Based on some guess I had about the oscilloscope voltage of the IR transmission increasing, I set the Threshold to 10x the value of what I wanted it to be (500 mV -> 5V). I tried to check if the oscillating lock point was due to the gain being too high causing control loop resonance to appear in the cavity length, but somehow by just touching the gain knob I was able to get the SHG to lock normally. Output green is 250 mW which is nominal.

I realigned the MZ to GRMC to get good mode matching. Not much issue. I checked the error signal and the base level has some large 1 Vpk oscillation at about 8.7 kHz. The spurious oscillation seems to be coming from the output of the GRMC mixer in the "cable management box" - the spectral peak is not present from the L input (GRMC DEMOD DDS2 DAC2) or the R input (GRMC REFL RF). It also wasn't present in IRMC ERR or OPO ERR. It was getting late so I didn't have time to check if it was bad connection or some other reason.

The PLLs managed to stay locked for 32 hours. So it seems not much problem there. But I did notice that on one occasion tapping the cables for the GRMC problem above cause a bit of noise to show up in CC.

I turned off the lasers.

Realignement of the beam after some issue with the polarizer alignement :

start : x=323.565  ; y= 64.3750; z=60

distance between blade and last mirror = 146mm

vertical cut of the beam ; x=314  ; y= 100; z=20

horizontal cut of the beam ; x=360  ; y= 180; z=20

max DC = 0.234V

y(166.500)=  9.14e-5  = y(146.943)

x(313.700)  = 7.14e-5 = x(333.480)

--> CENTER : x=323.59 ; y =156.721

In prepartion for installing the black out curtain and moving the Al bar, I removed all the optics form the table (Fig.1). They are now stroed in the big desiccator (fig .2).

I also tried rearranging the Thorlabs lens case but gave up since it was all mixed up.

During this work, I notice most of the the pedastals and mirror mounts I stored inside the box labelled cryogenic cavity had disapperead (fig.3).

After recovering PLLs I hoped that everything would go smoothly to recover squeezing so of course it didn't.

The alignment of SHG is sufficient but not great. I turned on DDS1 DAC1 (SHG/IRMC demod) and checked the phase for the error signal. It seems fine, not much (if any) change from previous phase and 4.88Vpk signal. However, when I locked, there is oscillation of the output power (IR transmission). Specifically, it oscillates at about 5.5 kHz, max ~ 1.9 V, min ~ 1.5 V (0.4 Vpk). For green, the power meter reads 210 mV which is roughly correct but the SHG temperature hasn't been optimized for a long time so it should be closer to 280 mV. I do not think the cause of this oscillation is mechanical since the lowest SHG resonance is about 7 kHz (Yuhang thesis fig 4.7). Likewise it shouldn't be temperature because temperature is slow to respond. There isn't any of this frequency in the demodulation LO from the DDS board (DDS1 DAC 1 into spectrum analyzer). It seems like a recent issue between June and now, possibly due to the lightning strike power outage in Mitaka a couple of weeks ago.

Marc and Logan adjusted the polarization to get rid of HOM splitting when they reconnected electrical things. Indeed I just did some other check - the BSF10C that goes to the transmission DCPD has significantly smaller reflection for p-pol than s-pol, and indeed at the set polarization the total power at the DCPD is minimized (SHG requires p-pol infrared). So the HWP is at the correct place (~ 195 degrees). Actually looking through some of my old photos, during Yuhang's New Years visit it was at 120 degrees - I have no idea why. Rotating the polarizer can reduce the oscillation but it also reduces the transmitted power, and besides it should have already been the correct setting anyway.

Peak optical power at the RFPD was about 3 mW (power meter MAX reading method), which is maybe a little bit low (3.6 mW before). The signal from the (amplified) DCPD on the oscilloscope (coming from TRANSMISS IN) is much larger now than what it was before, even with the gain on 0 dB. It was 550 mV from my old pictures at 30 dB, but is now 1.7 V at 0 dB. This 1.70 V on the DCPD is also the average value of the oscillating locked IR transmission as noted above. Another strange thing is that when I checked the lock threshold it seemed to be at a very large (i.e. not reasonable) value > 1.7 V. Previously it was set at 250 mV (a bit less than half of 550 mV). Did this have something to do with the power outage? I don't know how there is more DCPD voltage for probably similar input power, with the amplification turned down.

