Attached is a plot that shows the FSS_TPD_DC has a small drop in power right before lockloss at 23:19UTC (jan21).
The ISS_PDA_REL_OUT also drops, though I don't know what the power is relative to. The amplitude of the signal is about 0.0001 before the power drop, and the power drops by 0.001 (10x amplitude) before lock loss.
If this lockloss is the Noise Eatter going from OK to Not OK, while H1 is locked locked, it won't be for long, so a Verbal Alarm / DIAG_MAIN notice wouldn't help Operators.
Having a Verbal Alarm / DIAG_MAIN notice on the Noise Eatter would allert Operators to the Noise Eatter issue after lock loss, which I think would be helpful.