NAOJ GW Elog Logbook 3.2
[Hugo, Marc, Shalika]
laser diode 1A : 1.6mW
z = 20 mm gives distance between blades and last mirror on injection = 171mm
horizontal cut
Y: 103 mm cut (power halved) (power on dc 1.1) after hav power on dc was 0.55
X:374mm
z = 20 : Tue, May 28, 2024 2-14-49 PM.txt
z = 40 : Tue, May 28, 2024 2-21-29 PM.txt
z = 60: Tue, May 28, 2024 2-25-28 PM.txt
z = 80: Tue, May 28, 2024 2-29-48 PM.txt
verticale:
X: 313
Y : 191
z = 20 : Tue, May 28, 2024 11-32-57 AM.txt
z = 40 : Tue, May 28, 2024 2-01-31 PM.txt
z = 60: Tue, May 28, 2024 2-06-27 PM.txt
z = 80:Tue, May 28, 2024 2-10-41 PM.txt
after realignment horizontal
z = 20 : Tue, May 28, 2024 4-14-32 PM.txt
z = 40 : Tue, May 28, 2024 4-30-15 PM.txt
z = 60 : Tue, May 28, 2024 4-34-03 PM.txt
z = 80 : Tue, May 28, 2024 4-37-59 PM.txt
In the end we recovered normal incidence in both vertical and horizontal directions (within 0.026deg).
We wanted to prepare for linear polariscope measurement but the lockin amplifier showed strange error (undr) related to the chopper signal...
We'll start with polarization camera measurements in the coming days
[Logan, Marc]
We added one of the 2 remaining RF amplifier to the 500MHz clock output.
We soldered it to the NIM rack +12V and ground used for the LNVR. Note that we could have tried to use the +12V from the LNVR but were a bit worried to further damaged the clock..
The output of the clock is now +7.9dBm.
The output of the following splitter are
1 : -1.73dBm ; 2 : -6.19 dBm ; 3 : -5.09 dBm ; 4 : -5.33 dBm
Finally, we connected DDS2 to each output of the splitter and checked the signal output from DAC0.
We got about -9 dBm from all.
Next steps are fixing the clock crystal (was not fixed anymore) and the RF amplifier ; recover the electronic connections and recover squeezing.
The birefringence properties remained same. Comes from the fact that we do J/Co
after adding this, the jones matrix and info on birefringence was weird....I lost the crcuial info on the polarization properties rather than it working as a proper filter. So, not using this anymore.
Had actually picked it from a paper talking about improving noise in jones matrix.
after this the jones matrix of mirror is not [1,0][0,-1] anymore. the 1's have changed to 0.8 ish.
Added power factor of sqrt(Output power/ Input Power) in jones vector. This is to take account the power getting halved after BS
I removed the lens f=100mm and start to install the f=-50mm and f=50mm(not labeled yet) 10cm after the HWP.
put the condition on jones vector such that matrix is computed only for 2 orthogonal set of jones vector.
so the dot product is supposed to be 0. I have kept the range to be 85-90 deg. Therefore the dot product is required to be less than 0.2
after adding this, the jones matrix and info on birefringence was weird....I lost the crcuial info on the polarization properties rather than it working as a proper filter. So, not using this anymore.
Had actually picked it from a paper talking about improving noise in jones matrix.
Before the lens, w0 = -47,1 mm ; z0 = 26 μm , after the lens : w0 = 88.8 mm ; z0 = 49.2 μm
I moved the lens f=100mm after the characterization at d = 0.2 m (d = 0 is the same has the previous characterization)
!!!! The zero of reference did not correspond to any optic know, but it's at 101,5cm on the ruler, or the third a hole after the HWP.
[Logan, Marc]
Following the reinstallation of the mixers box we could not lock the GRMC.
We found out it was because the DDS2 was not generating any sidebands but was just outputing some low amplitude broadband noise.
We could see that this issue was somehow related to the 500MHz clock as the software could not recognize it from time to time.
Today we checked the DDS2.
All voltages supply are correct.
All output (filtered or not) are just noise increase.
Then we checked the 500 MHz clock output and got
Channel 1 : -8.6dBm
Channel 2 : -16.7dBm
Channel 3 : -19dBm
Channel 4 : -21 dBm
We checked the output of Wenzel output and got -8dBm.
It seems there are some issues with the splitter (ZFRSC-4-842-S+) that creates the low clock power.
We swapped the clock of DDS2 to channel 1 and got the expected output powers.
We'll check if the remaining amplifier can be used for this clock output or if we have to purchase a new splitter.
[Logan, Marc]
We added one of the 2 remaining RF amplifier to the 500MHz clock output.
We soldered it to the NIM rack +12V and ground used for the LNVR. Note that we could have tried to use the +12V from the LNVR but were a bit worried to further damaged the clock..
The output of the clock is now +7.9dBm.
The output of the following splitter are
1 : -1.73dBm ; 2 : -6.19 dBm ; 3 : -5.09 dBm ; 4 : -5.33 dBm
Finally, we connected DDS2 to each output of the splitter and checked the signal output from DAC0.
We got about -9 dBm from all.
Next steps are fixing the clock crystal (was not fixed anymore) and the RF amplifier ; recover the electronic connections and recover squeezing.
We took a Thorlabs IO-5-1064-HP from a box of miscalleneous optics for KAGRA filter cavity testing (next to Mitsuhashi Roberts linkage experiment, purchased end of FY23).
