Simultaneous Supernovae

[reinhold-willcox]

Describe the bug
We occasionally evolve two SNe in the same timestep. At the moment, these are evolved sequentially in the timestep, meaning the orbital changes are recalculated appropriately. But when this happens, the output shows strange things, e.g the companions in both SNe look like BHs. This happens any time the two stars have the same initial mass, and thus go through identical evolution (either through initial sampling, or if immediate RLOF equalizes the masses).

Label the issue
urgency_moderate - This is a moderately urgent issue
severity_moderate - This is a moderately severe bug

To Reproduce
Run a binary with an exactly equal mass ratio.
I've attached a screenshot of sample output below.

Expected behavior
It's unphysical for the stars to evolve identically, which they will always do if they have identical masses. If two stars have exactly the same mass, we should add a small perturbation to both of them to ensure they will follow different evolutionary trajectories.

Screenshots

This is the SN details for this one particular problem seed (apologies for the format)

Versioning (please complete the following information):

• COMPAS v02.15.17

[reinhold-willcox]

I know this issue hasn't gained much traction, but here is another example of a system which is nonsense because of the way we do simultaneous SNe. (Uses COMPAS v02.17.11, seed 206, no other options - i.e not using pythonSubmit).

Here is the detailed output plot:

Where clearly they have an identical evolution (blue line completely covers the red line).

Here are the relevant lines from the SN output:

INT,     #-,      SEED,                        206,             206,
INT,     -,       SN_Type(SN),                 16,              16,
FLOAT,   Myr,     Time,                        1.97456440e+01,  1.97456440e+01,
INT,     State,   Supernova_State,             1,               3,
BOOL,    State,   Unbound,                     0,               0,
INT,     -,       Stellar_Type(CP),            13,              13,
INT,     -,       Stellar_Type(SN),            13,              13,
INT,     -,       Stellar_Type_Prev(SN),       8,               8,
FLOAT,   Msol,    Mass(CP),                    1.277584e+00,    1.277584e+00,
FLOAT,   Msol,    Mass_Total@CO(SN),           2.286833e+00,    2.134421e+00,
FLOAT,   Msol,    Mass_Core@CO(SN),            2.286833e+00,    2.134421e+00,
FLOAT,   Msol,    Mass_CO_Core@CO(SN),         2.286833e+00,    2.134421e+00,
FLOAT,   Msol,    Mass_He_Core@CO(SN),         2.286833e+00,    2.134421e+00,
FLOAT,   Msol,    Mass(SN),                    1.277584e+00,    1.277584e+00,
BOOL,    Event,   Experienced_RLOF(SN),        1,               0,
STRING,  -,       MT_Donor_Hist(SN),           5-8,             NA,

Both stars are registered as USSN, but both stars are viewed as type HeHG (type 8) before the SN, but one star has gone through MT twice (and is appropriately ultra-stripped), while the other has not gone through any at all.

To re-emphasize: simultaneous SNe are both unphysical and lead to very confusing output in the SN file. To avoid this, we need to add in a mass perturbation any time the two masses are exactly equal.

[reinhold-willcox]

Closing this issue as it falls under the umbrella issue #805