Background: As a follow-up to Josh et al.'s finding that an EX air compressor is likely seismically injecting noise into DARM (alog 21436), a similar study was done for EY. 

Summary:

• it's likely an air compressor at EY is injecting seismic ground motion noise that couples to the optics table motion
• there is currently no evidence this couples to DARM

Details: 

The counterpart channel used in 21436 (HVE-EY:INSTAIR_PT499) was used in the same fashion to gauge whether the local air compressor was turning on. 

The induced motion is much different than observed at EX; it is not transient (on the scale of a second) but about three minutes long, as seen in H1_EY_AirCompressor_Glitch_Example.

This motion was first noticed at 29 and 58 Hz (as shown in this example spectrum, where blue is during an event and red is quiet time 5 minutes prior) there also appear to be shifted/additional harmonics in every other glitch instance, as seen in the plots in LHO_EY_AirCompressor.pdf. 

The attached pdf also demonstrates the correlation between the air compressor turning on (as defined in alog 21436) and 29, 58 Hz motion measured in both the ground and optics table. There is also a correlation with the mainsmon voltage monitor, as was observed for the air-compressor-induced motion seen in EX.

Both the daily hveto results and spectrograms of h(t) show no evidence this is producing transients in h(t), unlike the EX mechanism.