legacycentral bbs

Okay, the aforementioned situation is readily apparent now.

I took it out on a very dark road on Saturday night, just for a drive. The whole time I was out there, probably about an hour, it would fluctuate from very bright dash lights and headlights, to normal. There was no set pattern, and it didn't seem to affect the way the car runs at all. The only thing is, that whenever the dash/headlights are bright, if I declutch, the idle drops very quickly to 500rpm at best, and may try to stall, or even WILL stall. If the dash and headlights are at normal brightness, it idles down nice and slow and never goes below 750rpm.

So we're 100% sure now it's an electrical problem, but what? I'm probably going to do those other 3 ground points like you mentioned vrg3, but what should I try next if that doesn't do it? Remember, my battery and alternator are fairly new.

Hmm... I would say to start monitoring the voltage to see if you can find out any more. When my alternator died and I had to pick up a new one I also picked up a SunPro voltmeter at the same time for like $16. I wired it to my cigarette lighter and ziptied it to the cupholder just to monitor things temporarily; maybe you could do the same.

I'm curious if you're somehow getting excessively high voltage.

You should also check the positive battery cable.

Is there any corrosion or anything around or in the underhood fuse box?

Well, I wanted to figure out what the problem was this weekend, and I did!!

My battery is dead!

I was heading to a Orlando meet that the USMB organized (all 3 of us ) of all places, and when we decided to move our cars, mine wouldn't start I had noticed it getting more and more difficult to start lately, but that was finally where it happened. There were no jumper cables around, so I push-started it (which I feel bad about, since I jolted it pretty hard to start it) and told the guys I'd be back in an hour with my XT instead. I was so fortunate that my good friend was in the immediate vicinity with jumper cables and was good enough to follow me home. I owe him one!

I at first thought the alternator, but push-starting would have done no good if my alternator had conked out completely, so it has to be the battery.

I dug up the receipt, I bought it 11/25/03, so it will (it better!!) be replaced under warranty.

What do you all think about the undue stress this has put on my alternator? Again, it's only about 2 years old, so it should still be good.

I'd bet it's fine, but most replacement units suck sweaty ass, so you might want to get it checked out when you get the new battery as a precaution.

Steve

Good idea. I'll do that.

I'm glad my battery was free...

Coz that didn't solve my fucking problem.

I still have the brightening lights and the attempting to stall whenever they're bright like that.

I'm fucking pissed as hell right now. I'm sick of this shit.

I'm pulling my alternator tomorrow to test it. If it comes out fine, I dunno what I'm gonna do. Selling it would be too nice, driving it off a cliff would be cooler.

Perfect.

The alernator is perfect.

Just got done testing it, it works fine.

Now what?

I have a battery that isn't 2 days old, a 2-year-old alternator that has been tested and confirmed operational, and I still have this spiking thing, dash and headlights get bright and during this time, the engine will try to stall when the clutch is disengaged.

A side note: My head unit has been acting strange lately. The display dims significantly when the huge bass hits from my 3 working stock speakers and it has been shutting off and coming right back on rather frequently. I figured the wires at the back were loose, but could this have an effect on the whole car's electrical system??

Those who think they can help throw some ideas out, I'm off to get a voltmeter.

The dimming radio sounds like resistance in the power supply. This could be either on the ground or +12v side.

Have I already asked if/when you changed the battery cables?

No, you haven't, and no, I haven't

I checked and cleaned up the battery terminals when I put the new battery in, took a metal brush and scrubbed some of the corrosion off (wasn't much) and put some dielectric grease on them and then put them on the battery.

Rigged up the voltage gauge...

Very interesting behavior...

It sat at 12 volts once I hooked it up, and when I started it, it sat just a bit above 14. It sat there steady for a while, and then I turned the lights on. It immediately went down to 12. I turned them back off, and it went to about 13. It moved around a bit, from just above 12 to close to 14. Then the gauge jumped back to a little above 14 where it was before and the idle dropped to about 500rpm. I turned the lights back on and it went down to 12 again.

Now granted, most OEM voltage gauges in cars are probably designed to stay at one reading when things are normal, but I still don't like how much it jumped around.

Thoughts? Comments?

I'm gonna drive it to work and see what happens.

