Lockloss during SQZ comissioning during a suspect ZM4 move.

Took Calibration Sweep at 13ish H1 NLN Lock.

BB Start and End: 15:27 UTC, 15:35 UTC

File Names:

Simulines GPS End and Start: 1410190636, 1410192001

File Names:

File written out to: /ligo/groups/cal/H1/measurements/DARMOLG_SS/DARMOLG_SS_20240912T153646Z.hdf5
File written out to: /ligo/groups/cal/H1/measurements/PCALY2DARM_SS/PCALY2DARM_SS_20240912T153646Z.hdf5
File written out to: /ligo/groups/cal/H1/measurements/SUSETMX_L1_SS/SUSETMX_L1_SS_20240912T153646Z.hdf5
File written out to: /ligo/groups/cal/H1/measurements/SUSETMX_L2_SS/SUSETMX_L2_SS_20240912T153646Z.hdf5
File written out to: /ligo/groups/cal/H1/measurements/SUSETMX_L3_SS/SUSETMX_L3_SS_20240912T153646Z.hdf5

Calibration Monitor Screenshot taken right after hitting run on BB attached.

Thu Sep 12 08:14:19 2024 INFO: Fill completed in 14min 15secs

Jordan confirmed a good fill curbside.

TITLE: 09/12 Day Shift: 1430-2330 UTC (0730-1630 PST), all times posted in UTC
STATE of H1: Observing at 143Mpc
OUTGOING OPERATOR: TJ
CURRENT ENVIRONMENT:
    SEI_ENV state: EARTHQUAKE
    Wind: 14mph Gusts, 9mph 3min avg
    Primary useism: 0.19 μm/s
    Secondary useism: 0.29 μm/s
QUICK SUMMARY:

IFO is in NLN and OBSERVING since 01:49 UTC (13 hr lock!)

Just entered EQ mode due to 4 back-to-back earthquakes arriving from Mexico (4.6-5.1 Mag).

Planning on going into Calibration + Comissioning time at 8:20PST (15:30 UTC).

TITLE: 09/12 Eve Shift: 2330-0500 UTC (1630-2200 PST), all times posted in UTC
STATE of H1: Observing at 149Mpc
INCOMING OPERATOR: TJ
SHIFT SUMMARY:
Short 48 min lock, when the shift started, then let H1 try to lock itself. After 5 locklosses and an initial alignment we got to NLN.

1:47 UTC GRB-Short E511072
1:47 UTC GRB-Short E511073
1:48 UTC GRB-Short E511072
 
1:49 UTC Nominal_Low_Noise reached
1:49 UTC Observing reached


LOG:
No Log


PS here is a rainbow.

TITLE: 09/11 Eve Shift: 2330-0500 UTC (1630-2200 PST), all times posted in UTC
STATE of H1: Observing at 147Mpc
OUTGOING OPERATOR: Ibrahim
CURRENT ENVIRONMENT:
    SEI_ENV state: CALM
    Wind: 21mph Gusts, 15mph 3min avg
    Primary useism: 0.04 μm/s
    Secondary useism: 0.10 μm/s
QUICK SUMMARY:

IFO is locked ..... was locked for 48 min.
Everything was running smoothly until the lockloss of unknown cause.

 

TITLE: 09/11 Day Shift: 1430-2330 UTC (0730-1630 PST), all times posted in UTC
STATE of H1: Observing at 145Mpc
INCOMING OPERATOR: Tony
SHIFT SUMMARY:

Calibration Early Morning (8AM-10AM):

Lockloss during calibration with the probable reasons. Happened as soon as simulines ran an L1_SUSETMX injection. Either:

• It was the simulines measurement itself since this was recently changed, though reportedly everything about the excitations is the same
• It was the more sensitive IFO state during this measurement since we were running the cal sweep at a AS new offset (what offset).

So in short, no calibration work was done.

Squeeze Late Morning (10AM-12PM):

We got to NLN again at 17:17 UTC! Then a 6.3 EQ came though and we lost lock while the ground PEAKMON was reading 4.7 microns. We Initially thought we would make it but alas.

Sadly, no squeeze work was done either.

