Jim, Arnaud

Following yesterday's installation of the new interface chassis, the CRS capacitive actuators were tested this morning by injecting an impulse voltage to each plate while the CRS was damped. We confirmed that we could ring up the CRS with each injection, see fig1.
We also confirmed the problem with dac0 chan 14 (ai chan 15) from the ham2 hepi AI (U37) as reported here - we see less motion when actuating on that channel (first impulse on fig1) compared to the other channel.

The damping loop was tested and worked with a gain of -2e9 (fig2). This gain was integrated in the dampBP filter, so the damping can be engaged with a gain of +1. We won't measure the Q until the AI chassis is replaced.

Attached are monthly TCS trends for HWS & CO2 lasers.  (FAMIS 85474)

Adjusted each of the plots a little bit to see data (or see it a little better); a little laggy, perhaps because of lots of data in the plots?

HWS note:  for ITMx & ITMy SLED powermon are both slowly trending down.

For FAMIS #85191: All fan trends look fairly flat & normal.

This is for FAMIS #85575.
Laser Status:
    NPRO output power is 1.868W
    AMP1 output power is 70.75W
    AMP2 output power is 139.4W
    NPRO watchdog is GREEN
    AMP1 watchdog is GREEN
    AMP2 watchdog is GREEN
    PDWD watchdog is GREEN

PMC:
    It has been locked 7 days, 22 hr 41 minutes
    Reflected power = 28.68W
    Transmitted power = 105.1W
    PowerSum = 133.7W

FSS:
    It has been locked for 0 days 0 hr and 18 min
    TPD[V] = 0.5104V

ISS:
    The diffracted power is around 3.6%
    Last saturation event was 4 days 17 hours and 39 minutes ago

Possible Issues:
    PMC reflected power is high

TITLE: 09/30 Day Shift: 1430-2330 UTC (0730-1630 PST), all times posted in UTC
STATE of H1: Corrective Maintenance
OUTGOING OPERATOR: None
CURRENT ENVIRONMENT:
    SEI_ENV state: MAINTENANCE
    Wind: 6mph Gusts, 4mph 3min avg
    Primary useism: 0.01 μm/s
    Secondary useism: 0.16 μm/s 
QUICK SUMMARY:

GV20 work continues after left in soft-close overnight.  ISC LOCK in IDLE overnight.  Will unlock front doors.

Summary: In switching QOSEM whitening gains from the 1st to 2nd modification, there has been a slight increase in noise in the P and Y DOFs, all others uneffected. Whitening gain should be increased again by a factor of 2-3x.

This is one in a series of posts investigating the excess noise seen on the BBSS M1 QOSEMs. In short, at both LHO and LLO, the noisefloor of the QOSEMs appears to be around 8pm/rtHz at 100Hz, quickly increasing to 50pm/rtHz by 1Hz; where we expect the noisefloor to be flat at ~5pm/rtHz across this bandwidth. See more details in LHO alog 91466. 

Below in Figure 1, we plot QOSEM readout spectrums when the sat amp was configured with the 1st and 2nd modifications of whitening gain, as described in LHO:91998. The 1st modification dataset was taken from T0:1470725000, during which BBSS M1 damping was on. The 2nd modification dataset was taken from T0:1474137198, during which BBSS M1 damping was off, explaining the additional peaks seen in 2nd modification dataset.

This shows that the reduced whitening gain has not significantly impacted the QOSEM noisefloor, and further rules it out as the source of the excess noise. There is a slight increase in noisefloor in the pitch and yaw DOFs in the 2nd modification whitening, so there is a case that whitening is insufficient and should be increasing by a small margin again, perhaps a factor of 2-3x or so. 

Summary: ADC noise is very close to, if not limiting the QOSEM signals. It doesn't seem to be enough to explain the elevated QOSEM noisefloor seen, but whitening gain should not be lowered any futher than it already is.

This is one in a series of posts investigating the excess noise seen on the BBSS M1 QOSEMs. In short, at both LHO and LLO, the noisefloor of the QOSEMs appears to be around 8pm/rtHz at 100Hz, quickly increasing to 50pm/rtHz by 1Hz; where we expect the noisefloor to be flat at ~5pm/rtHz across this bandwidth. See more details in LHO alog 91466. 

