[Mesa-users] Building He layer in ns_h test suite problem

Tue Dec 5 22:59:44 UTC 2023

Hi Frank and David,

It looks like the change came about back when we were working on doing a better job building the test suite to be able to include all of the necessary steps to regenerate starting models from scratch. Prior to that, I believe ns_h had a starting model that already had an He envelope. But in the 14000s we were working on starting from ns models produced by the make_env test case, which makes an iron ns envelope. The commit message where Bill added the extra step to add an He envelope is here:

------------------------------------------------------------------------
r14754 | bpaxton | 2020-10-31 17:27:28 -0400 (Sat, 31 Oct 2020) | 2 lines

ns_h needs to have a thin layer of He4 on top of the Fe56 before start adding H1

So I think the choice of accretion rate is pure stellar engineering, and you're free to adjust it if you find something that works better for your models. 

However, I'll also note that testhub shows a dramatic decrease in the number of retries for this particular test case in July on the development version after the most recent public release (r23.05.1): https://testhub.mesastar.org/main/test_cases/star/ns_h?utf8=%E2%9C%93&history_type=show_instances&start_date=2023-07-01&end_date=2023-07-22&computers=cannon_conroy_intel&status=&commit=Search <https://testhub.mesastar.org/main/test_cases/star/ns_h?utf8=%E2%9C%93&history_type=show_instances&start_date=2023-07-01&end_date=2023-07-22&computers=cannon_conroy_intel&status=&commit=Search>

It turns out that the retries were caused by an EOS blending issue that can be alleviated by adopting the EOS settings in this commit:
https://github.com/MESAHub/mesa/commit/ab531f1015e9c978e8765b7731eccc6fda5928cb <https://github.com/MESAHub/mesa/commit/ab531f1015e9c978e8765b7731eccc6fda5928cb>

Cheers,
Evan

> On Dec 5, 2023, at 4:21 PM, David Neto via Mesa-users <mesa-users at lists.mesastar.org> wrote:
> 
> Hello Frank,
> 
> Thanks for the update. I just wanted to make sure I wasn't missing something.
> 
> Cheers,
> David
> 
> On Tue, Dec 5, 2023, 3:16 PM Frank Timmes <fxtimmes at gmail.com <mailto:fxtimmes at gmail.com>> wrote:
> hi david,
> 
> interesting observation.
> 
> this test suite case dates from ~2012 and version 4161 when mdot was 1e-11 msun/yr with a 0.75 h, 0.24 he composition. by version 12778, mdot was 1.7e-11 msun/yr with the same composition. starting with version 15140 the single inlist was split into multiple inlists with the first being the buildup of a he envelope with  mdot = 1e-9 msun/yr.
> 
> i cannot locate or remember a reason for this last value of mdot, but maybe someone can correct me. i doubt it was to stress test parts of mesa. 
> 
> fxt
> 
> 
> 
> 
> > On Dec 5, 2023, at 1:30 PM, David Neto via Mesa-users <mesa-users at lists.mesastar.org <mailto:mesa-users at lists.mesastar.org>> wrote:
> > 
> > Hello all,
> > 
> > In the ns_h test suite problem during the first part of the simulation (inlist_add_he_layer), was there a specific reason or motivation for the accretion rate used during this part of the simulation? With the default values, Mdot is 1E-9 Msun/yr, and you simulate for 0.5 seconds, which gives lots of retries and a fairly long compute time when run. If one cuts mdot in half and doubles the simulation time (so, building up the same outer helium layer mass), the number of retries and computation time drops dramatically. This is all in preparation for the second step, testing steady hydrogen burning.
> > 
> > I was just wondering if this choice of parameters in the test suite was to "stress test" parts of MESA or if there was another astrophysical reason for the choice of mdot and simulation time.
> > 
> > For reference, this is using MESA version 23.05.1 and MESA SDK version 22.6.1.
> > 
> > Cheers,
> > David
> > 
> > 
> > _______________________________________________
> > mesa-users at lists.mesastar.org <mailto:mesa-users at lists.mesastar.org>
> > https://nam04.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.mesastar.org%2Fmailman%2Flistinfo%2Fmesa-users&data=05%7C02%7Cdneto2%40groute.uic.edu%7C99d198890bd4419d878f08dbf5d78327%7Ce202cd477a564baa99e3e3b71a7c77dd%7C0%7C0%7C638374078195452444%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=sE7brap%2FzzU115tZodzvUdOjnPEslqVgdUueoQNP9cM%3D&reserved=0 <https://nam04.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.mesastar.org%2Fmailman%2Flistinfo%2Fmesa-users&data=05%7C02%7Cdneto2%40groute.uic.edu%7C99d198890bd4419d878f08dbf5d78327%7Ce202cd477a564baa99e3e3b71a7c77dd%7C0%7C0%7C638374078195452444%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=sE7brap%2FzzU115tZodzvUdOjnPEslqVgdUueoQNP9cM%3D&reserved=0>
> > 
> 
> _______________________________________________
> mesa-users at lists.mesastar.org <mailto:mesa-users at lists.mesastar.org>
> https://lists.mesastar.org/mailman/listinfo/mesa-users <https://lists.mesastar.org/mailman/listinfo/mesa-users>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20231205/08cc6f32/attachment.htm>