This entry reports the birefringence measurement of TAMA #1 with 1.5cm radius, 0.25 mm step size, 0.5s wait time and 0.488mW input power.

Due to time constraints only 3 measurements were possible.

Note that the AC and DC was realigned before input polarization calibration and with sample and that here 0 deg polarization angle corresponds to s polarization.

The analysis code was checked using the Shinkosha7 data and produced identical results.

The main result is that polarization, theta and delta_n are quite uniform but the main variation corresponds also to the maximum absorption (see entries 2723 or 2672) while the triangular patterns does not seem to affect significantly these parameters.

Matteo, Takahashi-san, Yuhang

Takahashi-san removed the jig and released BS earthquake stop this morning. Before closing vacuum chamber, we checked transfer function of BS and pitch/yaw coupling.

The transfer funtion shows in the first attached figure. We can see that the red line is close to reference below about 10 Hz. The deviation at low frequency is probably due to the air flow, while the deviation at high frequency is presumably due to acoustic coupling.

The pit/yaw coupling shows in the second attached figure. The green shows the coupling when we have magnet drop. The red curve is after we put new BS magnet, which got close to reference (blue curve). This figure shows that the BS mirror magnet replacement work is successful.

After confirming TF and coupling, we closed both PR and BS chamber and evacuated them. Now, we are using turbo pump to evacuate. Until the vacuum level reaches about 1e-8 mbar, we will open the gate valve between PR/BS and Input.

Matteo and Yuhang

To check what is the rotation angle of the Faraday rotator, we removed the HWP after it. We found the 2nd PBS reflected IR power decreased from 458uW to 192uW after removing HWP. This tells us that the rotation is arctan((458-192)/192) = 54 degree.

By adjusting the relative position between magnet and TGG crystal, we achieved reflection 213uW and transmission 226uW. In this case, the rotation is arctan(226/213) = 47 degree.

We put back HWP after optimizing HWP and found a transmission of 6uW from the 2nd PBS. After rotating HWP, we reduced it to 4uW.

In the end, we checked round trip loss of Faraday isolator using a IR HR mirror reflecting back the reflection from dichroic mirror. We found injection of 449uW, 2nd PBS reflection of 438uW, and total reflection of 390uW. The round trip losses from Faraday isolator is 11% now.

We found out that the TAMA 1 fast axis orientation was also quite uniform and similar to the aztec sample...

Following entries will report all measurements that were taken but here we report the various checked performed :

- with 14.7 mW input power and OD 2 the maximal power reaching the QPD was below 0.6 mW (maximum recommended by Thorlabs)

- the output voltage of the QPD with this optical power is about 1.4 V (consistent with the responisitivty of the QPD at 1064 nm)

- to avoid saturation of the lockin amplifier (at 1V) we placed 6dB attenuator and the lockin voltage was indeed divided by 2

- the spectrum of the signal show clearly the peak at the chopper frequency and harmonics (note that 3rd harmonic is stronger than 2nd)

- the beams alignment before the QPDs were tuned to always be inside the sensor with and without sample

- the 2 QPDs signals have the expected behavior when changing the input beam polarization angle

- the maximum of the QPDs with sample correspond to about 0.9 of the maximum without sample (consistent with sample reflectivity checked with power meter)

- for the last measurement (will be precised in the entry) we reduced the incident power to 0.588 mW, removed the OD and the laser linewidth to avoid as much as possible possible back-scattering and recentered the beam with and without sample before every measurement

Yuhang, Michael

We took the opportunity to measure the vacuum FI loss when the vacuum tanks were opened, using BAB. The vacuum Faraday isolator assembly can be seen in the photo (figure 1). The polarizers are two mirrors tilted at 55 degrees (Brewster polarizers). The black tube is the Faraday rotator. Attached to the end of the rotator is the half wave plate. From the reflection of the second polarizer, the IR beam hits the dichroic mirror, where it is recombined with green. Instead of propagating to the filter cavity, we used a mirror (furthest left in the photo) to reflect the beam back through the Faraday. Then, we measured the power in the following locations:

P1 in: Incident power at the first polarizer, coming from the squeezer table
P1 ref: Reflected power after P1 in
P1 tra: Power transmitted through the first polarizer, with respect to P1 in
FR/HW tra: Power after one pass through the Faraday Rotator and Half Wave plate
P2 ref: Power reflected from the second polarizer after coming from the Faraday
P2 tra: Power transmitted through the second polarizer, with respect to FR/HW tr
Di in: Incident power on the dichroic mirror
Di ref: Reflected power from the dichroic mirror
RT P1 tra: Power transmitted through the first polarizer after two trips through the Faraday.

