For the future evacuation and vacuum breaking, a log-log plot tool on MEDM was prepared.
This tool is available on the VAC_OVERVIEW screen as shown in Fig.1.
Scripts are in /opt/rtcds/userapps/release/vac/common/scripts/

-----
In the case that "gps time 2" is lager than "gps time 1", plot was made with data from "gps time 1" to "gps time 2".
Otherwise, a value in "gps time 2" works as duration since "gps time 1" for a positive value, and lookback time from "gps time 1" for a negative value.

I performed health check of IMMT2 (fig1-7).
All TFs have no significant change from the n-air health check, so they seem healthy.

Note:

TFs from each DoF EXC to Y have low coherence at 2-4 Hz.
It probably comes from the non-linearity of OpLev due to the large excitations.
Figure 8 shows the TFs when I introduced a notch at 1.5 Hz.
In that case, coherence at 2-4 Hz was recovered, so this low coherence is not a problem.

I performed health check of IMMT1 (fig1-7).
All TFs have no significant change from the n-air health check, so they seem healthy.

I modified INITIAL_ALIGNMENT guardian so that it can be used in the current situation.
Now, XARM alignment seems working well.

The changes are as follows:
1. Change request to IMMT1 from LOCK_ACQUISITION to ALIGNED
2. Skip to request MISALIGNED state to ETMY.
3. Skip engaging ISS

Note:

I'm not so sure the reason but PR2 good OpLev value moves a lot (more than 100 urad) from Thursday. Is it due to the change of the height due to PR2 work (klog29918)?
Also, POPFORWARD QPD2 signals are far from the center in vertical.
GRX transmission is also slightly worse, so I moved POP PICO to recover the GRX alignment.

Because the temp in the corner station seems to increase by ~ 0.6C from 11th June, I set the temp of the precision cooler near SR2 from 21.00 to 19.00 around 11am today.

By the way, the PSL PT data was lost in the PEM SEM page.

Sorry, it is my mistake.
I forgot to LOAD the corresponding guardians after changing the saturation check.
I loaded the guardians, and it seems working corerctly.

Guardian log of SR3 from entering ISOLATING to requesting TWR_DAMPED is as follows.
2024-06-14_23:37:20.588149Z VIS_SR3 new target: ISOLATED
2024-06-14_23:37:20.592504Z VIS_SR3 executing state: ISOLATING (390)
2024-06-14_23:37:20.593021Z VIS_SR3 [ISOLATING.enter]
2024-06-14_23:37:20.803851Z VIS_SR3 [ISOLATING.run] USERMSG 0: Now waiting RMS < 5: F0_DAMP_GAS F1_DAMP_GAS BF_DAMP_GAS IP_DCCTRL_L IP_DCCTRL_T IP_DCCTRL_Y
2024-06-14_23:37:40.903045Z VIS_SR3 [ISOLATING.run] USERMSG 0: Now waiting RMS < 5: F0_DAMP_GAS F1_DAMP_GAS IP_DCCTRL_L IP_DCCTRL_T IP_DCCTRL_Y
2024-06-14_23:37:43.513733Z VIS_SR3 [ISOLATING.run] USERMSG 0: Now waiting RMS < 5: F0_DAMP_GAS IP_DCCTRL_L IP_DCCTRL_T IP_DCCTRL_Y
2024-06-14_23:37:45.148265Z VIS_SR3 [ISOLATING.run] USERMSG 0: Now waiting RMS < 5: F0_DAMP_GAS IP_DCCTRL_L IP_DCCTRL_T
2024-06-14_23:39:17.014395Z VIS_SR3 [ISOLATING.run] USERMSG 0: Now waiting RMS < 5: F0_DAMP_GAS IP_DCCTRL_L
2024-06-14_23:39:43.405557Z VIS_SR3 [ISOLATING.run] USERMSG 0: Now waiting RMS < 5: F0_DAMP_GAS
2024-06-14_23:46:23.209870Z VIS_SR3 REQUEST: TWR_DAMPED
2024-06-14_23:46:23.224804Z VIS_SR3 calculating path: ISOLATING->TWR_DAMPED

As you can see, F0_DAMP_GAS wasn't calmed down. And then, checking error and feedback signals of SR3_F0 tell us F0_GAS_OUT was saturated before error signal reached around 0 (see also Fig.1).

-----
side story
Yesterday, Ushiba-kun enabled gas_notification for suspensions in the corner station. So this saturation should be informed on guardian and slack notification. VIS_params was surely modified. I doubt remaining bugs of gas_notification function at first. But I couldn't find a bug yet. And also, I couldn't find log of guardian load. Ushiba-kun might not load guardian after modifying parameters. If so, after loading guardian code, notifications will be enabled.

We performed the TF measurements for the OMC Stack3 (new) with vertical shaking before and after loading the additional weight (+155kg). The gaps between each stage were large enough and no
screw heads were touched (see pictures). 
After that, the TFs from vertical to the other DoF were also measured with the weight.
These data are uploaded on JGW-T2415845.

The TF of the new Stack3 without additional weight was a simple structure and good for comparing theoretical models.
The TF of the new Stack3 with +155kg weight had some peaks over 100 Hz. They might be couplings or mixing from the other DoF, caused because the position and angle of the weights were not perfect.

I adjusted the height of the SF and BF keystone so that the IM V1 OSEM could be in the linear range. Though the BF setpoint was changed by -1500, the IM V OSEMs were changed by only 176~275. The deviation of the BFD V (+910) was also inconsistent with the SF setpoint change (+1500).

