Digital SLR Astrophotography
by Michael A. Covington

Corrections and Updates

Last revised 2008 February 12
This page will be revised frequently as the need arises. Note however that it is not possible to track every change in the digital camera industry, and you should always use the World Wide Web and other current sources of information to help you make decisions.

Known misprints in the first printing:

The pictures on pages 176, 179, 186, and 189 were printed with insufficient contrast. Page 186 is the worst afflicted. To see these pictures properly on your computer screen, view the sample pages.

[p. 212]

Change:		DSLR Focus 95, 95
To:		DSLR Focus, 95

Other clarifications tentatively planned for second printing:

[p. 6, footnote]

Canon took this web site down in early 2008, apparently feeling that it was no longer up to date.

[p. 50]

In fact, the change came in response to a lawsuit. For now, the illustration at the top of p. 50 is not being changed because these telescopes are already in users' hands and were bought as a type of Ritchey-Chrétien. Meade's optical system is a good one; it just isn't a Ritchey-Chrétien.

[p. 78, caption]

If the lens is attached with a T-ring (as some older lenses are), misalignment (non-perpendicularity to the sensor plane) can arise there. To fix a misaligned T-ring, loosen and retighten the screws that hold the inner part of the T-ring in place relative to the outer part.

[p. 95]

Updates to DSLR Focus seem to have stopped in 2006, but I have seen no actual announcement of discontinuation. Other focusing software is being updated regularly to support newer cameras.

[p. 171]

in order to more closely match the preceding formula. The text is correct, but the original example caused some confusion.

[p. 179, figure caption]

[p. 187]

The illustration on p. 187 does not make clear why the boxcar filter is used. Here's a revised version with a better explanation:

Click here to download EPS version

Other notes

On the subject of calibration:

• It is somewhat unclear whether bias frames are needed with a DSLR. In theory, yes; in practice, especially with Canons, some kind of bias subtraction is already done in the camera. I suspect that Nikons in Mode 3 give a more "truly raw" image and are more likely to benefit from bias frame subtraction. A bias frame is a dark frame with the shortest possible exposure (1/1000 second or less), taken with the same ISO setting and camera temperature as the astrophotos and dark frames.
• Be sure to use many calibration frames — numerous darks and flats — for the same reason that you take numerous exposures of the celestial object: to keep down the grain. It makes no sense to take 10 exposures of a celestial object (to avoid grain) and then do the calibration with just one dark or flat frame (thereby reintroducing a lot of grain).

New cameras:

• The Canon 40D introduces live focusing, making it very easy to focus astrophotos, and has 14-bit instead of 12-bit output.
  
  It also adds the ability to open and close the shutter through USB commands, so that a separate parallel- or serial-port cable is not needed for full computer control. If for some reason you do want a parallel- or serial-port cable, you have to use a slightly different circuit than for earlier Canons; click here for details.
  
  This is an excellent camera for astrophotography and is what I am now using.
• The Canon Digital Rebel XSi (450D) is a less expensive, lighter-weight camera similar to the 40D, but I have not yet tested it.
• The Nikon D300 likewise adds live focusing. Contrary to early rumors, it does not eliminate the "star eater" (Mode 3) problem.