My Running Diary for locking.  Currently stuck at PREP DC READOUT TRANSITION (see below)

1524 Locking Begins (following notes from Ibrahim (84684) & Georgia's (84680) alogs)

• DRMI locks in under 2-min, then...
• 1st STOP for SRM/PRM Tweak Up @DRMI_LOCKED_CHECKED_ASC (wasn't sure if I would be fighting WFS @OFFLOAD_DRMI_ASC)
  • Spent about 4mins tweaking srm/prm while mainly watching pop18 (hard to see pop90 changes).  Basically want to both increase pop18 AND minimize oscilations in it (make it higher and flatter).
• 2nd STOP for SRM/PRM Tweak Up/check @CARM_5_PICOMETERS.  Spent about 4min here, but made even less tweaks than above.
• 3rd STOP for ASC tweaks at PREP ASC FOR FULL IFO.
  • Much different look from yesterday---less noisy signals.  Looking at the Locking ASC ndscope, there were no ASC signals railing (so no need to adjust Rail Limiters...which is good because this is still new to me).
  • Did have some signals which were not at zero, so adjusted:
    • SR2 Pitch to get SRC2_P at zero
    • SRM Yaw to get SRC1_Y to zero....but this one is  tied to SR2 Yaw
    • SR2 Yaw was adjusted to get SRC2_Y to zero, but it would affect SRC1_Y.  So I sort of split the difference to get them better and BOTH as close to zero as I could.
    • Thennnn...
• Went to ENGAGE ASC FOR FULL IFO -> with no issues 
  • Diag Main NOTE:  While waiting for ENGAGE ASC, happened to notice DIAG MAIN has notes about ALS x&y shutters being open
• Continued on...but then got stuck at PREP DC READOUT TRANSITION with the following issue for the TMS Servo:
  • "POP X PZT railed" comes and goes ont hte TMS SERVO guardian node.

Stepping out for a break.

Notes On Alignment:

1445-1523 Accidentally started "manual" alignment without selecting it in ISC LOCK (got through Input Align), so went back to down and at 1457 restarted a formal  Manual Initial Alignment (modified, where we only go to PRX_LOCKED for PRC).  Green Arms were much easier to lock this morning compared to yesterday morning---Since we were down over night, I tweaked up flashes in both arms by hand (especially making sure Y-arm's flashes were over 1.0).

• Input Align:  No Problem
• PRC:  Take to PRX_LOCKED and then tweak PRM to maximize ASC_A_DC_NSUM (went from 1.26 to 1.30...mostly in yaw)
• Mich Bright:  No Problem
• SRC: Needed an SR2 alignment by hand (per the handy operator alignment checklist sheet), but after that, No Problem
• 1523 Alignment complete

Now waiting for DRMI to lock.

TITLE: 06/01 Day Shift: 1430-2330 UTC (0730-1630 PST), all times posted in UTC
STATE of H1: Planned Engineering
OUTGOING OPERATOR: TJ
CURRENT ENVIRONMENT:
    SEI_ENV state: CALM
    Wind: 20mph Gusts, 12mph 3min avg
    Primary useism: 0.07 μm/s
    Secondary useism: 0.15 μm/s
QUICK SUMMARY:

A little breezy out (just under 20mph), but nothing like yesterday's sustained winds overnight.

Aim for today is continue progress through the states of ISC LOCK.  Today Team Syracuse will be onsite to help and Ibrahim will come in for an earlier eve shift for more coverage with the the Syracuse crew.

TITLE: 06/01 Eve Shift: 2330-0500 UTC (1630-2200 PST), all times posted in UTC
STATE of H1: Planned Engineering
INCOMING OPERATOR: TJ
SHIFT SUMMARY:

IFO is in DOWN due to PLANNED ENGINEERING (and ENVIRONMENT)

Technically not a shift day since there were no comissioners but Corey and I tried locking for the majority of the day and made it pretty far to where Syracuse Squad and I were yesterday. Serendiptiously however, the environment had other plans. We were hit by 4 6.0+ EQs across that made even green arms impossible. As those were winding down, we had a 3 hour wind event that gave us over 60 mph gusts. We still haven't recovered from the EWs nor the wind which is now sitting between 30mph and 40mph. I guess the good news for the commissioners is even if they did come in today, we would not have had any lock luck!

