[josephb, osamu, kiwamu]
We worked over by the 1Y2 rack today, trying to debug why we didn't get any signal to the c1lsc ADC.
We turned off the power to the rack several times while examining cards, including the whitening filter board, AA board, and the REFL 33 demod board. I will note, I incorrectly turned off power in the 1Y1 rack briefly.
We noticed a small wire on the whitening filter board on the channel 5 path. Rana suggested this was to part of a fix for the channels 4 and 5 having too much cross talk. A trace was cut and this jumper added to fix that particular problem.
We confirmed would could pass signals through each individual channel on the AA and whitening filter boards. When we put them back in, we did noticed a large offset when the inputs were not terminated. After terminating all inputs, values at the ADC were reasonable, measuring on from 0 to about -20 counts. We applied a 1 Hz, 0.1 Vpp signal and confirmed we saw the digital controls respond back with the correct sine wave.
We examined the REFL 33 demod board and confirmed it would work for demodulating 11 MHZ, although without tuning, the I and Q phases will not be exactly 90 degrees apart.
The REFL 33 I and Q outputs have been connected to the whitening board's 1 and 2 inputs, respectively. Once Kiwamu adds approriate LO and PD signals to the REFL 33 demod board he should be able to see the resulting I and Q signals digitally on the PD1 I and Q channels.
In an unrelated fix, we examined the suspensions screens, specifically the Dewhitening lights. Turns out the lights were still looking at SW2 bit 7 instead of SW2 bit 5. The actual front end models were using the correct bit (21 which corresponds to the 9th filter bank), so this was purely a display issue. Tomorrow I'll take a look at the binary outputs and see why the analog filters aren't actually changing.
|