40m QIL Cryo_Lab CTN SUS_Lab TCS_Lab OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  40m Log  Not logged in ELOG logo
Entry  Wed Feb 11 18:11:46 2015, Koji, Summary, LSC, 3f modulation cancellation 
    Reply  Thu Feb 12 23:47:45 2015, Koji, Update, LSC, 3f modulation cancellation 
       Reply  Sat Feb 14 00:48:13 2015, Koji, Update, LSC, 3f modulation cancellation 
          Reply  Sat Feb 14 19:54:04 2015, Koji, Summary, LSC, 3f modulation cancellation beat_setup1.JPGbeat_setup2.JPGelectrical_setup.pdf
             Reply  Sat Feb 14 20:37:51 2015, Koji, Summary, LSC, 3f modulation cancellation beat_pd_response.pdfbeat_nominal.pdf3f_reduction.pdfbeat_3f_reduced.pdf
                Reply  Sat Feb 14 22:14:02 2015, Koji, Summary, LSC, [HOW TO] 3f modulation cancellation freq_gen_box.JPGdelay_line.JPGcable_spec.pdf
                   Reply  Sun Feb 15 16:20:44 2015, Koji, Summary, LSC, [ELOG LIST] 3f modulation cancellation 
                      Reply  Sun Feb 15 20:55:48 2015, rana, Summary, LSC, [ELOG LIST] 3f modulation cancellation 
                         Reply  Mon Feb 16 00:08:44 2015, Koji, Summary, LSC, [ELOG LIST] 3f modulation cancellation 
                   Reply  Mon Feb 16 01:45:12 2015, Koji, Summary, LSC, modulation depth analysis modulation_nominal.pdfmodulation_3f_reduced.pdf
Message ID: 11029     Entry time: Sat Feb 14 19:54:04 2015     In reply to: 11028     Reply to this: 11031
Author: Koji 
Type: Summary 
Category: LSC 
Subject: 3f modulation cancellation 

Optical Setup

[Attachment 1]

Right before the PSL beam goes into the vacuum chamber, it goes through an AR-wedged plate.
This AR plate produces two beams. One of them is for the IO beam angle/position monitor.
And the other was usually dumped. I decided to use this beam.

A G&H mirror reflects the beam towards the edge of the table.
A 45deg HR mirror brings this beam to the beat set up at the south side of the table.
This beam is S-polarlized as it directly comes from the EOM.

[Attachment 2]

The beam from the PSL goes through a HWP and some matching lenses before the combining beam splitter (50% 45deg P).
The AUX laser beam is attenuated by a HWP and a PBS. The transmitted beam from the PBS is supposed
to have P-polarization. The beam alignment is usually done at the PSL beam side.

The combined beam is steered by a HR mirror and introduced to Thorlabs PDA10CF. As the PD has small diameter
of 0.5mm, the beam needed to be focused by a strong lens.

After careful adjustment of the beam mode matching, polarization, and alignment, the beatnote was ~1Vpp for 2.5Vdc.
In the end, I reduced the AUX laser power such that the beat amplitude went down to ~0.18Vpp (-11dBm at the PD,
-18dBm at the mixer, -27dBm at the spectrum analyzer) in order to minimize nonlinearity of the RF system and
in order that the spectrum analyzer didn't need input attenuation.

Electrical Setup

[Attachment 3]

The PD signal is mixed with a local oscillator signal at 95MHz, and then used to lock the PLL loop.
The PLL loop allows us to observe the peaks with more integration time, and thus with a better signal-to-noise ratio.

The signal from the PD output goes through a DC block, then 6dB attenuator. This attenuator is added to damp reflection
and distortion between the PD and the mixer. When the PLL is locked, the dominant signal is the one at 95MHz. Without this attenuator,
this strong 95MHz signal cause harmonic distortions like 190MHz. As a result, it causes series of spurious peaks at 190MHz +/- n* 11MHz.

10dB coupler is used to peep the PD signal without much disturbing the main line. Considering we have 6dB attanuator,
we can use this coupler output for the PLL and can use the main line for the RF monitor, next time.

The mixer takes the PD signal and the LO signal from Marconi. Marconi is set to have +7dBm output at 95MHz.
FOr the image rejection, SLP1.9 was used. The minicirsuit filters have high-Z at the stop band, we need a 50Ohm temrinator
between the mixer and the LPF.

The error signal from the LPF is fed to SR560 (G=+500, 1Hz 1st-order LPF). I still don't understand why I had to use a LPF
for the locking.
As the NPRO PZT is a frequency actuator, and the PLL is sensitive to the phase, we are supposed to use
a flat response for PLL locking. But it didn't work. Once we check the open loop TF of the system, it will become obvious (but I didn't).

The actuation signal is fed to the fast PZT input of the AUX NPRO laser.

Attachment 1: beat_setup1.JPG  1.066 MB  | Hide | Hide all
Attachment 2: beat_setup2.JPG  1.196 MB  | Hide | Hide all
Attachment 3: electrical_setup.pdf  22 kB  Uploaded Sun Feb 15 14:28:44 2015  | Hide | Hide all
ELOG V3.1.3-