[Takahashi, Ikeda, Hirata, Ushiba]

We checked the IMMT1 suspension visually. There were margins of the pico-motor range much enough to adjust both pitch and yaw motion. There were not any rubbing magnets in the IM. When we took the pictures with a fiber scope touching the EQ stop frame, the TM pitch jumped due to a weak joint in the X-Y stage supporting the frame.

[Hirata, Ushiba-san, Ikeda-san]

We checked IR beam position around PR2 HR side Mid-size baffle. (How to make the vertical line is same as klog:29282)
It looks that IR beam positon is about 3.5mm +Y side away from the center of aperture.(IR beam center is 68.5mm and aperture center is 72mm on the ruler.)

The readout of the water fluid has recovered.

[YokozaWashimi, Tanaka, Ozaki, Sudo]

Today we tried to evaluate the seismic isolation of the OMC stacks, using a 3-axial accelerometer (S2315344) and an impact hammer (G1910656).
This is a quick report.

1. Locating the ACC on the OMC vac-table (Fig.1, Fig.2)
2. Tapping each stack (top, middle, bottom of stack1, stack2) with the impact hammer (direction: Fig.3)
3. Picking up each pulse (-0.2/+0.8 sec) and calibrating its transfer functions from impact [N] to acceleration [m/s2] automatically (example: Fig.4, Fig.5)
4. Calculating averages of the events (Fig.6, Fig.7)

[Hirata, Ushiba]

Abstract:

We reconstructed POP forward beam path.
Though beam is still close to the knob of mirror mount just after periscope, the beam doesn't seem clipped.
Now, IR beam hits almost center of the both QPDs when the alignment to PR2 is good.

Detail:

1. Check of beam position at high power beam dump:

First, we requested aLIGNED state for IMMT1, IMMT2, and PRM.
Then, to confirm the good alignment has been kept from yesterday, we checked the beam spot on the PR2 HR target, which is almost center (fig1): good.

After confirming the alignment is good, we checked beam at high power beam dump (fig2).
Beam path seems far enough from the beam dump: also good.

2. Check beam spot on in-vac POM behind PR2:

After confirming the IR alignment to PR2, we checked beam spot on in-vac POM behind PR2 (fig3)
It is hard to say from the picture, but no clip happens at in-vac POM in my eyes.

3. Reconstruction of POP forward beam path:

After confirming beam spot on in-vac POM is not so bad, we started reconstruction of POP forward beam path.
At upper mirror on the periscope, the beam is not center but it seems no clip (fig4), so we keeep it as it is.
Then, we moved upper periscope actuators and hit the beam on the mirror center of lower mirror on the periscope (fig3).
After that, we moved FST1 in JGW-T1909623-v11, because it is hard to avoid clipping without moving the mirror.
Then, we aligned the beam to lower right of the FST1 (fig6) to avoid clipping at the knob of the mirror mount.
Figure 7 shows the current beam spot near the knob, which seems not so bad in my eyes.
After that, we realigned all the downstream and centered the QPDs.

Note:

During the work, on of the actuators of lower mirror mount on the periscope hits the periscope itself and cannot rotate to CW direction (we can rotate it in CCW direction).
So, we cannot move up the beam spot on FST1 without moving periscope mirror upward.
Since I am not familiar with the periscope installed for the POP forward, I didn't lift it up and keep it as it is.

[Kimura, M. Takahashi and Sawada (Hokuto)]
We restarted presuurization of IXC up to 9.8 x 10^4 Pa (~ atmospheric pressure) with G-2 class grade air on morning of 24th/Apr.
Detailes are as followes;
1. Re-start injection the G-2 grade air into IXC
 9:45 Start injection at 7.8 x 10^4 Pa
 10:23 8.3 x 10^4 Pa
 11:09 8.9 x 10^4 Pa
 11:45 9.5 x 10^4 Pa
 12:58 9.8 x 10^4 Pa.
 Stopped injection. 
2. After completion of pressurization to atmospheric pressure, repair of flange leak was done. k-log 29305
3. The amount of gas used for pressurization to 9.8 x 10^4 Pa from 2.3 x 10^2 Pas was 70 m^3.
The breakdown details of the gases used for the vacuum breaking of IXC are as follows:
 63 m^3 of high purity air (7 m^3 x 9 bottles, total 63 m^3)  and 7 m^3 of high purity nitrogen (7 m^3, 1 cylinder)  for a total of 70 m^3.

[Kimura and Sawada (Hokuto)]

 The leak at the IXC cross pipe flange (+Y side) was caused by forgetting to remove the tape that temporarily secures the elastomer seal.

 (See attached photos 1~6)
 The forgotten tape was compressed by the flange, and it is presumed that the leak occurred through a small gap created here.
The tape was removed and a new elastomer seal was attached to the gasket sealing surface to close the flange.
The claw clamps closing the flange (36 in total) was tightened diagonally with torques of 20, 40, and 60 Nm.

A modified drawing of 29285. We mis-understood that the nominal distance from the center of "the optical beam from PRM to PR2" and the center of the "PR2 HR mid baffle aperture" is about 6 mm according to JGW-T1910200-v4, while the correct nominal number of this center-to-center is 4.8 mm according to JGW-T2113078-v2. Hirata-san has confirmed 4.8 mm is correct center-to-center with CAD.

[Ikeda, Hirata, Takahashi, Ushiba]

Abstract:

We moved PR2 about 3.3 mm in + Y direction and IMMT2 beam spot about 6 mm in + Y direction.
Then, we centered IR beam on PR2 and check beam spot on PRM, which is also close to the center.
So, alignment to PR2 seems fine now.

