I reduced the amplitude of the OMC ASC dither lines today, to [33, 50, 100, 100] counts for the [P1, P2, Y1, Y2] error signals, which are injected at [575.1, 600.1, 625.1, 65.1] Hz.  Along with the ever-increasing ASC goodness, this reduced the noise around the lines.  The attached plot compares the sidebands

We didn't observe any ill effects from reducing the dither amplitude (the SIN and COS gains were increased to maintain the overall loop gain).  At some point we should measure the loops again to check that the bandwidths are still in the tenths-of-Hz range.

In other OMC news, someone merged the H1 OMC_CONTROL medm screen with the Livingston version last week; this change has been backed out.  It's not a bad idea per se, but there are some scripts and buttons used here that aren't used by L1, and vice versa, so things were a little confused.  Also, the DCPD gains had been reduced to 1, from 1000, around 2:30 in the afternoon today.  This change was also reverted (maybe from a bad burt restore?) to keep the DCPD SUM channel calibrated in milliamps (rather than amps).

For some reason the OMC-DCPD_SUM_OUT_DQ channel is no longer accessible via ezca read.  This caused problems in the OMC_LOCK Guardian.