Intermittent charging/battery light

[grvaughan]

Hi, my 01 Protege has been having an intermittent charging problem for the last couple of months. Sometimes the battery light will come on, and the battery will be low (about 12.3V) and take quite a while to charge. I replaced the battery & terminals when this all began.

Other times, the light won't come on at all and the battery voltage will be more like 12.7 and nearly fully charged. It's hard to understand because today, for example, the light was on this afternoon, but this evening even with the lights (and I turned on other stuff to load it up) it didn't come on.

The alternator was replaced (rebuilt) a year ago. I took it back to Midas and it checked out fine, even the load test, but the light didn't come on when they had it! They said it might be a short.

Anyway, I was wondering if anyone had any ideas and whether it's common for a rebuilt alternator to fail within a year.

Also, I'm still not sure what the battery light is measuring. Is it getting the voltage right from the alternator BAT terminal, or what??

[TStarGermany]

if and how quickly an alternator fails highly depends on the common load it's exposed to. driving lights, stereos, a/c etc pull a lot of power, the alternator will have to work more, therefor have a shorter life; nevertheless, an alternator shouldn't be toasted within a single year.

question: what happens if you turn on your car ,what does the voltage at the battery terminal read then ? should be around 13,8v.

it's also possible the alternator has not been rebuilt correctly (parts in the picture could be toasted) or you might have something like a bad grounding etc...

[TheMAN]

a weak/bad battery will also cause alternator and starter failures

a bad or improperly installed intake temp sensor will also cause the battery charge light to come on... don't ask me why it's like this, but it has something to do with how the ECU determines how the battery should be charged by the alternator

[grvaughan]

Is this something that should show up in the codes? I bought a code reader a while back and haven't seen anything like this.

The intake temperature sensor is in the air filter housing, right? I looked at it last night but it appeared fine (not that that would necessarily mean much).

[grvaughan]

Also, the charging is rather unpredictable. It was doing great a few days ago and then last night I had to get a jump after driving (w/ a fully charged new battery) about 12 miles with the lights on.

The only thing I've noticed is it usually does better when it's cold (we're in Houston and it's 70F+ much of the time).

I finally got around to checking the field voltage (the alternator's rather hard to reach), and there doesn't seem to be any, while the battery terminal is 12V, regardless of RPM.

I checked it with the engine off but the key turned to ignition, but does that work if the computer is controlling it, or is there only voltage when the engine is running? I couldn't detect any magnetism with the engine running, and wasn't sure if it was safe to disconnect the field connector and check the voltage with the engine running (or to start it like that).

I was about to conclude I needed a new alternator after just a year, but now I don't know. Any help would be greatly appreciated.

[grvaughan]

I'm confused - is the voltage regulated by the computer or by a voltage regulator in the alternator? The alternator diagram above shows a regulator.

I recently bought a charging system monitor from Amazon (about $12 - http://tinyurl.com/38hs4h) & it starts at about 12v with a fully-charged battery and heads down from there.

Every alternator problem I've experienced has always been at the end of a hot summer, so I suspect this is the problem again, even after just 1 year & 16K miles.

[TStarGermany]

the voltage itself is regulated by the "voltage regulator" on the alternator, that's its purpose (besides avoiding a too strong outgoing current or any return current (which could occur if your alternator's voltage is below the battery's one).

if you turn the ignition key (no engine running), ~300mA will be taken from the ignition lead (+) to induce a magnetic field...

besides corrosion there's not really another explanation why a fully functional alternator would not be able to provide current to load the battery.

try googling "alternator troubleshooting" ...

my best bet: get an working alternator out of a crashed 2.0 BJ for cheap and replace your old one. it's well possible the rebuilt wasn't done properly.

[grvaughan]

Yeah, thanks. I was thinking it's probably the alternator, but didn't understand whether the field voltage was coming from somewhere outside the alternator.

If the voltage regulation is all going on inside the alternator, then it's probably the problem. I'm still just getting used to all the sensors, etc. in cars nowadays. The last time I really worked on them much, cars still had carburetors! Now my kids are starting to drive (more & older cars to keep running) and my Protege is getting older, too (6 yrs, 122K miles), so I need to learn everything all over I guess.

It's too bad the alternator's awfully hard to get to on my Protege. Changing it out would be a piece of cake on my daughter's Camry!

[TheMAN]

the regulator is the ECU.... so it is not just any conventional alternator

like I said, you should check and make sure the intake temp sensor is working properly also... sounds to me it isn't

[grvaughan]

OK, a couple of weeks ago I took a good look at the intake air temp. sensor. It was showing 1500 ohms, which was pretty close to the 1200 ohms I saw recommended elsewhere as a bypass resistance, so I figured it was OK.

I was still suspicious of that though, because the wire to the sensor was rather bent. I took a good look at that, too but it was fine. Then I decided to just get the alternator replaced, despite what Midas had said.

I was apprehensive about going to a new place if they didn't know much about electrical stuff, but when I called the new shop (Integrity Tire & Auto, Richmond TX) they immediately told me the Mazda's charging system was unusual in that the voltage was regulated both in the alternator AND the computer. That explains the confusion - both answers are right - not that I understand how it works :)

Since the problem had been getting worse, I guess it was easier for them to diagnose it. They said it was 100% the alternator (my guess is the diodes weren't all replaced on the rebuilt).

So far it's working fine, after a week. Thanks to everyone for the help.