This photograph, taken in November 2001 during our brief but pleasant stay at COAA in the Algarve, is a tracked time exposure of the constellation Orion and surrounds. Camera Olympus OM-1; 28-80mm Hanimex zoom lens at 28mm; Fuji Super G 1600 film; exposure time 15 minutes. Scanned from a print.
The camera was mounted on a tracking platform, which compensates for the Earth's rotation. Briefly: the main axis points -- very accurately -- at the sky's north pole. Not quite at the Pole Star, Polaris, because it's actually a small distance away from the true pole position. A battery-driven motor then rotates the mount around this axis, once every day.
You'd hardly believe it was Orion, though. Mainly what you're seeing are the lights of Portimao, a coastal resort a few miles from COAA. Sky fog isn't very bad at the observatory, but there is some in this particular direction (though it's pretty much invisible to the naked eye).
The main fault was that we over-exposed the film; we should have gone for 5-10 minutes. It was also very fast film (1600), which didn't help. And the aperture was wrong, and....
Judging exposure times for wide-field astrophotography is not easy, with many variables all interacting with each other.
Anyway, we now have this picture. Can anything be done to rescue it?
Here's what we came up with. What you do is construct a 'flat field' consisting of the background glow only, then subtract it from the image. The kicker is making a good quality and accurate flat field: the method described in Correcting Camera Vignetting doesn't work very well here (the glow is unevenly distributed, and the brightness varies a lot), and other ways of constructing it were extremely laborious, too difficult, inaccurate, or all three. This method is quick, easy, accurate, and we haven't seen it described anywhere else.
|top home comments?||all contents © lee montgomerie and chris terran 2002|