Summary:  We do seem to have hysterisis in at least some of our suspentions after the Idaho EQ, similar to Montana, although we could get better data by moving the suspensions around more. 

Here are some plots to follow up on the Idaho EQ, in addition to what has been posted in 55867, the initial report in 55838 and suspension TF checks posted in 55875 and 55869 and checks that suspensions are free in 55857.

These plots are made starting with the same script I used in 2017 to check for hysteresis on the quads, 37799 and comments, and a follow up in 37972.

The first attachment is a time series just meant to help make it easier to understand the second plot.  I've taken several days of minute trends around the time of the EQ, removed the means of the top mass pitch osem read backs, and the drive applied to the top match pitch.  The torque I've scaled so that it is in units of urad displacement that we would expect for this drive.  The time line is a little more complicated than the Montana quake, because we transitioned the observatory to phase 3 (for COVID) and set the HEPIs to offline.  In the first attachment the full data is in black but data which is used in the second plot is color coded to illustrate the different time periods, HPI isolate before the EQ, HEPI off before and after the EQ, and HPI isolated after the EQ.

The second plot isn't as nice of an illustration of the hysteresis as they were in 2017, in part because we haven't exercised the alignment sliders as much in the aftermath of this EQ (we haven't tried to relock the IFO).  Based on these plots if we had to prioritize quads for more diagnostics or TMDS, ITMY looks the worst, followed by ITMX and ETMX. 

It is interesting that ETMY has the least evidence for hysteresis, since that is the suspension that was probably rubbing while HPI was offline 55848