Voltage and stability while running/cranking/idle?

[Freight]

I recently installed a voltmeter in my check panel(super easy) and have been monitoring voltages and am curious for everyone’s thoughts. These are all approximate, obviously

at the battery -

off(after sitting overnight) 11.6-11.8
cranking* 8.4-8.8
idle 12.1-12.4

*even sitting 4+ days I started, it will crank and start at that voltage. It’s slow but it will fire up after 3-4 revs

voltmeter(which agrees generally with multimeter on battery terminal readings) -

first started 12.1-12.4(battery light on)
idle(after rev to kill light) 12.4~
driving 12.7-13.2…only 13+ about 30% of the time. 12.8 average

While driving, engine speed seems to have no affect on voltage. It will fluctuate constantly even when RPM are stable for extended time, like on the highway. It’s not rapidly shifting, nor does there seem to be a rhythm or pattern, just constant fluctuation every few seconds. Checked the voltmeter I installed for accuracy and refresh rate, and it’s accurate and refreshes quickly enough.

i did notice that after starting the car, battery light out after a throttle blip, if I’m in neutral and hold the revs up at say 2k, the idle will slowly rhythmically bounce about 100-200rpms in sync with the voltage drifting .2~ amps. Chicken or egg, I’m not sure, but the voltage is never stable, and the idle is noticeably rougher before I blip the throttle to shut the light off. I don’t have any issues with headlights, or battery drain, it’s never died on me.

Does this sound like a battery problem, or a regulator problem? The fact the battery is surviving and performing even below freezing despite the low voltages implies it’s not the battery…to my unqualified brain. I’m gonna get a new voltage meter but does this sound like something else is amiss?




 

[Galahad]

Seems weird, the voltage isn't supposed to move around while driving.

I know the motronic system bounces between slightly rich and slightly lean for the cats, I suspect that's the fluctuation you're seeing but you shouldn't have that appear in the rpm or voltage.

couple things to check:

Take the regulator off the back of the alternator and the belt off the front, run some fine sand paper on the rings while you spin the alternator by hand, put it all back together. This solved the "rev to make the light go out" issue on my car. The car will idle worse with the light on since you're running off the battery. If there's significant scoring on the rings or the brushes that's a problem.

Measure the voltage at the battery while cranking and at the starter while cranking. If there's a big difference, get a new starter cable. If the battery voltage drops significantly while cranking you have a battery issue. If your voltmeter isn't plugged directly into the battery the reading may not match battery voltage during cranking.

Take a look at the ground cable between the engine and the frame, if it's frayed or corroded replace it. You could also measure the voltage between the engine block and the battery ground while cranking, that'll tell you the condition of the ground cable.

[Freight]

I measured the cranking voltage at the battery, off the terminals already and its mid-high 8v’s, dropped from the 11.5-12.2 depending on how long it’s sat. I’ll check it at the starter.

regarding the alternator, what do you mean by “the rings”?  I found a regulator locally and am about to swap the old one out for good measure, I assume it’s going to be obvious what your talking about?

[Galahad]

The regulator has carbon brushes that ride on some copper rings on the back of the alternator shaft inside the case, that's how they get the spinning electrons into non-spinning wires. It should be obvious when you remove the old regulator assembly.

[turbodan]

It should be charging at 14ish with the engine running, even at idle. Sometimes I also have to blip the throttle to get the alternator to wake up, that's not unusual. 

With the voltages youre seeing I would replace the alternator.

[Freight]

The regulator has carbon brushes that ride on some copper rings on the back of the alternator shaft inside the case, that's how they get the spinning electrons into non-spinning wires. It should be obvious when you remove the old regulator assembly.

 

 

New regulator in, everything was checked for smoothness. Voltage is now a lot more stable, but now when driving at night with headlights I see a stable 12.2/3v under load, and it’ll raise to 12.5 or so when I clutch in. Lights off and I’m getting 12.8v at idle, dropping to 12.4.

[Mike W.]

Yep, something's wrong. I doubt the alternator, possibly the regulator. Alternators are pretty simple devices, windings and diodes. But...

Before you do anything else, check the voltage at the alternator with the car running. From the B+ terminal with the fat red wire, to the body of the alternator. I'd say there is a better than 50-50 chance you have ~14 volts there. Not at the battery, check at the alternator. Do that first. Then check at the battery. If your readings are off by more than about a quarter of a volt, .25 volts, you have a wiring problem. More than likely at one or more of the terminals. Test to see what side by checking voltage between the alternator and the battery. But from hot to hot, then ground to ground. A common problem is the ground wire between the alt and the engine, check there both to make sure it's there and that it's in good condition.

[Freight]

Yep, something's wrong. I doubt the alternator, possibly the regulator. Alternators are pretty simple devices, windings and diodes. But...

Before you do anything else, check the voltage at the alternator with the car running. From the B+ terminal with the fat red wire, to the body of the alternator. I'd say there is a better than 50-50 chance you have ~14 volts there. Not at the battery, check at the alternator. Do that first. Then check at the battery. If your readings are off by more than about a quarter of a volt, .25 volts, you have a wiring problem. More than likely at one or more of the terminals. Test to see what side by checking voltage between the alternator and the battery. But from hot to hot, then ground to ground. A common problem is the ground wire between the alt and the engine, check there both to make sure it's there and that it's in good condition.

