craig.cahillane@LIGO.ORG - posted 14:53, Friday 11 February 2022 - last comment - 14:58, Friday 11 February 2022(61708)
PMC temperature loop trigger in bad spot
Jason, Jenne, Craig The PMC trigger which sets whether the PMC is locked or not has bad limits in our current state. The limits were 0.005 < H1:PSL-PMC_TRIGGER_OUTPUT < 0.1 volts. However, the H1:PSL-PMC_TRIGGER_OUTPUT was actually negative when the PMC was locked. This is causing the PMC temperature loop to constantly flicker on and off, since the trigger controls whether this loop is actuating. The temp loop is meant to offload the work of the PMC PZT. However, in it's current state the PMC temp loop drifts off into the distance, eventually causing the PMC PZT to rail and everything to lose lock (FSS, IMC, ISS first loop). The trigger output is negative because the reflected power incident on the PMC RFPD is too low, because the transmission is too high. This is actually a good thing, means the mode matching to the PMC is very good. Jenne went into the PSL and increased the amount of light on the PMC RFPD. This may not have actually helped, though. When the PMC is locked, the RFPD readout voltage is still very small, too small for the trigger to register. Jason lowered the minimum for triggering from 0.005 to 0.003, (and I lowered it further to 0.001) and now at least the trigger is not flickering. However, when we tried to offset this increase in optical gain by lowering the analog gain H1:PSL-PMC_GAIN, the RFPD readout dipped into the negatives again, indicating that we are just locking the PMC with too much gain and it's causing extra light to be reflected. Probably we are locking on the edge of loop stability causing some gain peaking. Additionally, the temp loop is still not doing it's job of holding the PMC PZT within the operable range. The temperature is still drifting off into crazy land and causing the PZT to rail and dump lock. In lieu of actually going into the PSL to solve the problem, I'll probably play around with the trigger settings, such as they are, and try to find something that allows the temp loop to work while not gain peaking. Right now I've got H1:PSL-PMC_GAIN = 3 dB H1:PSL-PMC_TRIGGER_OUTPUT = 0.005 H1:PSL-PMC_MIN_TRIGGER = 0.001 H1:PSL-PMC_REF = 0.1 H1:PSL-PWR_PMC_REFL_OUTPUT = 4.5 W H1:PSL-PWR_PMC_TRANS_OUTPUT = 111 W
One note: the PMC REFL PD is currently not very well calibrated, to the point that I don't trust its reading at all. This is because we tried to have the PMC locking PD do double-duty and provide the monitor for our reflected light level as well. This PD isn't quite capable of this, so we have to come up with another solution. Currently we're thinking about installing another monitor PD, likely with a pickoff from the light being directed into the PMC REFL camera; we have a good bit of ND filtering in front of this camera to keep from saturating it, so we have some extra light we could use for monitoring PMC REFL. Until we can make this change, I would not trust the PMC REFL reading in the PMC MEDM screen.