The way to make a PM-Alternator efficient is with "L-C-Tuning" plus minor "Shunt-Regulator" Regulation.

The Coils in the Alternator have Inductance, if You add Capacitance to the mix You can
get semi-reasonable "Low-Load" Voltage-Regulation.
This only means that the Voltage will not go to the Moon uncontrollably.
But it does mean that with careful "Tuning" You can keep the "Over-Voltage" in check.
Then You have to "waste" any excess Power which is in excess of what you're looking for.

There is no truly "efficient" way to Regulate a PM-Alternator.
It must be "Tuned" to the application.

Attached are 4 .RTF Text-Files which I had to put into a ZIP-package in order to attach them here.
They contain most of the research I did on regulating PM-Alternators.
.
.
.

 

OK, the TS is hoping to do better than the shunt zener diode on a British motorcycle. The bad news is that those designs are the result of drinking way too much warm beer. A shunt regulator is a poor choice in this application, as it always was. Designed by an accountant, not an EE.
The way to provide a stable output is to use a buck/boost regulator with a wide input voltage range. It has already been done and the biker is very happy with it, and so is his O.L. (that is a different part of the story, not allowed on this site.)
The original output is rectified and slightly filtered, and then feeds a commercially available DC/DC regulator that will accept about 8 to 30 volts input and deliver 12.8 volts for the lights and gages, and ignition module. It could be set a bit higher if the bike had a battery, which the one did not but the other did.

 

~30-Volts is just fine until the Headlight burns-out, or gets a loose connection, then instant smoke.

Not something that I would want to trust on a lonely Highway, or Trail-Riding at night

A battery makes a big improvement, as long as You check the Water-Level religiously.

There's just no getting around having to have some kind of continuous Load, ( Heat ).
.
.
.

 

Guys it is a permanent magnet generator. Look these up please.

Don't have to.
I know how they work.

The point of this thread is to design a switched mode shunting regulator. Before another person says this is impossible... this is already done with SCR circuits in motorcycle rectifiers. I'm trying to make a more efficient version of one.

The SCR simply shunts the excess voltage/current to ground as a shunt regulator with the excess power dissipated in the SCR and the generator windings.
Sorry, but that's the only way you can control the voltage of a permanent magnet alternator, and there is no way to do it more efficiently (whatever that means in this case).
If there were, they would do it with motorcycle regulators.

The only more efficient way is to use high input voltage buck regulator as MB2 proposed.
But note that the voltage of a permanent magnet alternator can get very large at high RPM when unloaded.

 

The power is sent in the opposite direction, to the generator. The inefficiency of the buck regulator itself and the rectifier that the generator interfaces with the regulator through is the only possible source of energy loss, according to the theoretical model. There is not a single other component in the circuit that is taking in energy. Assuming that the resistance of the wire does not come into play. myvikingjourney
MyKohlsCard.com Login

 

. Assuming that the resistance of the wire does not come into play.

It does.

 

The majority of most switching regulators, both buck and boost, store energy as magnetic flux in an inductive component. I am not aware of those sending power back to the source, except as power factor correction. And even the best of switching regulators fail to reach 100% efficiency, and none exceed that level. Worse yet, even if there were no losses at all in the regulation portion of the system, the control portion does consume power .

On the bikes that I mentioned in the earlier post, the regulated voltage fed the lights, and everything else. . On the one with the battery, it all ran off the battery, the regulator package charged the battery.
The cheating trick to avoid excess voltage was a higher voltage Zener diode, limiting the input voltage to a bit less than the regulator circuit could handle. So there iis a place for a shunt Zener diode, but not as a primary voltage regulator.
And as long as the battery is adequately charged, efficiency is not the primary goal. Another scheme to limit the high voltage power ar high RPMs could be a series inductance which would present a higher impedance as the frequency rose. I did not play with that concept at the time.