Subject: Minutes of APO User's Committee, 9/29/99


Submitted: Fri, 5 Nov 1999 16:38:04 -0700 (MST)

Message number: 393 (previous: 392, next: 394 up: Index)


minutes prepared by R. Walterbos
with input from Bruce Gillespie


Don York (HUC; morning only)
Ed Turner (PU)
Bruce Gillespie (APO)
Alan Uomoto (JHU)
Rene Walterbos (NMSU)
Mike Evans (UW)
Kurt Anderson (NMSU)
Jeff Morgan (UW)
Mark Klaene (APO)
Karen Loomis (APO)
Chris Stubbs (UW)
Frank Deglman (APO)
Jon Davis (APO)
Gretchen van Doren (APO)
Craig Loomis (APO)
Jim Gunn (PU; mirror discussion only)
by phone:	Michael Strauss (PU)
		Lew Hobbs (UC; part of afternoon)
		Jeff Brown (WSU)

1. Current status of Capital Improvements Fund projects (3-yr plan):
Listed in order as on original flow chart from Alan Uomoto.

- Primary mirror support: all work done, but now additional work
required due to M1 problem (see item 2).

- Secondary mirror refiguring: new mirror acquired.

- Secondary support truss: early (few $KS) improvements done while
ago. New secondary will eventually be installed with new top-end
(Kibblewhite's project), for improved stability and safety, fast
guiding capability, better positioning, and mirror
collimation. Reduced vibration and wind pickup, and provision for
artificial laser guide star launcher in front of secondary are other
goals. Secondary will be installed in temporary mount during shutdown.

- Altitude drive oscillations: addressed, but not clear if fully
suppressed. Better image quality with new secondary may show that
problem still exists at some level.

- Slow guider: done, but we will need more for the other ports

- Fast guider: not yet done, but preparationary work completed.

- Pointing improvements: not so much of an issue any more, although
original specs never met. Given slit view camera, 1-2" rms pointing
not essential.  We may learn from SDSS 2.5-m pointing and tracking
performance to make further improvements.

- Baffling: not done. Only partial baffle past tertiary done. Need
full baffling ahead of tertiary.

- Instrument rotator: problems solved, except few minor issues, but
will need additional rotators for other ports.

- Tertiary motion control: done.

- Collimation: Need $5-6K funding for additional software
license. Procedure for collimation and tools exists.

- Diagnostic subsystems: started in 1999, expect to complete in 2000

- Thermal management: to be improved further, with Kurt's 10-micron
camera. Study will be done by Kurt and undergrad.

- DIS: not yet done.

- GRIM2: Some upgrades done, but needs to be replaced. Experiment with
Casey's 1Kx1K camera in Q4, 1999 with user access expected in
2000. There was concern about some of the specifics of this camera,
including the need for liquid He cooling, and the high readnoise.

- Mirror cover automation: expected to be done within one month after
current shutdown, not later than end of year. Automation of lamps done.

- LN2 fill system: changed philosophy due to 180 ltr dewars available

- Networking: vBNS connection done. Improvements to internal network
done. May need higher throughput from site to Abilene or vBNS,
although according to Craig we only use 10% of current T1; mosaic CCD
camera may require better than T1 connection.

- Software: Echelle interface has been included, but complete overhaul
of software has not yet started. Craig and Russ Owen would be logical
project managers to pull effort together.

- Mass storage: done

See further down for list of future projects.

2. Mirror M1 status.

Cracks in back of mirror in 3 hex cells discovered during
shutdown. Problem likely to be related to epoxy used, or possibly
mirror support structure, given location of areas with most prominent
cracks (on outer edge of mirror, in opposite direction of each other,
on vertical line going through mirror center if mirror points to
horizon). At least one of these cracks was present in 1994. A third
smaller crack is also towards outside of mirror, but in "random"
location. Origin of cracks not known for sure, but concensus, after
consultation of many experts, is that remounting mirror is safe, with
some modifications to current support. Long-term solution may require
substantial more work on mirror mount.

Another expert on glass fractures will inspect mirror in early
october, before remounting it.

Plan is to go back in operation relatively soon, not yet clear when
we'd be back on the sky.