You were right. Alternator hot to body was 13.8 +/- .15

I’m realizing now that this makes no sense, but
hot to battery hot was .55mv
alt body to battery cold was .85mv
(I’m going to recheck that here in a few)

at a cursory glance I didn’t see a ground wire to the engine from the alt, and I thought the b34/b/35 ground the alternator through the mounting bolts?

[Mike W.]

Yep, something's wrong. I doubt the alternator, possibly the regulator. Alternators are pretty simple devices, windings and diodes. But...

Before you do anything else, check the voltage at the alternator with the car running. From the B+ terminal with the fat red wire, to the body of the alternator. I'd say there is a better than 50-50 chance you have ~14 volts there. Not at the battery, check at the alternator. Do that first. Then check at the battery. If your readings are off by more than about a quarter of a volt, .25 volts, you have a wiring problem. More than likely at one or more of the terminals. Test to see what side by checking voltage between the alternator and the battery on both hot and ground. A common problem is the ground wire between the alt and the engine, check there both to make sure it's there and that it's in good condition, but you have drop on both hot and ground sides.

You were right. Alternator hot to body was 13.8 +/- .15

I’m realizing now that this makes no sense, but
hot to battery hot was .55mv
alt body to battery cold was .85mv
(I’m going to recheck that here in a few)

at a cursory glance I didn’t see a ground wire to the engine from the alt, and I thought the b34/b/35 ground the alternator through the mounting bolts?

If your readings were volts, not mv, it all adds up. That would be 1.4 volts drop from the alt to the battery. 13.8 minus 1.4 equals 12.4, which is right about what you are seeing.

Some E28s ground thru the case to the block/mtg bracket, some are rubber mounted and may not. But I think, repeat think, they all had a ground wire.

So, remove, clean and replace all the related connections from the alt to the battery. And maybe install a ground wire. If that doesn't work run a new and additional wire, at least 10 gauge, from the alternator to the battery. I see you're from NY, salt and winters don't do wiring and connections any good. Your issues might be where the terminal is crimped onto the wire. Regardless, your alternator seems to be fine, it's from there to the battery that you have problems.

[Freight]

You were right. Alternator hot to body was 13.8 +/- .15

I’m realizing now that this makes no sense, but
hot to battery hot was .55mv
alt body to battery cold was .85mv
(I’m going to recheck that here in a few)

at a cursory glance I didn’t see a ground wire to the engine from the alt, and I thought the b34/b/35 ground the alternator through the mounting bolts?

If your readings were volts, not mv, it all adds up. That would be 1.4 volts drop from the alt to the battery. 13.8 minus 1.4 equals 12.4, which is right about what you are seeing.

Some E28s ground thru the case to the block/mtg bracket, some are rubber mounted and may not. But I think, repeat think, they all had a ground wire.

So, remove, clean and replace all the related connections from the alt to the battery. And maybe install a ground wire. If that doesn't work run a new and additional wire, at least 10 gauge, from the alternator to the battery. I see you're from NY, salt and winters don't do wiring and connections any good. Your issues might be where the terminal is crimped onto the wire. Regardless, your alternator seems to be fine, it's from there to the battery that you have problems.

Yep, land of salt here. My meter said mV, hence my confusion - I’m glad you understand the magic smoke better than I do. 

This afternoon I’m going to check all the grounds I can find, clean connections, etc. I did notice the body ground off the battery has some green corrosion on it. It’s still supple but suggests it’s time to replace it.

[Freight]

So I’ve gone through all the grounds in the engine bay, no change. Hood latch, main ground by battery, valve cover to firewall, head to main harness, block/motor mount to chassis. Did not see one directly off the alternator, though I pulled those wires and cleaned those connections as well. Where are the others located on the 535i? 

under the drivers kick panel by the brake pedal, and one under the rear seat? There’s one for the antenna too I believe, but when I fixed the antenna I went through all the wiring. I checked every ground I cleaned with the multimeter to make sure there was no resistance, and all came up with .2-.4ohms after cleaning(matched the resistance of touching the probes directly), so they’re good. 

what’s next? Ground wire directly from the alternator to the main ground beside the battery? Or would directly to the block work?

[Federico]

After checking the grounds, I'd add the alternator ground wire, kind of like shown here: E12 alt. I have 2x 10 awg ground wires on mine.
Also, make sure the block has its ground.
Finally, look up how to do a diode test on the alternator. When they start to go out they'll make an ac ripple that throws the regulator out of whack.

[Mike W.]

So I’ve gone through all the grounds in the engine bay, no change. Hood latch, main ground by battery, valve cover to firewall, head to main harness, block/motor mount to chassis. Did not see one directly off the alternator, though I pulled those wires and cleaned those connections as well. Where are the others located on the 535i? 

under the drivers kick panel by the brake pedal, and one under the rear seat? There’s one for the antenna too I believe, but when I fixed the antenna I went through all the wiring. I checked every ground I cleaned with the multimeter to make sure there was no resistance, and all came up with .2-.4ohms after cleaning(matched the resistance of touching the probes directly), so they’re good. 

what’s next? Ground wire directly from the alternator to the main ground beside the battery? Or would directly to the block work?

