Subject: Minutes, APO 3.5m User's Committee meeting, 11/1/99

From: strauss@astro.Princeton.EDU

Submitted: Thu, 4 Nov 1999 11:39:55 -0500 (EST)

Message number: 392 (previous: 391, next: 393 up: Index)

  Apache Point Observatory 3.5m User's Committee Meeting
		November 1, 1999

Attending: Jeff Brown, Lew Hobbs, Rene Walterbos, Bruce Gillespie, Ed
Turner, Michael Strauss, Alan Uomoto, Chris Stubbs. 

Minutes, prepared by Rene Walterbos, from the APO User's committee
meeting held at APO in late September will be distributed by the end
of this week.

**********Status of the Secondary and the Shutdown******************
  At that meeting, there was in place a detailed plan for redesigning
and reworking the primary mirror support system, given the troubles
found with cracks in the mirror cells.  During the extended shutdown
(to end roughly October 25), Jon Davis and crew were to gather the
information needed to carry out this redesign, and increase safety
margins in the support system in the event of power failures.

Gillespie: This plan was more or less on schedule, but just as we were
getting the telescope back onto the sky, the 3.5m tertiary mirror was
found not to be holding its position accurately.  There was damage in
the drive flex pivots; the suspicion was that the drives had bottomed
out (these flex pivots act as a "mechanical fuse").  There has in fact
been continued worry about the reliability of the mechanical limits on
secondary and tertiary.  At that point, serious problems with the SDSS
secondary were discovered. Given the very similar support structure on
the 3.5m secondary and lack of limit switches or other safety
mechanisms, we decided to stop operations until we understood the
problem fully.  A crash program was implemented to install limit
switches on the secondary and tertiary support structure (more on this
below), and we reverted to the previous secondary support structure
design (i.e., one different from the current design on the SDSS
telescope).  This reverting of design was straightforward, requiring
only a few parts to be machined.  This older design should be safer,
although it perhaps gives somewhat less exact positioning.  Finally,
we are considering updating the stepper motor system, to bring
commonality to the systems on the secondary and tertiary with better

  The current best estimate is that we'll get back on the sky for
commissioning on November 19, with science operations starting on
November 25 (Thanksgiving!). 

  With a real plan in place, we should have a real schedule for
getting back on the sky in the next 24 hours [MIGHT THIS BE UPDATED

Stubbs: The control system on the 3.5m secondary has been known to run
away, and put a huge strain on the secondary.  Limit switches that
physically turn off current to motors if the secondary is outside
allowed region of motion would be the right way to proceed.  However,
it is devilishly difficult to directly measure the tilt in practice.
So the right way to do it, is to measure the position of three points
around the rim of the secondary, and use these to decide when the
secondary is out of position.  It is not *perfect*, but should be
quite a bit better than what we've been working with up until now.

  A choice: to get a definitive, failsafe system that would absolutely
keep our brand-new secondary safe, we would like to keep the telescope
off-line for six months (!), and redesign everything from scratch.
  *Or*, we can get a system in place on the short timescale 
which would require failures at two levels before damage to the
secondary happens.  One cannot quantify the risk factors.
  *Or*, put the new secondary aside for the moment, and put the old
secondary, with all the deficiencies (and dangers) of the old mounting
system, back on the telescope.  All these mirror swaps would cause
another ten days lost to the sky.

  Chris is pushing for the second of these options; the new topend
from Ed Kibblewhite, which should be ready roughly 6 months from now,
would have a yet safer secondary support.

  One of the pluses of all of this is a faster focus mechanism.

As a group, we like Chris' approach, and recommend that it be adopted,
but feel that it would be good to get some external expert to give the

*****************Telescope scheduling:***********
  Due to the extended shutdown, eating into an appreciable fraction of
the fourth quarter, the suggestion was made that the telescope
schedule be redone.  However, at the last meeting of this group, we
decided not to reschedule the fourth quarter, but rather simply remind
institutions that they have the option of rescheduling within their
individual allocations.  Only Princeton has in fact done so.

  We will probably re-open just about the end of the week-long shutdown
for the controlled burn near the site. November 18-24 inclusive (bright
time), which is now scheduled as engineering time.  There is a
possibility that the burn will not happen, due to weather and/or
availability of forest service personnel.

  Sean Casey's planned run using his near-IR imager was cancelled due
to the shutdown; it will be rescheduled during the first quarter of 2000.  

We discussed the statistics on the use of the 3.5m, following the
plots at:
which were prepared for the upcoming Board of Governor's meeting. 

Notice that the fraction of time the telescope is closed seems to take
a jump upwards at about the time we became more stringent on closing
for dust (after the primary realuminization). It may, however, also be
due in part to increasing bad weather related to El Nino; it is hard
to separate the two effects. 

  The final plot here shows that the time allocation to each
institution is not quite matching the nominal allocation, due to the
specific vagaries of scheduling; up to 20% of some institutions.  This
may be due to specific styles of request, asking for impossible
allocations, which cannot be assigned.  

  It might be useful to start quantifying oversubscription rates for
each institution; a request has gone out to the TAC's at each
institution to give their numbers. 

  Frank Deglman has been hired as an observing specialist, so we're up
to full strength.  

  The next meeting will be held on December 13 at 11:30 AM East Coast

APO APO APO APO APO  Apache Point Observatory 3.5m  APO APO APO
APO  This is message 392 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