Remote connect

1. You should be inside NAOJ network or VPN access to it.
2. Connect to 133.40.117.62:3389

Bench part

Introduction of locking servo

Our locking servo has two servo control modes: auto and manual.

• In auto mode, the open and close of the loop will be decided by a logic circuit. The simple logic here is just a threshold(can be set by a potentiometer). As we know, the transmission or reflection of a cavity will go across a peak/dip when the PZT scan through the resonance. The threshold is trying to find this peak/dip.
• In manual mode, the close or open of the loop is decided by the operator.

There are also two servo operate modes: scan and lock.

• In scan mode, the output of servo will send out a ramp signal(the ramp signal amplitude and period can be changed by two potentiometers) and the loop is open.
• In lock mode, the loop will try to be closed. If the system is set well to be locked, the output of the servo will follow the error signal. Otherwise, the output of the servo will give a constant overload voltage.

For each servo, we can have two low pass filters, an inverter, a differentiator, an integrator(can be switched between 1/f and 1/f3) and a gain(can be changed by a potentiometer). And a sign for logic. The switch for remote control. The remote control signal can be sent via a 9-pin connector available on the board.

The 'perturbation in', 'EPS1' and 'EPS2' can be used to measure open-loop transfer function. EPS1 and EPS2 are two points before and after an adder. They are just after 'ERROR IN', so they can be also used to monitor the error signals.

The 'SQUARE SIG OUT' is used to synchronize signals when we are using 'scan' mode. Although it is not the same with the ramp signal, it has the same frequency with it.

However, the lock for Mach-Zehnder and two phase shifters are different. In this case, we don't use the 'scan' mode. We set the scan amplitude to zero. We also use not an auto mode but manual mode. For this lock, we just switch from scan sub-mode to lock sub-mode. After that change the sign for logic to lock.

Temperature control

TEC200C from Thorlabs is used to control the temperature of SHG/OPO. The precision of this control is 0.001kOm which corresponds to 3.8mK of temperature. The temperature of OPO needs usually to be changed when we change green power. Here I list some important numbers which are the range of change we usually have.

Green part

1. Lock SHG(use auto mode of servo, first put integrator to 1/F then switch it to 1/F3)
   • SHG transmission level is -1.23V (PD gives inverted output)
2. Lock GRMC(first make sure MZ is sending maximum power, then scan GRMC and change scan amplitude to a small value, then change offset to lock)
   • At the top of MZ fringe we have 80 mW in transmission from GRMC - if the power is lower check the alignment of all the cavities
   • GRMC transmission level is 326mV for s pol and 2.54V for p pol (PDA100A-EC with 0dB amplification)(156mW before green EOM)
   • GRMC reflection power is 100mW(MZ brightest fringe). Power reaching reflection RF-PD is 50uW.
3. Lock MZ(check variable gain out, make it close to zero and then lock)
   • once we decided the MZ offset, adjust the EP level to set error signal to zero

(Notice: the connection between GRMC and MZ is following this logic. First, press the reset button of MZ. Then there will be a slow scan of MZ (period of ten seconds?) although we put it in 'manual' mode. At the same time, GRMC is continuously in a fast scan (remember to put GRMC in 'auto' mode). When the slow scan of MZ make it output a large enough amount of green power, the GRMC will be locked. Then there will be a trigger sent to MZ to lock MZ, this trigger is sent through a cable connection different from other control servos.)

PLL part

Beat note level reference:

Beat note frequency:

Fiber couplings:

ML-CC: 1.4 mW → 0.19 mW, coupling: 0.19*2/1.4 = 27%

ML-p-pol: 3.6 mW → 560uW, coupling: 0.56*2/3.6 = 31%

CC: 3.3 mW → 0.5 mW, coupling: 0.5*2/3.3 = 30%

p-pol: 4.7 mW → 1.3 mW, coupling: 1.3*2/4.7= 55%

p-pol beatnote frequency is 141MHz while there is 50mW green.

p-pol beatnote frequency is 252MHz while there is no green. (Measured as 276MHz on 2019.7.18)

BAB on one of homodyne PD is 448mV while locking.

p-pol is 1.8V while locking.

Infrared part

1. Lock OPO
   • OPO p-pol transmission level is 2V
2. Check Alignment mode cleaner
3. Lock IRMC
4. Check homodyne shot noise level

Homodyne alignment

1. Send only LO to the HOM
2. Align HOM BS to balance the HOM (sub HOM channel = 0). Tweak the position of the two lenses before HOM to be sure the beam is not clipped but centered into the respective HOM pD
3. Remove steering mirror which send the beam into 2nd HOM PD and let LO enter into AMC. Align LO into AMC using the two last steering mirror before AMC
4. Block LO and send BAB into AMC. Align BAB using steering mirrors on BAB path before HOM BS
5. Remove BAB and put back HOM steering mirror. Check LO shot noise spectrum when SQZ is blocked. A tentative reference for this noise spectrum for ~1.9 mW of LO is -132 dBVrms/sqrt(Hz)

Coherent control part

1. Green power going to OPO is normally 50mW.
2. Reflection CC error signal is 144mV pk-pk. (between cc and green pump)
3. Power after 98% reflection is 325mW, while after filter it is 15mW.

Setting of spectrum analyzer for measurement of FIS squeezing

Setting of spectrum analyzer for measurement of FDS squeezing with Zero span mode

center frequency: 100kHz (frequency you want to measure)

RBW: 1kHz

VBW: 30Hz

Sweep time: 2s

Scan of green phase: 2Hz with 1Vpp

Cavity part

1. When BAB injection is 373uW and matching level around 94%, the filter cavity IR transmission is 440counts.

2. BAB to AMC has a peak of 4V(PD gain is 30dB) (128 mV higher-order mode exists)

3. LO to AMC has a peak of 11V(PD gain is 30dB)

Timing signal to DGS timing.png

DGS

To solve full disk problem:

1.Connect to standalone pc : ssh standalone

2. Check available space : df

3.Go to the data folder (either trend/second or full): cd /frames/trend/second

4.List the name of files in the chosen folder : ls -la

5.Delete all data starting with name 123 : rm -r 123

SR785 Spectrum Analyser

This spectrum analyser does not save to ASCII formats, only to proprietary .78C, .78D or .78W. To convert to .txt or similar, open up a command line terminal to the file location and use the following command:

SRTRANS /Oasc /D [filename].78D [filename].txt

/Oasc is the command option to convert to an ASCII format. /D keeps only data and strips all of the file headers.

It is helpful to remember some "wildcard" file matching notation to avoid having to convert a large amount of files one by one. Typically, the ? symbol will match to one character while * will match to any number of characters. For example:

SRTRANS /Oasc /D *Noise??.78D *Noise??.txt

Will convert all files that have the string "Noise" preceded by any amount of characters, and with any two characters at the end. If you don't have any other files lying around you can also use *.78D *.txt to convert all 78D files in the specific folder.