Eventually I could not lock SHG anymore. I'm not sure what is the problem right now. I will guess at some items to check:

• Main laser temperature and current
• Electrical connections, especially related to SHG servo
• Badly behaved cables
• DDS output for EOM
• Threshold and Gain knob vs output behaviour on SHG servo

I left both PLLs locked to see if they stay locked overnight.

Both PLLs maintained lock for 5+ hours. The weird ppol issue went away (maybe laser temperature stabilizing). Next up - checking stability of squezing coherent control.

I checked again. Both PLLs unlocked overnight. The laser temperature drifted down for the ppol (maybe 50 MHz equivalent) and also down for the CC (again maybe 50 MHz equivalent). 

I relocked both of them. The ppol one had a bit of weird behaviour where it does some kind of twitchy unlock relock behaviour - it skips maybe one MHz away from the lock point but then is quickly caught again. I reloaded the PLL settings (Analog Devices ADF4002 SDP board black) and turned on the Charge Pump Gain setting. I'm not entirely sure if that fixed it but I left for lunch, came back to TAMA 90 minutes later and both PLLs were still locked. Maybe I can turn up the timeout delay on the ppol too. I think fiddling with these settings too much may change the phase noise but I will check that after I recover squeezing. Perhaps there could also be a bit more optimization with ppol mutual polarization matching and laser to fiber alignmnet but for now it's serviceable. Anyway at the very least I think I have confirmed that there is no major stability issue with either PLL loop on its own. Now we see if that is maintained for CC1 and CC2 on squeezing.

I checked again the PLLs in TAMA. I don't know why but the CC fiber detector output voltages changed to 9.4 mV for ML and 4.4 mV for CC compared to last entry.  Because I suspected that I connected the polarization maintaining fiber poorly before, I double checked again that they were inserted correctly. I then saw that CC PLL beat signal Vpp (output from fiber PD to oscilloscope) could be increased from 4.40 mV (176.5 degree on HWP to CC fiber) to 11.4 mV (206 degree on HWP to CC fiber). The signal to CC BEAT PLL servo input is -0.5 dBm and the CC PLL MON channel is -27 dBm (same as ppol PLL MON). I could stably lock CC for a few minutes using the settings of fig 1 and CC PLL servo set to non-INV.

I left both PLLs locked before leaving and will check if they are still locked in the morning. Maybe not, because I turned the lasers from off->on before doing this test and then they drift a bit for about an hour.

QWP
average = 50
air r = 3 ; sample r = 5
center x= 323.565; y= 154.375 ; z=60

azi=0

air= Tue, Jul 30, 2024 3-05-22 PM.txt

QWP = Tue, Jul 30, 2024 3-12-20 PM.txt

azi= 15

air = Tue, Jul 30, 2024 3-28-48 PM.txt

QWP = Tue, Jul 30, 2024 3-34-30 PM.txt

azi = 30

air = Tue, Jul 30, 2024 3-51-01 PM.txt

QWP = Tue, Jul 30, 2024 3-57-21 PM.txt

azi =45

air = Tue, Jul 30, 2024 4-16-13 PM.txt

QWP = Tue, Jul 30, 2024 4-22-39 PM.txt

azi = 60

air = Tue, Jul 30, 2024 4-38-46 PM.txt

QWP = Tue, Jul 30, 2024 4-46-06 PM.txt

azi = 75

air = Tue, Jul 30, 2024 5-03-37 PM.txt

QWP = Tue, Jul 30, 2024 5-09-57 PM.txt

Chien-Ming, Hsun-Chung, Te-Hui, Logan, Michael

The wedge angle for the VOPO cavity mirrors was not so easy to see. It turns out the mechanical fixtures holding one of the mirrors had to be filed off because it didn't have enough clearance. The mirror was taken to the elec shop with First Contact. The Taiwan group will fabricate a new set of cavity mirrors which they will replace "next time" (October or November).

