Photographing an asteroid - failure and success

What's this mess?

I took this disaster of a photo while looking (in vain) for a passing asteroid.

A few days ago, it was a clear night in Melbourne. I took the opportunity to go out into the back yard and do some photography - as you do. What I was after was the Statue of Liberty nebula. (I eventually posted it alongside a more concrete Statue of Liberty.)

However, I'd heard about a close pass of an asteroid that was coming up, and thought I might be able to get a photo. The asteroid's name was (52768) 1998 OR2. Clearly a family name.

From what I can gather, it's an egg shape, roughly 2km by 5km. That's big, and if it collided with the Earth it'd cause a heap of problems, but it wasn't coming closer than 16 times the distance of the Moon.

It's also not a giant: the Chicxulub asteroid was up to 80km across. But it's large enough to be spotted with an 8-inch telescope. Mine is only a little more than half that, but I thought I'd have a stab in the dark, as it were.

It was harder than I thought.

First, I didn't really have a good idea as to where (52768) 1998 OR2 was. I'd read in a blog that it was going to be passing HIP 50745, which is an unnamed star. I didn't know where HIP 50745 was, how close the asteroid was going to be when it passed the star, or indeed when it was passing.

Second, I hadn't taken any unfiltered photos from Melbourne for ages, and I'd forgotten how it's done. I've avoided taking colour photos in Melbourne due to horrible light pollution. I use narrowband filters, which are very effective in blocking it out. 

Third, I didn't know whether I'd be able even to see the asteroid in the photos I took. I planned to take a half dozen or so exposures, stack them on top of each other and flip between them like a film. With luck, the asteroid would be obvious, moving from one frame to another.

As you can see, it was a fiasco. I had the wrong idea of where the asteroid was, and set my rig to find HIP 50745. It did that flawlessly, but the asteroid wasn't there.

Blissfully ignorant that I was ignorant, I took eight two-minute exposures, as I would at a dark sky site. The resulting images were washed out by light pollution which destroyed the contrast between light and dark.

When I reviewed the images, I couldn't see anything moving. I didn't know that this was because the asteroid wasn't there, or that it was lost in the light pollution.

At that point I gave up.

...and the success?

So I turned to the always-helpful astrophotographic community.

Christian Gow provided this. Have a look, it's brilliant. It compresses about 45 minutes into nine seconds, with each frame being a five second exposure.

Christian's scope is a Cassegrain, and coupled with an ASI294MC-P camera, which he then cropped, so his field of view is much narrower than mine. He also uses a separate guide scope, which means he could actually find and see the asteroid, and therefore track it. (I use an off axis guider, so I can't directly see my target, just the edge of the frame.)

Christian's exposure time was way less than mine. At the same time, he increased the camera's gain, and decreased the temperature to -23° to try to compensate for the increased noise. The result was great.

Each frame has a much darker background while retaining the stars, while the asteroid itself is close to a dot rather than a trail.

The time lapse starts with the stars still and the asteroid moving. But about half way through, it switches to track the asteroid. This magic can really only be done with autoguiding and a separate guide scope. Because Christian was able to see the asteroid directly, he could track it using the PHD software.

This is what success looks like. Thanks Christian!