With Yokozawa-san and Dan Chen

Here's a note of the ETMY suspension during the alignment work (a separate entry for the alignment work will be posted by Dan Chen later)

IP controls were described in 19353 and they were tested beforehand and they worked. However, today I tried to close the IP loops for L, T, and Y but the control signals keep integrating, eventually saturating the coil output.
In the beginning, I thought this was due to the fact that the IP was far from the setpoint. I put the IP close to the setpoint (well within 100 um/urad) using the IP stepper motors.
I tried to close the loop again but it didn't fix the problem.
I tried closing L, T, and Y loop one at a time and things started getting weird when I close the Y loop (when L and T controls are engaged).
L and T moved away from the setpoint but I think Y didn't respond. I need more time to investigate this issue.
But it seems like this could be caused by some severe actuation coupling, maybe one of the actuators is failing, or maybe the yaw readout is faulty somehow.
So the alignment proceeded without the IP yaw control.

I didn't engage any GAS filter loops and in the end, it worked fine.

For the BF, I improvised a yaw DC control and make the setpoint zero. (Separate klog will be posted later when I finalized the controls)
The BF yaw fluctuates a lot around the setpoint even with controls engaged (Without control it goes like 1000 urad peak2peak).
The peak-to-peak is around 20 urad at around 0.02 Hz with controls engaged (assuming that the yaw readout is calibrated).
This will be improved when the chamber is evacuated. Alternatively, we can make stronger damping around the yaw resonance frequency.

I didn't engage any MN loops.

I used BF yaw pico motors to move the chain below the BF in yaw such that the TM optical lever reads close to 0 in the horizontal readout.
As soon as I did that, we saw the reflected beam at EYA.
The optical lever at this point was well out of range so Yokozawa-san helped us to center the QPD.
After that, we align the ETMY yaw so the reflected beam spot goes close to the injected beam at EYA horizontally.
We used the BF pico motor for this. I decided not to use the setpoint at the BF because it's slow and we will control it at zero anyway in the future.
After an additional re-centering of the QPD, we moved and aligned the ETMY in pitch by putting OPTICALALIGN offset in the MN stage.
The good OFFSET value is around 1625.

We then removed the iris (handmade from aluminium foil) at EYA so the reflected green beam gets sent back to the center area.
And immediately we see a flash from the POS reflect PD. We blocked the beam several times to see if we can see changes in the PD channel and it did.
I recorded 30-second averages of several important readouts, i.e. the oplev readout, BF L readout, IP L and Y readout. See Fig. 1.
I decided to record the longitudinal positions of the BF because its position is rather arbitrary and it will eventually affect the oplev yaw readout if it's shifted (due to the oplev's horizontal config.).