Just a little micro-update. Last night I uploaded the firmware to the mega on my breadboard with the GPS TX line hooked up to the mega's RD. I also hooked an LED up to the GPS LED pin. The purpose of the test was to evaluate the USART interrupt driven receiving and NEMA parsing code. There were a couple of bugs to iron out, but in the end, it works. The LED lights as soon as the GPGSA frame indicates a 3D fix.

None of the rest of the circuit is wired up, though. I don't have a suitable VCOCXO at the moment, and even if I did, the power supply rails on my breadboard are far too noisy to try this. It'll have to wait for the boards to come back, likely sometime 8/5-8/7 or so.

But at this point, all of the major points of the design have been proven.