Reports until 17:29, Friday 25 October 2024
H1 PSL
jason.oberling@LIGO.ORG - posted 17:29, Friday 25 October 2024 (80885)
PSL NPRO Swap Day 4 (WP12155)

J. Oberling, R. Short

We continued on from yesterday.  We started by rough aligning our beam through Amp1 in preparation for beam propagation measurements to assess how well we installed our mode matching solution.  We had done a rough alignment previously so expected this to go quick, but it was more off than I expected.  We also checked our beam polarization going into FI01 (after WP01) and going into Amp1 (after WP02) using a temporary PBSC as an analyzer.  The beam should be vertically polarized w.r.t. the table top in both locations.  Going into FI01 the beam was definitely as vertically polarized as WP01.  But after WP02 roughly 50% of the beam was in horizontal polarization (~50mW beam, almost 24mW in horizontal polarization).  This is not good and not how we left this after our PSL recover work in March of 2023; wrong polarization through the amplifiers has a large effect on output beam quality (as learned with the O3 70W amplifier).  This immediately sent up red flags for us, referencing our slowly increasing PMC Refl (increasing since March '24, we swapping the PMC in July because of this).  Immediately we suspected something in FI01 was bad/broke/wrong, and could have been for a while (if something in FI01 had been slowly failing and changing the output beam polarization while failing, that would explain the slowly increasing PMC Refl, as changing the beam polarization from vertical through the amps negatively affects their output beam quality).  We have been using the old Faraday isolator from the 35W Front End laser, but we have 2 spare Newport FIs, identical to the one being used between the amplifiers.  Unfortunately we only have one other FI pedestal base, and it puts the FI at a 100mm beam height, not 4".  We looked through all of the spare pedestal bases from the O4 laser upgrade hoping to find one like is used for FI02 (puts the beam height at 4"), but no other spare FI bases were found.  We also measured the throughput of this Faraday at ~85%, which is lower than the ~93% we had at install.  Something here isn't right.

While checking the FI mount situation out, the alignment situation was bugging me so we took another look.  Checking back through our alignment again we noticed it was off on our alignment iris in front of L02.  So L21 and the EOM were removed and alignment checked again.  The alignment was fine on our first iris but a little off on our second.  M02 was tweaked to fix this, then the EOM installed and its alignment tweaked.  At a point where the EOM alignment looked good on our iris near L02 we decided to check power in and power out, and found we were losing ~6mW through the EOM (this is at a lower alignment power of ~74mW).  We put the power meter at the output of the EOM and adjusted the EOM alignment until the power out of it was maxed; best we could do was ~70mW out with 74mW in.  At this point we looked at the iris near L02 and the beam was a tiny bit high but for the most part looked OK, so we moved on.  L21 was reinstalled and its alignment tweaked.  Now to deal with the Faraday, but first lunch.

After lunch we swapped the Faraday with one of the Newport spares.  Since the pedestal base was a little too short we used a couple of shims under the Faraday to get it at the correct height.  Without making any adjustments to the Faraday we checked the beam on our temporary PBSC, still installed after WP02, and found almost no beam in transmission.  With no changes to the waveplates or tweaking of the Faraday the beam was now much better vertically polarized, lending credence to the theory that something had finally failed in the older Faraday we'd been using (we'll see if this was the source of our PMC Refl issue once we have the PMC up and running again).  We tweaked the angle of the input PBSC on the Faraday to minmize the rejected beam; lowest we could get was 0.40mW.  We then flipped it around and tweaked the output PBSC to minimize the transmitted beam, lowest we could get here was ~17µW; we saw ~11µW once, but we were unable to maintain this while clamping the output PBSC.  Since we found this at ~55 µW we called it a win and moved on.  The Faraday was flipped once again and we began setting up for beam propagation measurements.

These measurements are done with a leakage beam through M03, so we installed a rail (we have blocks in place to make this easy) and set up the WinCam beam profiler.  We then increased the power back to ~1.8W by rotating WP14.  After a little bit figuring out the basics of Varun's script, which we haven't ran since 2023, we were able to take one measurement before we called it for the day:

Our target is 165µm @ 1794.0mm, so we have a little work to do here.  We called it here for the day and will continue with mode matching tweaks and Amp1 recovery tomorrow.