Subject: Report on JHU Training at APO


Submitted: Mon, 03 Jun 1996 21:30:35 -0700 (PDT)

Message number: 68 (previous: 67, next: 69 up: Index)

The following was presented to the APO 3.5 m users committee
for information.  While I doubt that there are any surprises
for regular users, it might be instructive to see what first-
time users see.

Alan Uomoto
June 2, 1996


The first JHU training run (May 1996) at APO was a success. On the 
mountain were Marsha Allen, Rupali Chandar, Arthur Davidsen, Andy
Dring, Holland Ford, Scott Friedman, Gerard Gilmore, Tim Heckman,
Richard Henry, Bill Oegerle, Alan Uomoto, Rosie Wyse, and Wei Zheng.
The weather was just about  perfect for the entire run, and everyone
obtained scientifically useful data. The DIS was the primary instrument,
with GRIM2 taking two half-nights. We exercised the DIS slit viewing
camera, but did not have the opportunity to use the offset auto guider.

I have summarized comments from the participants below.

Alan Uomoto
May 31, 1996




The management and staff at APO are excellent, as demonstrated by the 
on-time availability of the telescope and requested instruments. In 
particular, the observing specialists (Karen Gloria and Eddie Bergeron) 
were helpful, experienced, and effective. We were also very impressed 
with the quickness of the engineering team (Jon Davis in particular) in 
putting the telescope back together after rebuilding the altitude 
drive. The telescope was in pieces at 5pm, and they had it back 
together and re-calibrated by 10pm.


The telescope points and tracks well and produces decent images (we saw 
1.1 to 1.5 arcsec during good conditions). The control system worked 
well, failing only when we tried to track through the zenith (a known 
limitation that will probably take us a while to get used to).


While the telescope is state-of-the-art, we were disappointed by the 
outdated instrumentation. The DIS is two generations old and suffers 
from problems that make it less effective than it could be. Lack of an  
imager with filters is a serious deficiency.


We are enthusiastic about the observatory and considered the ways that 
we could contribute (we understand that this is our observatory and we 
get out of it what we put into it). Many things were discussed, and 
these seemed like good short-term starting points.

(1)  We plan to build some air-spaced, reflective, long slits for DIS.

(2)  We talked about improving the DIS slitviewer and would consider 
leading an effort to do so.

(3)  Because the SDSS spectrographs will have a software user interface 
for the telescope and instrument, we thought this might be applied 
to the 3.5 m as part of the future software cleanup plan.



Specific things JHU users would like to see.  Some of these are
long-term wishes and some are unlikely ever to be done.  Others,
however, seem like they might be easy.


A mechanism to enable routine upgrades and maintenance of the 
telescope, instruments, and hardware should be implemented. It should 
be easy, for example, to get the DIS CCD noise problem fixed. The 
impact on science is large and the repair cost is probably small. How 
do we do this? Can the director receive funds/resources for this kind 
of thing? Do we assign tasks to institutions?


Put fresh coatings on the secondary and tertiary now (even before 
recoating the primary), and consider using protected silver. Silver is 
considerably better than aluminum everywhere redward of 3800 A (but it 
fails blueward of 3500). GRIM2 wins and DIS doesn't care (until its 
blue sensitivity problem is fixed). DSC loses in the U, but without a 
filter box it's hard to find projects that would keep it on the 
telescope a large fraction of the time.

Improving the telescope image quality by fixing the wind shake and 
secondary mirror should be done.


Remove the coherent noise from the CCDs.

Real-time video of the slit (leaky guider type, 0.03 to 30 Hz) would 
improve observing efficiency and might also be used as an autoguider. A 
lot of time is wasted centering objects.

More long slits.

Tools to mark the target with a cursor and automatically to move it 
onto the slit would be nice. This should include the ability to rotate 
about a particular point in the slit viewing image.

Can we install larger, better CCDs? The read noise is high for modern 
CCDs and the small format restricts wavelength coverage in high 
dispersion mode.

Separate shutters for red and blue channels.

There should be a method for the user to close the mirror cover and 
turn on the comparison lamps remotely.

The blue sensitivity of the DIS needs fixing.


Would be great if could implement the dithering macro from Remark. 
Would be even better if that routine worked -- it failed every time we 
tried to take more than 3 images at each position. Very tedious, 
especially since it failed randomly.

Note the plethora of bad pixels in the detector; new one would be 


While we understand the evolution of the instrument control system 
(Remark on the Mac, CCDOPS for the slitviewer PC, and IRAF on the Sun) 
and can forgive its idiosyncrasies, it is probably time to consolidate 
the good features onto a single reliable platform. A hardware 
independent X-windows implementation would be a good place to start.


On the APO homepage, note altitude in large letters and say "bring a 
hat so you don't get skin cancer."

Warn people to bring dressing gowns; this observatory is without 
private bathrooms or sinks (reference to SDSS dorms).



These are probably all known except, perhaps, for the telescope motion 
noted in the GRIM2 section. 


There was confusion about who was allowed to use twilight time for 
calibrations. While we did not have problems this time, it was easy to 
imagine situations where conflicts would arise. For example, the lens 
program used dusk for DIS calibrations in imaging mode. We used DIS for 
spectroscopy for the remainder of the first half after which GRIM2 was 
installed. DIS spectroscopy would have to spend dark time doing 
calibrations (this didn't happen, but we wondered).


The CCDs show coherent noise. The noise has an amplitude of ~15 DN in 
the Tek CCD, and its phase changes with time, making it a pain to 

Setup on the long slit takes too long. Three keyboards (and three 
computer operating systems) are needed, and it takes something like 10-
15 minutes to get started.

We need a better selection of reflective slits (1.5, 2, and 3 arcsec, 
for example). Too much light is lost off the slit jaws.


The first GRIM2 image after a slew showed the telescope to be moving 
(trailed images, about 5 arcsec, no wind). How long after "move 
complete" before the telescope stops shaking?

Images went way out of focus -- not clear what happened. Moved filters 
in and out, no difference, then changed f/5 to f/10 and back. OK after 

APO APO APO APO APO  Apache Point Observatory 3.5m  APO APO APO
APO  This is message 68 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