Jul 1 2021: B path alignment continues, SFI2 alignment helper apertures removed, first round of mode scanning (Georgia, Keita)
We used a fresh in-air fiber and damaged in-vac fiber to continue the alignment work.
We refined the alignment through SFI2 while the alignment apertures were still on. B:L1 was placed at the design location. Used B:M1 to center the beam on the POL1 side aperture (used an IR sensitive Nikon to monitor), then B:M2 to maximize the power in the IFO path outside of OPO platform. We iterated several times.
After this we removed the alignment helper apertures. They were bagged and put in one of the bins for VOPO parts.
We placed B:L2 at the design location, set up a mode scan rail in the IFO path and did the first round of mode scan.
I accidentally moved G path collimator, which was recovered later.
Numbers etc. will be posted later.
Below is the data from the B path profile and distance measurements, and the G path cavity scan:
Things look promising. Using 4 sigma width and pretending as if it's Gaussian, vertical overlap is already 99.8% and horizontal 99.2%. It seems like the beam is a tiny, tiny bit larger than the ideal mode and the waist could go ~10cm closer to SFI2.
Note that this was done with damaged in-vac fiber and feeble output power. We need to use a good in-vac fiber and repeat the measurement.
Fri. Jul 2, 2021, Replacement in-vac IR fiber looks good (Camilla, Keita)
Summary: We received two in-vac IR fibers that were previously (mis)used for green in HAM6 at LHO and LLO. I chose to use the LHO one. Microscope inspection looked good, the mode out of the fiber was good, and power throughput was good. We'll still have to measure the mode matching between IFO and OPO mode.
Microscope inspection:
Replacement in-vac fiber looked good on both sides except tiny things that might be particulates. It seems that oblique lighting is better for showing the core and the cladding structure while coaxial lighting is good at highlighting some particulates (dust?). There were particulates that we couldn't remove by using the fiber cleaner (it didn't do much anyway), but we didn't bother to clean it using a wipe as it didn't look terrible anyway. The first 2 pictures show the replacement fiber at the launcher end with oblique and coaxial lighting after multiple cleaning. Picture 3 and 4 are the same fiber at the end that mates with in-air fiber.
We also took pictures of the damaged fiber we removed. The damage was clearly on the mating end (picture 5), the launcher end was pretty clean (picture 6).
We also took pictures of the in-air Thorlabs patch cable but won't attach them here as they look good.
Power throughput:
100mW into in-air fiber collimator -> 3.9mW out of in-air IR fiber -> 3.7mW out of in-vac IR fiber launcher. In-air part is bad (more on this later) but in-air fiber to in-vac fiber is very good.
Realignment of CLF launcher. Somewhat large 2nd order mode power.:
After the fiber swap, the alignment changed by a large amount. However, 01 mode power was easily minimized by only adjusting the launcher.
The 2nd order mode was big-ish relative to 00, though (picture 7).
[0th, 1st, 2nd, 4th]-background = [2.08, 0.054, 0.564, 0.151], or
[1st, 2nd, 4th]/0th = [0.025, 0.27, 0.073].
2nd order mode power looks bigger than before to me. We'll leave it for now, if we have time we might adjust the launcher lens or F:L1 lens.
Poor coupling from the laser to the in-air fiber. Beam shape going into in-air fiber coupler is bad.:
I realigned the beam into EOM as I saw some clipping, then realigned things to maximize the coupling into in-air fiber but I was only getting 1.3mW out of 10mW.
On further inspection, the beam looked badly fringed or split. I removed EOM from the path and nothing changed. Pictures 8, 9 and 10 are the IR-camera image and two beamscan images (one where scan axes align with horizontal and vertical, the other where they're 45 degrees tilted).
We didn't have time to investigate the root cause of this, maybe the beam is clipping upstream, or maybe it's something more serious.
Anyway, because of this, there are more than one local maxima when you adjust the alignment. I skipped the 1.3mW local maxima and eventually got 3.9mW out of Thorlabs in-air fiber. That's good enough for alignment/mode matching of opo but we'll spend some time next week to investigate this.