Subject: DIS wish list - plse post to


Submitted: Mon, 11 Sep 1995 10:34:52 -0700

Message number: 5 (previous: 4, next: 6 up: Index)

Here are some suggestions for DIS improvements that I cast intot he sea
hoping that someone will find time to implement them...

1.  Pixel binning.  The only advertised choices at present appear to be 1x1
and 2x2.  Often one want to bin faint spectra along the slit (esp when the
seeing is poor), so 2x1 and 3x1 binning would be very handy.

2.  Partial chip readout.  I gather time-resolved spectra and need to keep
chip readout time to an aboslute minimum.  Reading out a patch of the chip
would be a great help.

       Table 1: Guassian fits to a uniformly illuminated slit
                      The trace of a blue lamp line is:
pixel   150-160  200-210  250-260  300-310  350-360  400-410  450-460  490-500
center    318.0   318.8    319.4    319.7    319.7    319.5    319.0    318.8
FWHM      1.22    1.30     1.40     1.33     1.30     1.32     1.16     1.25
                             and for a red line:
pixel    90-100  190-200  290-300  390-400  490-500  590-600  690-700
center    338.8   337.5    336.7    336.5   336.8    337.7    339.2
FWHM      1.44    1.57     1.43     2.0     2.2      1.6      1.39

3.  Better slit masks.  The "standard" 1.5" slit is bent - there is no
doubt about it (see table above).  It would really simplify life if a new
slit with a straight slit were fabricated.  We might even consider having a
wedge-shaped slit in order to match slit width to seeing conditions for
stellar objects.

4. Irregular slit projection.  Note the peculiar FWHM's for the slit in the
tables above, especially on the red side.  This is a result of
undersampling (which means that we can't tell if the slit focus is
  One onerous result of undersampling (for those of us who observe extended
targets) is that we cannot properly subtract sky lines by using portions of
those lines that lie near the edges of the chip.  Owing to both the
apparent slit curvature and the undersampling, night sky subtraction is
quite poor for objects more than about 50-100 pixels in size (depending on
which part of the projected lsit they are situated).  This seriously
impairs the utility of the long-slit functionality of the spectrograph.
  I suppose that the best way to fix the undersampling problem is to use a
straight slit that is carefully aligned to rows on the CCD.    Another is
to use a wider slit (ugh!).  Still a third is to use higher magnification
in the camera optics (expensive??)

--Bruce Balick 9/11/95

APO APO APO APO APO  Apache Point Observatory 3.5m  APO APO APO
APO  This is message 5 in the apo35-dis archive. You can find
APO  the archive in /u/strauss/apo/mailer/apo35-dis on
APO  To join/leave the list, send mail to
APO  To post a message, mail it to