Eventually the cavity mirrors were placed as they are for now. The other First Contact on the OPO was peeled off. Chien-Ming then tought me the procedure to align and mode-match the bowtie OPO. After a few false starts I eventually managed to get it to 80% with just LG mode remaining. Then it was getting late.

After achieving green resonance, we must get IR resonance, green/IR coresonance and then maximize amplification/deamplification by adjusting the lateral position of the wedged PPKTP. This is the same nonlinear gain measurement principle as before. We actuate on the relative phase of bright IR and green using a PZT phase shifter (in this case we will use on IR path but in TAMA we do it on green), allowing us to see the full range of nonlinear amplification/deamplification.

I returned the Vac Seal back to the TAMA South End dessicator.

Chien-Ming, Te-Hui, Hsun-Chung, Michael

We remembered that the OPO cavity was redesigned to have the input mirror on the right side rather than the left side. The green SHG fiber output was reversed as a result, pointing towards the door.

Two of the mirrors were fixed at the wrong orientation, with the wedge pointing vertical rather than horizontal. This would cause beams to be deflected up and down. Since the screw holes for the custom mirror mounts are only oriented for the horizontal direction, we had to glue them again. With permission of Takahashi-san I took Cosmotec VE-22 vac seal from the TAMA South End dessicator. We then glued the mirrors in the correct orientation.

I practiced a bit my First Contact technique on a laser line mirror. We peeled First Contact from one face of the OPO successfully, but for the other one the tab came unstuck before the polymer, so I had to reapply and properly soak the tab. The four OPO cavity mirrors also have First Contact applied.

We could see the alignment of the green beam inside the OPO using spare mirrors.

Te-Huei told me about the operation of the homodyne detector. It is somewhat based on Advanced LIGO design. We can as for TAMA separately monitor AC (10-10000 Hz) and DC output. Instead of SUM out we can monitor each PD directly. We also have some test diagnostics for the resonant frequency of the PD circuit. The one currently mounted has some spurious resonance at 20 kHz in the noise spectrum. There is a spare homodyne which doesn't seem to have this resonance if it is in contact with the optical table. The black coating on the case of the spare was ground off near the screws which apparently makes it better (I don't know why). The spare one doesn't have high quantum efficiency PDs installed, but according to Chien-Ming we can just easily remove and replace between the two.

The old TAMA spectrum analyzer seems to have some plastic film peeling off of the back of the screen, which causes it to be hard to view. Maybe this was caused by some temperature difference during the period the ATC room was not air conditioned?

Hsun Chung, Te Hui, Chien Ming, Logan, Michael

A rough account of various activities, I do not have all of the details.

Te-Hui and Hsun-Chung brought two of their homemade homodyne detectors and tested the dark noise. One of them is better than the other. The old spectrum analyser formerly of TAMA seems to have some issue with the window covering the screen - the bottom left corner of the screen cover has become somewhat opaque (cause unknown) and that part of the display cannot be seen.

The VOPO picomotor stage requires a Windows 10 PC to use the software. Hsun-Chung bought a secondhand W10 laptop from HardOff.

I took an oscilloscope from TAMA storage, as well as about 10 M6 x 15 screws and some 25 mm posts and 30 mm post holder. I also took two beam dumps and the +/- 25V power supply from the filter cavity cleanroom. Marc apparently found 6 Glan Taylor polarizers (100,000 extinction ratio, versus 300 for a typical PBS) that didn't have any designated use.

The green mode matching telescope (to go to the VOPO) was set up. We used some Thorlabs lenses which were quite dirty so we cleaned them with acetone.

