As a test, HAM3 Ry loop was switched to SUPERSENS6 blend, which includes a newly designed 30mHz CRS blend. It has been running stably for the last 20 minutes.

start_gps = 1468138518

Jennie Wright, Khanh Vu

This morning, we entered the calculated values into the sensing matrices on the JAC menu. We encountered a few points of confusion that are worth mentioning:

1. The calculated matrix contains 16 values because it is a 4x4 matrix, with four possible angular actuator offsets and pitch and yaw readout channels from two wavefront sensors. However, the matrices on the JAC panel contain only eight inputs. This design assumes that, ideally, a yaw actuation produces only a yaw signal and a pitch actuation produces only a pitch signal. In practice, the system is not perfectly decoupled, so we observe cross-coupling between pitch and yaw, leaving us with more than eight matrix entries.

2. For the PZT actuator, the relationship between pitch and yaw is reversed. A pitch offset produces a signal in the yaw channel, while a yaw offset produces a signal in the pitch channel. This is due to the configuration of the optical components, and this convention has been maintained in the control system.

We will continue working on the sensing matrix on Monday to determine how to account for these issues.

Jennie W

I edited the file /opt/rtcds/userapps/release/ioo/h1/medm/jac_macro.txt that changes the gifs on the JAC_OVERVIEW_NEW screen so that the screen displays the picture of the beam transmitting through the JAC and also reflecting off the front surface of JAC and hitting the WFS.

I have set these to work if we are putting in 200 mW or above into HAM1.

These values might need further tuned.

I used the values recorded during a 2W lock of JAC today and divided the REFL, WFS and TRANS power by 10 so these thresholds still make sense at 200mW.

The trigger PD on the REFL signal has such a small power on it that it will sometimes go negative so I left this value at -10.

If anyone notices that the this screen doesn't show a beam reaching the PDs when it actually is based on the epics values, please let me know so i can fix it.

TITLE: 07/24 Day Shift: 1430-2330 UTC (0730-1630 PST), all times posted in UTC
STATE of H1: Planned Engineering
INCOMING OPERATOR: None
SHIFT SUMMARY:
Comissioners got in today and we started trying to lock PRMI and align the input arm.

After a short while.... something happened. and there were some suggestions that a power outage had happened. 
Which also coincided with FW0 connection issue. ( The control room was running on FW1 and otherwise the CDS "one leg" as opposed to the zero leg which is good, cause FW0 is not connected.)

So we all started trying to gather data of all the things that would have been tripped or otherwise upset by a power glitch.
We even made a list!: 
    [Sheila] 10:20 PDT JAC REFL power drops -- i.e. we lost the beam out the JAC into the IFO.
    [Sheila] 10:20 PDT JM1 OSEMs show sharp change in behaviour " exponential droop" at the ~few [urad] level (JM1 is upstream of JAC)
    [Sheila] 10:20 PDT IMC PWR IN PD also shows a (faster) "exponential droop" from 2 [W] to 0.8 [W]
    [Sheila] 10:27 PDT Started making changes to laser power guardian (requesting 10 [W]) to get back to 2 [W]. This turns the IO rotation stage. OSEMs on JMs, RM1, seem to witness this
    [Dave] 10:20 PDT h1sush12 DAC2, channel 20 (the JM1 UL channel) shows a spike glitch, at the 0.01e6 [ct] ~= 1.5 [mV] output voltage level (28-bit DAC, so [DAC V] = [DAC ct] * 20/2^28 [DAC V/DAC ct])
    [Dave] 10:20 PDT ADC3 channels 0-3 (JM1's OSEM PDs) change mean behavior, similar "exponential" but up,
    [Dave] 10:29 PDT another sharp "glitch" feature change in the ADC3 CH 0-3 (JM1 OSEMs)
    [Dave] 10:29 PDT Reported "glitch" in the corner station power
    [Dave] 10:33 Frame writer went unstable

So I started looking for the MAINS channels to document the voltage change.