There is some weirdness with the power meter measuring more power after reflection from certain components, but it consistently happens in certain positions (P1 ref, P2 ref, Di ref), which makes us think it's more to do with the power meter geometry. It makes precise measurement of losses to be difficult. Still, we can draw two conclusions: 1) the polarizers have low transmission of the order of 1%, and 2) the round trip loss through the FI assembly is about 14%. We took several measurements at the positions that were causing problems. Also, the p-pol PLL unlocked at a couple of points (between columns 4/5, 10/11 on the table below)

Power measurement (units of uW)

Avg. RT P1 tra/P1 in = 0.857
RT P2 transmission = 2 uW

Takahashi R., Matteo, Marc, Yuhang, Michael

The BS chamber was opened and the magnet was glued.

Photos of the process are attached.

The BS vacuum tank was closed on Friday, and then the PR chamber was opened and the portable cover moved over PR.

Katsuki-san, Marc

We removed all the readout of the birefringence. The reason being that the photodetectors size was quite small wrt to the beam size. This means that performing the alignment without sample might not be so accurate for the case with sample.

First we replaced the optical breadboard from imperial holes size to metric ones.

Then we reinstalled the PBS, the ODs and tried to installed the new PSD (PDQ80A) that have quite larger sensor size but are also sensitive to visible light.

In order to have a larger beam on the PSD, we replaced the lens in front of the PBS by a 100 mm focal one.

Sadly, because the PSD sensor is 4 cm above the screwing hole, it was quite tricky to find proper post.

Therefore, we also installed 2 steering mirrors just after the ODs.

Finally, we installed laserline filters right in front of the PSDs.

We took time to put label on various cables.

The alignment was checked (ie we confirmed that the maximal power of both ac and dc is reached without sample).

I just started birefringence measurement of the TAMA T1 sample with s polarization at the input.

The 'bad news' is that the gain of the PSD requires to change the gain of both lockin amplifiers depending on the input polarizations.

Yuhang and Michael

When we went to place the incoupling mirror, we found that the screw threads for adjusting the incoupling mirror position were mismatched - the screw threads on components 1 and 28 (assiemepdopov3asm.pdf) have 1mm pitch while the micro adjuster screw has much smaller thread pitch. It will take some time to have the proper components ready.

This sample absorption has been remeasured as there was some discrepancy with measurement few years old.

This one agrees with the one I did few months ago : absorption is around 18 ppm

(Pin = 7.316W, Pt = 6.29W, R = 0.8342 ppm/cm)

There are however strange patterns on the XZ YZ measurements that were not present before.

We started evacuation of the BS chamber yesterday. I opened the gate valve between BS and NM2 today.

Fitting this data with a cosine gives input polarization equal to output sample polarization for angle about 41.34 deg and the phase about -6.5 deg

Marc, Matteo, Takahashi-san, Yuhang

We opened BS chamber today to check if there is any issue for the magnet.

We found one magnet falls down as shows in the attached figure.

Takahashi-san will help us prepare a replacement magnet. The replacement work will be done in the next next week.

We have done some more check before opening the BS vacuum chamber.

1. Check BS P and Y transfer functions. They are similar with reference as shows in the fig1 and fig2.

2. Check coupling between P and Y. A substantial coupling increase is found as shows in the fig3.

3. We tried to close BS local control loop. But failed.

The results of test 2 and 3 make us really need to open BS chamber to check. The opening work is scheduled to be in this afternoon.

Today I tried to tweak the input pump power and did several long Z scan to find the 2 surfaces of the sample.