One more question, will I drain the battery if I leave the gauge hooked up like that at 12 volts?

You might need to replace the battery cables. They could be fractured or corroded on the inside.

Hmm, I put a cheap gauge in my car when I installed the new alternator, and it stays pretty steady at 14 volts almost all the time. Only under transient conditions does it change.

Just above 12 volts seems lower than it should be.

Steve likes to talk about how some reman alternators suck... out of curiosity, who remanufactured yours?

Depending on the quality of the voltmeter, it might or might not drain the battery to leave it connected. I'd hook it to an ignition-switched line if I were you. Accessory (like the cigarette lighter) would work too but then you can't watch the voltage while cranking.

Drive around for a while and see if you can correlate any specific voltage behavior to the light-brightening behavior.

First off, I wired the voltmeter directly to the battery, since that was the easiest and cheapest option at the time. I had a feeling leaving the gauge directly hooked to it for 8 hours reading at 12 volts would drain it, so I disconnected it and reconnected it when I came home...

And this is what happened on the way home:

The lights stayed bright all the way to about a mile from home, and it pretty much stayed at 14 volts while they were bright. Then they started dimming and the voltage gauge would immediately drop to 12, sometimes a little bit less. As I turned into my neighborhood, the lights were bright and the gauge was sitting at 14, then I declutched and the lights got dim again, the gauge went to 12, and the car stalled. I cursed very loudly, restarted it, and parked in my driveway and just watched it idle.... When it was dim, it would sit at 11.5 volts, maybe even 11 at times, and idle pretty at 800 rpm. When it was bright, the gauge went to 14 volts and the idle would drop to about 500 rpm.

This really really sounds like the alternator to me, like the voltage regulator is sometimes not charging the battery and thusly taking a load off the alternator, explaining the higher idle. Even though the people tested it, even though it's relatively new, I really don't see what else it could be.

Anyone agree?

Yeah, I think I agree that it seems like the voltage regulator's misbehaving.

Do you think you might be able to connect the voltmeter between the alternator's sense wire and the alternator's case? Cuz that's the voltage the regulator is actually regulating. The sense wire is the white wire on the 4-pin harness connector (that only actually uses 3 pins). That might help make sure the regulator is malfunctioning.

How do you mean, like making a small 3-wire harness to connect the 3-pin plug from the car's harness to their respective pins on the alternator, and then splice the voltmeter into the wire I connected to the white wire on the harness?

I don't feel comfortable butchering the car's original harness, so please don't ask me to cut anything

Hehe, yeah, I certainly didn't mean you should cut your stock harness.

You could make a small 3-wire harness using quick disconnects, but I was thinking more along the lines of wedging a small wire into the connector and then connecting it, and then a fork terminal over an alternator mounting bolt or something.

I think I'll just make my own harness with the quick disconnects, yeah.

Now, the only thing I need to know is the polarity of these 3 wires, so I know which way to rig up the voltage gauge. I'm not sure, but I'm guessing that the ring connectors that I bolt onto the alternator is positive, and these 3 wires are negative. So if I want to do this correctly, I will want to take the wire coming out of the alternator that corresponds to the white wire on the harness and run it to the +12v on the voltage gauge, and then run the ground wire from the gauge into the harness where the white wire is, and then hook up the other two wires straight to their respective connectors.

Is that correct, or do I have the polarity backwards? I'm going to rig it up right now, but I won't actually connect anything or start it until I'm sure, so the faster I get a response, the better

Thanks a bazillion Vikash for the help you've given already btw

No problema.

Let me clarify. There are five wires going to your alternator:

Two fat white wires with ring terminals. These go onto the post. These carry the positive side of the charging circuit.

Three small wires going to the harness connector. The harness connector has space for 4 pins but only 3 are used. The pins are about the same size and shape as 1/4" quick disconnects. These three wires are:

A thin white wire that tells the regulator what voltage the alternator is putting out. The regulator uses this signal to determine whether to power the alternator's field coils or not. This is called the sense wire.

A thin yellow wire that carries ignition-switched power. This turns the regulator on only when the ignition switch is turned on. This is called the ignition wire.

A thin black-with-white-stripe wire that connects to the charge warning light on the dash. This allows the regulator to turn on the warning light when it fails to charge. This is called the light wire.

