Voltage regulation...

[spiderw31]

Ok, so a quick question on voltage regulation. I suspect I know the answer, but if anyone can confirm I'd appreciate it.

 

I've been having an issue for a bit with voltages. when cool / cold out, the voltage shoots up to 15-15.5+ (confirmed on a DVM). Now that it's been hot lately, I've been running the A/C and that causes voltages to run 12.5-13 after just 10 minutes of driving.

 

So my question is what performs the voltage regulation; is that internal to the alternator, or is that done by the ECM? I'm hoping ECM, but I'm thinking it's the alternator...

Edited June 16, 2015 by spiderw31

[primergray]

Does it drop lower than that with turn signals and/or headlights on? Just went through a lot of voltage dropping after 10-15 minutes or so...and cleaning up the connectors and the auxiliary positive terminal took care of it.

[spiderw31]

Oh yeah definitely, the more load I throw at it, the lower it drops but only at idle speeds. I expect voltage drops under heavy load, especially at low revs, but the low numbers I see are at speed.

 

I'm more concerned with the high voltages when cold. Last week the car randomly shut down when accelerating away from a stoplight (it started right back up though), and today it threw codes 43 and 53 along with an SES light and limp home mode.

 

I'd rather change the ECM than the alternator since the car is a 3.4, but I'm afraid I know the answer on that. It would be nice if it was just terminals, but mine are pretty clean so I'm not holding my breath on that...

[primergray]

Easier than ECM or alternator is taking the wires off the aux terminal, wirebrushing everything, then putting it back together...worth a try, no?

[spiderw31]

Easier than ECM or alternator is taking the wires off the aux terminal, wirebrushing everything, then putting it back together...worth a try, no?

Oh definitely, and it would be silly not to try! Just not expecting to much, though a pleasant surprise would be nice.

[primergray]

It's sounding like a heat/resistance thing to me...hoping someone else jumps in, but it's sounding exactly like what I was seeing for a few months until I did the aux terminal cleanup/retighten.

[primergray]

Is it going much below 12.5/13 after time and more heat? I was getting down to 10 or so, depending...and the terminal would be too hot to touch. Much hotter than general underhood metal.

[spiderw31]

That I couldn't say. the 12.5-13 range is when driving, so I don't know how hot it gets in there. It also only drops below that at a stop in gear, lights, blinks and brakes on, so again no good way to measure (by myself at least), so I don't know about heat or real voltage. I can say though that the gauge on the dash reads one full notch below 13 at that point so it's really low. That's right where the gauge reads ignition on / no start, which stupidly I've not actually measured yet, so I probably should do that when I get home today.

 

All this is making me wonder about the battery now though... it is getting up there in age, and I wonder if it just isn't properly taking a charge. Not sure if that explains the high voltages though (which were measured at the batt terminals BTW)

[spiderw31]

Alright, so the terminals are clean and look good. I was also able to simulate the load I see at a stoplight; I turned on the lights / fogs, turned on the A/C, cranked the fan to full tilt, and turned on the rear defroster.

 

Here is what I saw for voltages:

Car off = 12.8V

Ignition on, no start, minimal accessories = 12.5V

Car at idle, minimal accessories = 13.9V

Car at idle, A/C and lights on, fan high = 13.1V

Car at idle, same as above plus rear defroster = 12.1V

 

That last test is pretty close to a worst case, and produced levels on the voltage gauge similar to what I've seen at stoplights, so I think its a good result. Also the aux post was not hot, either right after the drive home, nor did it heat up at any point during the test.

 

Finally, I did have more issues on the drive home too. Coming to a stop light, revs started bouncing between 1500 - 500, volts bouncing between 13-14, oil pressure bouncing a bit too, and the car died shortly thereafter. It immediately restarted, but with a SES light, no A/C, and obviously in limp mode. Revs went up to 1500, and then 2000 (in neutral), and stayed there for maybe 10-20 seconds. Then they dropped down to 1500 again, and maybe 10 seconds later dropped below 1000, the SES light went out, and the A/C came back on. Car drove normally the rest of the 15 minutes to get home. Checked codes, and now I've got 23 in addition to 43 and 53. So thats saying IAT under/overvolt, ESC under/overvolt, and system overvolt. I'm really beginning to think this is the ECM now, and the remaining question is did the high system voltage when cold kill it.