Earthquakey, Windstormy Quiet Noon (12-4:30PM)

After this 6.3 EQ, we were hit by two more 5.8’s from Vanuatu while wind speeds steadily went over 30mph from 19:20 UTC to 20:20 UTC. We stayed in this unstable state for about 3 hours losing lock approximately 20 times pre-DRMI. Wind peaked at 38Mph but once it went below 28ish Mph, we were able to lock, and quite quickly as well. The successful acquisition run only took 50 minutes!

As though the tree has to fall when nobody is listening, I left the room for 40 minutes for the OPS meeting and I came back to a fully automatically locked IFO in NLN. We were in OBSERVING minutes later. I guess the wind also came down by 10mph avg in that time so scientifically, it was probably that.

Initial Alignment Weirdness: Ongoing Issue

During initial alignment SRC align and only since making the SR3 move, SRM, SR2, SQUEEZE_OUT and IFO_OUT saturate at times where they are not known to do so. Looking closely at the ASC AS_A DC_SUM_OUT, this happens when the counts are sufficiently high but a glitch occurs that misaligns SRM very badly. Ryan C found a temp solution by going into IFO_NOTIFY and then pausing at PREP_FOR_SRY for a few seconds.

Sheila and I investigated very briefly and found that the ASC trigger signal malfunctions sometimes by activating when it is much lower than its threshold for activating. More investigation and monitoring to come.

Other:

IMC Gain Redistribution during LASER_NOISE_SUPPRESSION worked! I’ve unmonitored its SDF (by instruction) and we’re testing/monitoring it. SDF Screenshot attached.

LOG:

Samantha Callos, Robert Schofield

August 30, 2024
CER ACs turned off and on at the following times:

CER Bank 1 (Upstairs)

• Off 22:44 UTC
• On 22:40 UTC

CER Bank 2 (Downstairs)

• Off 22:58 UTC
• On 23:04 UTC

The new Daikin heat pump that serves the clean side of the VPW has now been fully tied into the building and is running. One of the two cooling circuits appears to be DOA as it is completely flat. In the coming days Facilities team will pull a vacuum on the problematic lineset to assess whether it has a leak and if so, to what degree. In the meantime, the unit will still heat and, to a lesser degree, cool the spaces.

E. Otterman C. Soike T. Guidry

We've been experiencing locking issues the whole day, briefly reaching NLN twice for <30mins. Here's why:

Lock 1: Calib Injection cause

Lock 2: 6.3 EQ Cause

Lock Attempt 3: 38Mph Gusts + Back to back 5.8 mag EQs are preventing ALS from locking. We've lost lock an approximate 8 times now on the way to DRMI, only grabbing DRMI once for <1 minute.

Specifics (to be repeated in the summary alog) below:

Calibration Early Morning Comissioning 8AM-10AM:

Lockloss during calibration with the probable reasons. Happened as soon as simulines ran an L1_SUSETMX injection ran. Either:

• It was the simulines measurement itself since this was recently changed, though reportedly everything about the excitations is the same
• It was the more sensitive IFO state during this measurement since we were running the cal sweep at a AS new offset (what offset).

So in short, no calibration work was done.

Squeeze Late Morning 10AM-12PM:

We got to NLN again at 17:17 UTC! Then a 6.3 EQ came though and we lost lock while the ground PEAKMON was reading 4.7 microns. We Initially thought we would make it but alas.

Sadly, no squeeze work was done either.

Earthquakey, Windstormy Quiet Noon 12PM-Now

After this 6.3 EQ, we were hit by two more 5.8’s from Vanuatu while wind speeds steadily went over 30mph from 19:20 UTC to 20:20 UTC.

We've done one initial alignment, having issues there too at SRC align, which has a weird glitch that doesn't let SRM catch only sometimes also causing SQZ_OUT and SR2/SRM to saturate and even trip SRM WD once. Pausing ALIGN_IFO at PREP_FOR_SRY seems to help but for unknown reasons so far.

Where the EQ has calmed down, reducing the ground motion to a lockable state, the wind speeds have picked up past 35mph (which statistically means >2hr lock 95% of the time). ALS is barely locking and we haven't made it past Locking_ALS since 20:02 UTC (1hr 48 mins ago).

Sheila, Louis, Francisco

We changed the ASC-AS_A_DC_YAW_OFFSET from -0.15 to -0.3 and saw an increase in ASC-DHARD_Y power spectrum in the 10-30 Hz range.

