TITLE: 09/16 Day Shift: 1430-2330 UTC (0730-1630 PST), all times posted in UTC
STATE of H1: Corrective Maintenance
INCOMING OPERATOR: Corey
SHIFT SUMMARY: Lots of work today to understand our troubled IFO as well as some maintenance items. We perhaps have narrowed down the change from the power outage and have changed the input power into the PMC to bring us back to a similar place. We have run an initial alignment, which probably could have run automated, and we are now going through PRMI/DRMI for a second time. I accepted some SDFs in safe before we started locking, see screenshots attached.

BRSY was rung up during some work at EY today, and doesn't seem to be damping. I've contacted Jim, but we will keep an eye on it.

LOG:

• 2154 Started initial alignment, no issues
• 2223 Started main locking

On the night of Sept 10 the IMC REFL power slowly increased all night. To avoid this happening in the future, we decided to have H1_MANAGER check this and kill the lock before calling someone. This is the first instance in guardian, that I'm aware of, that we intentially kill the lock.

This takes place in the form of a decorator in the Low_Noise state of H1_MANAGER. The threshold is currently set to 35, and if it goes above then the node will kill the lock and move ISC_LOCK to IDLE before going to ASSISTANCE_REQUIRED itself.

We tested it while not in low noise with an offset in IMC_REFL and the decorator moved to the Relocking state of H1_MANAGER.

TITLE: 09/16 Eve Shift: 2330-0500 UTC (1630-2200 PST), all times posted in UTC
STATE of H1: Corrective Maintenance
OUTGOING OPERATOR: TJ
CURRENT ENVIRONMENT:
    SEI_ENV state: CALM
    Wind: 12mph Gusts, 7mph 3min avg
    Primary useism: 0.03 μm/s
    Secondary useism: 0.13 μm/s
QUICK SUMMARY:

Site Power Outage recovery continues (with H1 down, but it was  up last night/18hrs ago for almost 3hrs).  Saw H1 locking DRMI as I walked in but it lost lock...but it looks like this was intentional---running Guardian tests to return for being able to have Owl Shift calls.

BRSy is NOT In USE (It's been noisy since around the beginning of Maintenance today)---TJ says it might get a Beckhoff reboot in the morning.

We have been doing some things by hand to lock recently.  I've put a flag in CARM_5_PM to do these. 

This will take the TR_CARM offset to -52 (instead of the caclulated offset), waits, then increases the TR_CARM to 2.1, then sets the offset to -56, increases the DHARD P gain to -30 and Y to -40, and increases the input matrix element for REFLAIR 27 I to PRCL by a factor of 1.6. 

Of these things, I think that the DHARD gain increases should probably be a good thing that we would want to keep in the long run.  The PRCL gain we plan to check early in the locking sequence. 

R. Short, J. Oberling

Today we tuned up the RefCav alignment for the PSL FSS, since the TPD had been doing its usual wander and also had a drop after last week's power outage.  As usual, we began with a power budget:

• FSS In: 266.3 mW
• AOM In: 267.2 mW
• AOM Out, Single Pass: 170.1 mW
  • Single Pass Diffraction Efficiency: 63.7%
• AOM Out, Double Pass: 119.4 mW
  • Double Pass Diffraction Efficiency: 70.2%
• EOM Out: 117.1 mW

The AOM signle pass diffraction is lower than usual, so we adjusted the AOM alignment to improve it.  While we were doing this we kept seeing small drops in the AOM output that we couldn't recover with alignment, which was making us think that we were walking the alignment off.  Turns out the PMC transmission was slightly dropping.  We remeasured the power incident on the AOM and sure enough, it had dropped a little, to 265.1 mW.  The diffraction efficiency looked good, so we stopped moving the AOM and moved on with adjusting M21 to improve the double pass diffraction (always have to adjust this mirror if the AOM is moved).  The results:

• AOM Out, Single Pass: 189.6 mW
  • Single Pass Diffraction Efficiency: 71.5%
• AOM Out, Double Pass: 133.6 mW
  • Double Pass Diffraction Efficiency: 70.5%
• EOM Out: 131.9 mW

There was no clipping through the EOM, but we adjusted it to better center the beam through the input and output apertures.

