length_by_dist_to_bdy = 1
>>       
>>       conv_bdy_mix_softening_f  = 0.001
>>       conv_bdy_mix_softening_f0 = 0.003
>>       conv_bdy_mix_softening_min_D_mix = 1d-3
>> 
>>       xtra_dist_a_u_sczb = 2
>>       xtra_coef_a_u_sczb = 1d-4
>>       xtra_dist_b_u_sczb = 2
>>       xtra_coef_b_u_sczb = 1d-4
>>       
>>       xtra_dist_a_l_sczb = 2
>>       xtra_coef_a_l_sczb = 1d-4
>>       xtra_dist_b_l_sczb = 2
>>       xtra_coef_b_l_sczb = 1d-4
>> 
>>       xtra_dist_a_u_hb_czb = 2
>>       xtra_coef_a_u_hb_czb = 1d-4
>>       xtra_dist_b_u_hb_czb = 2
>>       xtra_coef_b_u_hb_czb = 1d-4
>> 
>> 
>> 
>> 
>> 
>> On Jan 25, 2015, at 4:33 AM, Alina Istrate wrote:
>> 
>>> Hi Bill, 
>>> 
>>> 
>>> 
>>> thanks a lot for your answer and sorry for answering a bit late but I was not feeling good the last few days. 
>>> 
>>> I attach to this email a plot showing the main sequence for 1.2 Msun  with different parameters: diffusion with Ledoux, diffusion with Schwarzschild, no diffusion (both Ledoux and Schwarzschild) and diffusion with Ledoux and overshooting as you mentioned. 
>>> 
>>> Just to make it a bit clear:  the track with "diffusion with Ledoux" had this parameters :    
>>>    
>>> 
>>>       overshoot_f_above_burn_h = 0.014    (this is the default value from the test suite, used in all the tracks except the red one - step overshooting)
>>> 
>>>       mixing_length_alpha = 1.6
>>>       use_Ledoux_criterion = .true.
>>>       alpha_semiconvection = 1d-3
>>>       semiconvection_option = 'Langer_85'
>>>       varcontrol_target = 1d-5
>>> 
>>> 
>>> 
>>> and the track with diffusion with Ledoux, step  overshooting (overshooting of 0.1 scale heights)
>>> 
>>>       step_overshoot_f_above_nonburn = 0.0
>>>       step_overshoot_f_below_nonburn = 0.0
>>>       step_overshoot_f_above_burn_h = 0.15
>>>       step_overshoot_f_below_burn_h = 0.0
>>>       overshoot_f0_above_burn_h = 0.05
>>>       overshoot_f0_below_burn_h = 0.0
>>>       step_overshoot_D0_coeff = 1.0
>>> 
>>>       mixing_length_alpha = 1.6
>>>       use_Ledoux_criterion = .true.
>>>       alpha_semiconvection = 1d-3
>>>       semiconvection_option = 'Langer_85'
>>>       varcontrol_target = 1d-5
>>> 
>>> 
>>> As you can see the zigzag behaviour happens just when Ledoux +diffusion  is used.  Again, the zigzag doesn't occur for 1.5Msun.  I am reading now the article suggested by Ehsan to get a better understanding of using convection criteria.  
>>>  
>>> 
>>> 
>>> 
>>> Cheers, 
>>> Alina 
>>> 
>>> 
>>> 
>>> 
>>> 
>>>  
>>> On 01/21/2015 06:40 PM, Bill Paxton wrote:
>>>> Hi Alina,
>>>> 
>>>> Here's one quick thing for you to try that I noticed looking at your inlist.
>>>> 
>>>> This is a common problem --- so it must be my fault for such poor documentation!
>>>> 
>>>> Overshooting works by taking the diffusion mixing coefficient at the edge of the convection zone and extending it beyond the zone.
>>>> 
>>>> But -- and here's the issue -- at the exact edge of the zone the mixing coefficient goes to 0.  So we don't want that.
>>>> 
>>>> Instead we want the value of the mixing coeff NEAR the edge, but not AT the edge.  
>>>> 
>>>> The "f0" parameter determines the exact meaning of "near" for this.  It tells the code how far back into the zone to go in terms of scale height.
>>>> 
>>>> The overshooting actually begins at the location determined by f0 back into the convection zone rather than at the edge where the diffusion coeff is ill-defined.   
>>>> 
>>>> So, for example, if you want overshooting of 0.2 scale heights beyond the normal edge, you might want to back up 0.05 scale heights to get the diffusion coeff from near the edge and then go out by 0.25 scale heights from there to reach 0.2 Hp beyond the old boundary.
>>>> 
>>>> In the inlist this would mean setting the "f0" to 0.05 and the "f" to 0.25 --
>>>> 
>>>> But it seems that people naturally just set f to 0.23 and f0 to 0.0 -- now can you see that we might get problems from this?
>>>> 
>>>> Try making these changes in your inlist and try things again.
>>>> 
>>>> Since mesa/star is maximally perverse, it probably won't get any better for your particular problem so we'll have to keep looking.
>>>> 
>>>> But it is a good idea to make this change before we try other things.
>>>> 
>>>> Let us know how it goes.
