Following on from 21:57 UTC Lockloss, as in Jim's alog 54795. We currently have microseism up at 0.80 μm/s which probably helped make this acquisition 6 hours long!

• 02:19 Initial alignment started. A few different changes were made during this acquisition by Jenne, Sheila and Keita:
  • Yarm initial alignment setpoints reset to improve future initial alignments.  alog 54798
  • SRY locking threshold lowered to 0.3 to help initial alignment get through SRC alog 54799
  • Dither convergence checker tightened to avoid locklosses at ENGAGE_SOFT_LOOPS alog 54803
• 04:10 NLN and Observing 
  • ALS EY SDF Diffs accepted and attached. These are from Jenne's work described in alog 54798

Keita, Sheila

We were looking at the response of the ADS loops to the impulse they get when turned on after coil driver switching.  We noticed that ADS DOF3 seems to have a much higher gain, and perhaps some gain peaking, based on a time series of the control and error signals.  Looking at the names of the filters one would think that DOF3 has 20dB less gain than the other DOFs. However, the filters called :0.1, the DOF 3 version has a factor of 50 more gain at DC.  This means that the total filter gain for DOF3 is a factor of 5 higher than the other ADS filters.

It would be nice to move this gain somewhere explicit.  Perhaps we could also consider lowering this gain to stop the gain peaking.

Jenne, Sheila, Keita, Camilla

At the end of engage_asc_for_full_ifo we have a converge checker for ASC and the ADS 3 loops (ITMs).  We have lost lock several times shortly after moving on to engage the other ADS loops in engage_soft_loops (see plots from Camilla here).  The initial alignment reseting that Jenne Keita and Camilla did (54798) seems to have helped us to come into lock with smaller ADS error signals, so that is an improvement.  In the attached screenshot ADS 3 comes on at -220 seconds when we are in the state_engage_asc_for_full_ifo, as this converges it brings the other ADS error signals closer to zero.  The convergence checker was waiting for ADS3's error signals to be below 6 before moving on to soft loops.  We lost lock a few seconds after engaging the other ADS loops, so for now we have reduced the convergence threshold from 6 to 2. 

This may mean that we sit a while longer in ENGAGE_ASC_FOR_FULL_IFO.

I've seen a few times lately that right after an initial alignment it seems like we regularly have to go to check_mich_fringes.  I had a look at the ADS DOF 10 signals that are used for MICH_BRIGHT_ALIGN, and it looks like we're basically skipping doing much of anything.  In the attached screenshot, we begin MICH_BRIGHT_ALIGN at the first t cursor, and exit the state for offloading 9 seconds later at the second t cursor.  Clearly the ADS has not converged, since neither the inmons nor the outputs are around 0. 

On lines 1022-1023 of ALIGN_IFO the convergence threshold is set to 10, but none of these signals ever get that high.  We need to lower this value to something like 0.5, so that the DOF10_OUTPUTs (ex. pink on bottom row) have a chance to actually converge. 

I will do this on Tuesday, when I know that I have time to test it, unless we happen to be doing an initial alignment when I'm on site before then (so that I can test it).  For now, operators will likely find that they need to do check mich fringes after an initial alignment.

[Camilla, Jenne, Sheila]

Lowered SRCL thresh.

I had seen this while sitting with Jim during an initial alignment earlier this week, and Sheila noted that she'd seen it recently too, but we're not getting the LSC POP DC power above the old ON threshold value of 0.4, so SRY never triggers and sends control signal to the SRM, so we keep getting stuck in the Acquire_SRY state of initial alignment.  I have changed the threshold in lscparams for SRXY locking to turn on at 0.3 and off at 0.2 (from the former values of on at 0.4 and off at 0.3).

Sheila notes that we should check (or at least think about) whether this is coming from a drifting of our alignment on the LSC POP diode, and if we need to think about using another diode for SRY or if we need to make sure we're centered on LSC POP.

[Camilla, Jenne, Sheila, Keita]

Camilla had several locklosses at ENGAGE_SOFT_LOOPS again tonight.  Normally the remedy is to go back and do another initial alignment, but Sheila pointed out that in all of the lock attempts the Yarm isn't arriving at the engage ASC states very close to the correct alignment.  Since this has been true even after fresh initial alignments, this is indicative of the inital alignment setpoints not being very good. 

We checked, and this is not at all related to the camera network/freezing issues of late.  Every time the camera setpoints come back correctly to the values that were last changed at the end of October.  So, this is likely just that things have drifted over the last ~3 months and the initial alignment setpoints just needed updating. 

Since we had been failing to even engage the soft loops, we hand-moved the Yarm soft using the script ....userapps/asc/h1/move_ARM_dev.py in steps of "YS P 0.03" to get the ADS DOF 5 Pit a bit closer to zero.  When it was off the limiter rail (but not yet all the way to zero), we let the guardian finish the soft loop alignment to the centers of the optics by letting it do ENGAGE_SOFT_LOOPS.  Once all of the soft loops had converged, we checked that the Yarm green was resonating the TEM00 mode, and saw that the green WFS and camera values were not close to zero.  So, we ran the script ....userapps/als/h1/scripts/setEndGreenQPDOffsets_YARM.py to change the WFS centering setpoints, and once that was finished we also set the camera values. 

