3-way Pigeon Flasher is working. The timing isn't quite what I would like, and the flash rate is a bit more voltage-sensitive than I would like, but it's good enough. Only, right now I have a bunch of components whose values I adjusted with parts in parallel. I need to get the equivalent values.
The input control circuits are working, but I have a layer of driver circuit to test before I can say the switching logic is working. Then I can put it together with the rest of the circuit and see if it all works together.
After that, it's all physical design, including the possibility of putting a lot of SMD capacitors in parallel to make up the main energy reserve for the big flasher (to reduce the voltage drop during the flash pulse). (It won't be cheap, but it will save other complications.) So far, I'm not doing well with SMD. I have some paste solder on order, and smaller tips for my soldering pencil. We'll see. These caps are about 2x2x3.5 mm, and 49.9 cents in quantity 100. I may end up using 20+ in each box, and the price break will make it cheaper to buy 100 than the 70+ that I would need.
Biggest open question is mounting the small LEDs (on the pigeon flasher). I may just put them through holes and epoxy them; they have neaar 180-degree radiation. Second biggest is mounting the big LED, which has a nice diffusion lens that makes it far more visible.
I will have another demand on my time. I have to brush up on new C++ developments.