Ibrahim, Ryan C, Gabriele and I did a bunch of work to help the locking effort of DRMI today. Here is a summary of some of the work I did.

We noticed that after being in DRMI locked for a while, we would see something ring up and cause a lockloss, close to 26 Hz in frequency. I checked the OLG of the PRCL loop and it has lost some phase at low frequency, with a zero crossing around 25 Hz. I doubled the PRCL gain and this averted the problem. Sheila mentioned offhand that this might be a phasing issue, but until that can be worked out, we need a factor of 2 gain in PRCL to remain stable in DRMI lock. Compare the blue trace on this plot (today before double gain) to the green trace (old reference) and the red trace (after the gain was doubled).

I then tried to work on the 3F transition. I tried driving a PRCL line at 88 Hz, and I phased REFLAIR RF27 to maximize the signal in I phase. I tried switching the signal over to PRCL 3F several times and this worked repeatedly, as long as the PRCL gain had the factor of 2 in the gain. I changed the phase by 90 degrees. This plot compares before (dotted) and after (solid). Here is the SDF.

I tried phasing REFLAIR RF135, and this is where I was not successful. I tried adjusting the phase to maximize the MICH signal in Q, driving a MICH line at 88 Hz. I went +90, which failed, and then I tried -90 which also failed. To see what I was getting wrong, I put the MICH 3F intrix values into the XARM input matrix on the LSC screen, and took a transfer function of MICH on 1F to the XARM 3 F signal. The phase was correct, but the gain was wrong by a factor of 2. I tried doubling the 3F intrix values and running the handoff again but this cause another lockloss. Therefore, I have reverted by changes to the RF135 phase to 115 deg, and I'll just give up on this for now, since I clearly don't understand what I'm getting wrong.

I'm not sure if we want to adjust the 1F phasing further or go with the double gain on PRCL, so I'm not going to make any guardian changes at this time.

The only permanent change I have made at this time is the updated REFLAIR RF27 phase, which has been SDFed. All other changes have been reverted.