Sister's 2002 Mazda Protege, nearly left her stranded. Check battery voltage, 11.7 volts, trickle charge overnight, have tested, battery is good.

Take out alternator, have it tested, it tests good.

Check voltage drop between battery positive terminal and positive post on alternator, perfect.

Put car back together and start up, alternator charging at 14.2 volts. A minute later, stops charging, reading battery voltage, battery light comes on. Disconnect battery, still only battery voltage.

WTF is going on?

It didn't die when yo udisconnected the battery? (that is odd, if so)

How long did they test it for when you had it out? Maybe the electronics on the regulator are overheating, and they didn't test it for long enough or with enough draw off it.

[This message has been edited by Stubby79 (edited 08-28-2012).]

quote Originally posted by Stubby79: How long did they test it for when you had it out? Maybe the electronics on the regulator are overheating, and they didn't test it for long enough or with enough draw off it.

That's what I was thinking. The regulator in the alternator is probably working at low temp, but once it heats up it stops. It's not uncommon, if you think about it. Heat is the killer of electronic components. Take a ICM for example. The one in our Bonneville worked fine until it got hot. Once hot it would stop and the car would stall. Once given time to cool off the car would start again. I think you have the same kind of thing going on here.

Edit: BTW, Hatchback or Sedan?

I love ours .

[This message has been edited by Khw (edited 08-28-2012).]

The weird part is, the voltage regulator isn't in the alternator, it's controlled by the PCM. I cannot find anywhere how to test the PCM output to rule out the PCM as a possibility, the factory service manual is a complete piece of crap, as is the wiring supplement. Here's the official Mazda wiring diagram for the charging circuit:



I checked all those fuses, too, all fine. I'd just replace the alternator, but it's $200, and the PCM is $300.

quote Originally posted by RWDPLZ: The weird part is, the voltage regulator isn't in the alternator, it's controlled by the PCM. I cannot find anywhere how to test the PCM output to rule out the PCM as a possibility, the factory service manual is a complete piece of crap, as is the wiring supplement. Here's the official Mazda wiring diagram for the charging circuit: I checked all those fuses, too, all fine. I'd just replace the alternator, but it's $200, and the PCM is $300.

Well, you could check the voltage at the wires that go to the PCM, first when the alternator is working properly and then again when it's not...the theory being that if they read the same before as they do after, then the problem is in the alternator. 'course, it might be all over the place as the PCM makes adjustments...

If i'm reading it right, it looks like the transistor in the altenator controls the field coil. At a guess, either it's not getting turned on (often) enough or it's overheating/faulty. The wire to it is the one I'd be most interested in checking...that looks like it should be an NPN transistor, meaning there should be positive voltage coming in from the PCM there(test for it, #2 wire/connector). It's probably pulsed on and off quickly, and might only be running on 5 volts(again, test for it). Could also be 12 volts. Don't ask me for certain...that's what testing it is for. Anyway, if I'm not totally off base on all this, then disconnecting the PCM wires and applying full voltage (12v or 5v, prefferably through a resistor cuz the transistor only needs the tiniest current to turn on) to it would make it go full-field(maximum output). That'd be one way to test if it's the altenator. Don't leave it like that for long, or the voltage will fry things.
'course use this information at your own risk, as it's coming from someone who hasn't studied electronics in 15 years.

By the way...it also means that if the PCM is already putting out a full 5 or 12(battery) volts on that wire, that means there's no point in you doing the same manually...the PCM is already trying to get the altenator to put out more than it is managing to, so the altenator (or the circuitry in it) is faulty.

[This message has been edited by Stubby79 (edited 08-29-2012).]

You could do something simple, like try a different battery. I've had batteries that were off and on with cell problems.

I see this A LOT at work where did you get the alternator from? More often than not, I have to replace them with either factory new OEM (stealership) alternators, or good ones I can find on ebay/junkyards. Reman ones just do not work in some newer cars with the computer controlled regulators. Took me 3 tries on a Toyota Solara to get it right. another was a mazda 6, went through 4 or 5 before convincing them to get a dealer part.

quote Originally posted by GreenPlatypus: I see this A LOT at work where did you get the alternator from? More often than not, I have to replace them with either factory new OEM (stealership) alternators, or good ones I can find on ebay/junkyards. Reman ones just do not work in some newer cars with the computer controlled regulators. Took me 3 tries on a Toyota Solara to get it right. another was a mazda 6, went through 4 or 5 before convincing them to get a dealer part.

I was thinking the same thing. sometimes you just have to buy a new part, not remanufactured.

First replacement alternator was a Duralast, bearings failed after a year. Second (this one) is a Remy.

Took it to a mechanic a few days ago, and they said it's definitely the alternator. Ordered a new one a few days ago, a Denso, it should be here any day. They wanted $440 to do the job, just paid the $45 for the diagnosis, and ordered the alternator for $160 + shipping.

Just to update, I installed the Denso alternator a few days later when it came, car has been working fine ever since.

excellent!