Yuhang and Michael

We removed First Contact from the mirror and inspected the cleaning. It seems like some residue of the rubber stain remains and there is a lot of scattered light (fig 1 and 2).

After this we applied alcohol solution from a bottle in the ATC cleanroom (isopropanol based, I think it was called "Clarity" or something like that) in an attempt to remove water soluble stains (fig 3). Then after removing we reapplied First Contact.

The BS issue has been checked again today by switching off the oplev laser since I was suspecting that BS PSD maybe broken. Indeed, after switching off BS oplev laser, the BS oplev signal is as Fig.1 which is the same as what is taken last Thursday. So the strange signal we are seeing is from the broken BS PSD.

To confirm the health of BS suspension, we need to take an another PSD and use it to characterize BS suspension.

Today, we have some progress about guardian deployment in TAMA. Thanks to the help from Yamat. The detailed information can be found in the #filter_cavity channel of gokagra in Slack. Here, I put a summary of my activity today.

1. Yamat helped us check simulink file and found no problem.

2. Yamat suggested to run caget K1:FDS-FC_GR_TRA to check potential issue on client workstation of desktop1 and k1grd0. Running it on desktop1, I got -27.7212 for K1:FDS-FC_GR_TRA. But running it on k1grd0, I got channel connect timed out for K1:FDS-FC_GR_TRA.

3. Yamat suggested to check environment values in guardian computer. To do that, I used 'env > tama_filter_cavity_k1grd0_env_out.txt' to save the environment variable of k1grd0 workstation. Then I used 'scp tama_filter_cavity_k1grd0_env_out.txt controls@192.168.11.110:/home/controls/Desktop' to copy environment variables from k1grd0 workstation to desktop1 workstation. To share with Yamat on Slack, I firstly uploaded the txt file to dropbox and then share link with him. (Note that I didn't just take screenshot because the environment variable information is large.)

4. After Yamat checking env output, he found that a variable called 'EPICS_CA_ADDR_LIST' seems to have a wrong IP address. He suggested to change this from '/home/controls/.bashrc or /kagra/apps/etc/client-user-env.sh.' and reboot k1grd0.

5. I found the channel 'EPICS_CA_ADDR_LIST' is in /kagra/apps/etc/epics-user-env.sh. So I did the modification and reboot k1grd0. But the problem is not solved.

6. I found after modifying from /kagra/apps/etc/epics-user-env.sh, the env output in k1grd0 is still as before and didn't change. Maybe this is why the problem is not solved. Now I am waiting for the answer from Yamat.

Yuhang and Michael

This is a followup to the issues discussed in 2769. 

The OPO cavity was aligned and we attempted to optimise the mode structure of the transmission spectrum. However, there were two second order HOMs we could not get rid of by altering the input steering mirrors, and inspection of the transmission camera showed a lot of scattered light around the main beam.

Today, when we looked at the setup, we saw a lot of dirt on the incoupling mirror, which was visible when shining a phone flashlight below it. So we took it out to apply First Contact on both sides. At this point, we also saw a mark from the rubber o-ring that was mounted on the mirror. There may be some inspection needed for the OPO as well. However, I do not personally recall seeing the scattered light problem when the OPO was being aligned without the incoupling mirror, the most I remember seeing was something like a diffraction pattern (attached figure).

We were also in doubt of the position of the OPO inside the holder apparatus. There is some freedom in the beam propagation direction with respect to the OPO placement inside the holder, and there was no specific instruction given in the assembly directions. Given that the beam size required is extremely small , it seems like there would be strict mode matching required here, and yet we also did not see any LG modes when scanning the cavity.

For now, we applied clear First Contact to both sides of the incoupling mirror. We also want to redo the OPO alignment. Most of previous works have ensured the beam propagation axis is very closely aligned to the OPO axis. But, the incoupling mirror position was not that finely adjusted last time.For example, if the beam is hitting the incoupling mirror off-center while not passing through the center of curvature, the wavefronts will not be perpendicular to the curved surface of the mirror, so there will be mode mismatch introduced. The micro adjusters for the incoupling mirror only adjust transverse x and y position, not tilt.  So we should optimise the beam position on the incoupling mirror, and then for locking make fine adjustments to the input beam steering mirrors.  

Today I did 2 long Z scans (between 20 and 100 mm then 100 mm to 110 mm) with 0.05 mm step size and median/average filters order 10 and could see the phase transition at Z = 37.6 mm and Z = 104.95 mm.

It means Z_center = 71.275mm.

The XY map started with 70 mm radius, median/average filters order 10 for a ~14h long measurement Pt = 6.541 W and Pin = 7.582 W.

It means that laser will be on up to December 30th for the 3D measurements.

