12:06 UT We had a problem with the air conditioning in the control building as of Dec. 25, so I turned off the correlator (the main source of heat) on Dec. 27 and we have not had any observations from Dec. 27-31. However, yesterday it was "fixed" (it appears to have been the thermostat acting up), so I got the correlator back on, and this morning I am working on getting the delays set. I did an initial setting based on a quick calibration last night, and was able to set all but Ant 8. For some reason, it had no coherence at all. Possible it was a huge delay--I will check on a geosat later. For now, I am doing another calibration observation. Everything else seems okay with the system.
03:45 UT The day's data seemed to go well. I did find and correct Ant 8's problem. It had a delay of 256 ns! This is clearly some weird glitch in the correlation that happens on a reboot. But after I set this large delay (which I did before the start of today's observations) Ant 8's data are fine. The only problem is Ant 13, which continues to have very bad high-frequency phase coherence. I suspect a pointing problem, but pointing on the Sun is very difficult now due to the soft limit on Dec on the old antennas, of which Ant 13 is one. I am off to the AAS meeting tomorrow (actually today, Jan. 2), so may have some difficulty keeping up with issues with observing. Hopefully there will be none!
01:29 UT I have returned from the AAS meeting, where I found I could not fully solve some issues that came up on around Jan. 7. The correlator stopped putting out all of its packets, and when I rebooted it there were some really large delays introduced. I did my best to determine them, but after some sleuthing today I found that several antennas (esp. 10 and 12) still had large delays, and worse, the Lo receiver delay (on Ant 14) did not get set for the 269 ns error I found from the correlator reboot. This meant that there were no good calibrator phases on any of the low-frequency calibrations for several days. As of right now, I believe I have measured the delays correctly on all baselines, so from today the data will be much better.