At both LHO and LLO, changes have been made to the QOSEM sat amp to reduce the whitening gain (see LHO:91853 and LLO:82137), in response to long saturation periods when the BBSS is rung up. In the figures below we plot noise budgets of the LHO BBSS, now including the new whitened ADC noise floor. The expected noisefloor of the GS 16 Bit ADC was taken from T1000322. Keep in mind that the true ADC noise floor often varies from channel to channel and unit to unit. Informally, I've heard they can vary by as much as factors or 2-3, especially towards the lower frequencies, however I dont have any solid reference for this.

Figure 1 shows the raw voltage noise of the QOSEM signals (Sum, XDiff, YDiff) after dewhitening, along with expected shot noise level and ADC noise. During this section of data, the BBSS damping loop was off, and the QOSEM LED sum feedback was turned off (see LHO:92011 for details). Here we see that the ADC noise is very close, if not limiting the QOSEM XDiff and YDiff signals, particularly if we consider true ADC could vary higher than shown. Certainly, whitening gain should not be reduced any futher. 

Figure 2 shows an open loop noise budget of the BBSS M1, now with the whitened ADC noise propagated in. Again we see that the ADC noise is very close to, if not limiting, particularly at the high frequencies. 

With this said, I'm not convinced we can blame ADC noise for this excess QOSEM noise we've been seeing. In prior budgets (such as LHO:91871), LHO QOSEMs were running with about 10x more whitening gain and the QOSEM noise floor was not sigificantly different. With that said whitening should certainly not be lowered any further. 

Travis, Gerardo, Jordan, Betsy, Tyler, Richard

This afternoon we began the process for opening GV20, following the procedure outlined in E2600314.

To prep, Tyler machined a 7/8" hole in the 8" diameter handwheel so we could drive the gearbox by hand. We then closed GV19, measuring the distance from the top of the clutch body to the top of the ball screw in the open and closed positions. 

Open: +37 9/16"  Closed: -10 5/16" (zero is the top of the clutch body)

Moving on to GV20, the motor had been decoupled from the gearbox yesterday, so we again confirmed the measurement from the top of the ball screw to the clutch body with the valve closed. 

Closed: -10.125" (zero is the top of the clutch body).

Then we torqued the loose set screws on the top clutch retaining nut. We confirmed there was only 1 set screw per hole, but they were not brass tipped set screws as expected.

We then proceeded to check the set screws on the drive shaft collar/ball nut. We had to rotate the drive shaft in order to access to these set screws, so we began turning the handwheel clockwise, per procedure at 1:48:40 Pacific. There was some increased resistance after~20 turns, and then at turn 31 the clutch popped (1:56pm Pacific). We stopped and called Jon F to discuss. We re-measured the ball screw position and found it at -10 5/16", so the ball screw had been driven down and that's what popped the clutch. We then turned the hand wheel counter clockwise with a scale on the ball screw and confirmed it was indeed raising. After 15 turns, the clutch reset and then at turn 31 (net 0) we stopped and re-measured the ball screw and it had returned to -10 1/8". We then checked the ballnut and confirmed no threads had been exposed. 

We then met with the rest of the group on Zoom to discuss before proceeding with the rest of the procedure. We decided to continue raising the valve until we could check the set screws on the drive shaft collar, then raise it 1" past the point where the valve un-cams then set it back down 1" to confirm operation and rotation is as expected.

So we continued raising the valve by turning the handwheel counterclockwise, stopping every 50 turns to inspect the ballnut and measure the height of the ballscrew to confirm it was raising. At turn 171 (3:22 pm) the valve state turned from red to yellow on the medm screen. We used this as the point of "soft close". The ball screw height was exactly -9" at this point. We noticed a small amount of oil dripping from the gearbox during raising. We then continued raising for the 1" test after turn 171 (expected 150 turns per inch, so 321 turns total) again stopping every 50 turns to check the ballnut. When stopping rotation of the handwheel, it rotate slowly backwards (clockwise), but would stop after varying fractions of a turn (maximum was ~1/2 turn, with caveat that I never fully released to wheel, so I could stop rotation if needed).

