My preference really is for the Bergey type construction with magnets rotating outside a stator, but I prefer to do it as Hugh'e brake drum type without slots and with the windings in the air gap.
It doesn't cog, has very small iron loss, Magnets are held on with centrifugal force. Has good cooling direct to the laminated core and alows the use of a fully wound 3 phase winding.
It uses more magnet than the slotted Bergey type core but less than an axial.
I don't say much about it here because you can't build it without machining facilities and you have to rescue the core from scrap motors to be cost effective.
There is absolutely no doubt that the Bergey style of construction is the most cost effective and the only way to compete on a commercial basis( or its equivalent as the commercial version of an induction conversion).
There is no one form of construction that will produce a machine with the properties to cover all wind conditions without external help.
Bergey have realised this and what they have adopted is what I have been doing for the last 10 years. They design the basic alternator to match the winds from about 15 mph upwards with good efficiency and use a boost converter to maintain output in the lower winds below cut in of the main winding.
That is how they manage to get 1600W from an 8ft machine and still have energy capture below 8 mph.
If you use this technique then much of the stator heat problem no longer arises as the efficiency is still high in the high output region.
For those without machining ability are largely forced to adopt the axial route, even motor conversions need some machining th make the rotor.
If you must use slotted iron cores then for home construction you are virtually forced to use a motor conversion.
Done properly it is entirely satisfactory but keeping magnets in place is more difficult than with the rotating can.
The problem for me is that you can't calculate the final result so you are forced to mess about with reconnecting standard windings whereas I would prefer to start from scratch and get what I want first time.
Zubbly is doing a wonderful job helping those that want to go that way and if you are not prepared to use some load matchiong scheme and you live in a high wind area then you will end up with a machine that performs well and is virtually indestructible, but if like me you live in a very poor wind area it may not be the best option.
Even with motor conversions you can not cover a wide range of wind speed at optimum efficiency without some means of matching and adopting the Bergey idea would improve performance.
Even the Bergey machine I think does suffer from iron loss and the low end performance is not as good as it could be, but as a commercial venture where you are up against cost as well as needing high maximum output to sell it, it is a good compromise.
What is best for you depends on your wind area, facilities and so many other things that it is impossible to come up with a single solution, but I don't think that for a 1 off it can be cost effective to try to get special cores or waste time making them unless you have a roll of low loss core material and a laser cutter.
The grade of core material does seriously affect the low end efficiency of a slotted core design and so does flux density and the size and dimensions of the slots. You are not likely to hit the best compromise. Even with motor cores the slots are not likely to be ideal for the best low speed results, you just have to use what you have.
Flux
