It sounds like something is wrong with whatever type of regulator this car uses to control the alternator, OR the alternator itself is bad, but I don't know why or how the alternator could work sometimes and not others which is why it feels more like a regulator.
Older cars the regulator was often a part of the alternator itself and it decided on its own when the system needed more current to put charge into the battery or power the car's electronics. With these ones I don't know, maybe the PCM uses sensors and a shunt to more precisely control the alternator's output to increase economy. The more amperage demanded from the alternator the harder it turns, since it is converting mechanical energy from the belt drive into electrical energy. It can take upwards of 3-6 HP to spin an alternator at full load so any time full load isn't required the regulator should ease off (I think), but I've never seen my voltmeter on my dash read less than 14.3-14.7 volts while driving.
The alternator should always be putting out some current, as far as I know. The running voltage should always be 14.3-14.7 the only thing varying would be how many amps is being drawn depending on load and battery SOC (state of charge).
So thats why it seems like its either a regulator of some sort or one of the voltage/current sensors could be providing bad data confusing the regulator too I supposed. Probably start by checking all wires from the battery and alternator area, I've found chipmunk chew marks under my hood on occasion.
If you have AlfaOBD it should have a bunch of electrical info it can plot and log during a test drive.
Older cars the regulator was often a part of the alternator itself and it decided on its own when the system needed more current to put charge into the battery or power the car's electronics. With these ones I don't know, maybe the PCM uses sensors and a shunt to more precisely control the alternator's output to increase economy. The more amperage demanded from the alternator the harder it turns, since it is converting mechanical energy from the belt drive into electrical energy. It can take upwards of 3-6 HP to spin an alternator at full load so any time full load isn't required the regulator should ease off (I think), but I've never seen my voltmeter on my dash read less than 14.3-14.7 volts while driving.
The alternator should always be putting out some current, as far as I know. The running voltage should always be 14.3-14.7 the only thing varying would be how many amps is being drawn depending on load and battery SOC (state of charge).
So thats why it seems like its either a regulator of some sort or one of the voltage/current sensors could be providing bad data confusing the regulator too I supposed. Probably start by checking all wires from the battery and alternator area, I've found chipmunk chew marks under my hood on occasion.
If you have AlfaOBD it should have a bunch of electrical info it can plot and log during a test drive.