Although the data was for 9 hours or so, there seems to be no Type-1 noise, but Type-2 can be found. So Type-1 should be triggered by neoLase/PMC/MC operation. 

There seem to be two types glitches, as Fig.1 to 3. Fig.1 (Type-1) was well recognized before. Fig.2 (Type-2) shows very fast spike and no effect on PMC. Fig.3 (Type-3) shows the step down, and no effect on PMC.

How about the master laser power for the 3rd FBL?

Today, we tried all of amplifiers with 125V of a new voltage converter, and with interlock disabling adapter.

What voltage was used for them?

An interlock socket was actually attached when you tried 2nd FBL?

(Tanaka, Miyakawa)

We started removing the 2nd amplifier from the rack, and we located it on the floor this morning. We checked it and we realized that it did not pass the temperature test. Meanwhile, the first one passed the interlock test and temperature test in the 19-inch rack. At this point, we concluded that the 2nd one had been broken.

Then we continued the test of the 1st amplifier with the seed laser, but unfortunately, we could not see any high-power beam at the output (we should have ~1W when the power amplifier was enabled, even though the current is 0A, but we saw only 10mW that comes from the seed laser). At this point, we concluded that the 1st one had also been broken. 

Then we removed the 1st amplifier from the rack, and we put the 3rd one on the rack. Fortunately, we succeeded in enabling the power amplifier at this stage with ~1W output!

While we were aligning the 1W light to PMC, the laser suddenly stopped with a high-temperature warning. We noticed that we had not turned on the chiller, and turned on the chiller immediately but we did not connect the water tube yet. There was a water leak in the PSL room, specially around the PCs, but fortunately, the water did not go in the direction of the optical table.

After wiping the water, we continued the 3rd amplifire, and we had another water trouble. The connection of a water tube to the ampifier box failed. Metal screw threads had been deformed at the box side, and the nut threads also became deformed. We struggled for an hour with this trouble, and after scraping threads with a metal driver, we finally could connect the tube to the box.

After fixing water-related things, we continued the test of 3rd amplifier again; however, we could not see the high-power light that was seen before. We were concerned that the laser was broken somehow, but after removing the power supply cable and USB cable and waiting several minutes, we could see the high power again. This situation was very similar situation when we concluded the 1st one was broken before. 3rd one survived anyway, so there could be some hope that the 1st one could be recovered.

Anyway, we had ~1W light at this point. We could lock PMC, and we performed a fine alignment of PMC.

We measured the OLTF of PMC, and the UGF was ~kHz, and then we measured the frequency noise using the PMC. Graphs were already uploaded by Tanaka-kiun.

We left both the FB laser off and the neoLASE off tonight because the interlock is not ready for FB, and PMC is aligned to the FB laser.

Fig.1 shows the PEM-VOLT_PSL_TABLE_GND_OUT_DQ during today's activity. The voltage level seemed to decrease from before to after the activities. Actually, neoLase was stopped now.

Fig.2 shows the data from last Friday to Tuesday evening. During this duration, IMC was locked by using neoLase. Then we can recognize many enhancements. Around -2days, there was a step down-like change, even though no activities at the site. It is interesting to find some relation with other channels.

After today's work, neoLase was stopped (Of course, FBL was stopped.). Let's see what will happen or not tonight.

Miyakawa, Tanaka

We succeeded in locking PMC (Miyakawa-san will report the details later.) with the 3rd Nufern fiber amplifier. Then, we measured the UGF of the PMC loop. Fig. 1 shows the result. Current UGF is ~4.2 kHz, it is almost the same as previous UGF, 4.2 kHz with NeoLASE according to klog35055.

Then we measured the frequency noise with the same procedure in klog35280. Fig. 2 shows the result, red curve is the noise with fiber amp. and blue curve is the noise with NeoLASE. According to each UGF, we can compared them below 4 kHz. The noise amplitude with fiber amp. seems to be half at 2 kHz.

Also, Some sturacuture in NeoLASE noise around 250 Hz seems to disapper in the noise of fiber amp. However, the noise in this region is able to be affected by sounds or voices. And also, the noise was measured in the daytime (7:05 UTC, Oct. 8, 2025). Therefore, I’m not sure that the noise in this region is the actual frequency noise or not.

It seems that the GND glitches of the PSL optical table are caused by the IMC trying to relock.

As shown in this plot, by zooming into a glitch, we can see that the GND level changes until the IMC lock is recovered. Two large spike like glitches appear in the PMC error signal at the beginning of this lock loss.

By zooming into the initial spike of the PMC error, we can see that the GND level started to change after the spike appeared in the PMC error.

We can also see that the IMC slow out and the PMC PZT drifted a lot preceeding the glitch.

I checked several such glitches. This behavior (preceeding drift) does not always happen (it seems to happen more than 50% of times, but not always).

I also picked a noisy period of laser and compared the spectra of the PSL GND with a quiet period.

The laser gets noisy after GPS=1444342694.

The following plot shows the spectra measured in this noisy period versus the ones taken in a quiet period (a few tens of seconds before the noisy period).

https://klog.icrr.u-tokyo.ac.jp/osl/uploads/35320_1760323593_LAS-Noisy-Spe.pdf

The IMC error signal gets about x10 larger. Also the PMC error gets a bit larger around 800Hz and above. However, there is no difference in the PSL GND noise. It may have difference at higher frequencies but the sampling rate of the PSL GND is 2kHz.

