I think Paris Rhone did an alternator where the fingers of the rotor went less than half way across the coil the latter held by a thin web so fingers rotated but coil remained static and direct fed without need for brushes, the Delco Remy had rear fingers supported on front fingers again fingers rotated but the coil was static.
Motor cycles had both internal (centre) rotating permanent magnet and outer rotating magnets often with no regulation, the latter often 5 coils (6th position had points) one for each function horn, brake lights, side lights, head lights latter often used output of two coils, the centre rotating magnet output rectified to DC so it could charge a battery with a Zenor diode to sink excess into heat.
Although above was brush less, what is considered as brush less is where a static outer coil generates current in a rotating coil which is rectified to DC with rotating diodes which feeds a second rotating coil which generates an output in the second stator, worked on 4 x 400 MW in Connah's Quay and 2 x 650 MW in Sizewell, with the Connah's Quay one there was also a permanent magnet generator at rear of all that, but it was just to run the lubricating oil pump, and also a load of brushes but these were for sensors over heat etc, not actually controlling output of alternator. Found same method used down to small 68 kW generators, smaller 3.5 kW and 7 kW tended to use brushes and output was from rotator not the stator reverse to all motor vehicle alternators I have worked on, however like the AC203 I am sure there is an exception some where?
Even at 230 volt some generators did not have a voltage regulator, two field coils one in parallel with a resistor and full wave rectifier set the off load volts, and one in series again full wave rectifier although much larger and a by-pass resistor set the on load volts, they actually worked quite well, used a lot for lighting sets.
Car dynamos were the same at one time, free wheel in the drive pulley allowed dynamo to motor when ignition turned on, once engine started the engine went faster than dynamo so it started to charge, the amount of charge was controlled by moving a third brush closer or further away to the common brush and an ammeter told driver charging rate. There was no regulator as such like the ignition advance and ****** it was down to the driver.
Lucas made a blocking diode that looked like an ignition coil and these were often fitted to old cars to stop the dynamo motoring. One thing I learnt working as a auto electrician is never say it's not done, as some one some where has likely done it, 8 volt batteries to improve lights on 6 volt vehicles, on the floor start buttons which swapped from 6 volt to 12 volt for starting, had to make sure all lights off first or it would blow all the bulbs (Mac 6x6 used for snow ploughing) 12 to 24 volt start relays on mini buses so 24 volt start and 12 volt run.
Twin 80 amp alternators with a alternator to battery inverter control panel to float charge engine starting battery while pulse charging domestic battery (narrow boats) and DC to DC inverters to charge bow thruster battery at front of boat, these are getting more common, Jaguar XE has DC to DC inverter to ensure voltage is stable to use while battery voltage varies depending on if computer has decided it is to charge or not.
We are in a time of change, caravans are presenting a problem, how to charge caravan battery without allowing start current to flow from caravan to car, and also ensure power available for anti snake device, and work fridge when voltage dependent relays no longer work as car switches alternator off when under load and on again on over run, so move to fitting DC to DC inverters in the caravan. However it means car and caravan need to match each other, which can cause problems, with some German cars not supplying power to anti snake devices.