[Writing on behalf of Jason and RyanS, who are actively working on the NPRO swap]
TL;DR: We're beginning the process of swapping for a spare NPRO.
This morning, Jason and Ryan went into the PSL enclosure and mounted a spare NPRO (they will remind me of which serial number later) on the PSL table, and cabled it up. They moved the PD that serves H1:PSL-PWR_NPRO_OUT_DQ over near that spare NPRO and used a spare pickoff mirror to direct a small amount of power from the spare NPRO to the PD. This test was to see if the spare NPRO exhibits glitches in the same way that the original NPRO does. We had decided yesterday at the commissioning meeting that if the spare NPRO, after having sat for ~2 hours, does not exhibit glitching behavior, that we would go ahead with the swap.
In the attachments, blue ndscopes are of the original NPRO, that has been in service for all of O4 to-date. The pink ndscopes are the spare NPRO, in this test setup.
With the original NPRO, you can see a signal when the door to the laser room opens and closes, and also you can see some glitches that are the same order of magnitude as the door. This gives me some confidence that if we do see the door with the spare NPRO, then we should be able to see glitches if they are there. With the spare NPRO and pickoff, the signal on the PD is much lower. However, you can still see when the door opens and closes. We let the laser sit for about 100 minutes, and didn't see any obvious glitching.
- Attachment 1: Orig NPRO, in main laser path, you can see a DC change when the door opens/closes, and you can also see glitches.
- Attachment 2: Orig NPRO, in main laser path. Same y-axes limits as Attach1. Orig NPRO sometimes is weird and fuzzy. Fuzz is almost same size as glitches.
- Attachment 3: Spare NPRO, in test path, but with otherwise all same electronics, same readout PD (different pickoff mirror, so lower signal). You can see a small change when the door opens/closes.
- Attachment 4: Spare NPRO, in test path. Same y-axes limits as Attach3. There is some drift due to (I think) temp change in the PSL enclosure, but no obvious glitches.
Since this spare NPRO is not obviously exhibiting glitching behavior, they are beginning the process of putting in that spare NPRO.
We anticipate that this work will take a few days. We're doing it now since we can't lock during the daytime (due to excess ground motion this week), and LLO is offline for another few days pumping.
Last night, Jason and Ryan put the new NPRO into the main path, as the first step of the swap (alog 80829).
It does see a few occasional glitch-like features (see attachment), but they are (a) much less frequent than the old laser, and (b) no more than 1/2 size. Another factor to consider is that overnight, the PSL environmental controls were left on, so this is a noisier environment than we normally have. Also, no external servos (eg FSS) are engaged.
I've tried to make the y-axis about the same as the orig laser (blue) plots from yesterday, so that it's easier to see that these rare glitches are indeed much smaller than the ones on the old laser.
EDIT: Sheila and I looked at this plot, along with some others that Oli had created a few weeks ago, and it looks like these small 'steps' in the power that are shown here may be of the same type that we had with the previous NPRO. I'm still hopeful that they are smaller and less frequent though. But, these 'steps' may be part of the glitches that we're looking at.
Attached is a picture of the test setup. The spare we are using is SN1661; there is an alog coming at some point regarding the testing of the spares and our rationale for choosing this laser. This closes WP 12153.