In recovering the RefCav, we decided to check beam alignment on the transmission PD, so I set up a LEMO splitter so we could see the DC voltage on a voltmeter.  The RefCav locked without much issue, and Ryan tweaked the beam alignment with the picomotor mirrrors.  Interestingly, the TPD voltage in MEDM did not match the voltage read by the voltmeter, so a recalibration there was necessary.  The TPD readings were not close to each other, at all, so we trended back to see the last time the TPD was calibrated.  Turns out this was at original PSL install in 2012, it hasn't been updated since.  Well then.  Alignment tweaking results:

• Initial TPD
  • MEDM: 0.508 V
  • Voltmeter: -0.320 V
• Final TPD
  • MEDM: 0.850 V
  • Voltmeter: -0.532 V

The negative sign is normal on these PDs from AEI, as they were designed to output a negative voltage; the sign gets flipped in the MEDM filter module.  I tweaked the beam alignment on the TPD, which increased the voltmeter voltage to -0.535 V, so it was decently aligned already.  We recalibrated the MEDM reading to match the voltmeter, and updated the offset to cancel out the small dark offset observed when the TPD was blocked.  Results:

• Gain
  • Old: -0.000336
  • New: -0.0002063
• Offset
  • Old: 130
  • New: 74.02

We then aligned the beam onto the RFPD and measured the RefCav visibility:

• Unlocked: 1.161 V
• Locked: 0.181 V
• Visibility: 84.4%

We were done in the enclosure at this point, so we exited and returned the enclosure to Science Mode.  Outside, we set up a network analyzer to measure the FSS TF.  With a common gain of 14 the FSS UGF was ~380 kHz.  We increased the common gain to 15 and the UGF was 445.6 kHz.  This is closer to where we generally like it, so we left the common gain at 15 (this can be reverted should there be an issue).  Ryan has a picture of the TF that he'll post as a comment to this alog.  With the lower, but now correct, TPD in MEDM, a couple of guardian thresholds required updating.  The TPD light on the Ops Overview screen now comes on if the TPD is less than 0.4 V and the "FSS Transmission Low" warning in DIAG_MAIN now comes on when the TPD is less than 0.3 V.  We'll monitor this over the coming days/weeks and adjust as needed.  This closes WP 12796.

LVEA was swept and everything looks good

Jason Ryan Jenne Daniel

We reduced the PSL power after the amplifier and before the AOM by a factor of 2.3. We then re-adjusted the power into the IMC to get back to 2W. This reduced the IMC reflected light power by ~2.7. This seems to strongly indicate that we have a heating issue in the path from the PMC to the polarizer. The following tests were run:

1. Normal PSL setup (first screen shot)
2. PMC input power reduced: the power was reduced with the waveplate before the ISS EOM (which is before the PMC)
3. EOM input power reduced: the power was reduced with the waveplate between the PMC and the EOM
4. Same as above but with enough power to lock th einterferometer at full power. The maximum IMC input power is now 70W.

The summary page finally created the omicron glitch plot for last night's 3 hours of lock. On Friday we halved the power on the IMC refl diode. Last night we locked for three hours. Here is the glitch rate from last night, which can be compared to our 10 hour lock just after the power outage which had a significantly increased glitch rate.. As a reference, here is what the glitch rate looked like a few days before the power outage. It appears the glitch rate is back to the normal level.

Camilla

Followed setup instructions from 80010, last done in 86445. Increased to 75mW injected in the SEED beam, we had 1mW on OPO_IR_PD_DC. Have counts of ~60 and 66 on ASC-AS_A and B and 7e-4 on ASC-OMC-A and B NSUMs all similar to the August 19th measurement. To turn on the OMC ASC, H1:OMC-ASC_MASTERGAIN to 0.02. Took H1:OMC-PZT2_OFFSET goes down to -50 to start and then ran a template /sqz/h1/Templates/dtt/OMC_SCANS/ Sept16_2025_PSAMS_OMC_scan_coldOM2.xml  ref 35,36.

PSAMS settings at nominal ZM4/ZM5 6.0V / -0/4V. Measured mode mismatch of TEM02 of 2.975%. In August 86445, this was only 2.174% however in that measurement, the OMC ASC has been left off so there would have been more in the misalignment peaks.

