03:17 UT Ant 12 is back! It will no doubt have terrible delays. Meanwhile, I am redoing my earlier calibration on 3C273 with parallel and crossed feeds, but now adding parallel and crossed feeds for the equatorial dishes. A total of four scans are being taken, in this order:
- one with the 27-m feed tracking parallactic angle (gives parallel-feed data for azel dishes 1-8, 12),
- one with the 27-m feed tracking parallactic angle + 90 degrees (gives crossed-feed data for azel dishes 1-8, 12),
- one with the 27-m feed rotated by to 90-degrees (gives crossed-feed data for equatorial dishes 9-11, 13),
- one with the 27-m feed unrotated (gives parallel-feed data for equatorial dishes 9-11, 13)
04:04 UT The first two scans worked fine, but I could not get $PA-STOP to work (should zero the FRM PA, but instead it kept sending commands to keep the feed perpendicular, overriding my FRM-SET-PA -90 command). I had to kill the schedule and reopen it. Now on the third scan as of 04:03 UT.
04:56 UT Those scans are completed. I then updated the delays (needed for ant 12, but also some very small changes to some others). All of these calibrations are "between" the ones needed for the Sun, so should not affect the calibration of those data.
03:20 UT I set up to do a long calibration on three sources (a repeat of my earlier special.scd, run on Jun 28) using pcal_hi-all.fsq, in order to fully test my feed-rotation correction. It will run from 07:25 UT to about 14:10 UT.
18:30 UT The feed-rotation correction works! The amplitudes are now flat over > 7 hours, and the phases behave just as expected. I wrote a memo describing the results. Also, there was an M-class flare today at 16:13 UT! Should be a good one for study--all antennas running, and even polarization measurement possible!
10:21 UT I used the July 3 data to determine Bx, By and Bz corrections, mainly for Ants 12 and 13, but tweaking a couple of others. I started a new run of special.scd, but this morning I found that I made the same classic error AGAIN, using errors in meters when I really wanted them in ns. Therefore, the corrections were only about a third of those needed (but maybe at least the correct sign). I discovered it and made the correction, and got back on the source as of 10:15 UT. Hopefully by 14:05 UT when the observations end, things will be a little better. I still was not able to correct Bz for Ant 12, though, so there is one more adjustment at least, for the future.
12:33 UT The Ant 12 (and other antenna) phases now look good! I need to let it run longer, but so far it seems the Bz error is small on Ant 12.
03:40 UT I have set up for another long set of calibrations like last night, but this time will do the 90-degree crossed-feed observation near 0 HA for 2253+161, so that I can remeasure the X-Y delays. I want to check repeatability, and also see if Ant 12 has changed due to the largish baseline adjustment. I also will update the baselines if warranted by these new measurements.
23:59 UT Several small flares late this afternoon. More on the way!
11:35 UT I have been trying for the past day or so to get a scheme working in which observing with the LO-frequency receiver on Ant 14 will automatically use delay centers stored in the Ant 15 slot. This is needed since there are delay differences between the LO- and HI-frequency receiver delays. After several false starts, I think I may have it now, and am taking data on 2253+161 with the LO-frequency receiver in place. I will take 10 min of data with the LO-frequency receiver, followed by 10 min with the HI-frequency receiver, and check the delays of both.
11:44 UT Argh. The dppxmp program was not running. NOW I am taking data...
02:05 UT Lots of small events today (Jul 09), but no large ones.
I found that all of the phasecals for this date were taken with the ant14 feed rotated by 90 degrees. Therefore, none of the phasecals should be used! For other dates around this time, there was essentially no phase slope anyway, so probably any slope can be ignored.
20:40 UT I just updated the 2-m antenna pointing coefficients based on the 17:30 UT solpntcal. I suppose this should be done much more regularly. We have had a number of flares over the past week, including a nice one around 19:30 UT today. Also an M-flare at about 01:10 UT on Jul 14 (but the older 2-m ants could not track that late).
02:20 UT We had a great, uninterrupted run of good data until today (Jul 28), but then today was a total loss. Apparently there was a power outage, and it took until about noon PDT to get the system working. Then I discovered this evening that the dppoffset was set to on (apparently the ACC was rebooted), so the IF attenuations were screwed up all afternoon. There are no good data all day. Hopefully tomorrow will be better...
13:27 UT I discovered another problem with Ant 14 -- the control system and bias computers were down. I have them back in service now, and I am hopeful that the phase coherence will finally be back.
14:16 UT <sigh> I discovered that Gelu had reloaded the software for Ants 3-7, which overrides their crio.ini with some standard (and incorrect) version, so I had to spend a couple of hours getting all of those back to their correct values. The main effect of this is to give the wrong voltage-to-power-conversion coefficients, and attenuation values, so the front-end levels are all wrong. The levels would have been wrong (with several settings as I attempted to figure out what was going on) during this morning's calibration, and also during the first 10-20 minutes on the Sun, on some antennas. The levels were also wrong all day yesterday.