Start with a jumper between the alt and the block. But you seem to be focusing just on grounds, the hot side can have resistance too, and according to your (interperted) meter readings, almost as much as the ground side. Measure voltage, car running, with the headlights on. That should put a load on it to accentuate any problems.

Measure point to point, you're looking for a voltage drop. From the alt to, where is it the starter where it goes next? The battery next IIRC. So that's 4 points for resistance. And it's no guaranty that the crimped on connections at the ends are still resistance free.

This is solvable, but it has many possibilities. But at least you know you don't have to throw an alternator at it.

[Freight]

So I’ve gone through all the grounds in the engine bay, no change. Hood latch, main ground by battery, valve cover to firewall, head to main harness, block/motor mount to chassis. Did not see one directly off the alternator, though I pulled those wires and cleaned those connections as well. Where are the others located on the 535i? 

under the drivers kick panel by the brake pedal, and one under the rear seat? There’s one for the antenna too I believe, but when I fixed the antenna I went through all the wiring. I checked every ground I cleaned with the multimeter to make sure there was no resistance, and all came up with .2-.4ohms after cleaning(matched the resistance of touching the probes directly), so they’re good. 

what’s next? Ground wire directly from the alternator to the main ground beside the battery? Or would directly to the block work?

Start with a jumper between the alt and the block. But you seem to be focusing just on grounds, the hot side can have resistance too, and according to your (interperted) meter readings, almost as much as the ground side. Measure voltage, car running, with the headlights on. That should put a load on it to accentuate any problems.

Measure point to point, you're looking for a voltage drop. From the alt to, where is it the starter where it goes next? The battery next IIRC. So that's 4 points for resistance. And it's no guaranty that the crimped on connections at the ends are still resistance free.

This is solvable, but it has many possibilities. But at least you know you don't have to throw an alternator at it.

Good point, I have really worried only about grounds. Electronics are hard for my brain to wrap itself around when they’re at the scale of a car. I did check all other grounds last night(except the secondary in the steering column), they looked good with no signs of corrosion but were cleaned and tightened for good measure. I did check the battery clamps and according to the meter, show no resistance.

I managed yesterday to kill the battery after having music on while checking the grounds and the pulling the fuel sender, so I tossed it on a tender overnight. When I got in the car this morning, the voltmeter only read 13.1 before starting. Foolishly I did not check the voltage directly across the battery. For 10-15 minutes the car was getting 12.9-13.1, before eventually settling back down to the 12.3ish it’s been during daytime drives.

today I’ll work through the hot side. 

[Freight]

Alright! I went over the hot side, disconnected and cleaned every point I came across, and all looked good but they must not have been. I am now getting 13.7v at the alternator, 13.5v at the battery. Headlights on, tested with the meter, are receiving 12.7v. In the cabin, window switches get 13.4v.

Seems everything is happy now!

[Mike W.]

Cool, a win.

You may like sorting out vacuum issues, I enjoy electrical. It just makes sense to me, always has.

[Freight]

I actually enjoy small electrical, things where I can see the entire circuit in front of me, tho I don’t really know much about it. I do a fair bit of soldering and repair work, but working on small devices and pcbs doesn’t seem to translate very well for me.

That’s how I feel about vacuum systems. They’re very linear, very simple, only a few things can go wrong. Just grab your diagram and work down the line methodically. Definitely time consuming but it’s fairly brainless. Not sure why it stresses people out.

[Mike W.]

Just grab your diagram and work down the line methodically. Definitely time consuming but it’s fairly brainless. Not sure why it stresses people out.

Speaking for myself, the lack of a diagram is a major contributor. Not that I haven't done some work there, but the lack of documentation is probably the biggest thing to me.

[Freight]

Just grab your diagram and work down the line methodically. Definitely time consuming but it’s fairly brainless. Not sure why it stresses people out.

Speaking for myself, the lack of a diagram is a major contributor. Not that I haven't done some work there, but the lack of documentation is probably the biggest thing to me.

Is there none for the e28? I haven’t looked at the vac system on my car yet cause it seems to be fine.

i recently sold a w123 merc turbo diesel and the vac diagrams for that car are an absolute dream. They’re available showing the routing through the car so there’s no guessing or trying to figure out where stuff’s located. That’s one thing I’ve found frustrating about the e28. The ETM is great, but the fact that there are no diagrams showing the locations in the car itself make things a bit more complicated and time consuming.

[Mike W.]

Is there none for the e28? I haven’t looked at the vac system on my car yet cause it seems to be fine.

E28s have so little vacuum it's pretty much irrelevant, I was speaking more in general, or maybe more E23. They have unreliable vacuum pods everywhere, many nearly impossible to get to. As a general rule I'm not crazy about vacuum operated things, they seem less reliable, especially when it comes to auto heat and A/C. Mechanical operation like a cable is ideal, electrical I still prefer. YMMV.