Subject: Minutes of April 13 APO User's Committee Meeting

From: strauss@astro.Princeton.EDU

Submitted: Wed, 15 Apr 1998 14:23:14 -0400 (EDT)

Message number: 271 (previous: 270, next: 272 up: Index)

	Apache Point Observatory 3.5m User's Committee Meeting
****************April 13, 1998**********************

Attending: Rene Walterbos, Michael Strauss, Bruce Gillespie, Ed
Turner, Jeff Secker, Alan Uomoto, Chris Stubbs

Not attending: Ed Kibblewhite 

*******Dust and the throughput of the telescope*************

The dust season has started.  We lost a full otherwise perfect night to
dust the other night (there was apparently a magnitude of extinction
to dust; it was *really* thick out there).
  We need as input the effect of dust on the mirror.  We really need
to get a robust measurement of total system throughput; the stuff that
has been done using DIS in slitless mode is believed not to be precise
enough for our purposes.  Multi-color photometry with SPICAM is the
way to go.

  Gillespie: Our real problem is gypsum dust, which can react with
aluminum and water, which is really bad news. 

  We could ask those observers who do careful photometric calibration (say
with SPICAM) to report the throughput they measure. 

  Walterbos: The real question is dust last year in June and July;
there is definitely real dust now (gypsum from White Sands), but there
is some question whether that which the dust meter triggers on in June
and July (when it starts getting humid) is as dangerous. 

  On dust, see Eric Deutsch's report:

  There is no obvious sharp features in the cumulative dust
distribution; so we wouldn't gain dramatically in number of nights
open with a small relaxation of closure conditions.  But with the
current cut, we're letting in 70% of all the dust we would have gotten
without any dust closure, based on a simple integration of the output
of the dust monitor. 

  Stubbs: The throughput goes to zero when we close the telescope!  So
let's keep the telescope open during even dusty conditions.
  Or, let's do a really careful study of this.  If gypsum can actually
damage glass, then we should be really careful.

  Another problem is pollen: trees within about 40 feet of each telescope
are being removed, which should help some.  This will also provide a
fire break.

Stubbs: Let's do images with SPICAM at a delta = +85 of an open
cluster in UBVRI, and with the guider as well, as a measurement of
throughput.  The advantage is that such a field is observable all year
at a constant airmass.  A full set of measurements takes only 5
minutes or so.  The real pain in the neck is getting flats for these
observations.  Perhaps put some extra filters into the guider, if we
can convince ourselves that the guider and SPICAM are really seeing
the same thing.  (Remember that the guider has 3 extra reflections in
its light path).  

  Strauss: despite the lack of precision of the slitless spectroscopy
throughput measurements, perhaps this is worth repeating to look for
gross changes in throughput, as we have a long baseline on these

  Stubbs: We have photometrically calibrated SPICAM images from a year
ago as well, with which we can compare. 

Baffling continues to be a problem.  Flat field varies by 5-10% on a
dark night in SPICAM depending on rotation angle.  Ouch! 

  There is a project led by Kurt Anderson to put conical baffles in,
which will reduce but not completely fix the light scattering

  Gillespie: We have advertised for a new observing specialist to
replace Charles Corson.  There are several candidates who are coming
for an interview soon.  We hope to have someone start by June 1.
  Karen Gloria wants to take a summer sabbatical, and so we're looking
for a temporary replacement for her as well.

*****************Engineering Status***********************
  There will be a summer shutdown, it is unclear how long it will be
(one week, several weeks).  The big-ticket item for this is to rebuild
drive boxes.  

Chris Stubbs: A series of engineering nights was used to carry out a
range of tests.  Data to determine the photometric calibration of the
guider were taken.  We made a series of observations to put together
our current image quality budget.  On one beautiful night, we got 0.8"
images in r', which seemed to be independent of whether or not we 
were pointing into the wind; fantastic!  This says that all the work
to tighten up the secondary support structure and the primary support
system has paid off; a real success for the three-year plan. 

  Checked for the wavelength dependence of FWHM.  The evidence is that
there are nights where we may be getting 0.5" site seeing; if only we
had our new secondary!

  We trailed star images to look at star motion in one dimension. 
This will give some indication of what can be achieved with fast

  We obtained polar field calibration data; see above. 

  We did tests to determine if the guider is confocal with instrument
plane wrt rotator angle; if it is, we can use the guider to focus. 

  We saw some funny business involving failure of the air supply for
the primary mirror support.  There is some concern that both the 2.5m
and 3.5m share a common air supply; when one goes, both will go. 

  We took pinhole images and flats as function of angle from the Moon:
conclusion, scattered light is a serious issue, and baffling is our
next high-priority task! 

  There is an increasing tendency of non-ARC-affiliated people to be
the observers at APO, through their collaborations with ARC
people.  This is a good thing, but they are not part of the ARC 
community, and therefore don't quite appreciate the culture; they 
expect KPNO-like support, and are occasionally somewhat nasty when
they don't receive it.  It may be appropriate to remember that it is
the ARC-member PI on any proposal who bears ultimate responsibility
for the observing run, and the conduct of the observer. 

The oversubscription rate is not being measured for APO; could
institutional schedulers give their total requests as well?
Anecdotally, oversubscription is a strong function of institution,
quarter, bright and dark time.  Moreover, some places just parcel the
time out without going through an institutional TAC, and therefore
oversubscription is an ill-defined quantities. 

  Stubbs: Let's spend $30K on a device for collimating the telescope
(see minutes of February '98 meeting).   The SDSS people do not appear
to be interested in such a device. 

  Previous month's minutes are approved. 

  Next meeting, May 11, 1998, 12:30 PM

APO APO APO APO APO  Apache Point Observatory 3.5m  APO APO APO
APO  This is message 271 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