[RobertISaar]

reasonable assumption. find another 16149396 PCM, throw in your MEMCAL(or if it has an interchangeable one in it, leave it) and you can rule out PCM/MEMCAL issues.

 

the voltages you're seeing aren't that unusual for results at idle.... the alternator can only produce so much current at low speeds. maybe I'm reading badly, but perhaps wedge something thin in-between the throttle-stop and the TB arm to bring engine speed up to the 1500/2000 range and retest. by 2000, the alternator is getting close to/at the engine speed necessary to produce all ~100 amps or so it is rated for.

[Garrett Powered]

I had a battery that "walmart" said they load tested "good" but when I replaced it with one from bi mart my car stopped doing that. exactly kind of what yours is doing. I know the battery was weak, I had let the car sit and the battery drained several times and got recharged. Walmart probably just doesnt want to help me out. My car seems to have a tough spell around this time of year though for some reason.

[pitzel]

On my car, at idle, basically the alternator doesn't put out anything. Needs something like 900rpm to actually turn the alternator on. Hot idle as commanded by the ECU is just a shade over 650rpm. So theoretically if it were to sit in that state long enough, with the A/C / rear defogger energized long enough, it would probably fully deplete the battery.

 

As for temperatures and voltages, that's a function of temperature on account of the silicon components used in the regulator. Its perfectly normal to be >15V in the dead of Canadian -40C winter.

 

Drainage more often than not on W-bodies, particularly early builds with the weird/stupid seatbelts in the doors, is due to a malfunctioning silicon relay that keeps a solenoid in the doors almost permanently energized. Find that relay (usually in the passenger-side of the under-dash), remove it, and problems go away. Alternatively you can go through a proper parasitic drain test and be led to the same conclusion.

 

I posted a video from a car electrical guru where they spent 2 hours figuring this issue out. Let me know if you can't find it in my various posts to these forums.

[RobertISaar]

anyone else hook up the otherwise unconnected "sense" wire on the 4-pin plug? I have it done for the 91SE, probably going to be doing it for the MC as well, I would think that it might have a chance of better voltage targets rather than going off of the regulator's internal reference.

[spiderw31]

reasonable assumption. find another 16149396 PCM, throw in your MEMCAL(or if it has an interchangeable one in it, leave it) and you can rule out PCM/MEMCAL issues.

 

the voltages you're seeing aren't that unusual for results at idle.... the alternator can only produce so much current at low speeds. maybe I'm reading badly, but perhaps wedge something thin in-between the throttle-stop and the TB arm to bring engine speed up to the 1500/2000 range and retest. by 2000, the alternator is getting close to/at the engine speed necessary to produce all ~100 amps or so it is rated for.

 

Yeah, I thought those numbers looked normal too, thought the one number that I'm really questioning isn't in that test list, and that is the voltage while driving with A/C on. That number seems low at ~13.2 (guessing based on where the gauge was while measuring last night). Then of course I question why in the world GM would setup the car to basically run off battery at idle. Every GM from that time period I've owned has acted that way, and it just seems stupid! At any rate, I do indeed have an ECM on its way.

 

I had a battery that "walmart" said they load tested "good" but when I replaced it with one from bi mart my car stopped doing that. exactly kind of what yours is doing. I know the battery was weak, I had let the car sit and the battery drained several times and got recharged. Walmart probably just doesnt want to help me out. My car seems to have a tough spell around this time of year though for some reason.

 

I was thinking from the way the car started over this past winter and the voltages I saw yesterday that my battery, while aging was still good. But your post got me thinking, and I realized the battery is a little over 6 years old (not the 4 I had been thinking previously)! I dunno if the battery is causing the trouble or not but it should definitely be replaced.

 

On my car, at idle, basically the alternator doesn't put out anything. Needs something like 900rpm to actually turn the alternator on. Hot idle as commanded by the ECU is just a shade over 650rpm. So theoretically if it were to sit in that state long enough, with the A/C / rear defogger energized long enough, it would probably fully deplete the battery.

 

As for temperatures and voltages, that's a function of temperature on account of the silicon components used in the regulator. Its perfectly normal to be >15V in the dead of Canadian -40C winter.

