I spent 4 1/2 days (on and off, not full-time) reading this thread. All of it. Ben, you're amazing! I have lots of projects around the house that have been started, but with jobs and kids and such, it's difficult to get anything completed. Or maybe I'm just a slacker.
I'm also not going to be someone who tells you what not to do, although I have excellent qualifications in not doing things. I do have a few random ideas about this project, which probably aren't new to you.
- As for heat, someone mentioned Webasto heaters. There are a number of small, efficient heaters available, which run on diesel, gasoline, kerosene, propane, etc. In Europe, many passenger cars can be ordered with an auxiliary heater option. If you want something mobile (i.e. not plug-in-before-leaving), given the energy density in those fuels, it's probably your best bet. You might even heat coolant with a plumber's torch...
- Another option would be to create proper heat storage: Take a box, put in a nice long electric heating coil (pipe-heater), add a few windings of copper coil, and pour concrete over it. Then insulate the thing. While at home, plug in the pipe-heater to heat up the concrete. On the road, run coolant through the copper into your heater-core. They use larger systems in the UK and parts of Germany to store "cheap" electricity during the night to heat the house during the day.
- You mentioned something about diodes for the turbo mode. Make sure you use a flyback configuration and size them right, if you're still planning on doing this. At no place in your application should you put a diode in line with the load. Even at the small 0.7V forward voltage drop, pulling 400A still makes it a 280W heating element. Come to think of it, I have never seen (or looked for) a diode that could handle more than 10A. As far as control goes -- but you probably already thought of this -- I'd use a microswitch to take the throttles "kick down" position and turn it into turbo.
- I see MPaul has a nifty OS controller going on. Neat. I wonder what it would take to create an OS charger. Microprocessors are cheap (as MPaul knows), so designing and building an intelligent charger is largely a matter of software these days.
- I'm still thinking about metrics. Like I said, microprocessors are cheap. Wouldn't it be fun to integrate the pwm controller, charger and metrics all in one... the charger would know how the discharge cycle was and could adjust the charging. Possibly charge gentler if you know there's enough time, and more aggressively when you don't. Even just "top off" if you know you just need to get back home for the overnight charge.
- Got the suspension worked out? Aren't the front-springs (motor/tranny weight) stronger than the rear ones? Could be an easy replacement.
Go Ben & Electro-Metro.
-JM