Triggered by Jeff and the quite large variation of the range, I tried to see if there was any correlation with beam motion, especially as sensed by the TRANSMON QPD.

The plot attached shows the variation of the BNS range over the last week, togeher with the best fit obtained with a linear combination of the 8 TRANS QPD signals (X/Y, A/B, PIT/YAW). 

The reconstruction is not paritcularly good, but my feeling is that there might be something interesting here. It's likely that at least some part of the range variation can still be caused by a slighly different alignment state of the IFO.

O1 day 103

model restarts logged for Tue 29/Dec/2015 'Maintenance-lite' day, no restarts reported

TITLE:  12/30 DAY Shift:  16:00-00:00UTC (08:00-04:00PDT), all times posted in UTC     

STATE of H1:   NLN at a solid 80Mpc

Outgoing Operator:  Ed

Quick Summary:

I talked to Ed while driving in to get the low-down of his shift.  Sounds like he had some issues with various things (i.e. ALS alignment, no Dark Michelson [known], OMC troubles @ DC READOUT, etc.)---making me scared of a lockloss!  At any rate, smooth sailing now with LVEA useism at 0.38um/s & minimal winds.

RED Timing error for H1OAF on CDS Overview.  I can not remember if hitting "DIAG RESET" will drop us out of Observing, so I will hold off on hitting that button (but will send an email to Dave/Jim).

TITLE: Dec 30 OWL Shift 08:00-16:00UTC (00:00-08:00 PDT), all times posted in UTC

STATE Of H1: Observing

SUPPORT: Kiwamu

INCOMING OPERATOR: Corey

SHIFT SUMMARY: IFO locked at 80-83Mpc. All the action was in my last log. Winds calm. EQ seismic bands nominal. µSei is below the 90th percentile mark. Looking good to hand off to Corey.

80Mpc

13:20UTC

Mid-SHIFT SUMMARY: No locking since the beginning of my shift. For having nothing  tripped after the Victoria quake which barely showed us anything on the eq bands, I had a heck of a time with optics that needed to be trended and re-aligned. Y-Arm gave me a ton of grief and then I re-visited PRM. SRC was a mess but I moved on to locking PRMI so that I could get the alignment there maximized. I then had to revisit IA to re-align  SRM and SR2 by hand so that I could move on to DRMI. IFO was locked at DC_READOUT_TRANSITION for over an hour and OMC wouldn’t engage and lock so I spent ≈ 1 hr on the phone with Kiwamu until the lock was lost. Since then locking has been less than successful despite good alignment (in my opinion). There was a small earthquake that gave me a little grief. The struggle continues. If I achieve the level of lock that I had before, I only hope that the OMC behaves as it should at that point. Fingers crossed.

TITLE: Dec 30 OWL Shift 08:00-16:00UTC (00:00-08:00 PDT), all times posted in UTC

STATE Of H1: Locking

OUTGOING OPERATOR: Jeff

QUICK SUMMARY: Recovering after Canadian quake.

Activity Log: All Times in UTC (PT)

00:00 (16:00) Take over from Corey
03:09 (19:09) Restarted VerbalAlarm 
07:40 (23:40) Lockloss – Possible EQ in Victoria, CA.
End of Shift Summary:

Title: 12/29/2015, Evening Shift 00:00 – 08:00 (16:00 – 00:00) All times in UTC (PT)

Support:  Kiwamu, Mike L.
 
Incoming Operator: Ed

Shift Detail Summary: Good Observing shift until 07:40 (23:40) when lost lock possibly due to EQ in Victoria, CA. Started relocking when Ed came on shift.  

   In general a good first half of the shift. IFO status and environmental conditions remain as they were at the start of the shift. There were a couple of EQs that passed without any apparent issues. 

   VerbalAlarms stopped working around the start of the shift (00:00 (16:00)). I restarted it (at 03:09 (19:29)) and notified the support group.  

