Date: Sun, 22 Nov 1998 07:46:44 -0700
From: "John W. Briggs"
Subject: 2.5m obslog 981121, MJD 51139
Saturday, Nov. 21, 1998 (MJD 51139)
Apache Point Observatory SDSS 2.5-m
Observing Log
	The encloser was open at sunset and the telescope in position
without incident, except that the west door overshot its limit switch, and
until I manipulated the limit switch, the interlocks saw the door as
closed.  At the end of the night I was able to close the door, but as it
came down, I noticed that the southwest cable was loose.  By the time the
door was near closed, however, the weight was evenly distributed.  I expect
further adjustment of the limit switch "feeler" will be required, as well
as possible adjustment of the cable.
	Lupton and Annis ran the camera and pursued various software
engineering tasks, including tuning the focus loop; Briggs ran the
telescope.  Image quality was somewhat asymmetric at the start of the night
and onward, but tolerable until the incident described below.  Early in the
session we noticed that the telescope moves significantly when the altitude
brake is turned off (such that stars jump to the left in the scrolling
display screens).
Annis reports on the rest of the session:
>We suffered from hardware problems most of the night.
>
>First, the secondary galil/setmir combination became confused.
>We think that the terminal server to the galil may be
>failing, as cut and paste of status commands to the galil
>revealed dropped characters. This may have been enough to
>get the galil into the strange state we found it in, where it
>a) emitted long strings of ^G and b) had an outlandish target
>position. The former seemed to be enough to hang the host
>sessions between the tcc25m and the galil. The only recovery
>method for that was to reboot the tcc25m. The fact that the
>galil had an outlandish target position, along with Russell's
>code, meant that even after we rebooted the galil,
>we had to home the secondary thus ending our imaging run.
>
>Next, we lost sdssad3 two times to the mysterious, silent
>crash we have been seeing, that a dsclat reveals is due to
>a bus error.
>
>Finally, we found that over a space of 10 frames the images
>had turned from donuts into dual-lobed donuts. We down-danced
>the primary, but that had no effect. We homed the secondary,
>and that seemed to cure the images.
>
>On the other hand, the tape drives worked perfectly, despite
>the 1" seeing conditions. We had one decent run, 2 hours long, named run
>297, >in 1.3" conditions.
	The 180-l dewar, which was not empty, was nevertheless exchanged
for a full one at about 7:00 AM Sunday.
			--J. W. Briggs.
********************************************************************
John W. Briggs
University of Chicago Engineering Center,
Yerkes Observatory
[Deployed at] Sloan Digital Sky Survey          Voice:  505-437-6822
              Apache Point Observatory          Fax:    505-434-5555
              Sunspot, NM 88349