Following the investigation in the entry #63491, this morning we perfomed a set of noise injections on both laser bench and external injection bench.
We started with the injection on laser bench by means of shaker installed few days ago #63577.

LB NOISE INJECTIONS
ITF locked

DAC channel: ENV_NOISE_CEB_EEroom

Freq. line 17.9 Hz, A=0.1 V, 09:59:55-10:13:57 UTCFreq. line 8 Hz, A=0.7 V, 10:37:20-10:42:46 UTCFreq. line 8 Hz, A=0.6 V, 10:50:58-10:57:20 UTCSecond part of the shift ITF unlocked (INJ standalone)Quiet = 15:35:00 (5’)  Freq. line 17.9 Hz, A=0.1 V, 15:40:20-15:48:00 UTCFreq. line 8 Hz, A=0.7 V, 15:49:51-15:58:36 UTCFreq. line 8 Hz, A=0.6 V, 16:00:04  - 16:06:01 UTCColored noise (1-15) Hz: 16:18:27  - 16:23:39 UTC

EIB NOISE INJECTIONS

ITF unlocked (INJ standalone)

(previous EIB injections #61713)

We injected a sine of 0.5Hz frequency (Henk Jan adjusted the cfg file to have the right amplitude by just enabling sine noise button)

The analysis of these injections will follow.

Triggered by a question from Romain Gouaty, here some little clarification about the IPC2 current setting as for the transmission of the beam to the ITF (and back).

Back in September 2021, we tuned the transmission of IPC2 from its maximum down to a value which ensured around 33W of input power to the PR mirror. This was done by looking at the power transmitted by the arm cavities (since the monitoring sensors we have in INJ are all upstream wrt IPC2 action). In the first plot, we have IMC_TRA_DC which is the power transmitted by the IMC cavity, and ITF_Input which is the channel calibrated from IMC_TRA by taking into account the known losses (7.5%) when the IPC2 is at the maximum of the trasmission. The tuning of IPC2 does not change these channels, while it decreases the power transmitted by the arm cavity (SNEB_B7_DC). We had a first adjustment of IPC2 on Sep 30 2021 and then a slight retuning on Feb 16 2022.

This estimate of the IPC2 Transmittivity depends on the matching conditions, so it is accurate up to around 1%.

The decrease of the power delivered by the Injection is mostly due to the drift of the amplifier (see plot 4, 1 year of Injection powers).

To be noticed that the effect of the attenuation by IPC2 occurs both on the beam sent to the ITF and on the beam backreflected by the PR. A figure taken from from Eric's presentation VIR-0094A-13 illustrates the issue, basically depending on the fact that there is a half-wavelength plate in between two polarizers. So the overall transmission of the backreflected beam from the IMC_TRA to SIB2 (to be added to all the other losses) is (IPC2_T)^2.

This morning we checked and tried to improve the matching of the input beam on the ITF after TCS activities and the change of the Chrocc power (see 56633).

We started from a mismatch of 1.9% in the west and 2.3 % in the north moved in total INJ L1z by 80000 steps in the backward direction, reaching 1.5 % for the west and 1.7 % for the north (see attached plot).

We could improve it further but for the moment we stop to allow the locking of the CITF.

The goal of the shift was to optimize the matching of the arms in a configuration which is as close as possible to the one at carm null (and then the NARM mismatch should have been optimized with TCS of NI CP).

We used the new PR parking position (-20 urad in TY) and adjust the input power on IPC2 in order to have the same amout of power in the faraday (back and forth) as the one we have at carm null.

The power was reduced by 32% so the input power into ITF is reduced to 10.5 Watt,

We did the following step for the meniscus lens in order to optimize the matching in the west arm which reaches about 1%

the outcomes of the measurement are shown in Figure 1. Unfortunately the planning of the afternoon and of the following days have been changed, thus this work will be reverted today.

After the swap from the fibered amplifier to the neovan there was a bit less power in transmission of the IMC (see plot 1).
We used the IPC2 situated after the IMC to bring the power back to what it was before the intervention. We took the power transmitted by the arms as a reference. (see plot 2)
We stopped increasing when we reached the same power on B7. It actually might still miss 0.5-1%.