Detail:

1. Move PR2 in +Y direction:

For reducing GRX beam light power to identify GRX beam spot center easily, we requested ALS_FIBX guardian to SAFE state (GRX power becomes about 0.8 mW).
First, we moved PR2 so that GRX beam spot on PR2 HR target became nominal.
Since PR2 target was set as projecting PR2 center along IR injection beam (PRM-PR2 line), GRX should be off centered by 3.3 mm (165.48 mm (distance between PR2 HR surface and target) * 0.02 rad (angle between input and reflected beams of PR2)).
Figure 1 shows the GRX beam at PR2 HR target before moving PR2.
Figure 2 and 3 show the GRX beam spot after moving PR2 by 3.5 mm in +Y direction with the traverser.
Now, beam spot seem shifted in -Y direction more than 3 but less than 3.5 mm according to the fig3, which is almost the nominal value.

2. Move IMMT2 beam spot by moving IMMT1:

After moving PR2, we moved IMMT1 so that beam spot on IMMT2 shifted in + Y direction.
First, we requested PROVIDING_STABLE_LIGHT state to IO guardian and wait until alignment becomes good, and HOLD_ALIGNMENT state to IMC guardian not to kick the suspension when blocking the beam to IMMT1T QPDs.
Then, we set a ruler in front of IMMT2 and check the beam spot on the ruler before and after IMMT1 moves.
Figure 4 and 5 show the beam spot before and after moving IMMT1.
IMMT2 beam spot roughly moved 6 mm in +Y direction.

According to the current situation reported in klog29285, we need to shift PR2 beam spot by 5 mm along +Y direction.
Roughly speaking, 6 mm beam spot shift on IMMT2 moves PRM beam spot by 4.5 mm (6*3/4).
Due to PR2 position changes (3.3 mm along +Y direction), PRM beam spot will change 0.8 mm (3.3/4) in +Y direction when the beam is centered to PR2. 
So, in total, PRM beam spot changes by 5.3 mm, which is the same as we would like to move.

3. Center the beam to PR2 by moving IMMT2:

After moving IMMT1, we centered the IR beam on PR2 by moving IMMT2.
Then, we aligned PRM so that PRM reflection goes to REFL table.
Figure 6 shows the beam spot on PR2 target with aligned PRM, which seems well centered horizontally.
After that, we checked the beam spot on PRM AR target (fig7).
The beam seems hitting almost center.

Note:

This time, we haven't changed beam spot vertically though it is obviously higher than the nominal position on PR2.

MEDM time machine works well.
It can show past values recorded in frame files on MEDM screens and can be launched from the MEDM execute menu.

After choosing a lock back time as relative time, gps time, local time etc. (see also the left windown in Fig.1), a new MEDM screen is launched. Channels not recorded in frame files such as _OUTMON are displayed as white boxes instead of record values as shown in the bottom window in Fig.1

After the alignment work (klog29299), we centered OpLevs and offload actuator outputs by using picomotor.
Followings are the steps we moved for offloading today.

IMMT1 pitch: -330 steps (400 -> 70)
IMMT1 yaw: +14950 steps (-2300 -> 12650)
IMMT2 pitch: +30 steps (-600 -> -570)
IMMT2 yaw: +9325 steps (-8500 -> 825)

[Kimura, M. Takahashi and Sawada (Hokuto)]
 We restarted presuurization of IXC up to 7.8 x 10^4 Pa with G-2 class grade air and G-2 class grade N2 on 23th/Apr.
Detailes are as followes;
1. Re-start injection the G-2 grade air into IXC
 8:34 Start injection at 3.2 x 10^4 Pa
 10:25 4.2 x 10^4 Pa
 11:34 5.5 x 10^4 Pa
 11:59 5.7 x 10^4 Pa
 13:06 6.6 x 10^4 Pa
 13:54 7.1 x 10^4 Pa 
 14:41 7.8 x 10^4 Pa.
 Stopped injection. 
2. Injection is resumed on the morning of April 24.
After completion of pressurization to atmospheric pressure, repair of flange leaks will begin.
The amount of gas used for pressurization to 7.8 x 10^4 Pa was pure air (7 m^3 x 7 bottles, total 49 m^3)
and pure nitrogen (7 m^3, 1 cylinder).

[Kimura, M. Takahashi and Sawada (Hokuto)]
 We presuurized IXC up to 3.2 x 10^4 Pa with G-2 class grade air on 22th/Apr.
The specifications of the high-purity air (G-2 grade) used are in klog-25912 for reference;
Detailes are as followes;
1. Start injection the G-2 grade air into IXC
 10:15 Start injection
 10:45 2.1 x 10^3 Pa
 11:40 6.5 x 10^3 Pa
 13:16 1.3 x 10^4 Pa
 14:50 2.4 x 10^4 Pa 
 16:10 3.2 x 10^4 Pa. Stopped injection. 
2. We will re-start injection on on 23rd/Apr. morning.

Reply to 19291. I confirmed that the designed value of open angle of the optical beams connecting PRM-PR2-PR3 is 0.02 rad.

In the attached figure, the "incident" and "exit" angles are at PR2, and are both 0.582 degs as shown. So the open angle is 0.582*2 = 1.16 deg, which corresponds to 0.02 rad.

Attached is fig2 for the original post.

[kimura]
  The parallel threaded seal joint with O-ring, which was the cause of air leakage in the valve drive of the GVitmy,
 was replaced with a regular tapered threaded seal joint.
After the replacement, a leakage test was conducted, and it was confirmed that there was no leakage.
Replacement with tapered thread seal joints is required for all KAGRA φ1000 gate valves, φ800 gate valves, and pendulum valves.
The following one gate valves are awaiting replacement.
1. GVetmy