Reran all O4b locklosses on the locklost tool with the most up to date code, which includes the "IMC" tag.
I used the lockloss tool data since the emergency OFI vent (back online ~23rd August) until today and did some excel wizardry (attached) to make the two attached plots, showing the number of locklosses per day tagged with IMC and without tag IMC. I made this plots for just locklosses from Observing and all NLN locklosses (can include commissioning/maintenance times), including:
Attaching plots from zooming in on a few locklosses:
| Time | Tags | Zoom Plot | Wide Plot |
| 2024-10-31 12:12:26.993164 UTC (IMC) | OBSERVE IMC REFINED | annotated plot | plot |
| 2024-10-30 12:16:33.504883 UTC (IMC) | OBSERVE IMC REFINED | annotated plot | plot |
| 2024-10-30 16:09:21.548828 UTC (Normal) | OBSERVE REFINED OMC_DCPD | N/A | annotated plot |
The 10-31 zoom plot notes the framerates of the channels: ASC, REFL and NPRO_PWR are 2kHz and GS13 is 4kHz, the others are 16kHz.
- In the "NORMAL" lockloss (nearly 100% of locklosses since O4a and O4b before OFI vent):
- The arm signals (ASC_AS_A/C) increase over 10's ms as the REFL signal decreases.
- ~250ms later the PSL channels change and the IMC looses lock.
- In the "IMC" lockloss (25% of locklosses since OFI vent):
- The arm signals (ASC_AS_A/C) suddenly (<0.5ms) decrease as the REFL signal increases.
- At the same time the PSL signals change.
- <10 ms later the IMC looses lock.
Since September 18th we've had 21 locklosses from NLN tagged as FSS_OSCILLATION, of these 20 also had the IMC tag. Since September 12th, we've had 49 locklosses from NLN tagged IMC, so roughly 40% of these IMC locklosses have the FSS oscillation tag, since the NPRO was swapped we don't have any tagged with FSS_OSCILLATION.
(Reminder, the FSS_OSCILLATION tag is an old tag, intended for a different problem, but it tags times where the FSS fast mon goes above a treshold.)
Updated plot attached of NLN locklosses tagged with and without the IMC tag.
