Subject: Echelle slit

From: Don York

Submitted: Wed, 7 Jul 1999 16:31:15 -0500 (CDT)

Message number: 11 (previous: 10, next: 12 up: Index)

I have been getting various types of spectra and doing various experiments
with the aim of understanding the flat fielding. In principle, things should
work well with the 1.6x1.6 slit. I had the 1.6x3.0 sec slit made because
several users thought it essential. I understand that the square slit is
back in and that is the way I prefer it. At the resolution we have, emission
lines in the night sky are not much of a problem and the continuum does
not affect most problems (sky subtraction was never a motivation for the
longer slit, though it appeared to be assumed so in a message posted with 
the night log a few nights ago. For sky subtraction, two square slits should
suffice and I am making such an aperture.

As the plan for the echelle was no moving parts, let us use the square slit
for now. The night assistants can change the slit for special purposes,
but please avoid this if possible. I hope to finish my flat fielding experiements in a few weeks and will advise of my findings at that time. 

I do not understand statements that bad seeing or poor guiding widen the
image, unless this comment refers to the fact that the long slit itself
leaves little projected space between the orders (which , in retrospect,
must have been the reference.) At any rate, with the long slit in place,
I have a number of exposures filling the slit. The flat fielding is not
obviously improved (for instance, with a filled slit for the flat lamps
exposure and a small projected star image). There are other peculiarities
in the flat fielding between flats and stars, and in fact there is even
some evidence of dependence on exposure time. All this is being looked
into. For now, S/N of 50 should be fairly easily achievable everywhere,
but building up multiple exposures to go further does not repeatably work.
(Remember that the effective saturation is 37,000 DN, or 120,000 electrons,
which means we should get S/N of about 1000/1 corrected. Of course, if
any of the orders are saturated in the star or the flat, the flat fielding
in those orders will not work. Sorry, 1000/1 above should read 300/1.

Please leave the square slit in for now unless you really must change it.
More later.

APO APO APO APO APO  Apache Point Observatory 3.5m  APO APO APO
APO  This is message 11 in the apo35-echelle 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