I have been trying to find reasons for our locklosses and wanted to share the known causes for our locklosses so far in February (39 locklosses from 1st Feb to today, 27th Feb).

The pie chart here shows that we can only account for 25% of our locklosses (caused by EQ, Tuesday maintenance and Commissioning activities). Some of the unknown locklosses we can blame the weather but for lots of them we had good observing conditions, see table here with environment stats (wind, seismic, microseism) highlighted in green to red scales.

Another interesting number is the time it takes for the IFO to fall out of lock. Looking at channel H1:LSC-POPAIR_B_RF18_I_ERR_16k_DQ we mainly have locklosses where light falls off the PD slowly~ 45ms or fast ~0.2ms. The pie chart here  shows 62% are slow (seem to be caused ground movement eta, all known locklosses are "slow") but 38% are fast, which are much harder to diagnose.

 

All the fast lock looses between Wed 12th and Sat 15th had the FSS oscillating before the lockloss as TJ found in alog 55081 but none before the 12th and we've had no fast locklosses since 15th. I wonder what changes. It does not look like the weather does. tagging PSL