113 Hz is the frequency of the line sent to the rampeauto perturb channel to test LO alignemnt with the technique proposed by Raffaele. I guess we forgot to switch it off last friday. 

I connected remotely and switched it off now. Sorry for the inconvenience!

First I increased green power from 40mW to 50mW. Carrier and CC resonance frequency is as follows. CC PLL frequency is same as last week.

CCFC error signal when AOM is scanned around CC resonance is shown in Pic. 1 channel 2. It is different from the error signal last week and seems very noisy. I also measured spectrum of CCFC error signal when CC is locked on resonance (Pic. 2). There is a very strong 113Hz peak in CCFC error signal.

[Aritomi, Yuhang, Eleonora]

To make one of CCSB on resonance inside filter cavity, we set CC PLL frequency 6.997043MHz following previous measurement. We also changed 14MHz and 21MHz accordingly in DDS board.

Binary number is as follows.

Then we split CC1 LO (14MHz) into 50:50 with Z99SC-62-S+ to use it for demodulation of CCFC, but CC1 was unstable due to low SNR. Therefore we amplified CC1 LO by 33dB and attenuated by 20dB before splitting. Since output of DDS board is -6.5dBm, CC1 and CCFC LO is -6.5+33-20-3 = 3.5dBm. Then CC1 locked stably.

We set CC PLL 7MHz as usual and scanned AOM around one of CC resonance while other CC sideband was off resonance. CCFC error signal was shown in Pic. 1 (Channel 1: IR transmission, Channel 2: CCFC error signal). It seems like usual PDH signal. Then we set CC PLL 6.99704303MHz and found CC resonance at 109.03598MHz while carrier is 109.03584MHz which means carrier is detuned by 109.03598-109.03584 = 140Hz. CCFC error signal was shown in Pic. 2. The error signal was just a dip and didn't change by demodulation phase.

On Monday, I found that the cavity remote locking was not working properly. Matteo told me that somehow the treshold for lock acquisition was changed last week.

I put it back to the original value of -0.5 V and now it seems to work quite good.

I also found the the transmission was not stable despite the dithering was engaged. The situation improveed after I increased the gain. (Pic 1) 

Current values:

INPUT ATTENUATION: 2.9

GAIN: 5.7

[Aritomi, Yuhang, Eleonora]

First we measured shot noise spectrum with sensor card in squeezing path to see LO back scattering. Then we reduced LO power from 1.86mW to 0.158mW by putting ND filter after IRMC. Pic. 1 shows shot noise with different LO power. Shot noise reduced by 10.5dB and 13Hz peak from LO back scattering reduced by 21dB as expected.

Then we measured shot noise with filter cavity. At the beginning we saw low frequency bump in shot noise spectrum, but after some alignment of homodyne, the low frequency bump somehow reduced a lot. Pic. 2 shows shot noise with filter cavity.

[Aritomi, Yuhang, Eleonora]

sqz_dB = 11.5;                    % produced SQZ

L_rt = 100e-6;                    % FC losses

L_inj = 0.20;                     % Injection losses

L_ro = 0.11;                      % Readout losses

A0 = 0.1;                         % Squeezed field/filter cavity mode mismatch losses

C0 = 0.1;                         % Squeezed field/local oscillator mode mismatch losses

ERR_L =   5e-12;                  % Lock accuracy [m]

ERR_csi = 80e-3;                  % Phase noise[rad]

phi_Hom = [0/180*pi, pi/2*90/90];      % Homodyne angle [rad] (you can input a vector of values)

det =  [100 60];                       % detuning frequency [Hz]

[Sato-san, Tanioka]

We installed a 4K shield which was modified by Namai-san and Ueda-san.
The procedures are as following.

1. Attached the cables on the shield as heat anchors
   At this moment, we found that the cable which was used for thermometer was disconnected. This was solved by bypassing the cable. But the displayed temperature seemed not correct. Calibration is needed.
2. Installed insulator on the shield.
3. Made holes on the insulators in order to pass through the beam.
4. Put the shield on the optical breadboard.
5. Check the cabling and the status of insulators.
6. Then clamped to the table.

[note]

• We found that we forgot to install the insulator under the optical table...
  I hope that the impact will be small enough.
• I forgot to prepare appropriate screws and washers.
  The ones installed temporary should be replaced.

[next step]

• Replace screws.
• Calibrate thermometer.
• Vacuum test
• Cooling test

Aritomi, Eleonora, and Yuhang

Last Friday, we had an issue with homodyne alignment. We found the frequency of peaks in shot noise is exactly the frequency of bench EW direction motion. At some point, we were thinking this is because of a bad way of aligning homodyne, but this seems not to be the case in the end. We figured out it should come from leaked LO backscattering.

