I used the earthquake down time from the weekend to look at the health of the BSC2-ISI sensors.
Sensors look OK (including the newly replaced H1 L4C - see alog 90840) except for CPS ST2 H1 (see red curve on the last plot) with large noise below 1Hz. This will need an investigation (I would start by swapping cables at the field box to see if it's an in-chamber issue).
Dtt Template lives under /ligo/svncommon/SeiSVN/seismic/BSC-ISI/H1/BS/Data/Spectra/Undamped/bs_all_sensors.xml
I had already done this test several years ago trying to diagnose the issue with BSC2, which, contrary to my comment in the alog Arnaud linked to, still has alot of flat looking excess noise in the 1-10hz. Arnaud's spectra here doesn't make me think this noise is due to a broken cps, the high frequency noise floor is unchanged, there's no evidence of glitching, unless it is some new failure mode.
Knowing that I would have few chances to try swapping in the near future, I did this over lunch when I went out to also unlock HAM1&2 hepi. I put BSC2 in damped, collected St2 CPS spectra, swapped the St2 H1 cps with a spare sensor, and got a comparison spectra. No change, see first attached spectra, refs are with the old H1 cps, live spectra are with the new sensor. I have left the spare sensor in for now, I will try to swap it back to the old sensor so we can preserve our ISI alignment (the new sensor reads a slightly different location that the old), but physically there should be no difference, the BSCs just use the free hanging position for the cps reference, unlike the HAMs.
Performance is no better with the new cps, as well, see second plot, refs are with the new cps, live is before the swap.
As dicussed with Jim on the phone it would be interesting to see the result of swapping H1 and V1 at the lemo input of the field cables - V1 noise is low enough between 10mHz-100mHz when the ISI is damped that the H1 noise would be obvious if it followed the cable.
