[Mesa-users] CCSN_IIp module: infall up to what radius?

Thu Jul 3 21:54:55 UTC 2025

Yes, yes, yes! The handoff is the issue. The infall was modest before that (1700 km to 1693 km in 10^5 seconds), but just switching from Lagrangian to Riemannian hydro before excising the core fixed everything.

Thank you so much, Jared! I think it's miraculous the way you spotted that bug with so little information.

Best,
Soham
________________________________
From: Jared Goldberg <jgoldberg at flatironinstitute.org>
Sent: Thursday, July 3, 2025 5:01 PM
To: Mandal, Soham (uqa7ur) <soham at virginia.edu>
Cc: mesa-users at lists.mesastar.org <mesa-users at lists.mesastar.org>
Subject: Re: [Mesa-users] CCSN_IIp module: infall up to what radius?

What is the initial R_center? Is it falling in at all?

There can sometimes be an issue with changing from lagrangian hydro (v_flag) to riemann hydro (u_flag) in the same timestep that you excise the core. This used to be done in make_pre_ccsn_IIp in old revisions but has been taken out of *_pre_ms_to_core_collapse in newer revisions. (Re-)improving compatibility between the pre_ccsn and ccsn setups is on my agenda for when I get some time...

If the handoff is the issue and it's not infalling, possible fixes include: add a step where you take a few very small timesteps where you just switch from v_flag to u_flag, without excising the core. Sometimes simply setting okay_to_remesh = .true. in inlist_infall works. Also, it may help to change to use_Skye = .true. in &eos, if it's false.

Another option might be to do the infall in lagrangian hydro rather than riemann hydro, i.e. commenting out the following in inlist_infall:
! change_initial_v_flag = .true.
! change_initial_u_flag = .true.
! new_v_flag = .false.
! new_u_flag = .true.

Also, 400 km comes from a slightly arbitrary engineering choice that (probably Bill Paxton) deemed "close enough to the ~100 km PNS radius, but also still within the acceptable scope of MESA's eos and opacity tables", probably also motivated by old work with the Kepler code.

~Jared

On Thu, Jul 3, 2025 at 4:31 PM Mandal, Soham (uqa7ur) <soham at virginia.edu<mailto:soham at virginia.edu>> wrote:
Thanks again, Jared! You're right, R_center in my model is not going below 1600 km in the ccsn_IIp setup. I was of the opinion that I should then just replace  "center_R_lower_limit = 4d+07" with "center_R_lower_limit = 1.6d+08", but was apprehensive that this might mean I'm doing something horribly wrong!

Will go through the works you mentioned and try to figure out why the stalling happens at such a large radius. Can you please also help me understand where the choice of R_center = 400 km comes from? Thanks a lot!

Best,
Soham
________________________________
From: Jared Goldberg <jgoldberg at flatironinstitute.org<mailto:jgoldberg at flatironinstitute.org>>
Sent: Thursday, July 3, 2025 3:41 PM
To: Mandal, Soham (uqa7ur) <soham at virginia.edu<mailto:soham at virginia.edu>>
Cc: mesa-users at lists.mesastar.org<mailto:mesa-users at lists.mesastar.org> <mesa-users at lists.mesastar.org<mailto:mesa-users at lists.mesastar.org>>
Subject: Re: [Mesa-users] CCSN_IIp module: infall up to what radius?

Do we then expect the inner radius to gradually halt at some value? For example, in my model, R_center won't go below ~1600 km. How does one determine that infall has ended and now it's time to inject energy?
In *_to_core_collapse, Fe infall velocity.  In ccsn_IIp, a central radius is specified (center_R_lower_limit = 4d+07). Is your R_center not going below 1600 km in the ccsn_IIp setup?

Exploding a star in MESA is a feat of engineering as much as it is a feat of physics, if not moreso.

Some people have calculated the evolution further in GR1D and STIR and PUSH and some other frameworks, then ported back to MESA (see works by Sanjana Curtis, Brandon Barker, others). It turns out that lightcurve modeling is often not so sensitive to when you halt the infall and put in the bomb, since you're using a thermal bomb that is the asymptotic energy + the binding energy at that time, and not doing a "self-consistent" explosion anyway (this is something many people have found through experimentation, including myself, though I don't know of a great reference that shows it). For the composition and explosive burning, it matters more, but then you run into the problem that 1D explosion models don't get mixing at the proto-neutron star boundary right, the explosion isn't actually 1D, et cetera.

~Jared

On Thu, Jul 3, 2025 at 3:17 PM Mandal, Soham (uqa7ur) <soham at virginia.edu<mailto:soham at virginia.edu>> wrote:
Thanks for that prompt clarification, Jared! So if I understand correctly, MESA choking at higher core densities sets an upper limit to the Fe core infall velocity stopping condition. In that case, one eventually has to calculate the rest of the core infall in the ccsn_IIp, after excising the proto-NS core.

Do we then expect the inner radius to gradually halt at some value? For example, in my model, R_center won't go below ~1600 km. How does one determine that infall has ended and now it's time to inject energy?

Thanks,
Soham
________________________________
From: Jared Goldberg <jgoldberg at flatironinstitute.org<mailto:jgoldberg at flatironinstitute.org>>
Sent: Thursday, July 3, 2025 2:38 PM
To: Mandal, Soham (uqa7ur) <soham at virginia.edu<mailto:soham at virginia.edu>>
Cc: mesa-users at lists.mesastar.org<mailto:mesa-users at lists.mesastar.org> <mesa-users at lists.mesastar.org<mailto:mesa-users at lists.mesastar.org>>
Subject: Re: [Mesa-users] CCSN_IIp module: infall up to what radius?

The *_pre_ms_to_core_collapse models end with an Fe core infall velocity stopping condition -- to get to smaller radii, you can increase that velocity (see fe_core_infall_limit). However, MESA often chokes at higher and higher core densities, which is why the rest of the core infall is carried out after excising the proto-NS core in inlist_infall in the ccsn_IIp setup.

~Jared

On Thu, Jul 3, 2025 at 2:22 PM Mandal, Soham (uqa7ur) via Mesa-users <mesa-users at lists.mesastar.org<mailto:mesa-users at lists.mesastar.org>> wrote:
Dear all,

I am running into an issue while trying to use the CCSN_IIp test_suite problem. So the inlist_infall continues evolution of the star till the inner radius falls below 400 km. I'm not sure how to compute this value (400 km) for any arbitrary model.

For example, in a pre-core-collapse model I ran, the inner radius won't go below ~1600 km (found by trial and error). Is there a way to compute the lower threshold of the inner radius and set the center_R_lower_limit parameter to that before the test problem is run? Any insight will be greatly appreciated!

Thanks,
Soham

------------------------------------------------
Dr. Soham Mandal
VITA postdoctoral research fellow
Department of Astronomy
University of Virginia
_______________________________________________
mesa-users at lists.mesastar.org<mailto: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/20250703/48aed464/attachment.htm>