Kara, Craig

- We lost lock in LOWNOISE_ASC after engaging DHARD_P FM4, "ELP17", a .  Took about 5 mins for the 1 Hz instability to ring up.  Will skip in future attempts.

- We had problems with losing lock around FIND_IR and ACQUIRE_DRMI due to ALS COMM tidal offloading/noisy DRMI weirdness.  Nothing changed in the tidal loop, so I checked the SEI_CONF settings, and all of our HAMs were stuck in ENGAGING_CONFIG_WINDY.  Not clear how long it's been that way, likely since the computer crash.
I added the SEI_CONF guardian to ISC_GUARDIANS medm screen so that we can see the "not all nodes arrived " warning next time this happens.

- Commented out the line about turning on DHARD P FM4, the guardian got to NOMINAL LOW NOISE by itself.  Noise was very high and perfectly coherent with DHARD P.  (Attachment two)

- INCREASE_POWER was taking upwards of 12 minutes waiting for the dither loops to converge.  At each 4 watt step in power, we  wait for the dither loops to converge to avoid locklosses, and the dither loops' bandwidth is on the order of 0.1 Hz.  The pitch of the X-arm, PIT4, was the main offending DOF.  Based on attachment one, I changed the convergence thresholds in line 3324 of ISC_LOCK as follows:

Convergence Threshold      PIT4     PIT5     YAW4      YAW5
-------------------------------------------------------------
Old                           2        2        4         2 
New                           7        7        6         0.1

I also found that running Hang's manual arm mover helped speed PIT4 convergence: python  /opt/rtcds/userapps/release/asc/h1/scripts/move_ARM_dev.py XS P -0.01

- Took two DHARD P swept sines, posted in attachment three.  Despite excitation tuning these measurements suffered from low coherence in the region of interest (~1 Hz), still not clear why we can't engage DHARD P FM4.  Saved in /ligo/svncommon/IscSVN/iscmodeling/trunk/ALIGOH1/ASC_loops/Measurements/DHARD/DHARD_P_OLG_SS_30W.xml.

- I turned off the SRCL FF and took another SRCL dither to Arm Power TF.  One of the SRCL dither points was at 19.71 Hz.  This was too close to one of the ASC dither lines, specifically the PRM pitch dither (PIT3) one at 19.65 Hz.  When the TF hit this point:
1) The dither loops started oscillating/misaligning, and the PRG and arm powers fell by 1%. (attachment four)
2) The DHARD P to DARM noise got much better.  (attachment five)
This explains the brief improvement in range (from 55 to 65 Mpc) we see at around 08:00:00 UTC today, and why the improvement went away (the dither alignment is taking the IFO back to where it was over about an hour).  Bad alignment gives lower DHARD to DARM coupling.