Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.astro.louisville.edu/software/sbig/archive/xmccd-2.1/BETA_VERSION Дата изменения: Sat Nov 10 10:34:26 2007 Дата индексирования: Mon Oct 1 21:07:56 2012 Кодировка:

November 10, 2007
===============

Please also read TODO for the most recent comments on the contents of this
directory. The files here will change whenever we have a new beta version that
has had an error-free telescope run.

The version in this directory is functional but also has parts still in
development. This version should probably be used instead of older ones
because flaws are being corrected here as soon as they are identified.

A few things to note, however, about this "beta" edition:

There is almost no error checking on the operation of the filter wheel. It
would be a good idea to check at some point at least once, to be sure that the
wheel is responding. I have seen instances on our STL6303 and STL1001 where a
newly installed wheel did not respond either to XmCCD or to SBIG's CCDops. A
procedure that seems to work reliably for us when installing a new filterwheel
is to check it for binding first, and press lightly on the motor to engage the
O-ring. Make sure that a white spot on the ring is directly over the pickup
before closing the case. CCDops has better error checking than XmCCD, so if you
have a Windows computer available, check your camera with CCDops when you use it
for the first time. That said, we obtain images with XmCCD that are are just as
good as those with commercial Windows programs that also rely on the SBIG
libraries for Windows.

The routines and user interface for filter control are new. In this version
the filter will increment with each button press. I have a built-in fixed delay
to inhibit multiple requests too fast for the wheel. You will have to change
the filter labels for your camera in the source code. Look at the top of
xmccd.c for the notes on how to make your own filter set appear. The finished
version will have a provision for a user file.

I am working on an implementation of time delay integration (TDI) or "drift
scan" imaging and pieces of it are in this beta version. It does not work
s is because there is only an approximate relative timing loop just to hold the
place for something that works better. Code for that has been written, but has
not yet been tested.

TDI will offer exposure times with the STL6303 of about 100 seconds on the
equator, increasing as the field is moved toward a pole. In effect, it uses
electronic shift of charge along a column on the CCD to track while the
telescope remains fixed. For this reason, the columns must be exactly parallel
to lines of constant declination. This requires both careful initial adjustment
of the camera rotation, and exacting polar alignment to maintain the parallelism
in different pointing directions. There are two inherent advantages over
conventional mechanically tracked imaging: very smooth seeing limited motion
with exceptional image quality; averaging over columns minimizing pixel-to-pixel
variations. The disadvantages are the fixed exposure time and a field of view
in declination limited by any distortion in the telescope (if a line of
declination is imaged to a curve), or by curvature in the declination lines as
the field approaches the pole.