[Takahashi, Washimi, Ito]

We started the recovery of the OSTM suspension.

1. Moved the OSTM suspension from the optical bench for the OMC to the workbench.
2. Removed two shield plates (X+ and Y+) from the suspension frame. The plates were stored in the OMMT chamber.
3. Removed the OSEMs with the support plates.
4. Removed the mirror (TM). The mirror was stored on the small table with the table KOACHs.
5. Removed the IM block and attached it to the hanging jig.

k1visetmxp, k1visetmynb k1visitmxp, k1visitmyp:.
k1visetmxnb, k1visetmynb, k1visitmxnb, k1visitmynb:
 Request from ushiba-san.
 Continued from K-Log#29795

 => Change 10 DOF to 5 each of 6 DOF

k1asc:.
 Request from Hirose-san (K-Log#29910)
 Only install & restart performed

 After the update, the DK of k1iopasc0 was red (error) from last week's update.
 There is no DAC1, so I tried to fix it, but it did not improve.
 k1asc_20240614_Final.mdl:NG
 k1asc_20230720.mdl:OK
 Delete PDA3 and 4 processing from k1asc_20240614_Final.mdl:OK
  => Time is running out today, will check next time

k1pemmanage: request from Yokozawa-san
 Only installation & reboot performed.

k1tmsx, k1tmsy:
 Only installation & reboot performed.
 

[Yuta Michimura, Haoyu]

The Simulink RT models of k1tmsx and k1tmsy have been modified to add outputs and filters for the polarization monitoring PDs at end stations.

We added filters for the s-pol and p-pol PDs (the filter banks are modified from the tms library file). See Fig 1, Fig 2 and Fig 3. The models were built and compiled successfuly yesterday (not installed yet). They were supposed to be installed today.

Channel assignments for PDs at TMSX/TMSY are:

EX0_MADC0_EPICS_CH28: TRX s-pol
EX0_MADC0_EPICS_CH29: TRX p-pol
EX0_MADC1_EPICS_CH28: TRX IR
EX0_MADC1_EPICS_CH29: TRX GR

EY0_MADC1_EPICS_CH28: TRY IR
EY0_MADC1_EPICS_CH29: TRY GR
EY0_MADC1_EPICS_CH30: TRY s-pol
EY0_MADC1_EPICS_CH31: TRY p-pol

[Takahashi, Washimi, Ito]

In this morning, we performed the weight test for the new stacks:

1. Measured the gaps for each rubber (top, middle)
2. Loaded the weight: 153kg for the Stack1, 159kg for the Stack2, and 155kg for the Stack3
3. Measured the gaps for each rubber (top, middle) again
4. Constructed the safety jigs

The measured gaps and spring constants are plotted here.
The spring constant's mean and standard error(σ/√N) is 533±26 N/mm. (see also the Mitaka test: klog29651)

As indicated in klog29820, the channel format in the DAQ block for IQ demodulation-related in REFL was html format, so it was repaired. 
Also, due to the increase in DAQ channels in the model update, the number of TRATE on the MEDM screen was 4179 and higher than 4096(Fig1), so the sample frequency for IQ demodulation-related channels was changed from 2048 to 512. It's now 2049.(Fig2, Fig3, Fig4)

Detail

• One way to check the text format of a DAQ block is to check the Properties of the DAQ block. If it is not in HTML format, a field called ClickFcn will appear in Properties. This editing was done by editing the channel in the ClickFcn field.(Fig5)
  In my case, I used the klog draft function as a memo, created the channels all together and copied and pasted them from there into the DAQ block. It should be very careful because of the possibility of being HTML format.

TF measurement of RFQPD

• At the HIGH signal, it is off-peak at the HIGH frequency. The LOW signal came between the peaks. It is the depth at the opposite target frequency at the HIGH and LOW signals. These are as designed. 
  However, the AM laser measurement did not reveal the depth at the LOW frequency of the HIGH signal. This may be buried by noise.
• Measurements with the AM laser resulted in Seg 1 and 3 being doubled relative to Seg 2 and 4 because the beams were elliptical. DC power per segment at QPD 3 and 4 in AM laser measurements(Fig1, Fig2). This is true for the LOW signal in measurements with the AM laser. But in the HIGH signal, the gain was almost the same in all segments. The reason for this is not known. 
• Also, the AM laser often turned off the OUTPUT signal output this time, as if the main unit was not in good condition.
• Destination of files with measurement results:  Dropbox/Measurements/IFO/ASC/REFL/RFQPD/characterization/TF

Did phasing for each segment

• For the HIGH and LOW signals, the following target frequencies were modulated into the AM laser at 1 Vp-p.
  • LOW signal: 11.248830066MHz(5.624365033MHz(Seed freq)× 2+100Hz)
    As a result, a demodulated signal was seen at 90 Hz.
  • High signal: 89.990010528MHz (5.624365033MHz × 16+100Hz+70Hz)
    Originally the demodulated signal was planned to be 100 Hz, but adjusted the frequency 70 Hz. 86 Hz signal was seen.
• In MEDM screen we installed 0 for R and 90 for D. In this state, we measured the phase difference of the I error signal of Seg 2-4 with respect to Seg 1. We dubstituted the difference in R. After phasing, The difference of each segment with respect to Seg1 is almost reduced to less than 0.2 deg.
• D has not been adjusted. The method for adjusting D is probably described in klog8359.

We got the acceptance by Ushiba-san. We updated the reference mirror center for ITMX.

I redraw the tcam-graffiti for the screen in the control room.