Hi Tom,
You say it's turning quickly, and the fly-balls are out. Gotta agree! I also see whirling blades in that picture - but by the looks of it fairly bright day for a picture. Your vantage point is rather far away, too. In this condition I would expect a digital camera shutter to be opening at about 1/500 or 1/250. Blades of my WT are usually sharp, unless I work the shutter to show a whirl, or it's pretty dark. So why are the blades so blurred in your pic? High RPM, of course.
As to why it would turn so fast, yet produce so little, I suspect the answer is that there is too much resistance in the entire system. If you've tried both batteries and a heating element, then it could be the total length of wire between generator and load. Your wire run is a combination of sizes and materials, so maybe you should calculate the resistance in a single line. Better yet, measure it with a sensitive ohm meter (trouble is you have to lower the tower to do that). Then multiply the resistance by the square of 25 amps as you show on the gauge, (625) and multiply all that by 1.73. (1.73 * R * I^2 ) That's the power being shed as heat through the wires.
Next thing to check is the continuity of both phases. Put the multimeter across two of the three phases (with the turbine running), each in turn, and see if the readings are very similar. You have to make some account for changing wind speed as you do this. In heating element loads I don't know how much this fluctuates with speed, but with battery charge it goes up and down a bit. Any OBVIOUS differences should alert you to an under-performing phase or bad wire. While testing this check for voltage from each phase to ground (or a tower leg or something) to find any "leakage" that may hint of chafed insulation.
You have a bunch of trouble-shooting to do, no matter what. I'm sure you can sort it all out! We're here to help if we can, too.