Problem(s) Encountered:

At around 11:30pm, we found that the spectrum of the last science target was showing almost no signal, just noise. We first checked that the dark hatch was open. Next, we reset Owl and the MIS software. The problem persisted. We checked a quartz frame and there was almost no light getting in. We then opened and closed the dark hatch. After this, we took a quartz frame and the signal was back to the appropriate level. We then tried a science target but the problem was back. (The guide probe had no problems moving to a guide star outside of the instrument fov.) We tried opening and closing the dark hatch again and then going straight to a star (to see if moving the finder in and out was part of the problem) but there was still severe light loss.

Solutions:

Found that there was a failure with the shutter controller electronics box.  When commanded, the shutter would open but almost immediately close.  Voltage should read ~5V but was showing around 0-1.5V.  Failure was intermittent, occurring roughly 3 out of every 5 attempts.  Opened the controller box and found no obvious issues.  Cleaned contacts and tested to find box worked every time over roughly 100 tests, both manually and Owl-controlled.  Worked flawlessly last night too.

We have a “spare” controller box, from the MkIII, in case this one goes belly up.  I’ll look around to see if another unit can still be purchased.

Interestingly, as mentioned, with load (the shutter), the voltage would go to 5V (on) and 0V (off) immediately.  If load is removed, the voltage, upon switching on, spikes to ~60V, then slowly lowers to ~9V over a period > 30 seconds.  This is the case with both shutter controllers mentioned above.