[mTakahashi, Yasui]

We tried a nitrogen purge to promote H2O removal.
9:37 9.3e-1 Pa
9:44 stopped the dry pump
9:58~ N2 purge
10:09 8.8e3 Pa
10:12 started the dry pump
13:14 24 Pa(SRM), 9.2e-2 Pa(OMMT)
13:15 GVommt open->14Pa(SRM), 12Pa(OMMT)
13:16 started the TMP
So far, the pressure has decreased faster than in the previous trial on April 24.
It took 1 hour 8 minutes to reach 4e-4 Pa from 1e-2 Pa this time, whereas it took 10.5 hours last time.  The nitrogen purge seems to have been effective.

[Tanaka, Saito, Hirose]

Yesterday, as preparation for klog36845, we performed the initial alignment of Xarm and Yarm. Then, we noticed that the feedback signal for the OPLEV control of the PITYAW on SR3 had changed significantly compared to the previous Yarm flash. We confirmed that it has been specifically moved to 4/23 PM1:50JST(fig1). When we performed Lissajous the PITYAW on SR3 so that Yarm would flash again, the optimal YAW value had changed from -9 urad to 55 urad(55urad: fig2).

[Tanaka, Hirose, Saito]

We constructed the PLL control system in which a feedback signal was generated using a mixer, a local oscillator, a SR560, and a Moku:Lab, and fed back to the PZT of the sub-laser. Unfortunately, the control system did not function successfully.
 

• Since the previous observation of the beat signal (klog:36835), the sub-laser had remained on with its temperature kept at 31.57 ℃. Under this condition, a beat signal was observed; however, its amplitude changed significantly, and its frequency drifted by about 5 MHz over approximately one minute. Because the frequency was close to 0 Hz and the range over which it could be further reduced was limited, we increased the sub-laser temperature to 31.60 ℃ to raise the beat signal frequency to about 112.1 MHz (Photo 1). Next, this beat signal and a local oscillator signal (112.1 MHz, 10 dBm) were input to the mixer to generate an error signal. The mixer output was then fed into the SR560, where it was amplified and passed through a first-order low-pass filter with a cutoff frequency of 1 MHz. The output of the SR560 was subsequently input to the Moku:Lab, where an integrator was applied using a PID controller (Photo 2). The output of the Moku:Lab was connected to the PZT of the sub-laser, and the overall gain was varied. However, the error signal did not decrease, and no change was observed in the feedback signal. The error signal amplitude was 1.090 V; however, when the local oscillator frequency was set to 90 MHz, it became 409.5 mV, and at 80 MHz, it became −154.0 mV, indicating that it varied with the local oscillator frequency (Photo 3). On the other hand, changing the temperature of the sub-laser did not result in any noticeable change in the error signal amplitude. This suggests that the signal being treated as the error signal may in fact be a different signal. It is also possible that the Moku:Lab was operated incorrectly.

[R.Takahashi, Washimi, Kimura, Yasui, M.Takahashi, Sawada]

We reinstalled the geophone pods. Additionally, we adjusted the tilt of the FLDACCs. The feedback signals to the proof mass(COILOUTF) became smaller: +12200 to +1400 for H1, +12500 to -500, and -4900 to +2200 in count, respectively.

I checked the IM transfer functions again at 10Pa. The DC gain was smaller than the reference a little (1~2dB).

I checked the TM transfer functions after closing the top flange. The Length Oplev was not aligned well, so the DC gain of the transfer function for L was smaller than the reference.

I checked the GAS transfer functions after closing the top flange. They were consistent with the references.

I checked the IP transfer functions after closing the top flange. IDAMP signals are not treated well (just summation of ACC+FLDACC+LVDT), so please check the LVDT responses in BLEND signals. The resonant frequencies of the IP were 63mHz for L, 74mHz for T, and 0.32Hz for Y, respectively.

I checked the IM transfer functions after closing the top flange. They are consistent with the references except for L and H1 (the DC gain is 3dB larger than the reference due to the replacement of the satellite box).

With H.Yamaguchi, M.Onishi, S.Matsuo

We transported the new Pcal-Y laser from Toyama University to Y-end.
For access to Y-end, we used the Mozumi entrance.

The laser is currently placed under the duct near the EYA chamber.

I checked the signals of the IM OSEMs during the pump down. Just after starting the pressurization from 2.0Pa, the signals in count rapidly increased in 1 min.

[Tanaka, Hirose, Saito]

We matched the polarization of the sub-laser beam to that of the beam coming from the interferometer. Next, we increased the power of the beam coming from the interferometer by opening the iris. Then, we adjusted the alignment so that the sub-laser beam overlapped with the beam coming from the interferometer at the mirror just before the RFPD. As a result, we were able to observe the beat signal between the beam coming from the interferometer and the sub-laser beam.
 

