Sheila, Dan Brown, Rana
Because DRMI acquisition has been slow, we looked at the phasing of REFLAIR which is what we use to acquire DRMI.
By injecting a 132 Hz line in PRM with notches on in MICH and SRCL, we measured that in DRMI the phase of REFLAIR 45 should be 112 degrees to minimize PRCL in the Q phase (currently 97), and for REFLAIR 9 the phase should be 8 degrees to minimize PRCL in the Q phase (currently -23 degrees).
We reverted these changes since we lost lock and had trouble reacquiring with the new phases. We didn't get a chance to measure the POP9I to SRCL element, but that would be the next step if we want to try to use these phases for DRMI acquisition.
Afterwards, we checked out the 1f in-vac diodes. We drove PRM and phased REFL & POP and minimized the Q phase signal via DTT ASD. The big change in REFL45 is because we wanted to change the sign to match the SRCL -> POP sign.
In most cases, this worked as expected. However, we were not able to get anything smaller than a ratio of 1:3 for REFL45. Seems like something is wrong with this chain. Maybe some strange optical effect or (more likely) a malfunction in the whitening gain or whitening filter. Since we don't use REFL45 for anything at the moment, it may not have been obvious before.
The table above is what CONLOG gives us when we ask for the Diffs.
do not be alarmed by the lack of Alarms
DanB, Craig, Rana
also, we started to investigate switching SRCL & MICH from POP to REFL in case that might reduce the REFL saturations; since REFL9 & REFL45 come from the same photodetector it may be a good idea to have the SRCL & MICH loops suppress the 45 MHz outputs. Perhaps closing the 45 MHz LSC loops will reduce some of the excess noise in the ASC loops.
Looks feasible in practice, but not sure it it makes sense in principle. I wonder if anyone has thought about or even tried this switchover before?