[Mesa-users] hydro_newton_step failed to accept error during calculation of 2Ms

[Francis Timmes]

hi amadeusz,

the inlist has many physical and numerical things going on at once;
pulsation, diffusion, rotation, using pressure instead of density and
the primary variable, and more. 

i commented all this stuff out to strip it down to the basics and mostly default values. 
it ran just fine to the given stopping condition and produced the expected results; 
see the attached two png files and inlist file "inlist_temp_revised".

my advice is to start with a simple inlist, like in the above example. 
then add in one physical effect or numerical choice at a time, and seek to 
understand how and why the model may change or not. this process will 
help ensure that by the time you build up to the complicated/sophisticated model 
in the original inlist, that you will know better what is going on and why.

it is perfectly ok to use MesaScript for this learning process;
see if you can have MesaScript generate the equivalent of the 
simple attached "inlist_temp_revised".

fxt





> On Mar 25, 2018, at 1:28 PM, Miszuda Amadeusz <miszuda at astro.uni.wroc.pl> wrote:
> 
> Dear Frank,
> 
> I appreciate Your quick response very much. It, with no doubt will allow me to resume my work a lot faster.
> As asked, i send You generated inlist, used later by MESA.
> 
> Sincerely,
> Amadeusz Miszuda
> <inlist_temp>
> ---------------------------------
> Miszuda Amadeusz
> Instytut Astronomiczny UWr
> ul. Kopernika 11
> 51-622 Wrocław
> 
> www.astro.uni.wroc.pl
> miszuda at astro.uni.wroc.pl
> ---------------------------------
> 
>> Wiadomość napisana przez Francis Timmes <fxt44 at mac.com> w dniu 25.03.2018, o godz. 22:09:
>> 
>> hi miszuda,
>> 
>> it appears you are using bill wolf's ruby-based MesaScript to generate the inlist(s). 
>> this is wonderful, and it would be constructive for help purposes to also send the actual 
>> (MesaScript generated) inlist that mesa is using.
>> 
>>> Additionally, sometimes in code, i get a warning: hydro_newton_step failed to accept 
>> 
>> this is within normal behavior. the warning usually means the attempted time step
>> was too large for the implicit solver to converge. mesa automatically decreases the
>> attempted time step and tries again.
>> 
>> fxt
>> 
>> 
>> 
>>> On Mar 25, 2018, at 1:37 AM, Miszuda Amadeusz <miszuda at astro.uni.wroc.pl> wrote:
>>> 
>>> Dear all MESA users,
>>> 
>>> I’m fairly new in the MESA society, currently on the beginning of my PhD in Astrophysics.
>>> As one of my first tasks, to get familiar with MESA code, was to calculate a 2 solar mass star
>>> evolution from ZAMS.
>>> 
>>> During calculations, I’ve noticed that star's position on PgPlot’s HR Diagram does not change 
>>> as it would be expected, but moves uniformly towards lower Teffs and luminosities. It stops only 
>>> at the point chosen to end calculations, in my case log_Teff_lower_limit=3.8
>>> Additionally, sometimes in code, i get a warning: hydro_newton_step failed to accept 
>>> Has anybody encountered such a problem?
>>> 
>>> I run r10108, 10.12-20180127 SDK on 10.13.3 macOS High Sierra.
>>> I attach my inlist, a screenshot of PgPlot and a calculated path.
>>> 
>>> Thank You for all the advices,
>>> Amadeusz Miszuda


-------------- next part --------------
A non-text attachment was scrubbed...
Name: trho_profile_000220.png
Type: image/png
Size: 8918 bytes
Desc: not available
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20180325/ba8e5d7e/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: hr_000220.png
Type: image/png
Size: 4312 bytes
Desc: not available
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20180325/ba8e5d7e/attachment-0001.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: inlist_temp_revised
Type: application/octet-stream
Size: 4420 bytes
Desc: not available
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20180325/ba8e5d7e/attachment.obj>