Once we got to turn 321 (3:37:40 Pacific), we stopped, checked the ballnut and measured the ball screw height of exactly -8". We then lowered the valve by turning the handwheel clockwise back 150 turns (3:49:00 Pacific), where it will remain overnight. By this time more oil had dripped from the gearbox, we put a foil tray underneath to collect any oil that may come out overnight. Pictures below are of the oil at the end of the day, after lowering the valve back to soft close position.

We will meet with the group tomorrow morning to decide the next steps.

TITLE: 09/30 Eve Shift: 2330-0500 UTC (1630-2200 PST), all times posted in UTC
STATE of H1: Corrective Maintenance
INCOMING OPERATOR: None
SHIFT SUMMARY:

IFO is in CORRECTIVE MAINTENANCE

VAC Update: GV20 is soft-closed with plans to open it tomorrow pending an investigation (waiting for update alog)

CRS Update: Jim has left HAM3 to damp using the ISI overnight. There was progress today - alog 92108

IFO was left in a single-bounce state in IDLE

OSB Doors were closed after conference

LOG:                                                                                                                                                                                                                                                               

Jim, Marc, Arnaud

wp 13661, CRS wiring D2600186

1- Pulled 2 new, 140ft db25 cables from TCS-R2 to SEI-C1 rack in anticipation of the laser driver swap, which includes new laser monitor channels and TEC driver channels. The cables are labelled CRS LASER 4 and CRS LASER 5 (as per wiring diagram D2600186). There is about 30ft of excess cable on the LVEA side, which was routed on the side of the TCS rack (no cabling left on the floor). See detailled alog from Marc here 92097.

2- Swapped the CRS interface chassis with the final version D2400073. Given the transimpedance amplfiier has a gain of 15.8k [V/A] compared to 4.99k with the previous chassis, we do see an increase of a factor of 3 in the hoqi signal (see fig1). 

The only issue we encountered today is the HEPI AI on SEI-C2 U27 that drives the CRS cap plates has dac0 channel 14 (AI chan15) is not responding as expected (only showing +6.3/-5.5 when requesting +/-10V, see picture attached.

Left to do: 

Take pictures of the new cables, and where they are plugged to for posterity
Fix the cts2mW calibration for both hoqi pds
Write down the SN of the chassis installed and the SN of the new cables installed 
Finish commissioning the CRS damping
Swap the Laser Driver chassis and commission the TEC loop
Send redlines to Dean on the CRS wiring diagram

TITLE: 09/29 Day Shift: 1430-2330 UTC (0730-1630 PST), all times posted in UTC
STATE of H1: Corrective Maintenance
INCOMING OPERATOR: Ibrahim
SHIFT SUMMARY:

Work today: 
ISIs on HAMs2 and 3 were both tripped today. But the IMs were trended back using the driftmon scopes found under the slider scopes button on the LSC-->Locking.adl 

Dave has an awesome CDS summary link here.

• VAC - EX GV 20 Manual open per procedure (will take all day) ----> Still in progress GV19 And GV20 are now both softclosed. 
• Jim - Restart CRSPROC. Some math is missing, add factor 8x that is missing from the contrast calculation for CRS. No daq restart.
• CDS/Jim/Kissel - Restart isiham2-5, seiproc models to switch out which CPS signals are used for reconstructing the ISI DIFF super sensors.
• CDS - isiham6 model restarts to add IPC connections for CRS laser driver
• CDS - Convert all usercode usages of arctan in crsproc
• CDS - observe.snap file updates for 10 models, safe.snap chan updates for: sfc2, susauxb2h34, iopsusb13, calcs
• CDS - virtual machine (proxmox VM) cluster work, hopefully not impacting for if it is, user interfaces will be down for the day
• Kissel - HAM2-3 driven measurements ???
• Fil - Install cable for NN at EY, vac rack to BSC10
• Fil - install camera at EX in air handler room  Kinda.....  this one is not working now.
• Fil - Investigate 3IFO monitoring signal loss, dust mons offline at times since tied together ---> ?
• VAC - LN2 delivery CP3 (MY) & CP8 (EX)
• FAC - Churn test of fire pumps ----> ?