As for the IFO, we had a successful manual initial alignment, followed by a totally automatic DRMI lock (4 times today since my morning arrival). We would stop at DRMI ASC, touch up PRM and SRM, then move to CARM_5_PICOMETERS and retouch alignment. Then, before FULL_IFO ASC, we would pause and look at the railing ASC error signals. We would touch the relevant optic DoFs (including running the move_ARM_dev.py to adjust any compound signals). In between doing this, we would touch SRM and PRM to maintain a wide POP90 and POP18 (high highs and low lows). Then, we would turn Full IFO ASC on and continue locking. Unfortunately, due to the environment, we were unable to turn full IFO ASC on before losing lock.

ISC_LOCK.STATES is not coming up when I click on "all" on the ISC OVERVIEW. Corey discovered this issue and I can only point to those who last edited ISC_LOCK as a cause of this bug. Will ask the syracuse crew tomorrow.

Other:

• Francisco had a tour
• There was (what my phone ID'd to be) a pacific gopher/western snake by OSB entrance. Maintain caution and distance if you find it somewhere since while not venomous/dangerous, was bluffingly lunging when approached within ~2/3 feet. The wind probably prompted its evacuation anyway.

LOG:

TITLE: 05/31 Day Shift: 1430-2330 UTC (0730-1630 PST), all times posted in UTC
STATE of H1: Planned Engineering
INCOMING OPERATOR: Ibrahim
SHIFT SUMMARY:

Mainly worked on getting H1 through the alignments in the morning, but had issues with a touchy Green Y-arm (earthquakes did not help).  And for the SRC, I missed that we had a misaligned SR2, but Guardian reminded me during a later alignment, and so got 

Once H1 locked DRMI, Ibrahim showed me POP18/90 tweaks needed to improve alignment and to continue.  Made it to ENGAGE ASC FOR FULL IFO a couple of times and had the opportunity to touch up ASC signals, but we also had locklosses at this point. 

For the afternoon, another set of earthquakes rolled in and then we have had a windstorm for the last few hours with gusts in the 60s for the last hour or so (quite a bit of dust out making it tough to see Rattlesnake).

LOG:

• 1330-1803 Beekeeper and Mitch on-site to address hives on x-arm
  • 1515 Moving to y-arm
• 1939 Family tour arrives (Francisco)

H1 was down over night, so the first action was to run an alignment of H1.  Unfortunately, there was an earthquake when alignment was started.

Alignment Notes:

• Green Arms:
  • Had no issues with the X-arm, but the Y-arm struggled.  Even with flashes at around 1.0, the Y-arm would hold lock only for only 10s of seconds and had trouble getting past ENABLE_WFS.
  • Once Earthquake Mode returned to "Calm," finally had better luck with locking the Y-arm
  • Have completed Intial Alignment Offloaded serveral times since 
• Input:  No Issues
• PRC (modified)
  • Continue the "modified alignment" for this cavity, where we only go up to the PRX_LOCKED state & then use the PRM to maximize H1:ASC-AS_A_DC_NSUM_OUT16.  
• MICH Bright:  No Issues
• SRC:
  • For the first alignment attempt, this completed fine (However, [not sure if it's related but...)  this was where I skipped taking ALS_Y to INITIAL_ALIGNMENT_OFFLOADED due to the locking issues noted above.)  When I tried locking after this, DRMI locking looked bad on the camera and with its flashes.
  • Once I finally was able to get the ALS_Y/Green Y-Arm aligned, and attempt SRC alignment, Align Ifo would get to the ACQUIRE_SRY step and immediately go to down repeatedly.

Currently have made it up to DRMI and have decent looking flashes and have even had short locks, but they are only lasting ~3 seconds.  

• NOTE:  I did run CHECK MICH FRINGES successfully, and then locked PRMI, but at PRMI_ASC noticed that ASC signals were not converging, so I went down and returned to locking DRMI.
• Will try another alignment next.

Sat May 31 10:13:43 2025 INFO: Fill completed in 13min 39secs

TITLE: 05/31 Day Shift: 1430-2330 UTC (0730-1630 PST), all times posted in UTC
STATE of H1: Planned Engineering
OUTGOING OPERATOR: Tony
CURRENT ENVIRONMENT:
    SEI_ENV state: CALM
    Wind: 4mph Gusts, 1mph 3min avg
    Primary useism: 0.03 μm/s
    Secondary useism: 0.16 μm/s
QUICK SUMMARY:

H1 is currently at Prep For Locking.  First order of business will be running through a "modified" manual alignment (where we do not do a complete PRC_Align_Offloaded).

One oddity, is for some reason I am not able to open up the "ALL" state window for the ISC_LOCK node (I can do this for other Guardian Nodes.). 

