Stall is when the angle of attack becomes too great for the type of aerofoil and in most cases occurs with an angle of attack of much over 12 deg.
At start up the wind is nearly perpendicular to the blade, the angle of attack is ridiculously high and it is hard stalled. As the blades speed up the apparent wind rotates as a vector to the real wind and comes into a reasonable angle of attack to the aerofoil. If you have things right you will be up to this speed before cut in and you will not be remotely stalled at cut in. As the emf exceeds the battery volts the alternator load comes on very rapidly and the prop will slow down relative to its intended tsr so although you start with tsr well above design at cut in you will reach the design figure rapidly as the wind picks up so if you cut in at 7mph then you will likely be at design tsr at 10 mph and you will be running near peak efficiency.
With direct battery loading things go very wrong in higher winds as the alternator is trying to maintain constant speed with its terminal voltage clamped at battery voltage and the prop will stall at quite modest wind speeds. If the alternator is less efficient then the prop will hold its speed up better than if you have a very efficient alternator so in this case the machine with small magnets will develop more power from the prop in higher winds, The electrical output will be higher for the same conditions even with lower efficiency as the gain from the prop is very considerable.
If you have to deal with a long line and high resistance cables then you will not get a high overall electrical efficiency so the larger alternator will then run as the smaller one does with a short cable run. If you can't avoid a long cable run then you can get better results with the bigger magnets and more efficient alternator.
For a short cable run then compared side by side the big magnet alternator will perform considerably worse and the maximum power out in reasonably high winds below furling will be well down. The issue can easily be solved by adding the equivalent of the extra cable resistance of a long line in as a series resistance. The performance of the big magnet machine will then be similar to the small magnet one ( and it cost a lot more)
Now the important bit, in very high wind areas the small magnet machine with thinner wire will be seriously limited by stator heating so you must make every effort to get it furling at a safe power output ( below 700W continuous). The big magnet machine will produce the same electrical loss overall but only part will be in the stator, the part in the resistance of the line will not heat the stator so you have much more chance to keep it furling safely in high winds and it will be much more durable. That is why it is a better choice for high wind areas.
With direct battery connection there is no way you can avoid some potential power loss in higher winds if you choose a very low cut in speed and for the same cut in speed the big magnet machine will some how have to have its high wind rotational speed increased to keep the prop from stalling or the performance will be way down on the small magnet version. If you have a long cable run then it will sort out but if you don't have a long cable run then you need to do something to raise the speed.
If you aren't worried about performance then you can of course run the big magnet machine stalled and it will then become fairly self protecting even if the furling is miles out and with plenty of wind in a high wind area this could be the safest option but you pay a heavy penalty in terms of total energy capture.
I don't agree that the stalled prop will try to furl earlier but this is a very complex area indeed and probably beyond simple analysis. In simple terms the thrust depends on the power extracted so the thing furls easiest at optimum tsr and will take more wind to develop that thrust if stalled. This is very much complicated by the seeking force, which is less when stalled and higher at best tsr and may be higher still during run away so what finally happens is in the lap of the gods.
You finally have to decide whether to use the cheap small magnet version, match the performance of the big magnet version with line resistance or have a near bomb proof but poor performing machine for high wind areas. Whatever you do stall in the higher wind region will limit the performance to a fraction of a properly matched scheme.
Not sure this will help Bruce as it may be a bit too detailed but this stall issue is complicated from the aerodynamic point of view and far worse when you include the alternator effect.
Flux