     Transition Summary:
Title:  12/29/2015, Evening Shift 00:00 – 08:00 (16:00 – 00:00) All times in UTC (PT)
	
State of H1: 00:00 (16:00), The IFO has been in Observing mode for the past 3.5 hours, with power currently at 21.8W, and range at 80Mpc.  Environmental conditions are good. The wind is up to a light breeze (2- 7mph). Seismic is flat at approximately 0.03um/s; microseism is also flat and centered around 0.3um/s.      

Outgoing Operator: Corey

Jeff, Sheila

We updated the Range integrand FOM, updating the reference which has been out of date for a while, and changed the scale.  

TITLE:  12/29 DAY Shift:  16:00-00:00UTC (08:00-04:00PDT), all times posted in UTC     

STATE of H1:   NLN for 3+hrs at 80Mpc

Incoming Operator:  Jeff B.

Support:  Kissel on-site fellow, Sheila (Locking Commissioner)

Quick Summary:

Fairly light Maintenance Day (which occurred only because H1 dropped out of lock).  Current lock looks noticeably less ratty (why??).

Shift Activities:

• ETMy Saturations:  17:13, 22:48
• MAINTENANCE ITEMS:
  • Per Ops Checksheet Tues Activity:  Checked that HEPI WD Counters were all zero (no need to clear)
  • 16:35-16:50 Bubba dropping off snow blower (out at woodshop building)
  • 17:00 GRB Test acknowledged.  This is the only 
  • 17:47-18:52utc:  Kissel performing Charge Measurement (~1hr)
  • 17:53-17:57utc:  Gray Reset PSL Watchdog at 17:55 (~5min)
  • 18:15- ~20:00 LN2 truck on-site (MY CP3)
  • 18:35-18:41 VPW Water heater check (Bubba)
  • 18:56 US Linen delivery (30-60min)
  • NO LVEA SWEEP was performed since there were no VEA entrances
• 17:45 H1Lockloss!
• 18:53-19:31 Initial Alignment
• 19:57-20:25 Kyle heading to MY (offsite at 20:48)
• 20:49 Undisturbed

LVEA is still too warm and not under control. So I have set HC3B to off (4ma).

Maintenance Summary:  17:45 Lockloss -->  20:49 Undisturbed

Not clear why we had the lockloss (LN2 truck came 30min after lockloss).

17:47 - 18:53:  Charge measurement, PSL WD Reset, LN2 at MY CP3

18:53 - 19:31: Initial Alignment.  Here are two notable:

1. ALS Alignment:  After GREEN WFS OFFLOAD, ETMy had output of 39 in yaw, so I tweaked ETMy yaw to get it down to 4 (is that worth doing?).  Also tweaked PR3 to take COMM Beatnote from 3.8 to 4.8.
2. Dark Michelson:  This was straightforward, EXCEPT for not being able to lock a Dark Michelson.  After tweaking BS & also SR2 (while looking at AS_C QPD) to no avail, restored those optics and moved on.  

19:32 - 20:34:  Lock Acquisition.  Notes below:

Since I didn't get to align BS in MICH_DARK_LOCK, I selected LOCK_PRMI in ISC_LOCK guardian, but this apparently doesn't work.  Sheila mentioned I should have selected PRMI_LOCKED instead.  Sheila looked into making the guardian script handle what I did better, but DRMI eventually locked on its own!  SOOOOO, I did NOT get to touch the alignment of the BS during Initial Alignment or PRMI.  

For DC READOUT (or around there), we received a "Wrong Mode" message.  It eventually went away on its own, but will need to figure out what an Operator should do if this message holds us up in the future.

20:41 - 20:48:  Ran A2L (per Tues Regular Mainteance) & it took ~7min.