3. Current shutdown report:

- tertiary realuminized. Reflectivity now 90%, was 88% with water spots.
improvement in blue could be larger.
- replaced pmss tubing.
- tested belloframs, replaced 6 axial, 2 transverse
- created tolerance for lateral links at perimeter seal
- cut access holes to lateral links
- removed old temp. sensors, added 3 new ones
- washed primary, was 84%, now 86%
- new 2ndry mirror installed
- new collimation tools: Shack-Hartman at Nasmyth 2, X-Y stage, video
- replaced broken flex pivot in 3ry and two in 2ndry
- installed mirror cover switches
- replaced T-vent tubes
- replaced 2ndry radiation shield, was foam, now aluminum.

4. There was extensive discussion of issue of effect of extended
shutdown on Q3 and Q4 scheduling. Concensus was that individual
institutions might wish to reschedule their time, but no formal action
was required, since shutdowns are treated as rain. Director would
inform institutional schedulers of the situation and leave it up to
them to make changes.


- Echelle: Don York presented detailed report on current performance
including some examples of spectra. The 50% sensitivity points are
located at 3800 and 7800 Angstrom. There is a ripple on scale of the
orders due to grove shadows, which should come out reduction. This is
unique to this spectrograph, since no other echelle gets all the light
as we do.

Other reduction issue:
	Aliasing: the orders are squashed together. The standard IRAF
	package defines mask by fitting to the data, but does not do a 
	perfect job (0.5-1% level). Two fixes:
	1. put in the 3" slit (which is in there now as well)
	   and jog star up and down slit. This doesn't have to be
	   done at an even rate. This will soften the edges of
	   the aliasing problem.
	2. software solution: There is Keck Package, Scott Burrows 
	   is the expert, Tom Barlow (?) maintains it. Don is working
	   on this.

Performance: Limiting magnitude now about 16. Throughput is good, like
Kitt Peak, but readnoise is too high (7 e-). Observers need to come
with good coordinates since acquisition camera is tiny. You need 10"
finder charts, not 1'. The read time on the chip is also limiting
efficiency, while the flat field lamp is not very bright in the blue
(4 minute exposure in blue versus 10 sec in red).


1. It would be good to get chip with lower readnoise (1 e- would allow
us to reach 19.5 mag in few hours) and smaller pixels (15 micron would
give 5 km/s resolution). Now maximum integration time is limited to 30
minutes due to cosmic rays. Binning to reduce readnoise is not an
option due to close packing of the orders. Dark current is also
somewhat too high now, cooling needs to be fixed to reduce operating

Cost to replace CCD:

$250K or, if we can get free chip and use existing electronics then
perhaps about $50K.

2. Other obvious improvement: AR coat the optics, which would increase
efficiency by 50%. This may cost about $6K per piece of optics, or
about $60K total.

- GRIM2: still operational, produces about 40% of published papers,
but could die any day. No longer a competitive instrument.

- DIS: upgrade essential, see below

- SPICAM: is functioning OK. Various new plans discussed (see below)


- Casey's IR camera (see also above under item 2).

Sofia spare, 1Kx1K Indium Antimanide, liquid He cooled cryostat, 27
micron pixels. J,H(?),K,L',M' bands, operating temperature 35K, pixel
scale not quite clear, readnoise 125 e-. 

Two concerns: cost of Liquid He cooling, high readnoise. Not clear if
multiple reads possible. Casey thinks that readnoise and cryo
requirements can be made better.


Narrowband tunable filter camera. Has nice dewar and detector, and
could possibly be retrofitted as general broadband imager, since
construction is modular. The FOV is small, but not clear how much of
this is related to AOTF rather than the camera itself. Run in November
by Chanover and Hillman had to be cancelled due to extended shutdown.

- SPICAM (Stubbs)

Planned upgrades include: moving it to one of the 45 degree ports so
it would be mounted permanently, this would require new instrument
rotator, and may have consequences for filter wheel access. 

Plan to implement fast guiding into SPICAM with lenslet array,
separated by 200 micron, each lenslet puts light in 4 fibters fed to
bench of counters. Output of counters is voltage which can drive
secondary for tip-tilt. Should work down to 18th mag, implementation
in focal plane and in SPICAM is not yet entirely clear. This would
allow for fast guiding only in spicam, not for all instruments. This
would also not provide closed loop focusing, but this could be
discussed further.

