This is promising. Obviously, it's easier to use this method to detect a known exoplanet (one that is big, with frequent transits, across a bright star) rather than detect a new exoplanet (even a big one with frequent transits across a bright star) ... but it's promising all the same.

Especially since it's a proof of concept, and I suspect small hardware upgrades would make it more effective (thus, more able to detect new exoplanets, in the right conditions).

I found these instructions on how to make a barn-door tracker a little better. http://www.garyseronik.com/?q=node/52

I suppose the micro-controller is nice, but if you only need one speed why bother.

Now I just need to get a DSLR instead of my Canon AE-1 Program.

You can just buy a Pentax DSLR which has a internally-moving sensor, and then buy the Pentax GPS unit that acts as an astro-tracker. Then the sensor moves with the stars according to GPS coordinates.

Also, if you're looking just for better image quality, a DSLR isn't anything special. It's just a camera with a mirror. You can get any APS-C or larger-frame sensor nowadays that aren't DSLRs.

You mean something like the mirror less offerings that Samsung and Sony have?

The precision of motion of a stepper motor is several magnitudes higher than a simple DC motor. For a project that needs weeks to execute I wouldn't save the few bucks for a micro.

> precision of motion of a stepper motor is several magnitudes higher than a simple DC motor

Depends on whether we're talking about open-loop control or closed-loop control. A stepper motor can only move in half-jumps or full-jumps. The size of the jump is intrinsic to the motor, and can be reduced with a gearbox. If we're doing closed-loop control of a DC motor, the resolution is limited by the feedback encoder, which typically has a resolution a magnitude or two higher than a stepper motor's.

I did the same thing last week. We used a rotating telescope with a dslr on the end. We also had a little gizmo that tracked arbitrary moving objects in the night sky and told the telescope which direction to move so as to keep the moving object in the same place.

Using 5-10 minute exposures, we were able to take the same kind of colorful astronomical pictures you would expect to see in NatGeo. It was truly amazing. You really don't pick up on the beautiful colours of the stars with the human eye.

Any details about the setup? I live in a suburb of a mid-sized city, but I can be in pretty dark skies in an hour or so. This seems like something that would be really neat to get into.

any publicly available pics? curious to see what a home setup can achieve.

Not sure you need tracking for this: if you let the camera still, you should obtain a stripe that will vary in luminosity as the planet passes, which basically integrates the sampling point for you. (This of course is limited to short period planets, and stars with sufficient luminosity to actually produce a stripe)...