Evan, Sheila
Today we blended CHARD P, with the same blend filters Craig and I used last night for yaw. This went smoothly and the attached screenshot shows the settings, which are accepted in SDF and shouldn't need to be in the guardian since they don't need to change. WE haven't made it to low noise to see what kind of improvement this gives us.
We also had a look at dithering SRM and demodulating DARM control to get an alignment signal for SRM. The SNR was not good for a 4Hz or 9 Hz dither, but for a 15 Hz dither there is clearly a signal, although the lock point was not good. I've added filters we could try to supress the length coupling in MICH2, SRCL2, PRCL2 filter banks, but we didn't get a chance to try this again.
Evan noticed that there is an instability in CHARD P at 0.2Hz when we tried to power up past 20 Watts or so last night, so we redisgned the boost that comes on at 17Watts to be gentler and have higher frequency zeros, (the new filter is MSBoost2 and in FM1). With this we seems to be stable at 30Watts, at least for 5 minutes (we broke lock by trying to go to 35W).
We had a bit of trouble with the laser today, nothing that can't be fixed easily ( noise eater and ISS oscialltions, difficulty locking the FSS). Evan changed the temp search ramp parameters for the FSS.
Also, we had to reduce the analog gain of the TMS QPDs even further in order to avoid saturation on some of the segments when going above 25 W. We used to set the gain to 9 dB once we achieved resonance; now we reduce it to 3 dB.
This is a problem that is better solved by picomotoring the TMS beams. The worst offender seems to be the Y-arm B diode; it has a factor of something like 40 in power between segments 2 and 4.