Originally, we planned to make simuline and pcaly bb injections with two different offset values for WFS given a thermalized and locked IFO. Having a non-thermalized IFO, we decided to make calibration measurements after changing the WFS offset, reverting the offset, and make calibration measurements again. We lost lock during the calibration measurement. However, we used DTT to see the data from the different offset values during NOMINAL_LOW_NOISE (as seen in AS_A_DC_YAW_OFFSET).

DHARD_DARM_PS_10-30_Hz shows the power spectrum from 10 Hz to 30 Hz for the channels of relevance. From this plot we see that the YAW dof is coupled to DARM such that the lines are visible. The increase in magnitude suggests that a big offset makes things worse, also seen in DHARD_DARM_TF_10-30_Hz, so we might try smaller offsets next time.

DTT template can be found in /ligo/home/francisco.llamas/A2L/DHARD_DARM.xml

Yesterday we added the remote access ioc (RACCESS) uid channels to the DAQ so they can be trended.

I have written a progam called who_is_logged_in which takes any gpstime format and lists who was logged into CDS remotely at that time.

The epoch for these channels is noon Tue 10 Sept 2024.

Example:

who_is_logged_in "17:00 yesterday"

Who is logged remotely into CDS on Tue Sep 10 17:00:00 2024 PDT

cdsssh    Number of users 5
cdsssh    elenna.capote                 (1 session)
cdsssh    erik.vonreis                  (1 session)
cdsssh    ezekiel.dohmen                (1 session)
cdsssh    gerardo.moreno                (1 session)
cdsssh    louis.dartez                  (2 sessions)

cdslogin  Number of users 1
cdslogin  david.barker                  (1 session)

The code uses minute trend data, so you will get the login list rounded to the minute.

If you ask for who is logged in now, it defaults to 10 minutes ago because minute trend data is not immediately available, e.g.

who_is_logged_in now

Who is logged remotely into CDS on Wed Sep 11 13:32:16 2024 PDT

cdsssh    Number of users 7
cdsssh    elenna.capote                 (1 session)
cdsssh    erik.vonreis                  (1 session)
cdsssh    ezekiel.dohmen                (1 session)
cdsssh    gerardo.moreno                (1 session)
cdsssh    jim.warner                    (1 session)
cdsssh    louis.dartez                  (2 sessions)
cdsssh    tyler.guidry                  (1 session)

cdslogin  Number of users 2
cdslogin  david.barker                  (1 session)
cdslogin  root                          (1 session)
 

TJ, Camilla.

We turned on the HWS lasers via medm and the LVEA chassis, they came back with the same powers we turned them off with (1.7mW for Y and 3.2mW for X). Since we realigned SR3 to the pre-April alignment yesterday 80028, both beams are on the CCD cameras, picture. The HWS plates are still removed from when we started alignment work in 78083. We will need to replace these before getting HWS signals, then restart HWS code.

CO2X chassis tripped off with the "RTD or IR alarm" error when I turned on the HWS SLED chassis (same rack). Turned back on with no issue (keyed off/on, gate button). We've seen similar behavior before. These CO2 chassis are planning an upgrade as per FRS6639.

I found the LVEA, clean receiving entrance and EE bay lights still on, turned them off but they would have been on during observing since yesterday, tagging Detchar.

There was an earthquake rumbling through the site from Papua New Guinea during commissioning this morning, so Sheila suggested we could try the IMC gain redistribution again, as we got several notifications of IMC splitmon saturations. To do this, we commented out the IMC power adjust decorator in the IMC_LOCK guardian, unfortunately  in the wrong state, so we again got bit by IMC_LOCK readjusting IMC fast gain while we were trying to change IMC fast gain and IN1 and IN2 gains. After losing lock and catching the error, the decorator was commented out in the ISS_ON state for IMC_LOCK, and the gain redistribution was added to ISC_LOCK's LASER_NOISE_SUPPRESSION state. Still waiting for the ground to calm down enough for the arms to stay locked, but we will try locking with the gain redistribution in this time.

I realized that there isn't an explict alog about this because Paul was asking, so I'm adding this trend to show that after the OFI repair our optical gain is better by about 2.1%, implying that the optical losses are reduced by 4.3%.  While the OM2 heaters is now off and was on before the vent, that didn't have a large impact on optical gain last time that we changed it in late June. 