Abe, Marc

Here are the vertical and horizontal angle of incidences used during the briefringence measurement of the spare ETMY.

vertical : 4.78 10^-3 rad or 0.27 deg

horizontal : 4.33 . 10^-2 rad or 2.48 deg

To compare the IR locking accuracy (CCFC error signal) for different FC green gain, I compared the 1/f^4 and 1/f filters of FC green lock. The 1/f^4 filter has much larger gain than the 1/f filter at low frequency. The nominal filter is 1/f^4 and the nominal gain is 1.5. For the 1/f filter, I used the FC gain of 5 to have the UGF of 14kHz. Fig 1 shows FC OLTF for 1/f^4 and 1/f filters. Both filters have UGF of 14kHz.

Fig 2 shows the FC green error signal (EPS1, 1Vpp fixed range) with 1/f^4 and 1/f filters. The FC green error signal with 1/f^4 filter at low frequency is below the dark noise and limited by the spectrum analyzer noise. This means that increasing the green gain just reinjects the dark noise.

Note that BS pointing and AA were engaged during the measurement, but Z correction was not engaged due to the problem. I found that 1/f filter with gain 5 is stable even without Z correction, but 1/f^4 filter is unstable without Z correction.

Then I aligned BAB to FC. The optimal p pol PLL frequency without green was 275MHz and BAB power before FC was 466uW. The maximum IR transmission of FC was 490.

Then I checked the CCFC error signal. I optimized the p pol PLL frequency to maximize the CCFC error signal. The optimal p pol PLL frequency with 20mW green was 230MHz and the CCFC error signal was 122mVpp.

Fig 3 shows the CCFC error signal with 1/f^4 and 1/f filters. We can see that increasing the green gain improves the IR locking accuracy above 10Hz, but not below 10Hz. The IR locking accuracy is dominated by the FC length noise below 10Hz and the laser noise above 10 Hz. By increasing the green gain, the laser noise can be reduced because the laser noise is common for both IR and green, but the FC length noise cannot be reduced because the FC length noise is different for IR and green.  

Michael and Yuhang

Today, we have successfully closed OPO cavity. The OPO no.28 component issue is solved by replacing it with a longer screw.

After cavity closed, we optimized OPO cavity alignment relative to its incident beam. However, we found the scan spectrum is as Fig.1.

1. We can see that there are apparently two higher order modes in the scan spectrum. By scanning OPO cavity with 10mHz ramp and looking at a designated camera , we found that the higher order modes are second-order Hermit Gaussian modes. (The cavity scan video is here https://drive.google.com/file/d/1Od2jI40D_3oZuZksuI474tunwtt7znIs/view?usp=sharing , note that it seems cavity mode is not centered in a circle. But this circle is just a mirror. The cavity mode looks to be well centered when mirror is removed.)

2. We can also see that there are modes overlapping with TEM00 mode. It indicates that we are having a cavity length which offers a non-ideal Gouy phase.

3. When we misalign pitch or yaw of the incident beam, we found the first order mode appears but the existing second order mode decreases. This is wired for me.

4. We have identified a potential beam clipping in the optical path before OPO. After fixing the clipping issue, we didn't get better OPO scanning spectrum.

5. We suspect that we may fix either crystal or in-coupling mirror too tightly, so that mirror has distortion and introduces higher order modes. After loosing a bit the screws, we didn't see improvement.

6. We suspect that there maybe some contamination. Especially, the crystal was in air for several months. But we are not sure if contamination is really true because the cleanroom in ATC is designed to be a class 10 cleanroom.

Personally, I think we need to open OPO again and adjust maybe crystal position.

Abe, Marc (with help of Michael and Yuhang to safely move the spare ETMY)

First we removed the spare ETMY.

Then, we used the razor blade and high power power meter connected to the lock-in amplifier DC to check the pump beam angle of incidence in vertical and horizontal planes. (analysis to follow)

We installed the surface reference sample and got R ~15.7 /W without tuning the alignment.

We tuned the alignment and got R_surface = 18.537 /W with Z_translationStage = 42 mm and Z_IU = 66 mm.

We also measured R_bulk = 0.6703 cm/W.

We reinstalled the spare ETMY with the 2 ears flat (ie same configuration as in entry 2755.

We tuned Z_IU to 5.4 mm taking into account the thickness after repolishing (14.3 cm instead of 15 cm).

We started to look for the 2 surfaces signal increasing the input pump power from 2W to about 7W (in transmission of the spare ETMY).

At that point we may have seen the expected signal with median and average filters order 10 but because the step size was 0.2mm we could not clearly see them.

For reference, the last time that the spare ETMY absorption has been measured the input pump power was about 10 W (see for example entry 1601)

As it was getting late on Christmas eve we stopped there and will start absorption measurement on monday.

