Dan, Koji

We succeeded in locking the OMC using AS45 and the OMC REFL beam, and took a number of noise measurements and transfer functions of the cavity length.  Data and plots are forthcoming.

On ISCT6, we took a pickoff of OMC REFL and steered this beam onto the ASAIR_A RFPD.  A lens was added to focus the spot onto the diode.  (Note that in order to recover the OMC REFL beam, we undid Keita's precautionary misalignment of the OMC REFL picomotors.  We'll need to steer this beam away from the OMC REFL QDPs again before we start locking.)  We used the OMC QPD alignment servo to keep the beam aligned to the cavity.  The dither alignment loops are still unstable and need to be fixed.

We used a pair of SR560s to take the ASAIR_A_45 I-phase signal and feed this back into the PZT driver.  We found that the aux input path for the PZT driver was working (yay!), and even better found that very little loop shaping was necessary to get a stable lock.  We were quite surprised that the flat-response PZT was giving us a nice 1/f loop; It looks like the capacitance of the PZT and the output impedance of the PZT driver are just right to provide a pole at around 2Hz, and the dewhitening filter on the board acts as a low-frequency boost.

We measured the cavity length noise from ~10mHz to 100kHz.  The resonances of the PZT mirror tombstone at ~8kHz and the PZT itself at ~80kHz were clearly visible.  There was a forest of lines around 1kHz that was *very* easy to excite by tapping on the HEPI crossbars.  The sensitivity of the OMC length to light taps on a structure outside the vacuum was rather alarming and needs to be investigated more closely -- does the noise couple into the ISI, the OMC suspension, the tip-tilts?  One good piece of news is the transfer function of PZT voltage --> cavity length is smooth around 20kHz, so there is a good place for an analog dither signal.