Wed Sep 11 08:12:51 2024 INFO: Fill completed in 12min 47secs

WP12069 New CDS NAT router test

Jonathan:

The new CDS NAT router was tested successfully. It has been left in place as the production machine. Details in Jonathan's alog.

WP12067 Add PEM ADC to h1iscey

Fil, Erik, Dave:

The additional PEM ADC was added to h1iscey's IO Chassis. At the same time the failed A2 Adnaco backplane in the chassis was investigated.

After replacing backplanes and swapping fibers the problem was traced to the Adnaco adapter card in the front end computer, specifically the PCIe slot it was in (slot7, left-most as viewed from rear of computer).

h1iscey does not have any Contec binary IO cards, so it was possible to skip PCIe-slot7 and temporarily only use three Adnaco adapter cards.

h1iopiscey model was modified to add the new ADC. A DAQ restart was required

WP12076 Add VACSTAT EPICS channels to DAQ

Dave:

A new H1EPICS_VACSTAT.ini was added to the DAQ. EDC+DAQ restart required.

WP12075 Add RACCESS UID EPICS channels to DAQ

Dave:

H1EPICS_RACCESS.ini was expanded to add the UID channels. EDC+DAQ restart required.

WP12059 Add CDS HW Status EPICS channels to DAQ

Dave:

H1EPICS_CDSHWSTAT.ini was added to the DAQ. DAQ+EDC restart required.

DAQ Restart

Dave:

The DAQ was restarted 12:01 0-leg, 12:05 1-leg for the above changes. EDC was restarted at 12:02.

No problems with the restart, no second GDS restarts were required.

h1digivideo3 cameras briefly offline

Dave:

While tracing the digital video servers ethernet cabling I discovered the WorkStation-VLAN ethernet cable in eth0 of h1digivideo3 was not fully seated and its cameras went blue-screen.

After Erik alerted me to the issue, I pushed the ethernet cable and it reseated with a satisfying snap and the images started streaming again.

Vicky, Camilla

Repeated 66946 PSAMs changes with SQZ-OMC scans with a new method of dithering the PZT around the TEM02 mode and minimizing it. With this we improved the mode mismatch from 4% to 3%. It will interesting to see if these settings are still better in full lock. Plots of OMC scan attached and same plot zoomed on peaks attached.

• Increased SEED power from 0.7 to 75mW. SQZ_OPO_LR to LOCKED_SEED_DITHER. Started with 0.07mW on OPO_IR_PD but we're misaligned on SQZT7. This increased to 1.25mW when we move the ZM4/5 alignment. Will need to fix table alignment to match IFO alignment when we're happy.
• Took IMC to OFFLINE (mis-aligns MC2), aligned SRM, misaligned SR2.
• Note, that we should have (but didn't) misalign the ITMs to make the REFL signal cleaner.
• When we bring ZMs back to IFO alignment, SQZ beam is on AS_AIR (a little high).
• Turned on DC 3/4 centering loops and then OMC ASC. Have power on AS_A/B/C. 70 counts on A/B_DC_SUM.
• Locked OMC with SQZ beam: OMC_LOCK guardian to DOWN. Changed H1:OMC-LSC_LOCK_TRIGGER_THRESHOLD from 1.0 to 0.3. Manually scanned using  H1:OMC-PZT2_OFFSET. Watched OMC-DCPD_SUM_OUT_DQ and OMC TRANS camera (exposure increased). Autolocker off. When we found the resonance, turned H1:OMC-LSC_SERVO ON, gain to 1, INT and BOOST on.
• Sending in 1.25mW, expect 70% reflected off SRM would be ~0.9mW, get 0.95mW on AS_C_NSUM, get 0.64mW transmitted though OMC.

Took OMC scans using tempalte /sqz/h1/Templates/dtt/OMC_SCANS/Sept10_2024_PSAMS_OMC_scan_coldOM2.xml  Unlock OMC and H1:OMC-PZT2_OFFSET to -50 (nominal is-17) before starting scan.

Over 1V and under -1V on ZM5 is bad at most ZM4 strains (tested at 4.0V ZM4). For each step we adjusted ZM4 and then fine adjusted ZM5.