We put it on a micro tilt adjustment stage and made sure it could be adjusted to the beam height from the Mephisto (72-73 mm). Actually the Faraday isolator that was already there was just put on a pedestal post instead of an adjustment stage, so it is likely that the vertical alignment through the FI was not good, generating a visible higher order mode and causing the improper beam profile.
Logan, Michael
We once again found that the beam profile is showing some strange behavior with the asymmetry of the beam divergence. By using the profiler camera we see that it is probably due to misalignment of the Faraday isolator.
We took a Thorlabs IO-5-1064-HP from a box of miscalleneous optics for KAGRA filter cavity testing (next to Mitsuhashi Roberts linkage experiment, purchased end of FY23).
We put it on a micro tilt adjustment stage and made sure it could be adjusted to the beam height from the Mephisto (72-73 mm). Actually the Faraday isolator that was already there was just put on a pedestal post instead of an adjustment stage, so it is likely that the vertical alignment through the FI was not good, generating a visible higher order mode and causing the improper beam profile.
Did scans from
LC1, LC1step, LC2, LC2Step, averaging
0-25V, 0.2V, 0-25V, 0.2V
then observed and put a constraint on ell and azi of 2 deg
second scan
0-1V, 0.01V, 0-25V, 0.1V
15-25V, 0.1V, 0-25V, 0.1V
third scan
0-0.06V, 0.01V, 0-8V, 0.05V
15-25V, 0.01V, 0-5.2V, 0.01V
15-25V, 0.01V, 5.2-8V, 0.05V
0-0.06V, 0.01V, 0-4.3V, 0.01V
4th scan
0-0.04V, 0.005V, 0.83-1.2V, 0.001V
15-25V, 0.005V, 0.83-1.2V, 0.001V
5th scan
14.99-15.02, 0.001, 0-2V, 0.001V
24-25V, 0.001V, 0.95-0.994V, 0.001V
Even though the voltages are being used from the file, it seems there could be a deviation, considering we did the calibration of LC 2 months ago. I can't see a zero crossing in azi and ell, also the file is a bit huge, so it will take longer than expected.
I am going to instead do scan using the voltage range estimated but without using the file, and doing first a rough scan, and then will later do a finer scan.
The voltage range estimated from characterization is 7-8 V for LC 1 and 0-25V for LC2 with condition of 2 degrees on azi and ell.
So, doing a scan from 6-9V with 0.2V step, and 0-25V, 0,2V step.
measurement file:
C:\Users\atama\OneDrive\LC-Experiment\Measurement Data\Coating measurement\BB1-E03\20240417\0 deg\AfterCompensation\Sun, May 5, 2024 2-44-03 PM.txt
(could see zero crossing in labview, still need to plot, measurement finished by the time I finished writing this elog)
Couldn't transfer voltage file , a bit too huge
decreased condition to 20mV, 2mV and 2 degrees.
Also, I had made a mistake in conversion from rad to degree in the previous one , at one instance
voltage file:
C:\Users\atama\OneDrive\LC-Experiment\Measurement Data\Coating measurement\BB1-E03\20240417\0 deg\CompensationVoltage\voltagecompensating_0.02_0.002_2.txt
measurement file :
C:\Users\atama\OneDrive\LC-Experiment\Measurement Data\Coating measurement\BB1-E03\20240417\0 deg\AfterCompensation\Sat, May 4, 2024 6-05-39 PM.txt
The measurement stopped because of windows update. I have paused windows update for next 5 weeks.
Started another one with new voltage set of radius 50mV, step 5mV. The condition on ell and azi was 5 degrees.The measurment is being saved with 1 average.
voltage file is here:
C:\Users\atama\OneDrive\LC-Experiment\Measurement Data\Coating measurement\BB1-E03\20240417\0 deg\CompensationVoltage\voltagecompensating_0.05_0.005.txt
find the measurment file here:
C:\Users\atama\OneDrive\LC-Experiment\Measurement Data\Coating measurement\BB1-E03\20240417\0 deg\AfterCompensation\Sat, May 4, 2024 5-10-32 PM.txt
We tried to calculate the beam profile but it seems that the beam divergence going in does not match the beam divergence going out.
i.e. approaching the beam waist, the beam shrinks ~ 100 um diameter per screw hole, but going past, it increases only ~ 50 um per screw hole.
We will look again for the source of this odd behaviour.
Logan, Michael
I showed Logan the use of the ATC clean room and the experiment setups we have there.
I showed and supervised the use of the Coherent Beam Master BM-3 UV beam profiler. Although I dropped and broke the included NG9 ND filter (0.5% transmission). The manual says to use below 5 mW and 200 W/mm^2 without the filter so it's fine in this case (3.8 mW).
We measured the beam profile from the s-polarized beam split off in the Taiwan/Korea experiment setups. I thought this beam was to be collimated at 2.3mm over some distance, but it doesn't maintain that size for long and turns out to have a waist of ~ 200 um radius about 30-40in from the PBS. So I also question a bit the Taiwan fiber SHG beam size, but we can look at that later. I tasked Logan with fitting the beam shape and redetermining the mode matching elements for the OPO.
We tried to calculate the beam profile but it seems that the beam divergence going in does not match the beam divergence going out.
i.e. approaching the beam waist, the beam shrinks ~ 100 um diameter per screw hole, but going past, it increases only ~ 50 um per screw hole.
We will look again for the source of this odd behaviour.
after this the jones matrix of mirror is not [1,0][0,-1] anymore. the 1's have changed to 0.8 ish.
The birefringence properties remained same. Comes from the fact that we do J/Co