H0:FMC-EX_MAINS_CHAN_1_VOLTAGE looked very strange, so did 2, and 3. They haven't reported Voltage correctly since JUN 18th! Perhaps from a Hardware Swap or a Beckhoff change? 

Thankfully dave has these channels in an ndscope template that still work to determine power glitches:
H1:PEM-CS_MAINSMON_EBAY_1_DQ  , and ebay 2 and 3.

Turns out there was a button pusher and an alog about how  that button no longer exists which you can read about here: alog 91246 . 

M6.0 Earthquake from Vanuatu Shook up End X so much I had to go down to EX and tickle the  H1:HPI-PUMP_EX_RMT_PRESS_PUMP_CTRL_OUTPUT  control loop until it worked again. 

LOG:                                                                                                                                                  

Keita, Sheila, Tony,  Jennie W

Before the JAC PZT problem, I did get an hour or so of alignment time in.  Summary: we now have light on LSC POP and POP X for the same PM1 alignment, and ITMX is back to the alignment that should point down the arm.  We have the expected power in LSC POP path, but a factor of 20 too small in both DC and RF signals in the popair path. 

These screenshots show MICH fringes with 10W input power, where I started and where I ended.  The idea was to move to the ITMX alignment that Jenne Driggers found using the arm beam here: 89738.  I watched the mich fringes and AS camera while moving the ITM, moved the beam splitter to keep the michelson fringes, and as Keita suggested moved PR3 to keep the beams on the ISCT1 refl camera.  This did cause the michelson fringes on the LSC pop diode to get smaller, when that happened I paused, went to the PR2 spot move guardian state, and adjusted PR3 to bring the fringes back on LSC POP.  After bringing the ITMX yaw alignment back I could see that there is now light on POPX and LSC POP A for the same PM1 alignment.  

When I walked ITMX pitch, I had to also adjust yaw several times as I went along to keep the mich fringes.  I also adjusted PRM to keep PRX alignment good as I moved along. Looking at this screenshot of the brief time when PRMI was flashing, the POP A LF flash was about the O4 power level (91211), as was reflair A, but popair has too little power.   The result of this was that PR3 started the day 56urad away from the O4 slider, but is not -15urad.  PR3 yaw started the day close to the O4 slider but is now -47urad.

I also adjusted the POP X dark offsets so that this QPD will be less confusing to read, SDF screenshot attached.

We tried walking PR3 in PR2 spot move to allow us to centering POP X without saturating PM1, this alignment is shown in this screenshot, but when we then aligned PM1 to put the beam on LSC POP, we were missing power there. 

Sheila had me look back to all the time that ALIGN_IFO's channel H1:GRD-ALIGN_IFO_STATE_N was in in MICH_DARK_LOCKED [43].
I made a Statecounter query.

python3 statecounter.py -chan "H1:GRD-ALIGN_IFO_STATE_N" -operator "==" -value 43 -trend "m-trend" -gpsstart "1368975600" -gpsstop "1479398400"

Which returned the output.txt file that now contains all the times that H1:GRD-ALIGN_IFO_STATE_N was in Mich_Dark_Locked. 
It was then easy to use those GPS times in an ndscope to check out the noise on other channels to compare them to today's Mich_Dark_Locked. 
Comparing that to what we can see in May of 2025 the ASC-AS_A_DC_NSUM has larger fluctuations than todays Mich_Dark .
ASC-AS_B is roughly comparable.

BUT H1:ASC-AS_C_NSUM does seem to have less fluctuations a year ago. 

Madi, TJ

This week, TJ and I went down to EX to re-measure the location of all components within the EX HWS path. The updated measurements are in the attached file. 

We each measured the distances, to give some indication of the uncertainty on each measurement. 

Shoshana, Arnaud

We cleaned up the SensInv filters in the CRS filter bank. Those invert the CRS sensor response so the signal is still calibrated in urad below 50mHz. 