LOG:
                                                                                                                                                                                                                                                                                                                                    

Yesterday I took several beam profiles of the SQZ beam on SQZT7 using the ThorLabs M² device at various settings for the ZM4 and ZM5 psams as we've done in the past to create a "grid" of values. The psam strain gauge values chosen for this set of measurements was:

ZM4 = [-3.5, -1.5, 0.5, 3.0], ZM5 = [-9.3, -7.0, -5.0, -1.3]

Between Sheila, Camilla, and myself, we were able to produce the attached figures comparing these profile results with the OMC scans taken last week (alog92065) to show the predicted mode mismatch between the SQZ beam and the OMC. The results indicate running on the very low end of the psams range with both ZM4 and ZM5 gives the best mode matching into the OMC with the current ZM2 setting.

Ryan Short, Sheila

We turned OM2 H1:AWC-OM2_TSAMS_POWER_SET to 4.6W we will let this heat overnight so we can measure squeezer to OMC mode scans tomorow morning.  

Sheila, Ryan Short, Camilla

Summary: Following up on 92044, we pico'd the filter cavity green co-alignment with the squeeze beam, we now have 90% of the best green transmission we saw last week at the same time as the IR is aligned to have 5% power in the 1st order HOM and 3% power in the second order HOM. 

I moved the H:M1 and H:M2 picos to improve the green flash height in the filter cavity;  if we try to move the picos when the filter cavity is locked in green it looses lock.  I can see that if I move far enough we start to loose power on the FC refl diode, I have not gone to the table to see if this is clipping on the in air or in vacuum path.  I was able to increase the green transmission from 8 counts to 12 counts.  Comparing this to a time from thursday when FC1 + FC2 were aligned to maximize green transmission, we had 19.4 counts on FC_TRANS_C for 7.3mW of FC laucnh power, so the filter cavity tranmission is 90% of what it was for that time (19:30 UTC Sept 24 2026) once we take into account the launch power drifting.  We don't understand why we needed to adjust these as we checked that the green and red were co-alinged right before we closed HAM7, 91497.  

VCO scan after this is attached, misalignment mode 5% of the 00 mode, mode mismatch is about 3%.  Ryan realized that we were saturating the FC trans PD with seed in our first scan, and may also have been saturating it in 92044. 

WP13657 ISI Sensor Correction.

This restart was originally slated for h1isiham[2,3,4,5] with [2,3] changes added to the common isiham23_o5 and the [4,5] with the isihammaster_2022 mdl file. But the latter is also used by h1isiham[1,6,7,8] so we ended up restarting all eight of the ISI HAM models today.

The h1isiham[1,2,3,4,5,7,8] restarts went first, none of these required a DAQ restart. They were restarted between 09:46 and 09:49.

WP13665 Add IPC between h1isiham6 and h1crsproc

Ethernet IPC channels were added to both models, going in both directions. DAQ restart was required.

WP13649 Add gain to h1crsproc

In addition to the above, add a gain to the model.

WP13656 Remove ATAN2.c function call from h1crsproc

In addition to the above, the ATAN2.c function call was replaced with the atan CDS part.

WP13612 h1seiproc model wiring changes.

WP13636 Proxmox Cluster Updates and re-keying

Please see Jonathan's alog from earlier today.

Restart Summary

Full list of models restarted today:

h1isiham[1,2,3,4,5,6,7,8] h1crsproc, h1seiproc.

DAQ Restart

The DAQ was restarted for the models h1isiham6, h1crsproc and h1seiproc.

There were no issues with the DAQ restart.

CHETA chassis move to LVEA and swap in Optics Lab.

In February 2026 the CHETA ITMX chassis was installed in the Optics Lab and it was connected to the MSR Slow Controls Bechhoff Chassis in the MSR via a long ethernet link which ran via Rm163, the computer users room and from there underfloor to the MSR.

Today Camilla moved the ITMX chassis into its permanant location in the LVEA. This will be connected to the Beckhoff network at a later stage. In the Optics Lab, the ITMY chassis was connected to the MSR chassis.

So at this time the ITMX chassis is not being read out, and the ITMY chassis is being read as ITMX.

When it comes time to connect ITMX LVEA chassis to Beckhoff, the ITMX channels will be redirected to it and a new ITMY set will be created for the Optics Lab chassis.

While this swap was ongoing this morning the DEV1 Beckhoff unit's slave count dropped from 19 to 7.