Date: Sun, 16 May 1999 06:09:13 -0600 (MDT)
From: hayes@apo.nmsu.edu
Subject: 2.5 m obslog 990515 (MDJ 51314)
			Saturday May 15, 1999
		Apache Point Observatory SDSS 2.5m
			     Observing Log
Observing Team:
		Jeffrey Hayes (APO)
		Ellyne Kinney (APO)
                Rob't Lupton (PU)
                Scott B (UC)
                Connie Rockosi (UC)
		David S. (UC)
Weather:  Clear with low dust.  Very windy: ~20 mph steady and gusts
up to 32 mph.
We performed an instrument swap after the sun was less than 30 degrees
up, taking the mosaic camera off and placing the spectrographic cartridge
in.  Plate 0143 was used.  The counterweights were set to 340 each.  NOTE:
the LCD display for CW 2 is not responding -- it reads from -001 to -004
for weights.  We set that counterweight by eye.
We could set the ALT and AZ fiducial okay, but the rotator, when moved by
the MCP would move very quickly 2 or 3 degrees and then stop.  I had set
the velocity to 50000, and this motion was at least 3 times that speed.
We replaced the tape encoding for the rotator with the older version, and
Robert Lupton rolled back the MCP code to the previous version from last
night.  With this version, we were on the sky.
By and large, a frustrating night as we could get stars down the fibers,
but when we switched over to guide, the guider would actually drive the
stars off the fibers.  It appears to be the same centroiding failure as
we have experienced on previous nights.
We started out with the field for plate 0143 and saw no stars, we tried a
few offsets with no luck.  We decided to try a nearby fk5 star.  Eric Neilson
from the PT team provided the coordinates (228.79733, 4.9393611).  We applied
the offsets from the boresight to fiber 10 (-0.080657,0.3407536) to these
coordinates.  We got no light for the star.  We tried some offsets with no luck.
At around 10:45 pm we lost the rotator axis as we pasted the meridian.  We
realized this happen because we use the flag /rotwrap instead of
/rotwrap=positive.  (Can positive be made the default?)
We then decided to try getting light from Arcturus (213.90958,19.185) down fiber
10 as a sanity check.  We saw some light that looked like a diffraction spike.
After a lot of fiddling, we were finally able to get the star centered in
fiber 10.  The offsets were -0.042,0.019.  We then adjusted the focus through
the range 3700-4900 in 200 step increments and decided that the original focus
setting was the best (3550).
We went back to the FK5 star and after a few minor offset were able to center
the star.  The cumulative offset at this point was -0.046,0.015. We then we to
to the field center for plate 0143 (227.206991,0) and after just three small
offsets got stars down the fiber.  We centered up the field using the star in
fiber 10.  We also tried a small rotation (.01 in each direction) but this
did not help the centering.  The final offsets for the fields were
-0.041,0.016. We then took data for an updated instrument block by making
offsets in a diamond pattern, and used Aronne's code to create a new
instrument block.
At 1:25AM we lost the rotator axis again.  We were not crossing the meridian
when this happened.  We were able to do and axis init and track back on the
field again.
We tried guiding of the field, but were unsuccessful guiding.  The stars would
quickly  move to the edges of the fibers and then wander away completely.
We had error messages complaining that the predFWHM was too large for the
scan size, so we tried using smaller FWHMs  (0.002 and 0.001) We also tried
increasing the scan region sizes.  Increasing the scan region did not seem to
help, but decreasing the FWHM to 0.001 did.  We were also getting centroiding
errors.  The most common message was center was not the highest pixel.
We went back to the FK5 star and tried guiding just using fiber 10.  We still
had no luck.  The guiding would fail before the star  wandered completely away.
The star would drift the the edge of the fiber and were getting the same
FWHM and centroiding errors.  We spent quite a long time trying the get guiding
to work.  We tried several variation in the instrument block and tried
retaking the offsets for the instrument block data.  We tried setting the
FWHM even smaller (0.0001 with and error that is was too narrow and 0.0003)
We then set the scan region for fiber 10 to the whole scan region and the
FWHM errors went away, but the guiding was still failing.
We went back to the field and tried retaking instrument block data.  We also
tried moving the rotation.  We were able to get all ten guide stars centered
in the fibers with a rotation of -0.139.  We retook the instrument block data
with the improved rotation, but guiding did not improve.
By 4am we were not focusing well on the stars.  We tried refocusing but none
of the new focus positions improved focus.  We decided the large airmass
may be effecting the images, so we tried to focus on a new FK5 star (coords
provided by Eric) 286.3525,13.86333.  It needed a large offset to center it
in fiber 10, -0.54, 0.0155.  We note that this offset is similar to the
one needed for Arcturus.
At 5AM we decided to close.
Jeffrey and Ellyne