By checking it, several strange behavior detected.
1. Mysterious TMSX signals
As shown in Fig.1., after the locked loss happened, the TMSX GR PD became lerger value (Four times larger than normal)
And SUM values of the GR QPD1,2 became minus value
After the 2nd locked loss, all PD, QPD SUM values became almost zero.
For those locked loss may not caused by the earthquake.
2. No change by turned on/off GRX shutter.
Those values didn't change the value even turned on/off the GRX shutter
3. gigE camera image
I backed the beam spot image of screen capture, the GRX beam image shifted from Fig.2. to Fig.3. (To left)
(Also GRY??)
4. TCam image
TCam image didn't change so muchas shown in Fig.4.
Anyway I requested DOWN state for LSC_LOCK guardian
More detail QPD signals can be found in Fig.5.
ALS-X input value didn't change during this lock
all QPD1,2 SEG1,2,3,4 values changed to -2000 ~-3000 (minus value was mysterious)
Fig.6. showed the Xarm ndscope
Some trouble in TMSX rack?
Fig.7. showed showed the PD, QPD, LVDT, ACC signals, and as a reference PEM channel and LSC_LOCK state.
At the same timing (-4h,52m), the behavior changed.
Fig.8. showed the very last stage of the 2nd locked loss (LSC_LOCK was ALS_DARM_LOCKED)
Also, the change of value occurred at the same time.
No significant change can be detected in PEM channel. (How about CAL channel?)
I checked all PEM channels around this lock loss.
K1:PEM-ACC_EXA_TABLE_TX,RX_Z_INMON and K1:PEM-MIC_EXA_BOOTH_EXA_Z_INMON had several offset?
(And MAG Y direction had some glitch??)
Conclusion :
We can recovered the TMSX signals and initial alignment and PRFPMI RF LOCKED succeeded.
And I forgot to turn off the floor light, sorry...
Detail :
When we arrived the Xend station, we checked both EX0 rack and TMSX rack.
No problem in EX0 rack.
In the TMSX rack, we noticed that "the display of the power supply -18 V was off" (Left side of kikusui power supply)
Fig.1. showed the front panel of analog circuits.
Two coil driver (+15V LED on, -15 V LED off)
LVDT driver / LVDT ACT DISTRIBUTOR (unknown from front panel)
geophone distributor (+15V LED on, -15 V LED off ?)
whitening filter circuits (all turned off)
Michmura-san logger (on)
function generator (on)
+18 V power supply (on)
-18 V power supply (off)
We called YamaT-san how to recover the power supply, but during calling, "The power supply for -18 V suddenly recovered"
(Komori-san said, he touched the AC adaptor, but no turned on/off work, his mysterious power??)
After then, LVDT and Geophone value became several values
(We didn't check those value is significant value or not)
But, the IRX. GRX PDA signal didn't recover.
Even I injected the LED light signal, the response of the PDA value in the DGS system didn't change so much.
(At this time, both IRX and GRX PDA was turned on (green LED lumped),
my guess was PDA working well, but several trouble in BNC-Dsub converter analog circuits)
We cannot know why the whitening filter circuits were turned off, we called Ushiba-san.
Ushiba-san replied that there would be the safety trigger (some logic to turned off automatically)
Today, there is no analog circuit expert, we need to ask then the logic of turning off the analog circuits.
Anyway, we turned on the three whitening filter analog circuits, GR QPD values recovered.
For IR QPD value, we need to check after PRFPMI RF LOCK recovered.
For Xend transmission s/p-pol, I would like ask someone to check values.
This is a mysterious point, after turning on the three whitening filter, the IR and GR PDAs recovered automatically.
Then, initial alignment and LSC_LOCK trial performed, and finally PRFPMI RF LOCKED state succeeded.
Summary :
We found the power supply of the -18 V was turned off (no display and no LED on circuits) : Mysterious point 1
The power supply of the -18 V was suddenly recovered : Mysterious point 2
Whitening filter circuits were turned off : Checking point 1 (Ask analog circuit experts)
When whitening filter circuits were turned on, the IR and GR PDA suddenly recovered : Mysterious point 3
LVDT signal, geophone signal, coil output, p/s pol signal were not checked by Yokozawa and Komori-san : Checking point 2
Fig.1. showed the time series
Blue : the timing of turning on the whitening filter circuit for the p/s pol signal
Orange : The timing of turning on the whitening filter circuit for the GR QPDs
green : The timing of turning on the whitening filter circuit for the IR QPDs
Obviously, the timing is same with the whitening filter circuit for the p/s pol signal.
We need to check some ground connection between BNC-Dsub converter and whitening filter circuit (and TMSX optical table???)
