Subject: Minutes of User's Committee Meeting

From: Michael Strauss

Submitted: Thu, 1 Feb 1996 10:16:51 -0500 (EST)

Message number: 43 (previous: 42, next: 44 up: Index)

  Minutes of APO 3.5m User's Committee Phone Conference

  Attending: Alan Uomoto, Bruce Gillespie, Ed Turner, Lew
Hobbs, Michael Strauss, Rene Walterbos, Al Diercks (sitting in for
Chris Stubbs)

		Table of Contents:
	Status Report for Telescope
	Status Report for Instruments
	Calibration issues
	Software and data
	Scheduled Shutdown
	New Observer Training
	Public Access to APO Web Site?  

	Status Report for Telescope
   1. Improvement in image quality reported over the last month, for
reasons which are not due to any obvious improvement in optics or
telescope. Maybe the weather is simply getting better!
   2. Pointing is also improving: 2" rms at the last pointing model.  
   3. The enclosure wheel in the worst shape is scheduled to be
replaced (see posting by Bruce to apo35-general #41). This will start
in a week, perhaps starting before the scheduled shutdown. The exact
time depends on L&F's schedule.
   4. There is a scheduled shutdown for engineering Jan 29-Feb. 4. The
list of tasks for this shutdown is included below. 

		Status Report for Instruments:
DIS: A filter fell out of the filter holder last week; has a chip in
it on one side. Clip design being looked into to make it more secure.

Slit-viewing camera: tests continue to minimize vignetting problem.

GRIM: Spectra and polarimetry has been done, but Bruce has not gotten
any feedback from the users about this. 
  The dark slide mechanism may be getting a bit sticky; it doesn't
always move out of the way when changing optical configurations. This
needs looking into. 

DSC: Back on the telescope. Still no filter wheel; Bruce will
communicate with Tim McKay to see what can be done at the site to
speed this up. 

Chaos: Successfully shined guide star laser out through finder scope
attached to the telescope. The resulting artificial guide star was not
detected, however. 
With a (real) bright star, Chaos was able to get image FWHM down to
0.2" at 1 micron. On that night, the seeing measured without the use
of Chaos was 0.8". 

AMBER: To arrive on site end of the month, on-line sometime in

		Calibration issues:
  3 pairs of emission-line spectra lamps are now permanently
installed; controllable by software. Also, there are two intensities
of quartz lamps for flat fields to choose from.

  Illumination of pupil by flat field lamp is less than ideal, giving
a very non-uniform flat field; this is especially a problem for DSC
(cf., apo35-dsc #16). Baffles have recently been installed on the
incandescent light source; the tests to see whether this improves
things have not yet been done. Further baffling is needed to reduce
scattered light from a variety of sources; this is high on the list of
tasks for the upcoming shutdown.
  For DIS and GRIM, the flatfields are pretty good (accurate to 1-2%),
especially if you rotate the instrument. For DIS, there was a small
shift of ~ 0.5 A depending on what side the arc lamp illuminated the
mirror. The lamps are now mounted in pairs on the upper ring, which
should fix this problem, but again, the tests to check this have not
yet been done. 

Because of the desire to minimize strain on the dome wheels, the snow
program (by which the dome is rotated if there is snow accumulation,
in an attempt to knock it off) will not be run. This may mean we will
find ourselves with clear nights following a snowfall during which we
cannot open the dome. 

Jon Brinkman is putting together a web page with all relevant
calibration data; keep your eyes open for it. 

There is some concern that the current way of doing flat fields,
through holes in the mirror covers, is as yet unproven. We don't yet
have enough experience to characterize how good the resulting flat
fields are. However, a proper flat field screen ("a white spot") is
likely to be quite expensive, ~$40,000. There is a need for proper
tests of the current system. 

		Software and data:
 Up until recently, data taken with the telescope has been kept on
disk for 36 hours before being deleted. It has happened only once that
this policy caused someone to lose their data, although there have
been some close calls. Nevertheless, the hold time has been increased
to 72 hours. Observers who foresee a problem with not being able to
back up their data in time can always request the on-site personnel to
make a backup tape for them.
  There followed much discussion about the suitability of archiving
*all* data from APO. The two rationales are first, so that observers
don't lose data, and second, to allow general access to data for (as
yet undefined) projects. It is not clear at the moment whether there
is a problem that needs fixing with the current situation. Many other
observatories have experience with this, of course.
  One concrete suggestion was that the data from each given night be
stored on a separate subdirectory, to reduce clutter. 

 Network has been behaving itself, more or less. There exists the
ability to operate the telescope via modem, but people have not taken
advantage of this yet. There will soon be two modem lines
available. It is the user's committee representative's responsibility
to make sure that instructions exist at each observing site for
getting a SLIP connection.

