Date: Tue, 6 Apr 1999 00:35:44 -0600
From: long@apo.nmsu.edu (Dan Long)
Subject: 2.5m obslog 990405 ( MJD 51274 )
                           MONDAY APRIL 5, 1999
                     APACHE POINT OBSERVATORY SDSS 2.5m
                              OBSERVING  LOG
   As you have probably already seen from Russell's comments, tonight did
not go well.  While we did manage to get the TCC to move the telescope at
some level, all axes were having trouble with oscillations.  This was known
in the afternoon, and Charlie Briegel and Claudio Rivetta were on line from
Fermi Lab and Russell Owen and Bill Boroski were on line from UW trying to
assess the problem.  It seems that for Azimuth and Altitude slew rate
velocities would induce oscillations which in many cases would not damp out
even at the end of the slew.  For the rotator a move of almost any velocity
( including hold ) would cause rather violent oscillations.  The
spectrograph should not be mounted until this problem is fixed to a high
degree of certainty.
   Other problems which occured were several MCP crashes, which when they
occured during a slew, allowed the telescope to run away, only being able
to stop it by pushing in a stop button.  This obviously needs to be
addressed before the telescope can be slewed without someone at the
telescope.
   For some reason, the fiducials could not be used as an absolute
reference.  Apparently the MCP is waiting for a particular fiducial and not
seeing the one it wants in order to have the "F" feature of the menu reset
axis position.
   Another curious thing which was not true during the last dark run was
that the altitude scale factor appears significantly off.  I set the
altitude position using the equator mark from the last dark run.  At this
time the altitude clinometer agreed closely with the altitude readout.
When I went to put the telescope to the horizon when we closed, the
altitude clinometer read the normal value of about .75 deg at stow, but the
altitude readout said the telescope was still at about 4 degrees.   During
the last dark run the telescope and clinometer agreed quite well over the
full range of altitude motion.
   Russell was also mentioning seeing large delays in response to the start
of a move, usually only in one axis.  I would tell him the telescope had
started moving from a slew command, and he would see the position remain
unchanged for 10 seconds or more.
   Many times when I had to intervene with the menu to stop oscillations,
Russell could no longer slew that axis, even though I still could through
the Menu.  We think it may be the case that if the axis is not left in hold
by the Menu, that the TCC will be unable to move it even if an AXIS INIT is
issued.  Apparently issuing a stop, "S" on the Menu, puts the motor
controller in a state not resettable by the TCC.
   A couple other notes.  First, the telescope seems fairly well balanced
with the engineering camera on and the counterweights at 200.  Second, I
noticed while balancing the telescope that I could get the telescope to
contact the turnbuckle rods at the top of the windscreen before the bumper
touched the PSS.  In normal servo operation they are no where near
touching, but something has changed such that the bumper will not prevent
contact with the windscreen when altitude is on the high side of center in
the windscreen.
   I closed up the telescope at about 22:00 once we realized it was not
going to be possible to take pointing data.  As a note to those working
during the day tomorrow, I don't think there will be any need to button up
the telescope in the afternoon on account of thermal effects since we
didn't even get a crude model tonight.  I'd certainly rather see a smoothly
tracking telescope with thermally distorted images tomorrow night, than
nice images on a poorly tracking one.
                                    Dan