Daniel, Sheila, Alexa,

We pinned down the wandering peak in the IMC error signal.  It was a beat note between the fixed frequency 78.2MHz and the IMC VCO.  Moving the IMC VCO frequency (using the tune offset) moves it, it dissapears when the 79.2MHz is off. It becomes very small when the cable from the RF pacth panel in R1 to the doubler for the fiber distribution is disconnected from the patch panel, and can be small when the able is connected at the patch panel but not at the input to the doubler, perhaps depending on where the cable is.  Disconnecting the output of the doubler has no impact.  It seems like a purely electrical problem.

We still do not know why the UGF has changed, but the of the IMC error signal measured at Imon is 1.5V at the moment.