The leaked LO mainly comes from the AR coating of cubic BS, then it goes to a sensor card in the squeezing path and the sensor card causes backscattering. About the leaked LO from homodyne, we measured power as following

And the sensor card is shown in the attached figure 1, which has a plastic shining surface. It will reflect a little bit scatter light which contaminates the shot noise spectrum. Besides, this sensor card is put in the position which makes it easy to sense the motion of bench in EW direction. So we got the spectrum as in the attached figure2. (In this case, we are using our homodyne as a very sensitive sensor of reflective surface motion)

We tried to characterize the effect by putting it farther and farther, but the contamination doesn't change. Until we put it after a Faraday isolator, which means we have an FI to isolate this backscattered light. We got a reduction of peaks by 15dB. See attached figure 3. But the backscattering noise didn't disappear.

But we need to notice that if the reflected surface doesn't move, we will not have this backscattering noise. For example, when we don't put anything in-between OPO and homodyne, although there should be light back-reflected from OPO to homodyne, we don't have backscattering noise.

Then we tried to send this leaked LO to filter cavity, we got the homodyne shot noise spectrum as shown in the attached figure 4. This measurement suggests that our low-frequency bump in FDS is from this backscattering noise. Also, the phase noise of squeezing will not introduce the increase of floor by almost 30dB because the anti-squeezing is only ~15dB.

Additional: we measured the backscattering noise from the sensor card when turbopump is on and off. We could see the difference from attached figure 5, which means it really introduces more bench motion. 

The scripts to open and close dithering loop has been changed from bash to python. This allowed to change the state of the loop input swiitch easily. This should make the loop engagment smoother in the case of integrators ( as we have for dithering) 

According to the investigations done by Shoda-san and Yuhang last week, it seems that the excess of noise appearing sometimes in the BS oplev is due to the PSD itself.

Today I have replaced it with a simiar one, with the same gain. The spectra of pitch and yaw are exactly the same of the old one (when no noise was present) and for the moment no excess of noise has been observed with the new PSD. I could close the damping loop without change any gain.  I will keep monitoring the spectra in the next days.

According to elog1868, there is a peak of bench appears at 14.2Hz when we excite EW direction.

Matteo and Yuhang

Today we tried to align homodyne to have flat shot noise until 10Hz. However, no matter what we tried(including aligning homodyne's BS, two lenses in front of homodyne, and adjusting the flipping mirror ) we always have a strong peak at 14.25Hz and its harmonics. We still don't understand why we have these peaks.

We checked the resonance of the bench and it is not the reason.

We checked we didn't send any modulation to IR phase shifter.

We checked we didn't send perturbation to IRMC. Also, IRMC error signal/reflection is very clean at low frequency.

We checked the main laser doesn't have strange behavior. We checked the noise eater. We checked PLL.

We will continue the investigation next week.

I checked the shape of spaer which was re-machined with two half inch mirrors at output port and folding port.
The beam can pass through the spacer and the reflected beam can come out.
Therefore the shape seems roughly O.K.

The next step is that gluing the fused silica mirrors to mirror holders for installation of the folded cavity.

In addition, I installed a post and a translation stage in TEM00 path for mode matching lens.
I will continue installation of the posts and stages for other beam paths.

Today, just after I align GR well (GR transmission is at the level of ~2550counts, with 12.28mW GR injected to FC) to FC, I checked BAB in FC transmission and reflection.

For transmission, the level is 340counts. Then I measured the injected power into FC, it was 0.333mW. By comparing the standard number we recored in WIKI, transmission is misaligned by ~20%. ((340-100)/(392-100)=82%)

For reflection, roughly also 20% is misaligned. (check the last attached picture for reflection in to AMC)

Last time we aligned BAB to FC is this Monday.

I tried to align FC again and the second time, TRA got misalign 60% while reflection is also roughly 20%. So reflection seems to be less sensitive to alignment.

I have updated the RT model and medm screen.

- The WD system is added.

- The commish message is also added and that can be seen from the sitemap.adl. Here, you can leave the comment when you do not want someone else to touch the FC remotely.

Shoda and Yuhang

We checked the coherence between each suspended mirror and CC2 correction signal.

The coherence below 0.5 is not mentions in the following form, check detail in the attached pictures.

Shoda and Yuhang

We checked the coherence between each suspended mirrors oplev signal and correction signal to FC length. The situation is summarized as follows.

The coherence below 0.5 is not mentioned. For the detail, please check the attached figures.

Yaochin and Yuhang

Since last week we had again the problem of filter cavity. We saw filter cavity transmission varied from 2000 counts to 3000 counts. In the end, we found out actually we are not operating SHG in the optimal temperature. Also, we had a temperature change due to the change of air-conditioner mode, so it varied.

After everything settles down, we fixed again the setting of the filter cavity lock loop.

Then we measured the open-loop transfer function again. Also, GR locking accuracy and IR locking accuracy. Especially this time we measured more accurately the low-frequency part of it. And compared with the measurement we did more one year ago(elog 690), the high-frequency part is the same while the low-frequency part is quite different. We should consider more why we have this big difference in IR error signal now.

 We should also measure coherence between IR error signal and each oplev spectrum.

I am sorry that what I wrote is wrong. The additional loss is 1-visibility**2. I think it is very clear for us that the efficiency of homodyne is visibility**2. This is written in Henning's thesis.