* fm9 is the reference filter (based on a fit of this measurement)
* fm1 is the running filter, which is a modified version of fm9 with lower Qs, to reduce ring down time during a filter reload or a step in the signal (see step response in fig2, red is fm1, blue is fm9). 

Summary:

JAC PZT outputs were turned off, this lead to hours of confusion.  It shows up in HAM1 suspensions because light was sparying all over HAM1.

I'm followed the pydarm deployment instructions here, to update the LHO pydarm install.  However, the deploy script is slightly out of date, and currently owned by Jamie, so used a modified version to point at environment-py39.py (a change in the setup allows for different python version installs).  I need to get this fixed longer term.

This is the 20260724.0 tag for pydarm, which pulls in fixes from the last year or so, aimed at allowing for an up to date test of calibration export functionality on Monday.

This is not the default cds conda environment, but the default you get when typing pydarm at a command line, or specifically invoking by running "conda activate /ligo/groups/cal/conda/pydarm"

Tony, Dave:

At 10:33 FW went unstable, restarting itself after writing a new frames.

I restarted the daqd process at 10:44 (systemctl restart rts-daqd) which so far appears to have fixed it.

J. Kissel

Seeing the "first light," calibrated SPI Optical Lever amplitude spectral densities in  and , and thinking about ADC noise limitations and the calibration models assumption that the spots are centered on the QPDs drove me to re-check the spot positions on the QPDs. 

Timeline of centering:
    2026-07-17 17:00 UTC (LHO:91124) -- Jennie turned on the SPI laser for the first time since doors went on and the ISIs were isolated, and re-centered *only* QPDB with M_C1 (and thought she had adjusted QPDA centering with M_M1, thought she had mistakenly moved M_M2, but confirmed later she did not; LHO:91219).
    2026-07-23 14:00 UTC (LHO:91218) -- Jim unlocks and restores function of HAM2 HEPI, such that now both HAM2 and HAM3 are fully functional, and remain ISOLATED with corner station sensor correction ON as their nominal condition.
    2026-07-23 20:30 UTC (LHO:91221) -- Jennie re-centered (only) QPDB again using M_C1.
    2026-07-24 16:05 UTC (this aLOG) -- Not knowing that Jennie had yet recentered any QPDs, and finding QPDA PIT's normalized beam displacement at +0.23 [rad], I recentered *both* QPDA and QPDB.

I attach a similar trend of relevant channels including both Jennie's centering work from 2026-07-23 and mine this morning. Similar to Jennie's plots but with thick lines, light mode, and with the y-axes labeled.
I saved this template to 
    /opt/rtcds/userapps/release/spi/h1/ndscope/QPD_Centering.yaml  (rev 35598).


Here're the DC values of the QPDs after this morning's centering (done with the precision one can achieve with a picomotor STEP SIZE of 1 [ct]):
$ gpstime now;z avg -30 H1:SPI-H23_OL_QPD_A_PIT_OUT_DQ H1:SPI-H23_OL_QPD_A_YAW_OUT_DQ H1:SPI-H23_OL_QPD_B_PIT_OUT_DQ H1:SPI-H23_OL_QPD_B_YAW_OUT_DQ -s

t = 
PDT: 2026-07-24 09:27:44.113664 PDT
UTC: 2026-07-24 16:27:44.113664 UTC
GPS: 1468945682.113664

Channel                        t-30 sec average       t-30 standard deviation
H1:SPI-H23_OL_QPD_A_PIT_OUT_DQ -0.0024822048842906954 4.722167842008142e-06
H1:SPI-H23_OL_QPD_A_YAW_OUT_DQ  0.0030039884615689514 5.884913671432033e-06
H1:SPI-H23_OL_QPD_B_PIT_OUT_DQ -0.0022898733150213957 8.094109742945136e-05
H1:SPI-H23_OL_QPD_B_YAW_OUT_DQ  0.008859152471025783  4.6337990240392655e-05

