Hi Zap
Your first method is correct. Using different batteries will only change the voltage if it is below the set point of the charge controller and that will have a minor effect on the measured power. Above the point where the charge controller is operating any battery will give the same result.
Method 2 will never give you an accurate result, there are factors going on with the rectifier that you can't allow for, but usually the answer is a lot nearer than you have got.
What do you call stator resistance? Assuming it is star connected you should be using line resistance ( between a pair of leads), this will be twice the phase resistance.
You don't say what type of alternator it is, if it is an air gap type you should get answers a lot closer than you have. If it has a slotted iron core you will have leakage reactance and you may not be able to make much sense of the results, but if it is showing reactance limiting at such low output you are in trouble.
Another factor may be the accuracy of the resistance measurement if it is done with a multimeter, even at 2 ohms you may be well in error, check it by the volt drop method unless you have a Kelvin bridge.
Try some results at higher speed and power, you are close to cut in and rectifier drop is significant so that may be affecting the results.
If it is an air gap alternator, check that you are using line resistance, check the resistance value and check at a higher speed and come back with your figures and perhaps I can help some more. If it is slotted iron cored I may not be able to make much sense of the results.
My air gap alternators behave as though the resistance is higher than the measured value by a factor of typically 30% but when you take this into account the second method gives results good enough to predict output for design.
For true power measurements method one is the right one.
Interesting stuff.
Flux