Chien-Ming informed me of some minor issues regarding the setup. For one, there is some elliptical polarization reaching the PBS. I set up the input Faraday to optimize for p-polarization at the input, as per the Thorlabs manual. It then outputs 45 degrees. However, Chien-Ming told me that there is some elliptical polarization at the PBS, since the Fresnel effect causes differing phase delay for s and p polarization at nonzero angle of incidence. But it's not a huge problem. To mitigate, we could either send 45 degrees in, s or p out (cheap), or use HWP after the FI (more $). Also my beam profile was precise (correct beam waist) but inaccurate (maybe 10cm offset from the actual position).

We put First Contact on both faces of the PPKTP crystal. The layer that I left on the linear OPO input mirror didn't peel easily so I put some more on as per the LIGO instructions. I also put in a spare PZT inner retaining ring in the linear OPO holder. The correct position is reached when it is screwed in to the limit of easy motion (see figure 1 of 2690)

with linear pol after correcting camera alignment

@220 deg (10 averaging, 10 iterations)

C:\Users\atama\Dropbox\LC-Experiment\Measurement Data\Coating measurement\AlGaAs\20240701\220 deg\Thu, Jul 25, 2024 11-20-44 AM.txt (The measurement took 30 minutes)

The alignment of camera was tuned to have the beam centered.

0-3.5V, 0.1V @220 deg with 100 avg

C:\Users\atama\Dropbox\LC-Experiment\Measurement Data\Coating measurement\AlGaAs\20240701\220 deg\Wed, Jul 24, 2024 6-05-45 PM.txt

Hirata gave me some drill bits yesterday. Unfortuantely. these drill bits are too small. If I can't find one in ATC I will order them online. If some has HEX5 bits for Makita drill, similar to the bit in the attached picture please let me know.

Both the TMP and scroll pump were in error state. After resetting them I turned them on. The pressure before was 1.27*10² Pa after about 10 minutes of pumping it reached 1*10^-2 Pa. There were no alarms or strange sounds both pumps look to not be damaged

Logan, Michael

Summary

The PLL coupling was measured and we attempted to lock. We could acquire both ppol and CC lock, but the CC lock is a bit weak and only holds for ~ 5 seconds.

Details

We attached a new power supply to the battery powered CC fiber photodiode. The cable was taped to the leg of the mini-table (with the QPDs) to make sure it doesn't go into the main laser path.

We measured the following fiber couplings:

Laser to fiber port
ML-ppol: 5.3 mW
ML-cc: 4.1 mW
ppol: 2.5 mW
cc: 0.51 mW

Fiber out to fiber PD
ML-ppol: 0.97 mW
ML-cc: 0.32 mW
ppol: 0.75 mW
cc: 0.17 mW

Fiber output efficiency (remembering that each fiber is 50:50 split before sending to duplicate PDs)
ML-ppol: 36.6%
ML-cc: 15.6%
ppol: 60%
cc: 66.7%

Fiber PD output (one laser at a time)
ppol (ML on): 27 mV
cc (ML on): 9 mV
ppol (ppol on): 22 mV
cc (cc on): 9.6 mV

Fiber PD output (two lasers on):
ML-ppol average: 49.4 mV
ML-ppol Vpp: 26.8 mV
ML-cc average: 10.9 mV
ML-cc Vpp: 9.2 mV
A DC level of 2.6 mV was subtracted out of the averages

While doing this procedure I noticed that maybe last time I hadn't screwed in the CC fiber correctly. This may have caused some weird polarization effect or something else. Maybe even all of the control problems... It's a bit weird that the ML-CC and CC output optical powers are different, but the PD voltage is the same. Maybe polarization is off a bit, I'm guessing for ML since ML volts should be > CC.

We could measure -8.5 dBm from CC BEAT (the fiber PD signal going into the PLL servo) and -38 dBm on CC PLL MON.

