Now that the 20D has shown that it's a capable imager for wide fields of bright (and even not so bright) objects, it's time to put the dedicated astronomical CCD back to work.

 

There is nothing any amateur astronomer needs more than dark skies. It's best to live under them, second best to be able to commute to them, and third best to deploy 21st Century technology to see through light pollution. The third option requires an astronomically optimized camera built around a thermoelectrically cooled CCD. These options are not mutually exclusive: if you can take the outfit on the road, better. If you can move it to a dark-sky cabin in New Mexico, better still. All in good time.

November 24-25, 2006. Tonight's outfit is exactly like yesterday's except that an SBIG (Santa Barbara Instrument Group) ST2000XM CCD camera takes the place of the Canon 20D behind the 5-inch refractor.

Tonight I spent only 20 minutes re-assembling the mount and the telescope and recabling the electronics. Then another 10 gave me decent polar alignment. That's good.

Then I lost two hours getting the computer to talk to the CCD. Damn thing wouldn't be recognized, wouldn't even come on. I tried different cables, different hubs, different ports. I unhooked the camera and carried it inside to troubleshoot it on another computer.

Then I noticed the switch on the power brick cunningly hidden beside a label that read "On/Off". It might as well have read:

Know your gear.

And so back to the yard. The first target was the Bubble Nebula in Cassiopiea. It would have helped had I aligned the guide scope and the main telescope first so I could use the former as a finder for the latter. (Digital setting circles or a full go-to system would be even better. More cables is a problem with the former. Maintaining a bank balance is a problem with the latter.)

I took an image, any image, and focussed the main scope. I matched a distinctive asterism in the frame against a deep star atlas of the general area and then slewed the telescope over to the field of the Bubble Nebula. The bright stars of the cluster Messier 52 formed a welcoming committee.

There were no stars in the area bright enough for the webcam to guide on (I should've known this from inspecting a chart). There isn't enough light coming through the 10nm hydrogen-alpha filter behind the small refractor for the ST2000's onboard guide chip (which is why I've put the webcam on a second, unfiltered guide telescope in the first place).

I plunged ahead. Fine, we'll image without guiding. I turned on the cooler, dropped the chip to -35C and made a few short exposures. The nebula showed up in just 1 minute of H-a data, but very weakly. In 5 minute exposures, the nebula was easy to see, but tracking was dreadful. When I stretched the dynamic range to find the fainter bits, I could see that there was something -- frost? -- on the chip. I warmed the chip and held it near ambient temperature for a few minutes before cooling it down again in stages to -25C. Seemed to help. Some. Maybe.

I made a few five minute exposures, seeing if they were identical or if something different might happen to some of them. Not much changed from one to another, so whatever is off is always off. Good. I hate intermittent problems. I finished with a dark frame in case something could be salvaged. (The result only looks good because it is very small. Just wait till I can guide decently.)

 

Bubble Nebula with M52, Cassiopeia

20 minutes, 5-inch A-P F6 @ F4.5
4, 5-minute subframes sigma combined
10nm Schuler Hydrogen-alpha filter
SBIG ST2000XM at -25C

I called it a night on the Bubble and went inside to wait for another target to clear the pines.

When I came back out, I first aligned the guide scope and the CCD on the Pleiades and then went hunting for IC443 in Gemini. Much easier. Why that subject? Because it's faint, spectacular, I've never imaged it, and right beside it is a star bright enough for the webcam to see.

The webcam locked onto the 3rd magnitude guidestar without trouble. An overstretched 1 minute exposure in hydrogen-alpha reassured me I was in the right neighborhood. I tweaked focus (not enough, as it turned out), and cooled the chip to -35C. I adjusted the aim just a little, went inside, and shot a series of ten minute exposures in H-a, finishing with a dark frame so I could subtract the camera's electronic noise.

IC 443, Supernova Remnant in Gemini
30 minutes, 5-inch A-P F6 @ F4.5
3, 10 minute subframes sigma combined
10nm Schuler Hydrogen-Alpha filter
ST2000XM at -35C

Every blessed frame is still trailed. Why? The guidestar is right there. The guider reports relatively small and well-distributed corrections. Out by the 'scope, I can hear the motors doing their thing. I'm thinking this may have something to do with trying to guide behind a slow USB 1.1 connection. A 2.0 adapter is already on its way.

I combined the images as best I could and did some really crude things to them in Photoshop to minimize trailing (copy a layer, offset it keeping only the darker pixels; stars become rounder while diffuse objects are hardly affected). Later I did the same thing to Bubble Nebula data.

The IC443 image is shown really tiny so you can't see its faults. It's red not only because it was in H-a light, but just because the deep red looks neat and further distracts you from the picture's flaws. Just you wait.

Do differently: Focus like a fiend. Align, align, align.

 

 

You're on Page 2

Cameras behind Telescopes:
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14
15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25
26 | 27 | 28 | 29

Text & Photos by David Cortner
Link all you like; please
get in touch for other uses.