In conclusion, the GND noise does not seem to be the cause of the laser noise. Rather the IMC servo is the cause of the GND noise.

[Kimura and Yasui]
 On Oct. 10, we took one gate valve out of the central room for modification. 
This gate valve will be transported to the VAT factory and modified into a gate valve with a viewing window.

Effect of Pylon Cutting on Structural Resonance (Preliminary Simulation Result)

I performed a preliminary structural simulation to evaluate whether cutting a part of the Ncal pylon foot would affect the resonance characteristics of the overall structure.

What was modeled

• A simplified CAD model was created including: pylon, vacuum chamber, rotating rotor, and a simplified structures on the chamber(motor, ferrofluidic seal).
• Material assumptions:
  • Pylon & chamber: Steel (generic structural steel)
  • Rotor: Aluminum + Tungsten
  • Upper structures: Aluminum
• Boundary condition: the pylon was constrained using six small cylinders at the base, representing the six anchoring bolts.

Simulation method

The CAD model was imported into COMSOL, and the eigenmode frequencies were calculated before and after the cutting of the pylon foot.

Main result (frequency comparison)

Conclusion

The impact of the cutting on the structural resonance is very limited. All mode frequency shifts are less than 0.4 Hz, which is negligible in this simplified model.

Important notes

• This simulation used a simplified model and the resonance frequencies should be treated as reference values.
• The upper structures on the chamber were hightly simplified.
• The pylon interior was assumed hollow (use of sandbags is under discussion).

Attachments

COMSOL report files are attached for reference.

Operators: Yuzurihara, Ikeda
Shift time: 9:00-17:00 JST
 
Check Items:
VAC: No issues were found.
CRY cooler: No issues were found.
Compressor: No issues were found.
Temp check: No issues were found.
GAS filter output: No issues were found.

IFO state (JST):
9:00 STANDBY
17:00 STANDBY
 

Fig.1. shows the clear correlation between glitches in PSL-PMC_MIXER_MON_OUT_DQ and the PSL room table ground PEM-VOLT_PSL_GND_OUT_DQ.

Its a good hint to remove these glitches.

Please check the sound around the PSL room related to the water pump, etc

Operators: Hirose, Ikeda
Shift time: 9:00-17:00 JST
 
Check Items:
VAC: No issues were found.
CRY cooler: No issues were found.
Compressor: No issues were found.
Temp check: At the 10:00 check, the PSL changed by more than 0.3 degrees, which is thought to be due to laser-related work performed in the PSL room the previous day. Additionally, the IYA also exceeded the range, but this was attributed to the effects of outside air intake. Hirose-san has already notified Miyoki-san and Hayakawa-san via email.
At the 16:00 check, it was stable.
GAS filter output:   No issues were found.

IFO state (JST):
9:00 STANDBY
17:00 STANDBY

(Tanaka, Miyakawa)

Today, we tried to use a voltage booster for the power supply. Until yesterday, we tried almost everything we could think of. One of the remaining common things was the power supply. We suspected the voltage was too low.

First, we measured the current voltage of the power supply in the PSL room. It was ~97V, now perfect but not terribly low. Then we prepared 120V (actually 119.7V) and used it for the 3rd amplifier on the floor. At the first trial, we thought it failed as same as before, but we noticed that the interlock disable adapter was not attached. We tried again with the adapter and the temperature showed normal ~25 degrees! instead of strange 3 degrees as before.

Then we tested the first amplifier on the floor with 120V, which was thought as broken, and the first one also showed normal ~25 degrees!

And surprisingly, the first amplifier worked with 100V again! We are not sure of the reason, but one possibility was that something was triggered to be alive again when it was raised up with 120V, for example capacitor or something, but just a guess.

However, the second amplifier was not recovered, even with 120V. One possibility is that the second amplifier is mounted in the 19-inch rack. The GND pin at the power supply adapter is connected to the chassis,  and the chassis is connected to the rack. We tried to float the chassis from the rack to put some insulation sheets, and the resistance changed from ~0.2 Ohm to a few M Ohm, but the situation was the same; the second one did not work. We expected that the second one would not be broken, and would not be different from the first one, so we wanted to use the second one as the default amplifier next. 

Exchanging the amplifier on the rack for another one will take for a while, especially to deal with water tubes, so we stopped the work today. Next week, we will start from exchanging the amplifier on the rack.

The seismometers' signal does not look to be saturated

Operators: Washimi, Dan
Shift time: 9:00-17:00 JST
 
Check Items:
VAC: No issues were found.
CRY cooler: No issues were found.
Compressor: No issues were found.
Temp check: T in PSL changed more than 0.3 degree, but this should be caused by the laser related works in the room.
GAS filter output:   No issues were found.

IFO state (JST):
9:00 STANDBY
  Commissioning
16:00 STANDBY

https://earthquake.usgs.gov/earthquakes/eventpage/us6000rfwz/executive

2025-10-10 01:43:59 (UTC)

7.265°N
126.755°E
58.1 km depth

If possible, a quick test using a temporary 120VAC supply from the breaker could be an option to check whether the fiber amp operates properly before the commercial UPS arrives. Can Kimura-san or one of tech persons take a 120VAC line out from the breaker??

I checked the lightning list.

There were many lightning strokes around this season, but not at 13:58 UTC.