*calculated with TEM02 / (TEM00 + TEM02)

I tried to get a better measurement of the ETMX charge per its OPLEVs this morning, I started with a higher gain on the ESD_OUTPUT filters than the last time (0.75 instead of 0.5, 1 worked until recently). This measurement had pretty low coherence like last week so I ctrl+c and increased the drive_amp by 20%. This measurement was a little better but still not as good as I would like it, and as we have had in the past, so I restarted it again with another 20% increase in drive_amp, I ran 3 measurements with this configuration. The prelimary processing looked decent but the full processing revealed pretty large error bars still.

This measurement is better than last weeks, but there is still room for improvement. I brought the drive_amp up from 11000 to 20600 with ESD_OUTPUT gains of 0.75 and I still wasn't saturating so I still have more room to push to improve the measurements and increase the coherence.

Here is a side by side of powers at the ports before and after the power outage. This is using last night's lock at 2 hours 55 minutes after the end of the max power, versus a lock before the power outage at 2 hours 55 minutes from max power.

It seems like the input power, POP, LSC REFL and circulating power numbers hang together. The OMC refl, as AS_C numbers also hang together. Sheila and I discussed that we would expect if the arm power increased the kappa c would increase, but it appears that arm power increased but kappa c decreased following the OMC refl and AS_C values.

Sheila, Elenna, Camilla. WP#12797 . IOT2L layout D1300357

We took the PSL input power down to ~100mW, locked out the rotation stage and then used the nano scan to take some beam profiles in the IMC REFL path on IOT2L. Elenna and Sheila also moved a beamdump to fully block the MC REFL rejected beam that was getting on the MC REFL camera. 

By eye the beam looked gaussian and the numbers show it is mainly symmetric. Maybe last measurements where when the WFS were being installed in 2013 6439. 

For positions B and C, we added a temporary steering mirror between IO_MCR_M7 and IO_MCR_L2. Distance between IO_MCR_BS1 and IO_MCR_M7 = 7"; Distance bewtween IO_MCR_M7 = 7" and temporary steering mirror = 1 1/4". 

Tue Sep 16 10:07:25 2025 INFO: Fill completed in 7min 21secs

Literally everyone (to name who I know/can recall would be unfair)

TITLE: 09/16 Eve Shift: 2330-0500 UTC (1630-2200 PST), all times posted in UTC
STATE of H1: IDLE
INCOMING OPERATOR: NONE
SHIFT SUMMARY:

IFO is in IDLE and DOWN for CORRECTIVE MAINTENANCE  buuuuut IFO was OBSERVING for ~1 hr.

Lockloss was intentional in order to avoid potential harmful lockloses and issues throughout the night. We actually got to NLN and OBSERVING though!

It's not exactly over yet - because we have only been locked for 2 hours. Problems may be there and dormant. Nevertheless, we got here from a outrageous outage recovery and a nasty ISC/Lock reacquisition that took 5 days to get back into observing.

The main things:

• we don't know our glitch rate yet and omicron is being slow/not showing anything. It's having trouble as confirmed by Jenne and DetChar.
• ARM_TRANS is different despite same power output, meaning likely losses elsewhere - this is why we're going IDLE for the night
  • X: 1716.84(then) to 1655.87 (now) - Dif = -60.97 % change: 3.7% more pre-outage|| Y: 1530.75 (then) to 1472.84 (now) - Dif = -57.91, % change: 4% more pre-outage

How we got to NLN:

Alog 86951 summarizes the lock acquisition. We just sat there for a bit at OMC_WHITENING as violins fell (and continued to)

After NLN:

• Mass acceptance of 127 SDFs (all attached, all accepted)
• IMC REFL still stable 3 hrs into MAX POWER
• Broadband PCAL run at 2hrs 2hrs 55 mins thermalized (alog 86955)
• I killed the lock via the IMC SERVO sign switch

LOG:

None

Start Time: 1442034629

End Time: 1442034940

2025-09-15 22:15:22,374 bb measurement complete.
2025-09-15 22:15:22,375 bb output: /ligo/groups/cal/H1/measurements/PCALY2DARM_BB/PCALY2DARM_BB_20250916T051011Z.xml
2025-09-15 22:15:22,375 all measurements complete