Because it did not succeed at first, I reinstalled the bulk reference sample and got R_bulk = 0.8342 cm/W

Finally, by taking 10 averages and mean I could see the surfaces and started the measurement at Z_center = 62.75 mm

Using the value of this long Z scan the absorption is consistent with the previous measurement :

AC ~ 1.5e-4V

DC~4V

Ptrans = 6.290W

Pinc = 7.315W

gives absorption ~ 22 ppm/cm

Measurement is on-going with 0.5mm step size and 30 mm radius.

Do not touch the sample this week !

Katsuki-san, Marc

Today we removed the AZTEC sample in order to prepare the absorption and birefringence measurement of the TAMA T1 sample.

We reinstalled the steering mirror in the injection part to dump the sample reflection (see fig1)

We installed a flipping mirror before the birefringence readout to redirect the pump beam to the high power power meter (see fig2).

We reconnected the absorption photodiode to the lockin amplifier.

We installed the surface reference sample and with z = 42.3mm and zIU = 66 mm we got R_surf = 18.72 /W

We installed the bulk reference sample with z = 40.95 mm  and zIU = 65.7 mm and got R_bulk = 0.6679 cm/W.

We installed the TAMA T1 and using DC values could find Y_center = 121.92 mm and X_center = 326.74 mm.

We increased the pump laser current to 6A (beforehand we changed the HWP angle that control the power to 9 deg that corresponds to the min pump power) which gave Pt ~6.1W.

We realigned the DC at the previous z_center (62.4mm) and did long z scan but could not find clear surfaces absorption neither on the AC nor the phase.

This seems to confirm that this sample absorption is quite low ( previously measured at 18 ppm).

Because it is the first time using such high power with all the birefringence optical component we turned off the laser today and will start the absorption measurement tomorrow morning.

Katsuki-san, Marc

Today we checked the relationship between the input polarization angle and the AZTEC sample polarization angle.

We confirmed that thanks to the realignment the maximal power of both photodiode is reached without sample.

Further analysis to follow.

Here are the figures.

During the week-end we took birefringence measurements of the AZTEC sample rotated by 180 deg and back to same orientation as the absorption measurement (ie 0 deg).

Due to some mistakes it was not possible to perform too many measurements (once labview is exited it reset the lockin amplfiier to absorption measurement)..

To avoid confusion I report here the measurement with 180 deg rotation.

Recently, it is found that GRMC loop cannot be closed. To check what is the problem, I disconnected the slow scan of MZ and put MZ PZT offset to a value that GRMC transmission is the highest. After that, according to the original setting, I have done following checks

0. GRMC has a good alignment.

1. PDH signal has 316mV pk-pk checked from EPS1.

2. GRMC has loop sign of INV, which is as design.

3. The RF source phase is reloaded. The phase of RF source is 125deg. When it is changed to 35deg, the signal around resonance becomes flat. This indicates the RF signal phase is still a good one.

4. There is a switch which has +/- sign. This doesn't decide the sign of control loop. But when we use this type of servo for CC1/2 controls, we need to flip this switch. I tried to flip this switch, but it doesn't help to close loop.

5. GRMC transmission is checked to have 1.13V peak. This is two times smaller than the value written by Pierre.

6. Loop gain is 3 as usually used.

7. Threshold for peak identification is -0.55V. This is as required.

8. The GR power reaching AOM is measured to be 44mW, whose nominal value is 50mW.

These checks show little issues but they should not prevent the GRMC locking. More investigation is required.

After the earthquake on 7th this month, we confirmed the normal operation of vacuum system, mirror suspension system. Especially, the filter cavity alignment was recovered and mirror's oplev shows usual spectra.

However, we found already that the driving of BS has problem. The problem is when we try to move yaw or pitch, the beam reflected by BS moves diagonally.

To confirm if there is any issue for magnets, I took a photo of BS magnets. Attached photo shows this check. Two red boxes indicate the location of four coil-magnet actuators. An arrow points to the magnet which seems to be broken. This magnet may be knocked during earthquake and get tilted. So the coil cannot drive it properly.