14:23 UT At long last, we have new batteries in the Ant 13 solar power station. Presumably Ant 13 will no longer lose power every night, so it should be available every day barring some other cause of downtime. The batteries at Ant 12 will also be replaced in a day or so, and the Ant 2 Az drive will be fixed this week also. That will solve the known problems aside from the continuing DPP crashes (and occasional schedule hang-ups).
13:25 UT Today I updated the antenna pointing, reset the FEM attenuations (note that ants 2 and 7 had screwed up crio.ini files, so I uploaded the right ones), and updated the DCM master table. This was all done AFTER the morning calibration (which was short anyway, because there was a bug in the GPS satellite coordinates today!).
21:19 UT Oddly, after today's SOLPNTCAL the offsets were the same as before. I checked the database, and indeed the antenna coordinate change did not happen. I am not sure why, since when I reran the exact same sequence of commands, the update went fine. In about 20 minutes I will see the second SOLPNTCAL at 21:30 UT, and can verify if the update worked.
10:49 UT I updated the delays and restarted this morning's refcal. There is less than 10 min to go in the morning refcal, so it should not be used for today's calibration. Hopefully the calibration from tonight will be okay for the refcal. Only ant 10 needed a significant adjustment, but I updated all of them.
22:19 UT I did an analysis of the EQ coefficients, and find that instead of 8, we should be using 4 for most baselines. Just now I changed the setting on the fly, right in the middle of the scan. Once this file is done, I will look at it and see how it compared in terms of solar amplitude and phase.
22:37 UT The change causes absolutely no phase difference, while the amplitude drops as expected. Although the EQ coefficient dropped by a factor of 2, the amplitude drops by a factor of 4 since it is applied to the voltage, which gets squared for power.