J. Oberling, K. Kawabe

This afternoon we went into the PSL enclosure to inspect things after last week's power outage.  We concentrated on the IOO side of the PSL table, downstream of the PMC.  Our results:

We did a visual inspection with both the IR viewer and the IR-sensitive Nikon camera and did not find any obvious signs of damage on any of the optical surfaces we had access to; the only ones we couldn't see were the crystal surfaces inside the ISC EOM, we could see everything else.  I looked at the optics between Amp2 and the PMC and everything there looked normal, no signs of anything amiss.

While the beam was not perfectly centered on every optic, we saw no clipping anywhere in the beam path.  The irises after the bottom periscope mirror were not well centered, but they've been that way for a while so we didn't have a good reference for assessing alignment in that path (these irises were set after the O4 PSL upgrade, but there have been a couple of alignment shifts since then and the irises were not reset).  For reference, the beam is in the -X direction on the HWP in the power control rotation stage and in the -Z direction (but centered horizontally) on the PZT mirror after the power control stage.  We do have a good alignment reference on the ALS path (picked off through mirror IO_MB_M2, the mirror just before the ISC EOM), as those were set as part of the HAM1 realignment during this year's vent.  By my eye the first iris looked a tiny bit off in yaw (-Y direction) and pitch (+Z direction), while the second iris looked perfectly centered.  We found this odd, so Keita used the IR-sensitive camera to get a better angle on both irises and took some pictures.  With the better angle the beam looked well centered in yaw and maybe a little off in pitch (+Z direction) on that first iris, so I think my eye was influenced by the angle from which I was viewing the iris.  The second iris still looked very well centered.  Edit to add: Since the ALS path alignment looks good, to me this signals that there was not an appreciable alignment shift as a result of the change in PMC temperature.  If the PMC was the source of the alignment shift we would see it in both the main IFO and ALS paths.  If there is a misalignment in the main IFO path, its source is not the PMC.  Upon further reflection, a more accurate statement is:  If the PMC is the source of an alignment shift, the shift is too small to be seen on the PSL table (but not necessarily too small to be seen by the IMC).

The other spot of note is the entrance aperture for the ISC EOM.  It's really bright so it's hard to make a definitive determination, but it could be argued there's a very slight misalignment going into the ISC EOM.  I couldn't make anything out with the IR viewer, but Keita's picture shows the typical ring around the aperture a little brighter on the left side versus the right.  Despite this, there is no clipping in the beam, as we set up a WinCAM beam profiler to check.

The WinCAM was set behind IO_AB_L4, which is the lens immediately behind the bottom periscope mirror (this is the path that goes to the IMC_IN PD).  The attached picture shows what the beam looks like there.  No signs of clipping in the beam, so it's clearing all the apertures in the beam path.  I recall doing a similar measurement at this spot several years ago, but a quick alog search yields nothing.  I'll do a deeper dive tomorrow and add a comment should I find anything.

So to summarize, we saw no signs of damage to any visible optical surfaces.  We saw no clear evidence of a misalignment in the beam; the ALS path looks good, and nothing in the main IFO path looks suspicious outside of the ISC EOM entrance aperture (a lack of good alignment irises makes this a little difficult to assess; once we get the IFO back to a good alignment we should reset those irises).  We saw no clipping in the beam.

Keita has many pictures that he will post as a comment to this log.

I don't really think this is related to the poor range, but it seems that one of the cps on HAM3 has excess high frequency noise and has been noisy for a while.

First image is 30+ day trends of the 65-100hz nad 130-200hz blrms for the HAM3 cps. Something happened  about 30 days ago that cause the H2 cps to get noisy at higher frequency.

Second image are rz location trends for all the HAM ISI for the last day around the power outage. HAM3 shows more rz noise after the power outage.

Last image are asds comparing HAM2 and HAM3 horizontal CPS. HAM3 H2 shows much more noise above 200hz.

Since finding this, I've tried power cycling the CPS on HAM3 and reseating the card, but that so far has not fixed the noise. Since this has been going for a while, I will wait until maintenance to try to either fix or replace the card for this CPS.