I think we need continuous operation of the NeoLase to stabilize it. Otherwise, the temperature in the PSL room cannot be stabilized anymore. So interlock is better to be set as soon as possible for security.

(Uchiyama, Tanaka, Miyakawa)

Abstract:

We concluded that the beam jitter comes from the master of neoLASE.
The interlock system for neoLASE was successfully tested.
GRX and GRY locked stably without mode hops.

Detail:

We introduced a direct long path to another camera from the output of neoLASE. This path has less number of mirrors (4 mirrors) than the original path (11 mirrors). The direct path uses the transmitted light of the first high reflectance steering mirror after the output of neoLASE, so there are no overlapped mirrors between the original path and the new direct path.

We compared both camera screens side by side, and we watched the similar motion at the same time. So we concluded that the jitter comes from the laser source of the neoLASE. We used the master laser only, and no power amplifier was on, so it comes from the master laser. Meanwhile, we tested ON/OFF of the Noise Eater, different crystal temperatures, and different currents, but no significant change was seen in any case.

Additionally, we connected new cables of the interlock signal to the amplifier of the neoLASE laser. Then we pressed the emergency button as a test to enable the interlock while the neoLASE was ON for both master and amplifier, and then both amplifier and master became off. The interlock system seems to work for neoLASE and also the existing FB laser (it won't work for a while).

We had no way to improve the beam jitter if it is the master laser of neoLASE, except for keeping lock long term to be more stable. Then we tried the IFO lock acquisition, starting from the initial alignment for arms. After some work for alignment, GRX and GRY were locked well, and no mode hop was seen. Then we tried PRM alignment, but POP90 (sideband power) started reducing when ADS for PRMI was engaged. Time was up today. We will continue tomorrow.

(Uchiyama, Tanaka, Miyakawa)

We still suspected the slight mode hop, so we changed the laser crystal temperature from 37.0 to 36.0 degrees, which was the same temperature as December reached to DC readout with the neoLASE laser.

Power estimation at the PMC REFL and TRANS was not consistent with the neoLASE laser, so we calibrated the input power and PMC REFL power at DCPD and RFPD, with the actual measurement at the beam dump of PMC REFL. The measured power was 21.16W with 9A current. We changed the conversion gain at K1PSL_PMC_REFL_RFDC from 2.360 to 2.030 and K1LAS_POW_FB from 6 to 7.5. 

In the morning, we compared the free run RIN (relative intensity noise) at ISS PD between the FB laser and the neoLASE laser, and the neoLASE had 10 times higher noise below 10Hz to 1Hz (see Fig. 1).

Then we compared the RIN between upstream and downstream before PMC. The up location was before the AOM, and the down location was the reflected light of PMC (see fig.2). Fortunately, the upper location showed less RIN below 10Hz, which means that the laser itself is OK for intensity noise and noise is added somewhere. The high frequency of blue line is limited by the detector noise because very low power was injected.

We suspect the beam is clipped at the AOM, and actually, we found a clipped shape after the AOM. We moved the closest mirror position to eliminate the clip, and the intensity fluctuation looked a bit better. At that time, we could engage the ISS servo with the integrator for a while, but it did not last long, so we continued the investigation of RIN at many places.

For safety and convenience, we used only the master laser of neoLASE for the investigation. RIN  became worse and worse from upstream to the downstream (see Fig. 3). However, the mode was not matched at all if only the master was ON, and all the beam sizes during measurement were very large. This behavior of growing noise might be due to the clipping at each optics.

At the very last stage, we found a clip at the RFPD in 20W operation. We tried to find the best position on the PD, and the DC power increased by ~5%.

We should find good and safe locations to measure RIN with 9A next.
 

How about put a mechanical metal iris to reject higher modes on the edge area of the beam? Burning and scattered light are of concern.

I am wondering the effect of the scattered light on each optics, especially for the beam dumper to waste 30W to obtain 20W from 50W.

Can you show how to treat this 30W?? Is the manner the same as the last case around 2024, Oct ~ Dec?

Maybe this one?

In addition, I found a strong IR beam hit on the first steering mirror for the neoLase, as shown in photo-1, and 2. The difference of phot1-2 is just the focus for the visible light and IR. 

This camera is for visible light and has an IR filter. So IR beam spot  in this camera should be strong or so concentrated as the IR beam spot on PR2.

IR light on the surface of the AOM and the input mirror of PMC.

This is the PBS to waste 30W to the beam dumper. You can see the brightest spot(1) on the border of the cube type PBS. Secondly brighter spot(2) is on the AR surface of the PBS in the direction of the beam dumper. Also, you can see the spot(3) on the AR surface in the direction of neoLase.

However, there seems to be an undesirable scattered spot between (1) and (2) inside the PBS.

In addition, I am also wondering the reflected IR on this PBS from (3) is well misaligned to the neoLase side or not.

I expect some amount of light that goes back to the neoLase will be rejcted by an isolater in side the box. My guess is right??

In this photo in October 2024, there was as light shields just around the output of the neoLase.

In addition, the beam dumper was the black-hole type. However, at present, maybe a power meter type where the ligh hits on the "flat" surface??

IMC trans power fluctuation using neoLase at present seems to be double of the FBL (Fig.1). However, in December 2024 the IMC trans fluctuation using neoLase was ~1/2 of the FBL(Fig.2). So totally the present neolase has 4 times larger fluctuation at low frequency.

Fig.3 shows the relation of PMC and IMC trans. They are well correlated with each other. So neoLase itself should be stabilized.

Does the "power-meter-type" beam dump have water flow system? If so the fluctuaion might affect the optics around it.

How about comparing with the temperature drift in the PSL room?

> Does the "power-meter-type" beam dump have water flow system? If so the fluctuaion might affect the optics around it.

No at present, according to the photo.

I checked the recent PMC/IMC trans and temps of PT05/06.

Norm-PMC trans trend seems to be smooth, but sometimes there are lock losses? While, IMC Norm- trans seems to have jumps. Some of these jumps were related to PMC lock violations. Some of them were not. These jumps are just because of the normalization errors, because non-normalized output was smooth.

Anyway, the alignment to PMC is now degrading maybe because of the temp drift from the time when the adjustment was done.

Fig.1 shows the IMC/PMC trans and temperatures of PT05/06. Although the temps seem to become well-stable, the fluctuations of PMC/IMC are the same.

If this situation cannot be improved, we should consider using the fiber laser again with a new amp.

IMC trans shows sudden enhancement, while steady reduction of PMC trans. Why?

After klog#34935, no IR can be seen around the AOM hole.