• First, we checked the polarization of the sub-laser by inserting a PBS after the FI and found that it was mostly P-polarized. Since the beam coming from the interferometer is S-polarized, we installed a HWP after the FI and used the PBS to adjust the polarization to S-polarization. Next, the iris used for alignment had been partially closed; by opening it, the power of the beam coming from the interferometer incident on the RFPD increased from about 2 μW to about 15 μW. Then, with both the beam coming from the interferometer and the sub-laser beam incident on the RFPD, we found that their spot positions differed at the mirror just before the RFPD. We adjusted the alignment so that the sub-laser beam overlapped with the beam coming from the interferometer. At this time, observing the DC output of the RFPD with the oscilloscope on Moku:Lab, we found an offset of about 12 mV; the beam coming from the interferometer contributed about 29 mV and the sub-laser beam about 83 mV. Next, we attempted to observe the beat signal using the spectrum analyzer on Moku:Lab. When we set the sub-laser temperature to 31.57 ℃ to match the main laser frequency, a beat signal was observed (Photo 1). The amplitude of the beat signal fluctuated significantly, and its frequency changed by about 3.5 MHz over the course of approximately one minute.

[Kimura and Yasui]

 Attached is the residual gas distribution during the SRM build-up test using the Qmass installed in the OMMT.

Since the pressure inside the Qmass analysis tube exceeded 1×10² Pa, the Qmass measurement was interrupted at 12:45 p.m. due to the interlock mechanism.
 

[Kimura, Yasui]
1. Build-up test
We had the build-up test by closing the GV between the duct and the pumping unit.
The slope of the pressure rise rate was approximately 0.6, which is the same rate as yesterday.
During the build-up test, I checked the rate of the pressure decrease with the TMP alone.
According to our logbook, when we tried to swich the vacuum pump from TMP to IP on 7 January 2026, we closed the GV between duct and pumping unit. At that time, the pressure value immediately droped from 2E-5 to 6E-6 Pa. Sine it took 1hour 47minites to reach the pressureof 6E-6Pa from 2E-5Pa this time, there seems to be something affecting the TMP performance.

2. Pressurization
We pressurized SRM again.
13:27 duct-side GV open
13:33 start pressurization 2.0Pa (cylinder pressure: 11.8 MPa)
13:40 3.0E3 Pa
13:57 1.0E4Pa
14:05 1.4E4pa
14:23 2.3E4Pa
14:37 3.0E4Pa
15:00 4.1E4Pa
15:08 4.7E4Pa
15:20 5.5E4Pa
15:30 5.9E4Pa(cylinder pressure: 1MPa->replace with a new cylinder(15.8MPa)
15:40 6.1E4Pa
16:00 7.3E4Pa
16:10 7.9E4Pa
16:20 8.7E4Pa(cylinder pressure:10.5MPa)
We will continue to pressurize it to atmospheric pressure on 30 April morning.
 

[Tanaka, Hirose, Saito]

We performed alignment to inject both the beam coming from the interferometer and the sub-laser beam into the RFPD. While the DC components of each beam were confirmed, the AC component of the interference signal could not be observed.
 

• We aligned the system so that both the beam coming from the interferometer  and the sub-laser beam were incident on the RFPD. The incident optical powers measured with a power meter were approximately 2.04 μW for the beam coming from the interferometer and 7.95 μW for the sub-laser beam. Observing the DC output of the RFPD with the oscilloscope on Moku:Lab, we found an offset of about 10 mV; the beam coming from the interferometer contributed about 18.09 mV and the sub-laser beam about 47.63 mV.　Next, we attempted to observe the AC component using the spectrum analyzer on Moku:Lab. To match the main laser frequency, we set the sub-laser temperature to around 31℃ and tuned it in that range to check for the AC component, but none was observed. Possible reasons include insufficient optical power incident on the RFPD or a low degree of interference between the beam coming from the interferometer and the sub-laser beam. We plan to change the ND filter and further investigate the interference condition.

[kimura, hSawada, Yasui]

We had the build-up test again. The slope of the pressure rise rate was approximately 0.5 for the first 30minutes and then rose to about1.2. 

During the buildup test, we checked the Q-mass data and found that the N₂/O₂ ratio was not 4, indicating that it differs from the composition of the air.

We also had the helium leak test, and a small leak ( ~1.0E-10 Pam^3/s) was detected on the +X-side-flange.

Finally, we replaced the dry pump with a repaired one to improve its performance.

The log-log plot of the SRM pressure after TMP startup indicates that the pressure is still decreasing.

[Ikeda, Sawada, Kimura, Takahashi]

We checked the leak of the geophone pods. First, the background level of the test chamber, which was taken from the CLIO site, was measured by the He leak detector. The level was smaller than the detectable limit, 1.0e-13 Pa ‣m^3 /s. After that, we measured the leak level of the H1, H2, and H3 geophoen pods one by one. In any case, the leak level was smaller than the limit.

I compared the vacuum evacuation speed this time with that at 28 February 2025.
Figure 1 shows the time trend of the vacuum pressure in 2025.
It takes 1 day and 7.5 hours to reach the pressure of 1.0e-4 Pa from 1.0e-2 Pa.

Figure 2 shows the time trend of the vacuum pressure this time.
After 1 day and 7.5 hours passed from the time that the vacuum pressure was 1.0e-2 Pa, the vacuum pressure was 1.9e-4 Pa.
Even after 2 day and 15 hours, the vacuum pressure was 1.2e-4 Pa.

So, the vacuum codition seems still worse thatn before.

I checked the signals of the IM OSEMs during the evacuation. The signals in count were increasing slightly during the pump-down with the DRY pump, due to the buoyancy. Just after switching the pump from the DRY to the TMP, the signals decreased rapidly in 10 min. The pressure change was from 10 Pa to 0.02 Pa. In this pressure region, the thermal conductivity of the gas changes a lot. I think this signal drop is due to the temperature change of the LED in the OSEMs.