The power is fluctuating but from the plot there was:

43.4/41.4 = 4.8 % missing on IMC_TRA
1.075/1.03 = 4.4 % missing on B7
1.1/1.05 = 4.8 % missing on B8

We increased by
1.075/1.03 = 4.4 %
1.085/1.045 = 3.8 %

To be noted that the beam going back towards B2 sees the IPC2 two times, there is a quadratic effect on the power, so about 8 % increase with respect to what it was this morning. 

After having increased the laser power in order to have around 40 W injected to the ITF (#53212), this morning we measured the mismatch of the input beam to the arm cavities.

We procedeed as already described in #52880, giving a kick at the input mirrors.

• open dampers: F7_TX and F7_TZ to 0 and ZCorMi to 0
• unlock arm
• kick to the NI: Set ADD zCorMi to 5 and then again to 0 (just with go, without ramp) (p.4 ScNICDMir) and wait for around 3 min. 
• close again the dampers and relock the arm.
• extract and analyze data with the matlab code in /users/optics/Commissioning/180323_matching

The gps for the North arm is 09:51:04 UTC and we measured 3% mismatch (attached fig), which is not far away with the value reached after the last tuning (#52880).

The gps for the West arm is 10:33:05 UTC. However we didn't manage to extract a value for the mismatch because the code wasn't able to recognize the mismatch peaks. We still have to investigate this issue.

All the controls has been closed and seem to work as before.

The setting of the previous working point required the use of many motors:

F0 trolley 1: 620 steps CCW

F0 trolley 2: 20 steps CCW

F0 trolley 3: 1230 steps CW

F0 fishing rod: 800 steps CCW

F4 fishing rod: 900 steps CCW

F7 fishing rod: 1000 steps CCW

MA_TZ balanc (for pitch): 1900 steps CW

MA_TX balanc (for roll): 250 steps CCW

We can consider those adjustments quite relevant, but all well inside the range of the motors, which remain usable for future adjustments.

During the maintenance, we went to the laser lab in order to perform a measurement of the power on the B2 beam, and to make a visual inspection searching for possible ghost beams.

To perform this measurement, all the interferometer mirrors except for PR were misaligned.  

As soon as we opened the minitower, we could notice with the viewer that there was a lot of dust hitting the beam, so we decided not to install the calorimeter to make the B2 power measurement, since it has a fan which could possibly scatter the dust on the optics.

We continued with the visual inspection in order to spot possible ghost beams. The main scatter light source seemed to come from a mirror mount being part of the IPC3 periscope (figure 1, 2). This could be light scattered from the IPC3 itself, as we have a lot of power in the beam. Notice that in this configuration (only PR aligned) there is about 4 times more light than in LN3. We couldn't do anything in order to dump this light, as a custom shaped absorbing glass is needed.

Moreover, we noticed a ghost beam along the RFC_refl path which was hitting the chamber door. In this case, we installed an absorbing glass to dump it (figure 3).

This afternoon we relocked the IMC cavity to 16 W, with the automatic alignment in drift control and the BPC loop closed (look at the fig.).

We then shutdown the slave laser diodes and the amplifier, leaving only the ML turned on, because of safety reasons due to the weather conditions.

The missing plot

Today after the cavity has been aligned, by hitting the center of the NE and NI mirrors (using the dithering lines), and the B7 diodes have been centered the input beam has been aligned by changing the offset on the error signal of the PR vertical control.

The power drops of a factor 4 for a PR vertical displacement of ~ 100 um. The control of the beam in the cavity alignment can be done with the vertical position of PR.

Today was connected the two last electro optical modulators to the associated radio frequency amplifier.

As shows on the attached plot, all the frequency modulation are now available and set in order to be resonant into the IMC (except for the 22 MHz).

6.270777 MHz

8.361036 MHz

22.304000 MHz

56.436993 MHz

131.686317 MHz

This afternoon we installed the baffles BD1 and BD2 (final version in SiC) on the SIB2 bench, see attached picture 1 where the baffles are circled in red. Those baffles are diaphragms with 40mm clear aperture, 90mm external diameter, 4mm thickness made by Ceramtec. No AR coating was foreseen on the surface of the baffles as they are mainly intended as protection when the beam is slightly misaligned.

