Sub-Mch SN Ia records

[brettppark]

Hi everyone, just a quick clarification question regarding the unresolved HeSD issue mentioned in #1350. Looking at the code (e.g., COWD::EvolveToNextPhase()), it seems that even when the physical conditions for a sub-Chandrasekhar explosion are met and the m_HeShellDetonation flag is triggered, the code bypasses ResolveHeSD() and falls back to calling ResolveSNIa(). Does this mean that sub-Mch explosions are physically occurring in the simulation but are simply being mislabeled and recorded as standard SNIA? If so, I assume the current SN Ia output data is actually a mix of both near-Mch and sub-Mch events rather than purely near-Mch ones. Could anyone confirm if this understanding is correct? Thanks!

[ilyamandel]

Hi @brettppark ,
I believe there is indeed an issue with HeSD detonations bypassing ResolveHeSD(). @nrsegovia and @jeffriley may be able to confirm. I know @SimonStevenson also looked into this. Perhaps this should have a dedicated issue if it's still unresolved.

[jeffriley]

I think there is an issue, but my belief is a result of code inspection - I don't have the results of any test runs. I was planning to resolve any issues in the COMPAS v4 - that's probably 2-3 months away by the time we debug and validate the new version. If there are particular runs that confirm there is a problem, and we need a fix before v4, then let me know the commandline, yaml file, grid file, etc. to reproduce the problem and I'll see if we can fix it before v4.

EDIT: if this can be reproduced, maybe open a new issue as @ilyamandel suggests.

[ilyamandel]

@nrsegovia , @SimonStevenson , @brettppark -- can you reproduce this unresolved HeSD situation? If so, it would be great to create an issue; even if it's not sorted before v4, this will be good to have as a future test.

[nrsegovia]

Sorry, I have been out of the loop for a while. I will try looking into it. Are there any simple examples that can be run for testing purposes?

[ilyamandel]

I haven't looked at this at all, sorry!

[brettppark]

Yes, I have reproduced it.

I tested this with a local patch in COWD::EvolveToNextPhase(): following @nrsegovia suggestion from #1350

STELLAR_TYPE COWD::EvolveToNextPhase() {

    STELLAR_TYPE stellarType;

    if (m_OffCenterIgnition) {
        stellarType = STELLAR_TYPE::OXYGEN_NEON_WHITE_DWARF;
    }
    else {
        stellarType = m_HeShellDetonation ? ResolveHeSD() : ResolveSNIa();
    }
    return stellarType;
}

With this patch, events that were previously recorded as standard SNIA (SN_Type = 64) are instead recorded as HeSD (SN_Type = 128), consistent with He-shell detonations previously bypassing ResolveHeSD() and falling through the standard SNIa branch.
Here are the yaml file I used and a grid file containing systems that are recorded as HeSD with the patch:
config1.yaml
HeSD_Z0.0100.txt

COMPAS version v.03.29.00 is used.

[nrsegovia]

Thanks @brettppark for the files. I was able to reproduce both the problem and the fix (single line in COWD.cpp). If no other problems need to be fixed, it can be pulled from https://github.com/TeamCOMPAS/COMPAS/tree/HeSD_tests .