	Scheduled Shutdown:
The following are the list of projects planned for the week-long
  (these assume a working telescope, which means that the enclosure
wheels have to have been fixed):
   1. The guider misbehaves sometimes when the telescope goes through
zenith. This needs to be fixed. 
   2. The DIS slit-viewing camera is almost working; it needs more
work to minimize vignetting. 
   3. There is work to be done on the baffling to minimize stray light
problems in the telescope. 
   4. The r0 telescope needs to be made working and calibrated. 
   5. Measurement of piezoelectric properties of the secondary
      pistons, for use in tip-tilt (for poor-man's adaptive optics).
   6. Design optical breadboard at back center folded Cass, so that
one can put a detector there to measure optics off-axis.
   7. Complete measurement of throughput of all instruments. 

Washington is sending 5 people for this week; APO will have several as
well; so in one week, should make some real progress.

		New Observer Training: 
general policy is that before someone is allowed to observe, they
should be trained on site. For someone with extensive observing
experience, they should just come; if it is a student, or someone with
limited experience, training means coming with someone with extensive
experience to train them, spending three nights on the mountain. There
are no resources on the mountain to train unexperienced observers!
Sounds clear, but in practice there are variety of questions which
come up.
  Can an experienced graduate student teach a new grad student?
  What if the three nights are cloudy? 
  Should people have to be trained separately for each instrument?

Question: what is this really for? To learn how to observe, and to
have a mental picture of how the observatory is set up.  No-one could
think of examples in which people have observed remotely without the
necessary experience? In any case, we don't want to raise a generation
of astronomers who don't have real telescope experience ("Nintendo
astronomy"). We did not make any progress in clarifying this issue

		Public Access to APO Web Site?  
  Ed Turner says it is always a good idea to be as open as
possible. Also, our "private" forum is not very private at all; *many*
machines already have access to it, and a good hacker could penetrate
without much trouble. There are potential political dangers with
making the web site open to all; the information therein could be used
against us. Moreover, there is potential for hackers to read the
information contained therein, and break in, operating the telescope
themselves. However, a determined hacker will be able to do this
anyway; we may be fooling ourselves if we believe this forum to be
really private.
  Ed Turner will post a notice to apo35-general asking for further
opinions on this issue. 

The next meeting of the user's committee will be Monday, February 5,
at the usual time. Among other things, we will discuss the list below
of high-priority items which need attention at APO, as compiled by 
telescope scientist Chris Stubbs. 

I: "Known to be deserving of attention"

1. secondary.
2. enclosure wheels.
3. 20 Hz problem
4. oscillations in primary mirror cell.
5. windshake
6. baffling, scattered light.  

The first three are pretty well understood, the latter ones require more
diagnostics and characterization.  I think that (apart from 
the administrative tasks of getting 
the secondary procurement going and deciding whether to go with L&F
for the wheels) understanding items 4 and 5 should take priority.   
II: "Performance Enhancing Upgrades"

1. Optimize current guider setup 
2. Field lens for guider?
3. Tertiary rotation automated
4. Closed loop focus using guider.
5. Guider frames used to monitor conditions.
6. Tip-tilt of secondary implemented.

Jeff Morgan and Eric Deutsch are working on (1).
The next things to try are (4) and (5),  
followed by (2), then (3) and finally (6). 
III: "Instrument Issues"

for all: Full characterization of performance, throughputs, etc. 
         Read noise, linearity, bad pixel files.

         Generate Checklists for nightly instrument setup 
     A. DIS
1. slit viewer
2. replace detectors
3. UV throughput problem 

     B. GRIM
1. instrument block anomalies
2. documentation (specifically, drawings)

     C. DSC
1. filter wheel

     D. Others

1. get status report on Echelle.
2. Poll users on what we should be trying to achive in the long term, 
other instruments?

Obviously we first need to characterize fully the current performacne, 
in terms of achieved throughput, etc.  I very much hope we can
work towards getting the slit viewer done in the near future.

IV. "Computing"

1. Implement tape backup of all data prior to deleting it!
2. Reliable acquisition and storage and easy access to image quality
   parameter data: temperatures, met data, image FWHM from guider.
3. Get Higher net throughput, redundant connectivity.
4. Identify and Characterize bugs in existing code.

Numbers 1 and 2 are top priority here.
V. "Operational"

1. Generate a list (based on the above) of tests that can be
   done during inclement weather, and during daytime as available.
2. Identify and eliminate bottlenecks in observing process.
3. Streamline pointing model generation after tertiary rotation.
4. Develop repeatable and reliable collimation procedures
5. Develop a routine of ongoing performance monitoring tests.
6. Establish procedures for laser guide star activity.
These are all somewhat less critical than the others, but we should 
be paying attention.


APO APO APO APO APO  Apache Point Observatory 3.5m  APO APO APO
APO  This is message 43 in the apo35-general archive. You can find
APO  the archive on
APO  To join/leave the list, send mail to
APO  To post a message, mail it to