Previously there was apparently some problem with the DDS software where the ppol PLL local oscillator frequency could not be changed. This was for whatever reason fixed with the power reset. We could see the ppol frequency change on the spectrum analyzer. The laser temperature was a bit off. For the correct mode, it should be of a relatively large amplitude (although I couldn't see image sidebands), and if you rotate the temperature knob clockwise the frequency should increase. Initially it was the other way around (same for CC). I rebooted the ppol PLL software (Analog Devices ADF4002, black SDB board), loaded the most recent settings (20240625) and wrote to register. It didn't lock initially but after changing the phase detector polarity to negative it could lock.

For the CC PLL I adjusted the temperature to the correct polarity and loaded the software (Analog Devices ADF4001, black SDB board) and settings (20240625). I could not lock initially. I changed the phase detector polarity to negative and nothing. I eventually managed to lock by i) changing charge pump gain from 0 to 1 (I guess this is the equivalent of changing the loop gain of an analogue control loop), ii) changing timeout from 3 to 35 PFD cycles (I guess this something like increasing a lock loss threshold or control bandwidth), iii) changing the CC temperature to move the sidebands within 10 MHz of the main peak. Then we could see CC PLL lock, but only for about 5 seconds at a time. It would then lose lock and maybe or maybe not reacquire. So maybe the signal is a bit too small. Also because I wrote "I guess" in the description of settings I changed, I don't really know what they do. It's probably in Marco Vardaro thesis, 2018 era elogs, the Analog Devices manuals or Yuhang's distant memory.

On the CC temperature point, during the New Years visit Yuhang told me that the CC servo has some weird issue where it would only lock if the sidebands initially were brought 20-30 MHz away from the main peak. For whatever reason this is not the case right now (although it could be a possibility if I recover the good electronic settings). I could not observe any independent impulse/glitch noise however we noticed that when searching for some lost cable, tapping some of the cables causes noise in the CC spectrum. However, we didn't make too much of an effort to isolate the suspicious cable today.

I also turned on the homodyne PSU (+19V 0.066A, -19V 0.062A) and found another +25/-25V PSU which I had absent mindedly forgotten about. We should use this one for ATC experiment instead of the 18V one.

2 combinations of voltages of 2LC which produce polarized light with ellipticity<=2.5deg and with difference in azimuth angle>=40deg were sorted out. The voltages set applied to both LC simultaneously is 104. I was previously doing a scan from 0-3.5V with 0.1V step, which for 2 LC were 1296 sets of voltages. So now we have ~10 times less voltage sets. This will result in faster evaluation.

They are saved here:

C:\Users\atama\Dropbox\LC-Experiment\Measurement Data\Scan voltage\LinearPolvoltage.txt

The linear Polarization was generated from the most recent calibration in elog 3611 (100 avg:C:\Users\atama\Dropbox\LC-Experiment\Measurement Data\Polarization states\20240613\Fri, Jun 14, 2024 10-54-07 PM.txt)

The Labview "Complete Integratino-FileChoice" which can run voltages from a file was updated to match the needs similar to "Complete Integrated Dual Step.VI"

The file was chosen in labview and was run with 10 averaging for 10 iterations.

The AlGaAs was rotated past 45deg of incidence and measured using linear polarization (obtained above) as input:

@220 deg:

C:\Users\atama\Dropbox\LC-Experiment\Measurement Data\Coating measurement\AlGaAs\20240701\220 deg\Tue, Jul 23, 2024 4-52-36 PM.txt

At this position, the alignment in height was not changed, but it seems the beam was not centered at the camera (perhaps the power was 1.7mW and now it is 1.4mW)

[Hsun-Chung, Marc, Te-Huei]

We need a voltage supply delivering up to +/-20V for the homodyne detector that is being set up in ATC.

We found the TEXIO PW18-3AD (decigo 2011) of elec shop that was previously used to test some circuit but does not seems to be used since sometimes now.

We confirmed that we are able to generate +/-18V (enough to operate the homodyne) and brought it to ATC.

In case somebody wants to use voltage supply in elec shop we also brought another one from TAMA there.

[Hugo, Marc]

Today we wanted to restart birefringence measurement with PCI and longer average.

We found out that the motorized rotator is broken (sounds but no motion).

We ordered 2 new ones.