We can go down the exotic rabbit hole if we want, but we gotta go all the way...
Well, kinda. There's one secret sauce you're missing if you go this route: a permanent-magnet generator head. The hardcore wind turbine people and a few others are using PMGs, but I haven't seen any on any non-hybrid vehicle.
I agree it's rare, but
these guys have been doing it for the better part of 30 years. If you have the cash, and are starting from scratch on a complete system, you might want to check these guys out. If you're improvising in a pinch and don't have any wrecked Priuses handy (along with a PCB fab, electronic parts supply house, and a ton of electronics experience), you can stick to the simple Rotor/Stator alternator approach.
The difference in efficiency is enormous - a standard claw-pole generator is 40-60% efficient, a PMG is 90+ % efficient, thanks to not needing energy to generate a field. The hybrid cars also use their PMGs as drive (and starter) motors, so get the efficiency gain on both sides of the equation.
Permanent magnet alternators are more efficient at the point of generation, but automotive alternators have rectification built-in while PMG's used in Hybrid cars don't - you lose some of that efficiency again with the external controls that are required to make use of the power. This is another area where gains are possible for both Brushless-rotor and PMG type of generator heads. Automotive alternators use 6-diodes to form a three-phase rectifier, this changes the three-phase alternator voltage into rippling DC voltage. PMG heads used in hybrid vehicles are driven by variable voltage and frequency drives and synchronous rectification (active, not passive rectification).
As an exercise in the difference between Passive and Active rectification:
Let's assume a hypothetical 100Amp 28.8Volt alternator. I'm not going to go into the complicated AC math and work in individual component variations and margins - let's keep this only mostly confusing, not entirely confusing
.
Passive (your "off-the-shelf" alternator): A diode has a voltage drop of between 0.3volts and 0.7volts (depends on the chemical make-up). In order for the alternator to pass 100Amps, that diode is "consuming" 100Amps * 0.3Volts = 30Watts. Two diodes are required to make a complete circuit, so you are actually losing ~60Watts per phase on that alternator (180Watts total) just for the service of changing AC to DC power - and this is if you have the most expensive and exotic diodes. If we assume this was a 28.8Volt alternator, that's a 6.25% loss right at the back of the alternator.
Active: Changing to synchronous rectification, MOSFETS or IGBT are used in the place of diodes and driven by very low power circuitry, and this can lead to a significant reduction in power consumption. MOSFETs for example instead of having a given voltage drop, have a given resistance. If we replaced that 100Amp diode that has a 0.3Volt drop with a
1kW MOSFET that has a 0.0023Ohm R[SUB]dsON[/SUB], then put 4 in parallel to get the power rating (28.8V * 100Amps = 2.88kW, okay so the engineer in me added some margin, in parallel the math is: 1/(1/resistance)+...
n...+(1/resistance)), the total resistance across the MOSFET array when it is "ON" is only 0.00575Ohms. The voltage that would be "consumed" is only 100Amps * 0.000575Ohms = 0.0575Volts, and that works out to 100Amps * 0.0575Volts = 5.75Watts (11.5Watts per phase, 34.5Watts total) which works out to only a 1.2% loss due to rectification. The control circuitry is less than 1Watt total so I'm omitting it from this for simplicity (only a ~4% error).
That's a 145Watt difference, a 5% difference in output efficiency (take that 70% efficient alternator and now it's 75% efficient, 80% goes to 85% - again, an over-simplification). This was also just assuming we addressed the Stator coil (output), there is more to be gained in the Rotor too. If you use a self-exciting alternator (one-wire), you get the power to control the rotor's variable current from the small PMG in the alternator, so you get a little bit more total efficiency - it is possible to sneak it up to nearly 90% and not have to build your own voltage regulator.
I'm not suggesting that everyone go out and change their stuff (it's ain't broke, so don't kill yourself trying to fix it), but for people who are not electro-phobes and have the time and patience to learn, there is a ton of efficiency that can be gained with
off-the shelf active rectification stuff that's becoming available. If you want to go really crazy, you can tear down an alternator and replace the rotor and stator with a higher cobalt content magnet steel, and you can liquid cool the stator and rotor to avoid flirting with the Currie Temperature. In general: cold makes better magnets and better conductors (coils), hot makes worse magnets and makes conductors (coils) more resistive (harder to pass power through).
