Time to start a new project

I've come to the conclusion recently that I want a hobby project that is a little bit more physical and direct then the tools projects that I've been working on to date. And at the same time I want to combine some interests of mine: Electronics, software engineering, photography and astronomy. So I'm going to design, build and use a digital astronomy camera.

And not just any camera. Here are some design goals:

• Reasonable pixel count for the sensor (say, 4MPixels),
• Lightweight sensor module,
• Wireless and/or wired network operation,
• Focus mode,
• Support for auto-guiding, and
  
• Open source all the design files, firmware and software.
  

This is a non-trivial set of goals. I expect that the project will run me on the order of $3000 in parts and countless hours of labor. That's in line with a similar device purchased off the shelf (except for the "labor" part) so not too wacky a plan. The project may at times suffer from "Creeping Featurism", but so long as it remains open source and the price doesn't escalate too far, so be it. No managers are telling me it has to be done now, or ever. It's for fun, so I'll have fun features, gosh darn it! But I have some justification for the major features.

Reasonable Pixel Count

Well, this goes without saying, doesn't it? But I'm specifically thinking of around 4MPixels. Smaller cameras are already available for embarrassingly cheap prices, especially if you get down to the "VGA" range. People have even used low-end webcams at this level. I would like my device to be worth building, so I really think it needs a few more pixels then that. But too many pixels and the device becomes crazy expensive. The sensors for large pixel counts get pricey pretty fast. Since the design is open source, one can alter the design and substitute a larger sensor if one has the financial means. The 4MPixel sensors seem to around that sweet spot where it is worth building the thing, but it doesn't break the budget.

Lightweight Sensor Module

The sensor module attaches to a telescope in the place of an eyepiece (or in addition to the eyepiece for eyepiece projection) so it is way out there beyond the center of mass. I know that my Nikon-80 on the end of a 1 meter long telescope is enough to throw the telescope balance way out of whack. A light sensor module (much lighter then a DSLR camera body) will be easier to mount.

Wireless and/or wired network access

The ideal is to be able to access the camera via a wireless network. Not too crazy, people can access their printers that way these days. Wires on a moving telescope are a particular irritation because of potential binding. (Also, I have another motive here. I have access to some larger telescopes, where a wire to the camera could get quite long. Think 20" refractor.)

Focus mode

This is a mode where the camera makes a live full resolution display of a small region of interest. One can use this display to adjust the telescope focus on a nearby star. Getting accurate focus is a persistent problem with astrophotography, and is downright exasperating with DSLR cameras.

Support for auto-guiding

This is where the camera not only collects the light to form the image, but also provides a sense of the motion of the image. This allows computer controlled mounts to be adjusted by software to keep the target centered. Getting this feature implies some sort of microprocessor that can do image processing, and also obviously implies that the sensor exposure be collected as a stack of images.

Open source all the design files, firmware and software

Mechanical drawings, board schematics and layouts, FPGA designs, firmware and host software will use open source tools wherever reasonable version exist, and the designs will be released under GPL. Everything will be published, so that you can make a camera of your own.

Onward!

So that's my plan in general. I'm getting used to this blogging business, but I intend to try and write up my thinking and progress as a sort of journal for this project. Hopefully others will find this of interest.