Date: Thu, 15 Apr 1999 15:22:40 -0600
From: long@apo.nmsu.edu (Dan Long)
Subject: 2.5m obslog 990414 ( MJD 51283 )
Apologies for not pulling the trigger on this message before I left last night.
                          WEDNESDAY APRIL 14, 1999
                     APACHE POINT OBSERVATORY SDSS 2.5m
                              OBSERVING  LOG
   For the first time since Sunday night, the enclosure was opened for a
night's observing.  Unfortunately telescope and weather problems conspired
against getting a pointing model.  We were open from shortly before sunset
until about 23:00 when the dew point depression fell below the limit.  We
were not able to re-open.  Jeff Hayes and Ellyne Kinney were around and
helping for the first half of the night.
   I noticed the 15 degree stop status is "neither" when it is actually in
the retracted position.  The proximity sensor appears to either be out of
adjustment or broken.  The extended position is properly reported.
   While we were open I first adjusted the balance.  The telescope was a
bit front heavy as it appraoched zenith.  As the counterweights are nearly
at the bottom of their travel with the engineering camera on, I figured
that adding the clamshell weights to the top of the truss might be more
than the counterweights could balance, especially when no instrument is
mounted.  I decided to put a couple clamshell weights on the back of the
telescope.  There were two tapped holes in the back of the PSS which were
conveniently the same size as the screws which hold the clamshells
together.  Right now they hang a little below the back of the PSS and might
possibly interfere with the doghouse door.  A better way of attaching them
so they could be entirely above the bottom of the PSS would be nice.  In
fact an adjustable counterweight system such as the one use to trim the
balance on the 3.5m would be very nice.
   The next step was to set the fiducials.  The altitude fiducial was no
problem.  While moving in azimuth via the Menu, the Menu program stopped
updating.  I had to reboot the MCP to get a prompt back.  The valid rotator
fiducial when crossed this evening, did not result in setting the position
of the rotator.  When I tried setting it again later in the night, it did
indeed set the position, but it was not correct.  When the rotator was
really at zero, the Menu's position showed 1:41:30.
   I noticed you can still contact the windscreen with the truss when
moving the telescope by hand within the stopped windscreen.  It hits the
top of the windscreen at one of the turnbuckled rods.
   Russell Owen signed on from UW and was running the TCC as we attempted
yet again to obtain a crude pointing model.  Unfortunately several
telescope problems prevented us getting very far before the weather shut us
down.
   First I noticed while looking through the finderscope that we were
offsetting in altitude whether Russell issued an X or Y instrument plane
offset.
   The altitude ran away while trying to offset in azimuth.
   The rotator ran away while again trying to offset in azimuth.
   I was seeing a large ~.05 degree lurch in altitude each time Russell
used a computed offset ( supposedly kinder to the servos since they were
dropping out of closed loop on some uncomputed offsets ), in either
altitude or azimuth.
  On about half the slews we made ( which were all to the same star ),
there were .02 to .04 degree errors in one axis or the other.  The error
appeared to remain constant as the axis tracked.
   After all this I decided to call Charlie and see what might have changed
since Sunday.  He reverted to a previous version of the code to see if we
could get back to what seemed to be a more stable system on Sunday.
Unfortunately as soon as I got off the phone we had to close due to
humidity, and were not able to open the rest of the night.
   When closing up, we had a problem with coming up out of stow in
altitude.  Apparently we were right at the down limit for the windscreen.
Sadly we were unable to go up either.  Each time the altitude was reset
there was some logic condition where the windscreen would power on and off
at about 2 Hz for a couple seconds then everything would stop.  I even
tried lifting up on the telescope a little while someone else reset the
amplifier in hopes that the short time the windscreen was on, it would move
up off the limit.  This didn't work.  I ended up having to move the
windscreen up with the manual control to get it off the limit.  I'm not
certain, but I think that if there were some hysteresis in the limits this
probably wouldn't happen.  Once the down limit was reached, down could be
inhibited until the windscreen moved up say half a degree or so.  This
would prevent things from bouncing back to the limit when it gets powered
up as soon as it comes off the limit ( if this is indeed what is causing
the bouncing ).
                                    Dan