Strange auto meridian flip 'solve and sync' behavior


My setup is:
EQMOD 1.28b
SGP beta using Pinpoint
PHD2 2.4.1b

I’ve been having problems with automatic meridian flips and I found some strange behavior with the solve and sync in the auto meridian flip sequence. I use Framing+Mosaic to create targets, and SGP slews and centers on them flawlessly and accurately using the reference image and pinpoint.

The problem seems to only be in the automatic meridian flip sequence. After much trial and error I finally noticed (by watching cdc) that the scope was being synced back across the meridian (sgp is set to flip 1 deg past) in the initial solve and sync step, so the slew command that should have flipped the pier slewed the scope to a point still east of the meridian. Both EQMOD and cdc were set to the J2000 epoch. After the failed flip the sgp target ‘slew now’ and ‘center on now’ flips the pier as it should and ‘solve and syncs’ it right back to the reference image.

The strange part is that the way I got it to work was to set EQMOD+cdc from J2000 to JNOW

When I changed the epoch the auto meridian flip ‘solve and sync’ didn’t shift the coordinates and the slew command subsequently resulted in a meridian flip.

Here are the sgp log(s).
The failed flip is in this log at 11:11pm -

The successful flip is in this one at 12:13am -

Unfortunately I wasn’t able to confirm the fix because I left the ‘pause before flip’ checked… (GNS didn’t wake me up for that? I’ll have to investigate that more too). Any further testing will have to wait for the weather…

Any thoughts or suggestions that would explain this?