>>>> 
>>>> BTW: concerning the bug, please tell Pablo he has my permission to find and fix it.  ;)
>>>> 
>>>> Alternatively, if you don't want to wait for Pablo, comment out these lines in star/private/diffusion.f90
>>>> 
>>>>             if (ierr /= 0) then
>>>>                write(*,*) 'solver returned ierr /= 0', idid
>>>>                stop 1
>>>>             end if
>>>> 
>>>> -Bill
>>>> 
>>>> 
>>>> 
>>>> 
>>>> 
>>>> On Jan 20, 2015, at 5:52 AM, Alina Istrate wrote:
>>>> 
>>>>> Hi Bill, 
>>>>> 
>>>>> I tried Rob's suggestions and the only  difference made was by varcontrol_target  set to 1d-6 , the "zigzag" behaviour  is still there but at a much smaller scale. 
>>>>> 
>>>>> Also, semiconvection  is on . 
>>>>> 
>>>>> Moreover I encountered a strange error  and  my office mate Pablo Marchant  thinks is a bug so I listed it here (the error seems to come from this module  num/private/mod_dopri5.f). 
>>>>> 
>>>>> 
>>>>>  ################
>>>>> STOP 1
>>>>>  exit at x=        0.1283E+14
>>>>>   more than nmax =        1000 steps are needed
>>>>>  solver returned ierr /= 0          -2
>>>>> 
>>>>> 
>>>>> I  attached the  slightly modified inlist from test_suite/1.5M_with_diffusion   (MESA version 7184) 
>>>>>  
>>>>> Thanks a lot for your help!
>>>>> 
>>>>> Cheers, 
>>>>> 
>>>>> 
>>>>> Alina 
>>>>> 
>>>>> 
>>>>> 
>>>>> 
>>>>> On 01/19/2015 06:25 PM, Bill Paxton wrote:
>>>>>> Hi Alina,
>>>>>> 
>>>>>> Rob's suggestions are good ones.  There are also ways to increase resolution selectively near convective boundaries as an alternative to using mesh_delta_coeff to get a general increase in resolution throughout the model.
>>>>>> 
>>>>>> Did you turn on semiconvection along with Ledoux?   Pure Ledoux without semiconvection has similar problems to pure Schw without overshooting.  It produces sharp changes in composition that can give numerical instabilities.
>>>>>> 
>>>>>> If you'd send your inlist's etc, I'd be happy to take a look.
>>>>>> 
>>>>>> Cheers,
>>>>>> Bill
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> On Jan 19, 2015, at 8:35 AM, Alina Istrate wrote:
>>>>>> 
>>>>>>> Dear all,
>>>>>>> 
>>>>>>> 
>>>>>>> I am running the test suite 1.5M_with_diffusion with a 1.2 Msun  and I
>>>>>>> noticed something strange.   When including the Ledoux criterion for
>>>>>>> convection the HR diagram during the first 2 Gyr  (I stop the run then)
>>>>>>> looks really strange,  showing a strong "zigzag" behaviour (this does
>>>>>>> not happen for 1.5 Msun).  Did anyone else encounter this or has an idea
>>>>>>> what produces it?
>>>>>>> I attach here the inlist  used as well as an HR diagram.
>>>>>>> 
>>>>>>> 
>>>>>>> 
>>>>>>> Cheers,
>>>>>>> 
>>>>>>> Alina
>>>>>>> 
>>>>>>> 
>>>>>>> p.s  Mesa version  used is 7184
>>>>>>> 
>>>>>>> 
>>>>>>> 
>>>>>>> 
>>>>>>> 
>>>>>>> -- 
>>>>>>> Alina Istrate 
>>>>>>> PhD student
>>>>>>> 
>>>>>>> 
>>>>>>> Argelander Institute fur Astronomie
>>>>>>> Auf dem Hugel 71
>>>>>>> D-53121 Bonn
>>>>>>> Germany
>>>>>>> Room 2.024
>>>>>>> 
>>>>>>> <inlist_1.5M_with_diffusion><HR_1.2_zig_zag.1.2_zig_zag.pdf>------------------------------------------------------------------------------
>>>>>>> New Year. New Location. New Benefits. New Data Center in Ashburn, VA.
>>>>>>> GigeNET is offering a free month of service with a new server in Ashburn.
>>>>>>> Choose from 2 high performing configs, both with 100TB of bandwidth.
>>>>>>> Higher redundancy.Lower latency.Increased capacity.Completely compliant.
>>>>>>> http://p.sf.net/sfu/gigenet_______________________________________________
>>>>>>> mesa-users mailing list
>>>>>>> mesa-users at lists.sourceforge.net
>>>>>>> https://lists.sourceforge.net/lists/listinfo/mesa-users
>>>>> 
>>>>> -- 
>>>>> Alina Istrate 
>>>>> PhD student
>>>>> 
>>>>> 
>>>>> Argelander Institute fur Astronomie
>>>>> Auf dem Hugel 71
>>>>> D-53121 Bonn
>>>>> Germany
>>>>> Room 2.024
>>>>> <inlist_1.5M_with_diffusion>
>>>> 
>>> 
>>> -- 
>>> Alina Istrate 
>>> PhD student
>>> 
>>> 
>>> Argelander Institute fur Astronomie
>>> Auf dem Hugel 71
>>> D-53121 Bonn
>>> Germany
>>> Room 2.024
>>> <inlist_1.5M_with_diffusion><HR_comparison.pdf>
>> 
> 
> -- 
> Alina Istrate 
> PhD student
> 
> 
> Argelander Institute fur Astronomie
> Auf dem Hugel 71
> D-53121 Bonn
> Germany
> Room 2.024

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20150128/16ab3540/attachment.html>


More information about the Mesa-users mailing list