The biggest nuisance with the scheme I have described is that for good high wind efficiency you need to keep line losses to an absolute minimum. Ok for a short run but for a long wire run it does become an issue.
Alternator efficiency problems are the same whatever voltage you choose but the line loss is worse with low volts and buck schemes will come out better if you cant keep the line loss negligible.
For conventional machines we need line loss( it's better there than in the alternator) but once we go for high efficiency schemes loss is not needed to match the prop and the limiting factor will be how much expense you are prepared to throw at it to gain a few %.
With a 24v system with the same size of alternator as normally resistive matched, If you can avoid significant line loss and match the alternator winding at 25 mph, you should see over 70% and that is with no increase in cost. If you want much better then alternator cost goes up in big leaps for every few % you gain.
With a well designed system with boost converter at 24v you can achieve 70% from cut in to furling with the same alternator cost as the resistive match.
At 12v things are not so good, with the scheme as I have shown it, you will be boosting from about 6v and the diode losses will have a significantly greater effect. If you include the rectifiers in the boost circuit you may do a little better but then you no longer have a scheme that can work without the converter if it should fail.
Once we decide this is the way to go and stop worrying about failure then going for a buck scheme will give better efficiency on low voltages and also reduce the effect of line loss.
Starting from scratch I would not choose a buck-boost scheme, but if you have an alternator already and it was too slow in low winds and too fast in high winds then as long as it was efficient enough then a buck-boost scheme would be perfectly satisfactory.
The actual converter circuitry like everything else will be a compromise between simplicity, reliability, the bits you can get and and your ability to manage it.
Refinements such as synchronus rectification are logical moves but for our voltages they are not a big issue. A 3.3v computer supply is a different matter.
It is not realistic price wise to attempt to build alternators with upwards of 90% efficiency and if you keep conventional rectifiers I don't even think you can do it.
It's back to the greedy thing, if it's possible you want it.
Moving from an efficiency of 40% to one of 70% will make a dramatic reduction in stator loss and it seems to me to be about the most cost effective region to aim for.
If you go for motor conversions you may get a higher top end efficiency at a cost effective price but you will have to trade a bit of low wind performance as you can't recover the iron loss.
The snag with all these things is that there is no one case that will meet every individual need and even if I gave specific circuits people in various parts of the world may not be able to get the specified bits. It's not quite like magnets and wire.
All I can really hope to do is give general ideas of the various schemes . I may produce a reasonably documented boost circuit based on the ideas I have given, but even then anyone not involved in electronics may not get it to work. Ok if it works first time, but fault finding without experience and test equipment makes it a doubtful starter.
Probably general advice to steer those capable in the right direction is much more useful. .
Flux