The Xarm green didn't want to catch the 00 mode, and since the Xarm ADS hasn't been a problem we elected to go forward with locking.  We lost lock later in the sequence, so did an initial alignment to confirm that we can reacquire with these updated Yarm initial alignment setpoints.  I have on the board to check the Xarm setpoints on Tuesday after maintenance.

Attached are screenshots of the new values, and the old values. 

Alog 50968 has a sketch of the procedure on how to update initial alignment references.

Since the annulus ion pump fix on GV20 at EX after many weeks of no pumping, many AMU peaks have fallen into the noise, but N2 still dominates. Before and after RGA scans attached.

TITLE: 01/30 Eve Shift: 00:00-08:00 UTC (16:00-00:00 PST), all times posted in UTC
STATE of H1: Lock Acquisition
OUTGOING OPERATOR: Jim
CURRENT ENVIRONMENT:
    SEI_CONF state: WINDY
    Wind: 11mph Gusts, 8mph 5min avg
    Primary useism: 0.06 μm/s
    Secondary useism: 0.94 μm/s
QUICK SUMMARY: Locking now, Jim has completed an initial alignment. We have high microseism and high-ish wind to compete with.

TITLE: 01/30 Day Shift: 16:00-00:00 UTC (08:00-16:00 PST), all times posted in UTC
STATE of H1: Lock Acquisition
INCOMING OPERATOR: Camilla
SHIFT SUMMARY: Environment is getting tougher
LOG:
21:57 Lockloss, no obvious cause. DRMI was bad, before that Jason had to help lock the ISS 

23:09 Starting initial alignment

   Added 100ml to the Crystal Chiller. Add no water to the Diode Chiller. 

   Closing FAMIS #10546

TITLE: 01/29 Owl Shift: 08:00-16:00 UTC (00:00-08:00 PST), all times posted in UTC
STATE of H1: Observing at 116Mpc
INCOMING OPERATOR: Jim
SHIFT SUMMARY:

Nice quiet shift with a couple EQs which turned out not to be issues.
LOG:

• 9:50 Short GRB alert, confirmed, consulted with LLO/Virgo, & 15min stand down.
• 14:01 EQ Alert (6.0 from Solomon Is):  No Action Needed.
• 14:54 Electrical work for VAC group at MX. (Ken)
• 15:10 EQ Alert (5.1 from Jamaica):  No Action Needed.

H1's locked for almost 10.25hrs & all is well environmentally & operationally.

TITLE: 01/29 Owl Shift: 08:00-16:00 UTC (00:00-08:00 PST), all times posted in UTC
STATE of H1: Observing at 117Mpc
OUTGOING OPERATOR: Camilla
CURRENT ENVIRONMENT:
    SEI_CONF state: WINDY
    Wind: 3mph Gusts, 2mph 5min avg
    Primary useism: 0.04 μm/s
    Secondary useism: 0.47 μm/s

Microseism continues inching up; winds are low.  (Wow, those EQs earlier in the day are gargantuan!)
QUICK SUMMARY:

H1's been locked 6.25hrs & all is well.

TITLE: 01/29 Eve Shift: 00:00-08:00 UTC (16:00-00:00 PST), all times posted in UTC
STATE of H1: Observing at 116Mpc
INCOMING OPERATOR: Corey
SHIFT SUMMARY: Two candidates, two days in a row!!! https://gracedb.ligo.org/superevents/public/O3/
LOG:

• 01:49 NLN and Observing, see alog 54786

All quiet since we have acquired lock. Microseism seems to be slowly rising with secondary useism: 0.39 μm/s. Locked 2h10.

• 00:31 Initial alignment repeated as couldn't lock PRMI, wondered if we completed the 1st initial alignment too early.
• 01:18 Paused at ENGAGE_AC_FOR_FULL_IFO to wait for signals to converge with Jenne and Sheila as previously had a lockloss from ENGAGE_SOFT_LOOPS believed to be because signals hadn't converged fast enough.
• 01:49 NLN and Observing
  • SUS-IM1, 2, 3 SDF diffs accepted and attached. tagging SUS

(Kyle R, Gerardo M)

Removed and replaced the annulus ion pump for GV20 at X-End.

We used the crane to suspend the annulus piping while the system was decoupled from the ion pump flange.  The instrument air compressor was turned on to feed the aux cart safety valve.

No issues to report during the swap.  After the swap the new flange was leak tested using helium, we sprayed very aggressively and no leak was detected, leak test was performed with a background of 8.0 x 10^-10 torr *L/sec at the leak detector.  After the system passed the leak check, the annulus ion pump system was pumped for 3 hours with an aux cart, on that time frame the pressure dropped enough for the ion pump to be able to sustain it, after the 3 hours the hanged turbo and aux cart were decoupled, system is back to nominal.

The instrument air compressor is off.

Note for vacuum team, on the ion pump magnet frame 1 of the 4 bolts is broken, an attempt to remove it with some needle nose pliers was unsuccessful.

Done under WP#8515.

Back to NLN, Jenne and Jeff have updated ALS references, SDF diffs in attached screenshots.