[Mesa-users] Failure during star_relax_entropy

[Rob Farmer]

Hi,
Yes equ(i_dlnE_dt,k) is the predicted change in energy in cell k. Without
seeing your code, the only thing I can think of is the ppisn code does
several net changes, so I would check that your eos code can handle
different composition vectors and isn't assuming some isotope is/isn't
there.

In your eos call could also check all the results you've computed to make
sure your not returning a bad number with is_bad(x) from utils_lib.

One option is to put -ffpe-trap=invalid,overflow,zero in FCbasic in
utils/makefile_header to track floating point exceptions. Then do a clean
recompile of $MESA_DIR and your work folder. This can help flag when the
error occurs, but you may run into many other issues first before you find
the one you care about.

Rob


On Wed, 26 May 2021 at 23:36, Jeremy Sakstein <sakstein at hawaii.edu> wrote:

> Hi everyone,
>
> I am trying to run a modified version of the PPISN test_suite for 30-40 [image:
> M_\odot] models in version 12778.
>
> The models evolve fine until the pulsations finish and the relaxation
> routines are called. The models evolve through star_relax_composition
> perfectly but then fail during star_relax_entropy.
>
> The quantity "average relative error" hovers around 0.5 or so and drives
> the time-step to small values, causing the code to terminate.
>
> I have been trying to figure out what is causing this, but so far I have
> had no luck. Does anyone have any suggestions for what I can do to try to
> figure out the cause?
>
> Often, the process terminates with the error:
>
> do1_energy_eqn s% equ(i_dlnE_dt,k)        3179                       NaN
>
> I traced this back to $MESA_DIR/star/private/hydro_equ_l.f90, but after
> tracing back all of the calls, I am still not sure what this routine does,
> what s% equ(i_dlnE_dt,k) is, or why it would end up as NaN. I think that s%
> equ(i_dlnE_dt,k) may be dlnE_dt in cell k, which may explain why this is
> the problematic variable since I am changing the EOS.
>
> I was hoping someone would enlighten me/suggest some tests I can run to
> try to figure out what is causing this behavior.
>
> I am attaching an example output in case it is useful. I can also send my
> inlists and run_star_extras but they require a small change to one private
> file ($MESA_DIR/star_data/public/star_data_def.inc) to add more controls
> and the models take a long time to reach this point running on 16 cores so
> I figured I would see if anyone had any suggestions for me to try myself
> first.
>
> Cheers,
>
> Jeremy
>
>
>
> --
> Jeremy Sakstein
> Assistant Professor of Theoretical Physics
> Department of Physics & Astronomy
> University of Hawaiʻi at Mānoa
> Watanabe Hall
> 2505 Correa Road
> Honolulu, HI, 96822, USA
> http://www.jeremysakstein.com
> _______________________________________________
> mesa-users at lists.mesastar.org
> https://lists.mesastar.org/mailman/listinfo/mesa-users
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20210527/a558b08e/attachment.htm>