 

As I said the Saar, needing the car to be off idle to actually keep the battery charged if you are running a decent load seems pretty stupid to me, but yeah that what GM seemed to do back then! The other thing is when my car was pushing 15.5+ volts this winter, it was nowhere near -40, I got that reading at about +15-20F, so that seems pretty high. When we were down near 0F here, I was probably topping 16V.

[RobertISaar]

that seems more than a bit odd to me.... the regulator should heat up fairly quickly at those temps/output, enough to get into normal voltage ranges within a minute or two.

 

http://imgur.com/Cm1E5ZZ

 

that is from a TSB GM released in 97.

[spiderw31]

that seems more than a bit odd to me.... the regulator should heat up fairly quickly at those temps/output, enough to get into normal voltage ranges within a minute or two.

 

I know, right? When at its highest the needle on the gauge was just a little below the red zone, and in order to pull it down at all I had the headlights on, rear defog, fan turned up, even turn on the interior lights a couple times. Basically everything I could was turned on. I usually could start turning things off after about 10 minutes or so of driving, and be back to normal after maybe 15 minutes, but even then the voltage probably was about 14.8 or so. Yeah, not exactly right!

[RobertISaar]

...... you may be looking at a new alternator then, that certainly isn't normal. I'd be looking at sooner rather than later, at least don't head into the coming winter like that.

[spiderw31]

Yep, I plan on it, and sooner rather than later too!

 

I barely got the car home tonight, and it barely even runs at all now. I pulled the battery cables and reset the codes, and I barely got the car running again with an instant code 53. Considering the engine barely ran at all, I really doubt it was over voltage. Not that I could test it anyways, since it was nearly impossible to keep it running. It was also making a nasty high pitch screaming from the right fender area that sounded like angry electronics. Hey, at least it will be immediately obvious if the ECM replacement fixes the issue!

[spiderw31]

Well, someone just gave me a great idea... pull the belt, and see what happens. Wouldn't you know it, the car starts right up and runs with the belt off the alt. I guess I just found my issue! I was hoping it was the easier job (the ECM), but oh well. New alt, here I come!

[primergray]

Ugh...had one try to overcharge and lockup on me before...but that was on the Monte, so about 250 times easier to change. Not reachable for a persuasive hammer tap as a second opinion? No matter...sounds like you've got your answer, even if it's the least pleasant of the possibilities.

[spiderw31]

Yeah, I had a spare good alt on the shelf ready to go in if needed, but not the most fun job in the world. I suppose it's a good thing I have to get in that area though, because I just discovered that my inner CV boot had been sliced through by the edge of the heat shield, so I'm fixing that while I'm in there. Hey, at least that gives me a lot more room!

 

I got the disassembly done, right to the point of lowering the cradle today before it started to rain, so I'll be continuing tomorrow. I do need to figure out a way to temporarily seal up the axle mount though so it doesn't pour out all the transmission fluid while I have the cradle lowered. Oh, and we'll also see how upset I am at myself for installing the rear bolt on the alt years back...

Edited June 20, 2015 by spiderw31

[primergray]

BTW, where in NoVA are you? Annandale here.

[spiderw31]

BTW, where in NoVA are you? Annandale here.

 

Man I totally bailed on this thread. Sorry about that! Anyways, I'm out in Leesburg. Good to see another W member at least roughly close by!

 

I thought I'd put closure on the issue: I tried like crazy to get the alt out the bottom, but failed. When I had done the whole top end of the mother several years ago, I (stupidly) put the rear bolt back on the alternator when changing it (which also was stupid since the old alt wasn't dead; I was trying to head off a failure). Well, that bolt sure comeback to haunt me, cause there was no way I was going to get at it! So, since I had a leaky lower intake along with several other random repairs needed on the top end, I decided I'd be far enough into the motor just to do it the way I did before. And so I putted the t-belt, rear cam carrier, and the rear head. Crazy I know, but really it didn't add too much time considering all the other odd jobs I sorted along the way. Also I got to see the condition of the inside of the rear bank, and only very minor were; it looked really good inside, so I was definitely happy about that! I dropped the old alt (the perfectly good one I pulled years back) back in, and put everything back together. Oh boy did she run great... for all of ten miles, and then the original issue was back . Turns out after all that the issue was indeed the ECM! So rebuilt ECM from Rock Auto is now in, and so far the car has been fine all day. Thats far, far better than it has been since all this started, so I think it's safe to say the issue is resolved.