The alternator and the voltage regulator both ground through the body of the alternator. There are no ground wires going to the alternator.

So -- you want to connect your voltmeter's negative lead to the alternator chassis, and you want to connect its positive lead to the sense wire. Do not touch the fat wires with ring terminals.

You will then be measuring the voltage that the regulator is trying to regulate. Does that make sense?

Okay, yeah it makes sense.

That's pretty similar to how I rigged it, just that I was thinking I would put the gauge into the circuit that goes from the sense wire to the respective alternator pin, but I was wrong. So I'll want to run the positive lead from the gauge into the sense wire on the harness and there is a small bolt on the alternator body at the back of it, can I take that out just to put the gauge ground on it and put it back in?

Also, I tried to insulate my 3 homemade wires from each other on their connections to the harness with electrical tape, my quick disconnectors for my ignition and charge light wires could otherwise easily come in contact, and actually, the sense wire could come in contact with both of them. I did a rather sloppy job of isolating them to themselves with electrical tape, should I recheck them perhaps and make 100% sure they're not touching (i.e. would something REALLY bad happen if they touched) or do I not really have to worry about it??

And one more thing, when I test it like this, if I get similar readings to what I got before, that does truly confirm that my voltage regulator is bad and I need a new alternator right?

Thanks Vikash, you rock!

Well, it probably won't be the end of the world if there's a short between those wires... if the light wire touches one of the other wires it could fry the regulator but that would probably be it. It's certainly something to avoid.

Also, you want to be sure you're making solid connections because otherwise the quality of the connections could ruin the results of this little test.

So, yeah, insulate the terminals carefully.

That small bolt on the back of the body should be okay for the ground. It's possible that removing it would make something inside the alternator fall out but I doubt it. If you can get a fork-shaped terminal instead of a ring-shaped one you can slip the terminal under the bolt without removing it completely.

If you get similar readings to what you had before, I would suspect the voltage regulator, which is built into the alternator. I can't be absolutely sure, though...

Good luck. You're welcome. =)

Vikash, I hope you see this before I get home from work tonight and try this, but I was thinking about what you told me to do, and it seems to me like you're making me bypass the regulator. There will be no wire attached to the sense wire's corresponding pin on the alternator, so you're taking the current that would be going to it, sending it to the voltage gauge instead, and then immediately grounding it. I thought it was the regulator we were trying to diagnose

Am I misunderstanding what you're saying, or how the whole thing works? It wouldn't surprise me , but I do actually kinda find this cool how I'm, like, facing my fear of electricals and stuff...

Thanks again Vikash.

No, wait, I didn't mean to connect the sense wire to the meter instead of to the alternator. Connect it to both. Just splice into the sense wire but leave the existing circuit intact. Like, wire the voltmeter in parallel with the regulator. Kinda.

Ahhhhh, okay. Thanks for clearing that up. That makes more sense

I'll post back with the results tonight.

Well, I did everything exactly as I was instructed to, and here are the results:

The car and gauge behave in the identical manner as they did before

So while Vikash doesn't claim that the results make it 100% clear that the voltage regulator is indeed bad, it's conclusive enough that I'm going to get up early tomorrow morning and go get a new alternator and put it in in time for me to drive Patti to school for my 11:30 class.

On a side note, I felt it appropriate to add this tidbit of information: Upon installing this current alternator, I dropped the washer that goes above the two positive ring-connectors down into the alternator body. I never saw it again. I dunno if it caught up inside and screwed something up or what....

I will report back tomorrow to conclude this saga.

Okay, just got finished putting the new alternator in.

I didn't rig up the voltage gauge to it, but I can tell things are operating as they should. The lights stay bright, the car runs great...

Sort of...

But as you recall, I stated a couple posts ago, that whenever the lights got bright and the voltage read at 14 or 15 volts or whatever, the engine would refuse to idle properly and try to stall, and when the lights were dimmer, and the voltage would read at 12 volts, the engine would sit pretty at a 750 rpm idle.

So using common sense and pure logic, you all can deduce the following:

My engine now refuses to idle properly and tries to stall ALL the time, instead of just occasionally

So yeah it solved the problem, but it exacerbated another of my problems...

Hmph. Did you reset the ECU after changing the alternator?