- Closed loop focus: DIMM provides useful info, but not yet
implemented. Shack Hartman tests give some predictive input as
well. Autofocus loop may be essential though. The plan is to use
Nasmyth off-set guider for auto focus.

- DIS: Stubbs: current plan calls for much reduced effort in improving
DIS, focusing on CCD replacement as most fundamental part, but not on
redoing optics, or fixing UV throughput problem.

New chips not yet available, but will most likely be SITE 1K's, 24
micron pixels, less than 8e- readnoise, hopefully as low as 5 e-. An
alternative would be 15 micron 2K chips, if we can find them. The
Lincoln lab CCDs originally envisioned will not pan out.

Time estimate for replacing chips: 1 year, Stubbs and Gunn will
coordinate effort.

Stubbs will distribute plan on short notice for feedback on planned
DIS upgrade time table and specifics.

Gunn is still interested in addressing the UV response problem.

- FTS (Stubbs) 

Fourier Transform Spectrometer, with LLNL. Resolution from 10-few
thousand.  Good for extended bright objects, eg. globular
clusters. APO is their main telescope choice now (need Nasmyth focus

- PNS (Stubbs)

Planetary Nebulae Spectrograph, optimized for [OIII], using Ken
Freeman's trick to obtain kinematics for lots of nebulae quickly, by
180 degree flip. Mannery is involved and a graduate student
(Ph.D. project).

- MOSAIC Camera: is still long way down the road. The Lincoln Lab CCDs
were not optimum, with a CT problem which can hopefully be addressed
in software mode.

Tyson may be interested in hosting BIGCAM at APO (now at CTIO, but
need there may be less).

- Fabry-Perot & Lyot (Woodgate's GSFC instrument)

This is being tested as a UW collaboration, with some supplemented DD
time. They will come in December, 5 full nights. FP imager would
become available to community. Remote operations is a concern.


New funding available through center of excellence grant. Redesign of
telescope top end part of this effort.

- LLR (Stubbs) Lunar Laser Ranging (also done at McDonald). Best test
of equivalence principle of General Relativity.

- Chicago/JHU instrument

Currently planned as medium resolution (R~10,000) spectrograph, high
throughput in 7000-10,000 Anstrom region, CCD detector, point source
(3") aperture device, which should be better than echelle in this
wavelength regime. Niche: resolve out OH lines in IR to do science



+ indicate hjghest priority
* indicate secondary priority
rest no clear concensus, or priority not directly up to (like new instruments)

+fast guider			+baffles, calibration	+M1 support
<-----------------------------  *new top end ---------------------->
+DIS UV response		+DIS detectors		+limits&interlocks
				+DIS slitviewer		*telescope eng. monit.
				+/*software rewrite

New Instruments:		*Image quality		new axis controllers
	- IR imager		  - M1 AO		aluminization chamber
	- IR spectrograph	  - closed loop focus	
	- narrow-band imager	  - thermal
				  - DIMM
				  - fast guide
				*echelle upgrade
				APO/NMSU network
				general network
				guiders and rotators
				(sytem efficiency campaign)


There was lengthy discussion on the need for an aluminization
facility, and the way to meet the cost between the 3.5-m and SDSS. 

With the WSMR support gone, the O&M will already go up next year. The
aluminization facility would add another substantial increase. On the
other hand, there may be a window given the needs in the SDSS
project. Stubbs noted that if we take care of the mirror,
realuminization may only be required once every 3 or 4 years, in which
case the cost over the expected life time of the observatory is not so
high. The risk of transportation is another factor, mostly because
this is not so well under our control, once the mirror is on the
truck. This will require Board input.

9. Site issues

Bruce Gillespie discussed the needs for improving the phone system,
staff retention issues, enhanced staff requirements for bringing
visiting instruments to the site, storage, and new vehicle.


The monitoring programs do cut significantly in the nominal half night
shares. Walterbos suggested to allocate monitoring programs together
in half-night blocks, if at all possible, and not squeeze them
inbetween, in front of, or after two regular half night programs. This
is especially a problem for twilight use, and short summer nights. 

APO APO APO APO APO  Apache Point Observatory 3.5m  APO APO APO
APO  This is message 393 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