Mazda pickup - charging system problem?

The vehicle: 1997 Mazda B2300 pickup, 161,000 miles. Original alternator, second battery (current one is about 3 years/60,000 miles old). Never had any reason to suspect electrical system trouble until tonight, when the following symptoms suddenly appeared:

When the engine revs go up, the voltage registered on the dash guage drops, and the headlights & panel lights dim. Rev it high enough and the battery warning light on the dash illuminates. Reduce the revs and the symptoms go away.

At idle, or even at 3,000 RPM cruising down the freeway, all lights appear normal brightness, and the voltage guage reads about where it normally sits. Turning on the cabin air fan causes a noticeable drop on the guage, but that by itself causes no other symptoms.

I drove about 20 miles home while nervously monitoring this condition tonight, and it didn’t seem to get any worse. After stopping at my parking space, I let it idle for a few minutes - no problems. I revved the engine while in neutral, and by something like 3,500 or 4,000 RPM the indicated voltage had dropped noticeably, and the battery warning light had illuminated.

Most things I read detailing alternator issues (my first thought) talk about low voltage symptoms at idle, which resolve themselves at higher revs. My symptoms are pretty much the opposite, it seems. I can’t say I’d be surprised to find out that it’s still the alternator, though.

Any ideas? I’ll be going downstairs later tonight to poke around, probably armed with my Haynes manual and ancient analog voltmeter.

Sounds like the regulator inside the alternator is starting to go south.

The only other thing I could think of is that the bearings in the alternator or something else on the same belt are going, so that at higher revs the belt slips. Or maybe it’s the belt itself. You didn’t mention any belt squealing sounds, though.

Thanks - that seems as reasonable a hypothesis as any. I’m pretty sure that the voltage regulator and alternator on this vehicle must be replaced as a unit, but I’ll look into it. Haynes also specifies some more detailed checks on the alternator that I’ll try, armed with my voltmeter.

I didn’t hear any belt noises, even with the hood open. The belt is about 3 years old, and replacing it is on my to-do list, but at the moment I’m inclined to doubt that it’s the problem. There’s an automatic spring-loaded tensioner for the belt (serpentine, drives all the accessories), and it’s crude position indicator is still within the “OK” range. I make a point of looking at it about once a week, and nothing has changed recently.