$ gpstime now;z avg -30 -s H1:SPI-H23_OL_QPD_A_PIT_OUT_DQ H1:SPI-H23_OL_QPD_A_YAW_OUT_DQ H1:SPI-H23_OL_QPD_B_PIT_OUT_DQ H1:SPI-H23_OL_QPD_B_YAW_OUT_DQ H1:SPI-H23_OL_QPD_A_SUM_OUT16 H1:SPI-H23_OL_QPD_B_SUM_OUT16

t0 = 
PDT: 2026-07-24 09:54:54.983607 PDT
UTC: 2026-07-24 16:54:54.983607 UTC
GPS: 1468947312.983607

Channel                        (t0-30 sec) average       (t0-30 sec) standard deviation
H1:SPI-H23_OL_QPD_A_PIT_OUT_DQ  -0.002713  3.4e-05
H1:SPI-H23_OL_QPD_A_YAW_OUT_DQ   0.003265  4.2e-05
H1:SPI-H23_OL_QPD_B_PIT_OUT_DQ  -0.002033 10.0e-05
H1:SPI-H23_OL_QPD_B_YAW_OUT_DQ   0.009682 5.3e-05
H1:SPI-H23_OL_QPD_A_SUM_OUT16   25.846691 0.003820814636259141
H1:SPI-H23_OL_QPD_B_SUM_OUT16   29.265472 0.004081774857250754

WP13433 TW1 offload

As Tony mentioned, I did a quick NDS1 restart at 07:57 to get it to serve the past 6 months of minute trends from the archive following this week's copy.

At 08:09 I started the deletion of the old data from TW1's SSD-RAID. This typically takes a few hours. At the start of deletion disk usage was 93%.

TITLE: 07/24 Day Shift: 1430-2330 UTC (0730-1630 PST), all times posted in UTC
STATE of H1: Planned Engineering
OUTGOING OPERATOR: None
CURRENT ENVIRONMENT:
    SEI_ENV state: CALM
    Wind: 5mph Gusts, 2mph 3min avg
    Primary useism: 0.02 μm/s
    Secondary useism: 0.10 μm/s 
QUICK SUMMARY:
 

The Vacuum pressure is still being pumped down.

CDS team needed to reboot the NDS server again so the NDScope screen shots may stop beifly.

SUS: First stage of PRM seems to be pushing a bit harder than I'd expect given that it's misaligned.

This is a late entry from last Friday.  We have removed the relay tube between HAM5 and HAM7 and replaced it with the adapter/viewport assembly on the HAM5 side.  The assembly is being pumped down independently by a small turbo/aux cart.  The GV should not be opened until the pressure in the viewport assembly is approximately the same as the corner pressure and the aux cart is valved out.

Vicky, Austin, Camilla. Follows yesterdays alog 62089. Summary: aligned but weren't able to see any baffles and beam is smaller than ideal.

Austin and I measured distances between optics (all measurements attached and key ones below): 

We started with translation stages at: Collimating lens @ 1.44; HWS:L2 @ 12.39. Later Vicky and I moved the L2 on it's translation stage upstream ~ 1.5" (starting location attached). 

Vicky and I realigned the beam to remove the majority of the ITM reflection. Once we were finished and ETM was misaligned we could still see faint rings reflected off ITMX (that moved when ITMX adjusted) photo here. So more alignment could be done to get rid of those. We adjusted the CCD camera to try nd focus better but didn't see a difference so replaced in original position. 

We adjusted the collimating lens and L2 to try and get the retro-reflected beam at the iris after ALS-M11 D1800270 to be a similar size as the outgoing beam - it is ~ double the size which is the best we could get with the 40mm collimating lens. The waist of the beam is still just after L3. 

Adjusted wave plate to maximize beam to LPM from 270 to 292. This might have been the cause of reflected beam streaks we see on the camera. In the final image. 

Final beam after yesterday is attached. It still seems to small to me so maybe we need to go back to the 50mm lens. Once we are happy we need to dither the ETM to measure the magnification.