11:26 UT Since 2018 March 8, we have not been taking regular observations due to installation of the notch filters in all of the FEMs. This gives us full 1-18 GHz frequency range. Installation of the filters has proceeded at a pace of about 2 per day, and as of this morning only 2 remain to be installed (ants 12 and 13). Some occasional observations for calibration and checking have continued, but they will not be useful for science because the system has been very far from optimized (both signal levels and delays). This morning I have done an FEM-INIT and DCM table calibration for ants 1-10 (ants 12-14 are also done, but not relevant), and am doing some calibrator observations on both HI and LO receiver for setting delays on those antennas. One thing I notice is that the delay phase vs. frequency now has a much steeper "U" shape, similar to what we were seeing on Ant 5 (which already had notch filters installed). I believe it will be necessary to correct for this phase shape (which should be stable) in the dppxmp program.
17:51 UT The changeover to the notch filters is done, and I have performed the gain leveling needed for the FEMs and DCMs. Unfortunately, the wind is high today, so I will just do some solar observations for the time being. When the wind dies down, I will observe a calibrator to set delays, and then try to set up X vs. Y calibration observations. Hopefully by tomorrow we should be able to start regular observations with the new system. I will have to develop a new calibration procedure, though, to do both LO and HI receiver observations regularly. We will also need new analysis procedures to deal with it.
15:35 UT I obtained a good X vs. Y delay calibration yesterday, and find that it is possible to get results for all bands except band 2, which is expected since the science channels in band 2 are entirely within the notch of the filter. As long as we are not running the 300 MHz correlator, we should just skip band 2. I also realized that since Kjell removed and then replaced all of the front ends, the pointing of the dishes had changed, so I updated the parameters based on the SOLPNTCAL observation. Finally, I changed the solar schedule to include both LO and HI receiver observations on the refcals. We will need to change the analysis procedures to deal with it--may be rather complicated.
13:26 UT We had several days of successful observations and calibrations. I changed the pcal_lo.fsq and solar.fsq to skip band 2 (added band 11 in the first case, and replaced band 2 with band 5 in the second case). However, last night we suffered some sort of partial power or network outage affecting ants 1-8 and 10, all of which go through the central shed. I hope it is just a circuit breaker that needs resetting, but observations are interrupted until it is fixed.