Additionally there was an M6.0 Japan earthquake 6hrs ago and we now have M6.1 Tonga/New Zealand earthquake currently inbound.

TITLE: 05/31 Eve Shift: 2330-0500 UTC (1630-2200 PST), all times posted in UTC
STATE of H1: Planned Engineering
INCOMING OPERATOR: Tony
SHIFT SUMMARY:

IFO is in DOWN for PLANNED ENGINEERING. We were able to get to POWER_10W, losing lock at ~18W.

An extremely productive day in which we babied IFO, carefuly and cautiously watching POP signals and ASC whilst moving suspensions.

The steps we did on shift - Each of these were done multiple times throughout the shift with small variations. See alogs 84680 for more details.

Manual Initial Alignment: Mostly automatic but we had to stop at PR_X and move PRM by hand while lookign at AS_C signals. Once this was maximized, we could move on to MICH Bright. Once done, we moved onto normal locking.

ALS Locking: Manually done but arms easily catch. We were able to get far enough (~ENGAGE_SOFT_LOOPS) in alignment to set the green references. After our following LL, green arms were automatic.

PRMI/DRMI: Mostly automatic to lock but had to touch up with at OFFLOAD_DRMI_ASC, CARM_5_PICOMETERS and PREP_ASC_FOR_FULL_IFO

ASC: At this point, Georgia instructed to pull up the ASC signals and watch the error signals as they converge. I learnt how to do this and turn off the rail limiters. Slowly, I iterated between moving PRM and SRC to maintain good POP18 and POP90 and moving specfic PR and SR mirror DOFs to zero the error signals. After this was done, CHARD_Y was still insufficinetly far from zero so I was shown how to use the move_ARM_dev.py and what acceptable step magnitudes looked like. We walked CHARD_Y back to zero whilst looking at all the other signals includiong POP (iterating and remax/minning when appropriate). Once alignments looked good, we went through ENGAGE_ASC_FOR_FULL_IFO successfully. We were able to move automtically until DAMP_BOUNCE.

ADS: We still had a pervasive ADS PIT3 Pitch (shakes IX and points with PRM) in DARM (now lower than I had seen it in months). Craig manually went through A2L gain just for P IX Center (alog 84683). This managed to temporarily dissappear the ADS noise and we held our breath and went onwards.

POWER UP: We clicked POWER_10W and made it successfully. We waited 30s before moving forth to POWER_25W. We got to around 18W before losing lock. Before this, we saw the PR_GAIN dip and the CHARD_P ASC loop walk away and oscillate unstably. Lockloss.

We're going back up now and so far so automatic. Syracuse crew, particularly Georgia, were extremely instructive, teaching me how to lock and move forward as comissioners have done this week.

Other:

Erik showed me how to cycle the control room internet, which turned off twice randomly for unknown reasons today. They're monitoring and I'll be on-site tomorrow to respond if anything happens in that regard.

LOG:

Ibrahim, Syracuse Squad

• After Ibrahim and Corey did an initial alignment we tried going straight to PREP_ASC_FOR_FULL_IFO but lost lock aroung CARM_5_PICOMETERS as the RF18 buildings dropped suddenly
• Next time we stopped at OFFLOAD_DRMI_ASC and touched up SRM and PRM, AND stopped at CARM_5_PICOMETERS and again touched them up. I tried engaging INP1_P, PRC2_P, CHARD_P with the same steps as guardian all at once, but we lost lock. I guess we did have to manually step them closer :') though maybe the gian was different in guardian than what worked earlier today
• Next lock we got further through ASC
  • Stepped IM4 and PR2 in pit, closed INP1 and PRC2 loops using guardian shell
  • Stepped IM4 and PR2 in yaw, closed INP1 and PRC2 in yaw using guardian shell
  • Stepped SRM in pit and yaw, closed SRC1 and SRC2 in pit and yaw using guardian shell
  • Stepped CHARD P and CHARD Y using the move_arm_dev script, touched up PRM, stepped CHARD P and CHARD Y some more, closed CHARD P and CHARD Y by hand using Sheila's gain values (https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=84667) (20 for pit, 3 for yaw)
  • Closed ADS DOF 3 using guardian shell
  • Stepped the PRC2 gains from 250 to 500 by hand
  • I did these steps from the guardian:
    •             ezca['ASC-AS_A_RF36_Q_YAW_TRAMP' ] = 10
                  ezca['ASC-AS_A_RF36_Q_PIT_TRAMP' ] = 10
                  ezca.get_LIGOFilter('ASC-AS_A_RF36_Q_PIT').switch_off('OFFSET')
                  ezca.get_LIGOFilter('ASC-AS_A_RF36_Q_YAW').switch_off('OFFSET')
                  ezca['ASC-AS_A_RF36_Q_YAW_OFFSET' ] = 0
                  ezca['ASC-AS_A_RF36_Q_PIT_OFFSET' ] = 0
    • This slowly killed the lock over 5 mins for some reason? The RF18 and PRG dipped. I won't do that next time.