Aritomi, Yuhang

There was an earthquake on 12th Dec. which caused mis-alignment of suspended mirrors. We worked on the re-alignment on 22th Dec. During the re-alignment, we found that the BS picomotor can be only moved in pitch direction but not in yaw direction. Fortunately, we were able to bring back the filter cavity alignment even in this situation.

Since BS picomotor couldn't be moved in yaw direction, we looked into the BS oplev spectrum. The result is shown in Fig.1 and we can see that the oplev signal is totally off from the nominal value. Note that input oplev was different because we took pre-amplifier from input temporarily.

For the time being, we can still work with this condition. But maybe we should correct BS mirror from touching in the near future.

[Aritomi, Yuhang]

We continued the investigation of the 20kHz peaks in FC green error signal. We checked the IRMC error signal and there are no 20kHz peaks in the IRMC error signal. So the 20kHz peaks should not come from laser.

Then we measured SHG OLTF and found that there is a peak around 20kHz as shown in Fig 1. In this measurement, the SHG gain was 1.3. We decreased the SHG gain from 1.3 to 1 to avoid the 20kHz peak, but the FC green error signal is still noisy.

Regarding the Z correction, we found that the dc_damp2 filter was completely different from the one written in elog2608. We changed the dc_damp2 filter to the one written in elog2608. We set the dc_damp2 filter gain as 10 in foton and 4 in medm. However, FC is still not very stable.

Marc, Michael

To facilitate the removal of the mirror, we used the knob controller of X direction to move the mirror.

After that, we found out that the Zaber reading got good so we started the measurement with 45 deg input polarization angle.

We found that FC green error signal is very noisy. Fig 1,2 show the FC green error signal before/after lock. Fig 3 shows the spectrum of the FC green error signal after lock. There are two peaks around 20 kHz in the FC green error signal. It seems that the two peaks come from SHG. The SHG error signal is also shown in Fig 3. The SHG gain was 1.4. We decreased the SHG gain from 1.4 to 1, but the peaks around 20 kHz did not change.

During this investigation, we checked green power and OLTF of FC green lock. The measured green power is as follows. The green power seems reasonable. The UGF was 14kHz, which is also reasonable. Note that FC gain was 1.6.

We also found that the Z correction is not very stable. We quickly checked the filters in filter bank of Z correction as follows.

The bad filters immediately caused FC unlock when they were engaged. The not bad filters didn't cause the immediate unlock, but FC was not very stable for long term. We need to investigate the Z correction.

Michael and Yuhang

Some cables need to be connected to the new OPO, such as PZT cables, thermometer cables, and Peltier cables. The old connection was not firm enough and we decide to use a new way to connect cables. This new way is suggested by Tomaru-san and it uses a copper tube as interface. After putting the two sides of cables inside the copper tube (Fig.1,2), we use a plier to clamp two cables so that the cables are fixed (Fig.3,4). Then we use Kapton tape to isolate the copper tube from environment (Fig.5).

After fixing, we found the connection is firm enough. We also tried to drive PZT and found the driving is successful.

Michael and Yuhang

Today we found a new issue about new OPO component No.28. The issue is described in the attached PDF file. The issue is that a counter bore hole is made from a wrong side of no.28 component.

We also took photos about no.28 component with wrong and correct holes, as shown in the attached Fig. 1 and 2. (The one with correct hole actually has wrong thread as reported in elog2724)

For solution, we found that a longer screw seems to be able to solve the problem. We will look for some longer screws and clean them. Then try to use them for the new OPO.

Again issue with the translation stage so we have to remove the mirror to reset the motors reading from Zaber..

I deactivated the windows update for 1 month

I checked data especially of the measurement with 45deg input polarization angle.

It seems there was some issue with the calibration of the DC signal (the DC with approximately 45 deg input polarization is higher by a factor 2 than the expected maximum value when we inject pure p polarization).

We will redo this measurement again today.

To realize the CCFC+green OLTF with green only, we need the following green filter.

1 + G_CCFC/G_green = 1 + P_CCFC/P_green*G_f*P_f 

where G_CCFC, G_green are the OLTF of CCFC and green, P_CCFC, P_green are the cavity pole of CCFC and green, and G_f = 1000, P_f are gain and pole of the CCFC filter. The pole P can be written as

P = 1/(1+i*f/f_p)

where f_p is the pole frequency. f_p for P_CCFC, P_green, P_f are 57 Hz, 1450 Hz, 30 Hz, respectively.

This ideal filter is complicated to realize with combination of zeros and poles. Instead, we can use a simple filter which is the combination of 2 poles and 2 zeros. The pole and zero frequency are 30 Hz and 1000 Hz, respectively, and gain is 1000. As you can see in the attached figure, the simple filter is similar to the ideal filter. 

Abe, Katsuki, Marc

This entry reports the birefringence measurements of spare ETMY without roll rotation (ie with the 2 ears flats).

delta_n and theta seems quite consistent over all measurements.