~1205 hrs. local 

Performed the daily manual overfill of CP3 (minutes after NORCO truck left site following LN2 delivery to CP3's dewar) -> Dewar vapor pressure ("head" pressure) after NORCO delivery has remained near nominal value of ~15 - 18 psi. -> CP3 exhaust line was exhausting GN2 when I arrived as if delivery (increase in LN2 column height in dewar) hadn't resulted in an over fill on its own though the exhaust line pressure was elevated when I arrived (1.5 psi) -> I reviewed the response of the LLCV %open from the previous NORCO delivery and have decided to leave the current manual %open value of 15% as is for now -> I may come in tonight to verify that CP3 isn't over filling on its own and adjust the manual LLCV %open value if needed.

J. Kissel,

With only one more week to go on O1 "proper," (see JRPC Wiki), I think we've made the right gamble on not flipping the bias sign (as decided in the last two weeks, see LHO aLOG 24241, and 24383) -- the acceleration of the H1 SUS ETMY ESD's relative actuation strength accumulation in longitudinal is sufficiently negative that we will stay under the desired 10% systematic error in the calibration for at least another week or two (without correcting for the change, which we will do in the future). As usual, H1 SUS ETMX, (though inconsequential) shows a completely tolerable level of charge, and hasn't really moved since we've begun turning OFF the EX ESD Bias Voltage when its not needed. See attached trend plots. 

The new question is now "when and in what order do we make changes to the DARM loop?" We know we want to do the following (see G1501372):
    - Clean up the QUAD's longitudinal hierarchical control filters -- converting them all the distributed, and then chosing a better frequency distribution such that we don't need nearly as many violin mode notches, and which ever we do need don't impact the shpae of the over-all super actuator. 
    - Jack up the boost on the DARM filter to get more low frequency suppression.
    - Reduce the EY Bias Voltage by 2 to reduce the rate of charge accumulation.
    - Flip the EY Bias to reduce whatever chare has accumulated over the course of the run, such that we hit zero at the start of O2 and stay there.
    - (maybe move to a true DARM drive?)
Each of these tasks, once complete, will require a re-assessment of how close we are to saturations at each isolation stage. Further we should be sure to perform these changes slowly and methodically such that we understand the impact at every step and ensure that we retain IFO robustness (and *maybe* try to keep the DARM loop model and calibration up to date, if it's not too arduous). Part of slow and methodical is asking in what order we perform the above steps, and how much time should we use to reassess robustness in betweeen each step. Given that O2 prep is slated to begin mid-July (see G1501561), I think we can do these 4 (or 5) tasks over the course of a month and still have plenty of time to assess robustness. Further, since 
    - it's unclear how much up-time we'll have during the commissioning period to measure the longitudinal strength change via PCAL, 
    - there are also plans to make changes to the tracking of longitudinal actuation strength (see, e.g., Integration Issues 1138, 1170, and 1155)
    - making changes to the DARM loop as described above will mean that we'll need to get a new DARM model up and running before valid kappa's can be produced
I suggest we continue to make these charge measurements once a week throughout the commissioning period. Though we've (sadly) shown that the optica lever charge assessments are not exactly proportional to the longitudinal assessments from PCAL, the assessment is still close enough that it'll do.

My previous entry about the PRMI branch in the ISC_LOCK guardian isn't as clear as it could be, so here is a newer version.

If you find that DRMI won't lock and seems to be missalinged (you see a lot of higher order modes on the AS camera, POP90 and POP18 flashes are low) you may want to lock PRMI to adjust alignment manually:

1. with ISC_LOCK in the state LOCK_DRMI_1F (or any earlier states), request LOCK_PRMI
2. Once PRMI is locked, manually adjust the alignment of BS and PRM to increase the POP18 and POP90 buildups. (POP18_I_NORM_MON around 50 counts and POP90_I_NORM_MON around 80 counts, should be stable).
3. Once you are happy with the buildups, request DRMI_LOCKED.  Either it will smoothly make the transition or it will break the PRMI lock and go back to trying to lock DRMI. 

The transition from PRMI to DRMI only works sometimes, so this is probably best used only when the alignment is bad for now.  (Its not a reliable way to speed up DRMI locking).