I'm trying to guardianify as much as possible, so I edited ISC_LOCK and lscparams in these lines:

• lscparams [225] [229] changed for the CHARD P and Y gains
• Commented out the PRC2 gain increase and and AS_A_RF36 changes ISC_LOCK [3551-3564]
• I added lines that increase PRC2 gains to 500 just below those in ISC_LOCK
• I commented out the CHARD gain changes [3585] [3586] in ISC_LOCK
• Commented out the ADS DOF3 gain increase in ENGAGE_SOFT_LOOPS: ISC_LOCK [3706, 3707]

The next lock, Ibrahim drove the IFO

• Tweaked the SRM/PRM alignment at OFFLOAD_DRMI_ASC and CARM_5_PICOMETERS
• Turned off the limiters and tweaked the IM4, PR2, SRM, CHARD_Y pointing at PREP_ASC_FOR_FULL_IFO
• Ran ENGAGE_ISC_FOR_FULL_IFO guardian state, which has some lines commented out as listed above
  • We survived, hooray!
• At DARM_TO_DC_READOUT, things looked dippy, POP18 was looking bad, ADS PIT3 was far from converged. We opened PIT3 (switched off the input, output, cleared the history) and engaged the human servo (Craig tapping the alignment for nice POP18 and PRG).
• The alignment looked good here so we reset the green references - SDF screenshot below in case this is needed to be reverted.
• PIT3 seemed way off the PRM pointing, so Craig adjusted the ITMX P2L SPOT gain from 0.85 to -1.2. We shouldn't have to do this, but it allowed us to close the PIT3 loop.
• Powered up to ~18W before the alignment went crazy and we lost lock.

The next lock

• We had to aggressively adjust BS and PRM pit in PRMI
• We checked that Craig's A2L gain change (see comment to this alog) worked
• We ran ENGAGE_ASC_FOR_FULL_IFO with no stepping - hooray!
• ENGAGE_ASC_FOR_FULL_IFO was taking ages, we by-hand added the 10dB extra gain to ADS PIT3 and YAW3, which was commented out earlier...
• We lost lock after engaging the soft loops this time

Same problem with the MSR GC switch at 17:36 this evening. Erik was onsite and power cycled the unit, which cleared the problem.

It had only been running 5 hours 8 minutes since the restart at lunchtime.

We will keep monitoring and we are planning on how to replace this unit.

We often have massive glitches in the PRG when we transition from CARM_TO_REFL to RESONANCE (states 409 - 410 in ISC_LOCK_STATE_N).
PNG 1 shows this.
This evening this glitch caused at least one lockloss.

I believe the problem may be from lines 2700 to 2705 in ISC_LOCK.py.

# Transition between TRXY and REFLAIR
ISC_library.intrix['REFLBIAS', 'TR_REFL9'] = 2.0  #put back from 3.0 to 2.0 on 09/10/2018
ISC_library.intrix['REFLBIAS', 'TR_CARM'] = 0.0  
ezca['LSC-PD_DOF_MTRX_TRAMP'] = 0.2
time.sleep(0.1)
ISC_library.intrix.load()


We switch between these two intrix values (from TR_CARM to TR_REFL9) in only 0.2 seconds.

I'm going to up this to a full 2 seconds, and see if these glitches go away.

Mr Todd, Craig

Sheila reminded us about measuring mode matching using intensity noise: alog 70985.

We were wondering about the assumption that the RF sidebands are small, to see if we can explain the noise lump around the arm pole at 45 Hz.

We looked at the latest mod depth up down test at around GPS 1426014687.
We compared the REFL A LF RIN noise at a nominal mod depth time,
a 9 MHz 3dB low time, 
and a 9 MHz 3dB high time.

We do tend to see a difference in the REFL A LF RIN noise between these times.
The noise seems elevated when the 9 MHz is lower.
The difference seems smaller between nominal and 9 MHz higher.
I don't believe it's from a change in the raw power on the photodiode, which went from 7.25 mW to 7.16 mW, which is around a 1.5% change in power.

Mr Todd will make a better plot of this change in noise.