Images 2 and 3 show BD2 and BD1 respectively, installed on the bench to replace the temporary aluminium versions.

IMC cavity relocked.

The IMC cavity has been unlocked to allow PC team to take some measurement with the phase camera installed on the EIB.

We took advantage of the intervention in th INJ tower (entry 32942) to replace the diaphragm SIB1_BD5, previously realized in bare stainless steel, with an AR-coated part in SiC. This ensures a much higher Laser Induced Damage Threshold (around 30KW/cm2), a low TIS (nominal TIS<100ppm) and low reflectivity (R<0.5% AOI<20deg).

A suitable counterweight was added at the same location of BD5 ( but bottom side of the bench) to compensate for the weight difference between the part in stainless steel and the part made of SiC. The counterweight can be partially seen in the second picture: a cilindrical post just below the BD5 diaphragm. Paolo checked that the balancing was acceptable.

We would like to thank Michele Bazzi for his prompt fixing of the support of BD5 when we discovered that the SiC version was some 0.2mm larger than expected.

Last week I've measured the TF between the Laser amplifier and the IMC TRA DC, PSTAB PD1 AC and PSTAB PD1 DC signals when the PSTAB analog loop is on (the high frequency loop). The noise has been injected in a band from 0 to 200 Hz with low amplitude in order to not saturate the analog loop for ~ 1500 sec. The measurements are shown in Figure 1. It is worth to be noted that for PSTAB PD1 AC we have an analog high pass filter, visible in the curve.

The fit of the curve for PSTA PD1 DC is shown in Figure 2 and the results is :

the gains are

The DAC is a LAPP DAC in DAQ box in EEroom (Dac1955_INJ).

This morning, we continued with the investigation about the noise in the output of the DACs in the EEroom, 8 channels DAC, see logbook entry 32713.

We tried at first to measure it with a LAPP ADC, see figure 1 where for the DAC output a temporary signal has been used (INJ_IMC_QF_I_H_TEMP).

We were not able to see the auto-oscillation described in 32713,  and it can be due to the low pass filter present in the ADC signals together with the (very) high frequency of this oscillation (~100kHz).

If instead we measure the signal with an oscilloscope or a spectrum analyzer we can see clearly the oscillation.

See figure 2 and 3 where a 200 Hz sinusoidal signal with 0.3 and 0.5 V of amplitude generated by the DAC is shown.

In order to understand if it is due to the fact that the oscilloscope is single ended while the DAC output is differential we also measure it with a spectrum analyzer with differential input.

As it is visible in figure 4 the auto-oscillation (~100kHz, ~2Vpp) is still present even with a differential input.

This behavior is present in many channels, so it seems is not a failure of a single DAC channel.

This DAC can not be then used to drive the Pstab low frequency loop, it is not safe for the laser amplifier electronics, and may be it would be better to do not use even for the Galvo controls.
 

We took advantage of the recent intervention in the injection tower to install two baffles related to INJ, both between the output of the SIB1 telescope and the output port of the injection tower (flange towards PR tower).

The baffles are intended to handle stray-light coming from PR to SIB1 and light coming from SIB1 going to PR. Due to the possibly high intensity at this location of the ITF, they are made of mirror-finish (to reduce residual scattering) AISI304L plus 8um-thick DLC layer (to adsorb YAG laser light) plus AR coating (to decrease natural reflectivity).

Damage threshold for this solution is of the order of 0.5KW/cm^2, residual TIS <1000ppm, R<0.5%.

Picture 1: SIB1 -view towards PR tower. At the end of the bench there is BafINJ_a (mounted on a pillar) and BafINJ_b (inside the flange).
Picture 2: Close view of BafINJ_b
Picture 3: Close view of BafINJ_a (back side). The "cloudy" aspect is a feature of DLC coating when applied to very smooth surfaces. This appearence is not detrimental to scattering nor reflectivity, as resulted from several measurements carried on in Optics Lab at EGO.
Picture 4: SIB1 towards PR. BafINJ_a (foreground) and BAFINJ_b (background).