Lockloss details:

• t -40 seconds, mode3 is damping
• t -35 seconds, mode2 is ringing up
• t -25 seconds, mode9 rings up, no filter change
• t -15 seconds, I turn the output of mode2 off, and it remains off, and the output looks like zero, but there was still a signal on the output channel
• t -14 seconds, I changed the filter on IX mode 3
• t -12 seconds, IX mode3 output increases
• t  -6 seconds, OMC_DCPD signal gets noisy
• t  -2 seconds, I set IX mode2 gain to zero, it ramps for 10 seconds, output is still off, still a small signal on the output
• t at 0, lockloss

I don't know if this is by design, but I found that the violin damping continues for almost 2 seconds:

• t at 0, lockloss, damping on all modes continues
• t +1.2 seconds, damping gains on modes  3, 5, and 9 start to change
• t +1.3 seconds, damping gains on modes  3, 5, and 9 reach "0"
• t +1.5 seconds, the output on modes  3, 5, and 9 reach "0"

The mode2 switch must get switched in the DOWN script, because the mode2 output was re-enabled, and what doesn't make sense is that even with the mode2 gain at "0," mode2 output damps for about 5.5 seconds, and the output signal isn't small, it's +/-20000 counts!

• t +1.5 seconds, the output on modes  3, 5, and 9 reach "0"
• t +2.0 seconds, mode 2 switch changes, and the output of the damping filter is enabled
• t +7.5 seconds, the output on mode 2 reaches "0"
• t +20 seconds,  all switches change, the output on mode 2 reaches "0"

Many plots that go with the timeline.

The cause was human error, the response of all the modes continuing to damp may be intentional, the response of mode2 may be understood, but was unexpected when I looked at the timeline of the signals.