Sheila, Georgia, Craig, Jamie, Keita

Summary:

We picomotored the POP QPDs so that we are now on both QPDs, we had some difficulties doing this but we think that we have recovered and reset the QPD offsets to a reasonable place. 

Details:

After we saw the coupling of HAM3 motion to PRCL and DARM that Georgia is writing about, Keita pointed out that scattered light in the POP QPD path could easily couple to the LSC POP path.  Since the beam was falling off POP B which we don't use in loop, I attempted to use the pico's while we were locked at low noise. 

• enabling the picomotor driver adds noise to DARM at high frequencies repeatably. 
• We cannot open the PRC1 loops while the rest of our ASC loops are closed, this causes an instability and lockloss.
• We sat at 2W without ASC (recycling gain of about 46) and I pico'd the POP sled very far to get both beams roughly centered on their QPDs (this was a long walk). 
• I had to use the pico motor labeled POP A centering to center POP B, and vice versa. 
• One problem we ran into in reclosing the ASC loops is that the ALS Y arm guardian was turning off the top mass offloading for ITMY and ETMY occasionally, we haven't fixed this problem tonight but it should be fixed. 
• Once we were able to walk our selves back to the old soft offsets we moved the POP A offsets around until we got a recycling gain of just below 48, after powering up this is reduced to 47. 
• We will need to do the final picomotor steps another time, preferably once we have reset all of our green alignment references as well.  
• Georgia remeasured the noise coupling from HAM3 ISI, and it is slightly worse now than it was when the beam was falling off the QPD.  That could be because we have a slightly different alignment now. 

Side note about PRC2 cut offs:

• Since PRC2 ASC noise is the highest ASC noise that we have now, Gabriele and I looked at these loops earlier.  We tried to reshape the loops, but this turned out to be unstable.  We edited the ELP2Hz filter to be a 6Hz cut off, which I have engaged manually and they seem fine.  We could add these to the guardian, but we haven't done that.