Subject: Update on DIS gratings

From: Russet McMillan

Submitted: Sat, 23 Mar 2002 02:36:17 -0700 (MST)

Message number: 563 (previous: 562, next: 564 up: Index)

Hello all,

The responses to my earlier solicitation for opinions on DIS grating
availability after the upgrade were many and varied (surprise!).
Most respondents seemed to feel that the suggestions I made sounded
good, but the gratings they eventually ended up using would depend
upon the throughput and S:N with the new detectors.

We have now completed much of the characterization of the new DIS, and
in some senses the throughput is not yet as good as we could wish.
The blue field flattening lens has a red coating on it which causes a
sharp dip in response starting at about 4600 Angstroms.  The red chip
has some additional read noise in the form of periodic spikes every
few hundred pixels.  Both of these issues are expected to be resolved
in April, along with an improvement of the focus/tilt of the blue
detector, and at that time further characterization will be needed.  A
third problem, that the optics designed for the old detectors cause
some vignetting on the new larger-format detectors, will not be
addressed until the summer shutdown (and then FURTHER characterization
will be needed!)

You can read some of the initial results of the characterization at .  Further work is still
ongoing.  Also, check the plot of vignetting at the top of .

The significance of all this for the choice of which gratings to use
is that there may still be reasons for users to want any of the six
available gratings.  Although the medium resolution gratings permit a
full spectral range to fall on the chip, only the central portion of
the chip has the best throughput.  For 80% of maximum throughput or
better, the relevant portions of the spectrum should go between
columns 400-1600 (on either chip), and for 90% or better the spectrum
should fit between 700-1300.  This means that some observers pursuing
the very faintest objects might still want to use the low resolution

Note that the sensitivity and S:N are still an improvement over the
old detectors, and there are other options for changing resolution
without changing gratings; the chips can be easily binned in either X
or Y to consolidate light along the slit or in the dispersion

Given all these considerations, we will not be making any gratings
unavailable during the next few months, but we do hope to encourage
users to stick to a limited set.  The default sets we have arrived at
are the high resolution pair and another pair including the low blue
and the medium red (both these gratings are 300 lines/mm, providing a
dispersion of 3.1 Angstroms/pixel on the detectors).  One pair of
gratings provides the best resolution we can offer, while the other
provides a compromise between full spectral range, good S:N, and
reasonable resolution (all better than with the old detectors and the
low OR medium gratings).  S:N can be further improved by binning, if

For the time being, we are continuing to use the old low, medium, and
high names for the gratings, even though the two low gratings, for
instance, do not have the same number of lines/mm.  We are discussing
new terminology now and would welcome suggestions; simple names for
pairs (e.g., Pair 1), or specific names for individual gratings,
(e.g., Red 300 l/mm?)

Here's the procedure that will be used for DIS observations during the
next quarter:  

1) Regardless of the grating requests included in your Q2 proposals,
   we will assume that one of the default grating pairs will meet your

2) If you are happy with either of the defaults, you don't need to
   send us advance notice.  We would still appreciate a note one day
   before your observations to let us know you're aware that the
   observing time is coming up, and to tell us what grating centers
   you will likely use; if you change your mind at startup, those
   numbers can be altered.

3) If you are NOT happy with the defaults and want another grating
   pair, you must inform us at least 3 days in advance, with
   confirmation 1 day in advance of your observations.  

   a) Please tell us why you need a non-default setup, and specify at
   MOST one red grating and one blue grating for your program.  Check
   the list of available gratings at
   to be sure you're requesting gratings that we actually have, and
   that they will provide the resolution you need.  Gratings may be
   mixed and matched (as with the default BL+RM combination), but red
   gratings cannot be placed on the blue side or vice versa.

   b) If your request arrives late, or if you request non-existent
   gratings, it's possible the setup will not be ready at your
   scheduled start time.  Grating change and re-calibration requires
   approximately half an hour per grating, one hour per red/blue pair.

   c) If your non-default grating request conflicts with the default
   request of another observer on the same night, you will have to
   give up time to allow the grating change to occur without
   detracting time from the other program.  If you have a non-default
   grating request which DOESN'T result in any conflict, you shouldn't
   lose any time.  If both the first and second-half observers have
   non-default grating requests which result in a conflict, the time
   required for the change will be split between both programs.  This
   is similar to the policy for planned vs. unplanned instrument
   changes, and should be your incentive to try the default gratings
   if they can work for you.

At the end of the second quarter (or perhaps when the optics are
upgraded), we will review this policy.  At that time it may be
possible to retire some gratings from use or make some other
simplifying change.  For now, all the gratings are still available but
we encourage users to try the default sets.

Thanks to everyone who gave us their input on this question, and we
hope you'll all be happy with the new detectors!

				  Russet McMillan.

APO APO APO APO APO  Apache Point Observatory 3.5m  APO APO APO
APO  This is message 563 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