The end remark, on the webpage
http://www.allaboutcircuits.com/vol_6/chpt_4/8.html
is

By rectifying the three-phase AC power produced, it is possible to have the alternator power its own field coil with DC voltage, eliminating the need for a battery. However, some independent source of DC voltage will still be necessary for start-up, as the field coil must be energized before any AC power can be produced.

This reminds me a project I did to start-up an alternator by its own output voltage.
Theoretically, the alternator output voltage should be zero if its field coil has no current. But I noticed that in a real situation a very small voltage (a few volts, in case of 220Vac generator) is produced due to the very weak magnetic residue. The solution was to add a relay. When the relay is off, it connects the alternator output to the field coil. This produces a positive feedback and the output voltage increases rather quickly. But when the field controller circuit is supplied by enough voltage (also from the output) to run properly, it turns on the start-up relay (to disconnect the direct connection to the output) and lets the coil driver (usually a thyristor) take the current control of the field coil.

I did this just once in the far past for a special request since I design controllers other than for motors and generators.


I wonder if this can be repeated now with some actual generators to eliminate the need of a start-up battery.

Kerim

 

240V AC type generators are designed to have a small amount of residual magnetism in the rotor. Vehicle type altenators dont hold residual magnetism. Also the vehicle type altenator regulators dont cope well with no battery conected.