Dan+Kiwamu+Koji

Today we finished the in-vacuum punchlist for HAM6.  Once the SEI and SUS teams signed off on the chamber, we started to work on some alignment servos to stabilize the beam into the OMC.  Our earlier attempts to engage a simple OMCQPD-to-TT feedback had seen a lot of angle-to-angle coupling.  I tried to perform some coil balancing on OM1 using the OMC QPDs as a readback.  The coils are clearly unbalanced, if the tip-tilt is driven in a pringle mode you can see motion in pitch and yaw on the QPD, and driving in pitch will generate some signal in yaw.  But I wasn't able to reduce the coupling using the perturbcoilbalance_fourosem.py script.  Will try again later this weekend.

Kiwamu helped us to commission a temporary servo to stabilize the beam at HAM6.  The beam coming from the IFO into the chamber has a lot of pitch motion at 0.5-0.75Hz, it's not clear where this is coming from, it has to be the ITMs, SR3, or SR2.  Once ASC-AS_C (most awkward name ever!) was in the ASC model we tried using SR2 to damp the motion, but the M3 stage has really weak actuation, and the M2 stage is complicated.  We tried using the M2 stage of the BS, but again we didn't have enough gain.  We wound up feeding the AS_C signal back to IM4, using the DC3 ASC filter modules.  This turned out to be solid as a rock, Kiwamu engaged a RG filter in pitch and an integrator in yaw and the spot on AS_C is now completely motionless.  But we've essentially taken output-port noise and impressed it on the input beam, so once we start locking the PRC we're going to have to use something else.

I tried to tweak up the OMC-QPD alignment servo, there was some improvement but the pitch-to-yaw coupling is still not good.  What we need to do is measure a proper sensing matrix.

In the attached PDF I show spectra with and without the ASC loops engaged.  The references are before the servos are engaged (top: AS_C, bottom: OMC-QPD).  The AS_C motion at 0.5Hz is totally crushed.

After the beam was more or less stabilized, Koji and I went to the chamber to find OMC TRANS.  We were able to see the OMC flashes on a card with an IR viewer, with the IMC input at 10W.  We steered the beam out the viewport emulator and clamped down the mirror.  This was an awkward job with the ISI unlocked.  We used an analog camera to see the mode structure of the flashes, these were pretty high order and many were fairly symmetric, which might indicate trouble with the mode-matching.

While we were on the table we exercised the beam diverters.  These were a little sticky, sometimes it took a few tries to open them, especially going from the closed -> open state.  We were careful to leave them in the open position.

HAM6 is ready to have the doors installed.  The only thing on our punch list we haven't accomplished is an in-air lock of the OMC.  With the motion of the beam in the chamber this might not be something we can do before pump-down.