Overcharging despite new regulator - PeachParts Mercedes-Benz Forum

Skippy · #1 01-15-2014, 06:20 PM

Patient: '83 300D, 330,083 miles. Voltage regulator is new from 28 September, 2500 miles ago. Battery was replaced at that time also. Prior to that the car had an overcharging problem which fried the battery and headlights.

I did some checks today after tightening my squealing alternator belts. 12.58 VDC with the engine off. Engine running at idle with no load gradually came up to over 17 volts. Turning on headlights and blower pulls it down to about 12.5 at idle, but revving the engine gets erratic results ranging from mid 14's to 16. I checked AC voltage and got 36 VAC at the battery terminals with no load and 28 with headlights and blower on. I've also noticed that the battery light in the dash never illuminates.

I'm thinking its time for an alternator, but wonder if there is something else at play here. Could this be some sort of oddball wiring problem? I'm open to suggestions.

whunter · #2 01-15-2014, 06:59 PM

The alternator diode bridge is damaged = replace the alternator.

.

Skippy · #3 01-15-2014, 07:31 PM

That's kind of what I thought. I just wanted to hear it from someone else. Thanks.

I've heard that a bad diode will lead to the battery discharging through the bum diode when the car sits. Mine doesn't have that issue. ???

whunter · #4 01-15-2014, 08:19 PM

Quote:

Originally Posted by Skippy

That's kind of what I thought. I just wanted to hear it from someone else. Thanks.

I've heard that a bad diode will lead to the battery discharging through the bum diode when the car sits. Mine doesn't have that issue. ???

Why is it over charging?

It is named alternator because it generates alternating current = AC voltage.

it is routed through what is commonly known as a voltage regulator - WRONG name - What you have is a Rectifier which changes the AC voltage to DC voltage.

The most common is the Zener Diode Tree, a series of diodes and resistors in some cases, the Avalanche Diode Tree does the same job.

Somewhere in this circuit a diode has failed = allowing excess DC voltage to pass, the rectifier is still changing AC to DC, but its not cutting it down enough.

.

fahrvergnugen · #5 09-21-2014, 07:39 PM

So the diodes are actually on the alternator, and not the regulator?

I was hoping by just replacing the voltage regulator I would fix the diagnosed bad diode (via OReillys).

dieselbenz1 · #6 09-21-2014, 08:16 PM

If the alternator was changed at some point make certain you match the regulator ( it's actually a rectifier as pointed out before ) with the alternator not all are created equally. Better yet get the old regulator number and order an identical. Miss matches can turn ugly and costly real fast. Is it a bosch alternator?

fahrvergnugen · #7 09-21-2014, 09:55 PM

I don't believe it is a Bosch; it was a free one, rebuilt by someone in the past. It was sitting on a good engine for a number of years. No doubt I killed one of the diodes when I yanked off the ground to stop the 'new' starter from continuing to crank when the engine had already started.

The regulator/brushes may be fine; my question is are the diodes on the alternator, or on the regulator? All I know is I have a bad diode; replacing the regulator may be unnecessary.

dieselbenz1 · #8 09-21-2014, 10:03 PM

Diodes are in the regulator sounds like a mismatch if you can get the make and model of the alternator then you can order the correct regulator.

fahrvergnugen · #9 09-21-2014, 10:44 PM

IF the diodes are in the regulator, I had planned to buy this...

Voltage Regulator (80 AMP) - Bosch 30091 | FCP Euro

It would be nice to bump what I believe to be a 55 amp alt up to 80 amps...

Why do you say it is a mis-match? The alt came off a OM617, same as mine. It has the same plug on the back.

fahrvergnugen · #10 09-21-2014, 10:50 PM

Looks like the diodes are IN the alternator itself, not on the regulator.

http://caunter.ca/volvo960/vp14805.pdf Look at page three.

dieselbenz1 · #11 09-21-2014, 10:53 PM

Alternators and regulators are specific to one another. If the alternator is not bosch why would you attempt to put a bosch regulator where it does not belong. Changing to 80 amps won't do a thing if the alternator is rated at 55 amps.

tangofox007 · #12 09-21-2014, 11:00 PM

Quote:

Originally Posted by fahrvergnugen

Looks like the diodes are IN the alternator itself, not on the regulator.

That is correct. Claims to the contrary are greatly exaggerated.
Here is a pretty good summary of basic alterntor operation:

http://alternatorparts.com/understanding-alternators.html

Mxfrank · #13 09-21-2014, 11:21 PM

Quote:

Originally Posted by whunter

Why is it over charging?
it is routed through what is commonly known as a voltage regulator - WRONG name - What you have is a Rectifier which changes the AC voltage to DC voltage.

The most common is the Zener Diode Tree, a series of diodes and resistors in some cases, the Avalanche Diode Tree does the same job.

Somewhere in this circuit a diode has failed = allowing excess DC voltage to pass, the rectifier is still changing AC to DC, but its not cutting it down enough

.

Wrong on theory, maybe right on conclusion. The alternator diodes in these cars (actually, most cars) are neither zeners nor are they avalanche diodes. They are simple half wave rectifiers. Although they are mounted to a mechanical fixture called a bridge, they are, confusingly, not arranged as a full-wave bridge in the electrical sense. The voltage regulator is a separate piece of electronics, which controls output voltage by regulating the duty cycle of the field. Let me explain what's happening here.

First of, the voltage regulator is a completely separate set of electronics from the diode assembly: it's the little can you mount on the back of the alternator. The diodes are internal to the alternator. There are nine, see the photo, nine diodes inside the alternator. This is because there are three A/C windings, creating a three phase output. Each phase has a negative and a positive diode, which accounts for six diodes. These diodes are probably fine in your case (you wouldn't be able to make these high voltages if even one diode was bad).

Which leaves us with three small diodes, which are appropriately called a trio. These three extra output (+) diodes feed the voltage regulator's internal sense connection (see photo). When the alternator is running, the voltage regulator uses these diodes for both power and sense. By power, I mean that the VR is directly powered by the trio. By sense, I mean that it uses the trio voltage to test the output voltage of the alternator. Based on the sense, it will control the duty cycle of the field coil (rotor) keeping the voltage in at spec.

When the car is first starting, the alternator has to be "bootstrapped" before it will charge. What does this mean? If you understand the previous paragraph, you understand that the voltage regulator is powered by the alternator itself when everything is running. But at start up, the alternator isn't generating power. To get things started, a bootstrap current has to be fed to the VR from outside. This is provided, and I know it's confusing, by the dash lamp. One side of the dash lamp is connected to the battery via the ignition switch. The other side is in series with the VR, actually feeding power to it via the same connection as the trio.

Since the trio isn't providing power at rest, the dash lamp finds ground through the VR. Once the engine starts and the alternator starts generating current, the dash lamp will have 12V+ on both leads, and so won't light.

What's wrong here? Assuming the dash lamp is good, you have 16V at the B+ and at the trio output, otherwise the lamp would light. The most likely explanation is that there's something not right about the VR.

There's one other possibility, and it depends on how the dash lamp is wired (I don't have specific info for your car). In many cases, the engineers will add a resistor in parallel with the lamp. If this is the case, the lamp can be blown, but the alternator will still bootstrap. The trio and B+ could be out of sync due to a bad trio diode, and there's no lamp to light. If the lamp never ever lights, even when the key is in position 2, check your bulb.

Your AC readings are garbage. Because these diodes are half wave rectifiers, the output of the alternator isn't flat DC. It's a ripple wave form, which can't be read correctly by many AC meters. Just worry about DC.

What I would do is take the VR out, and reach inside the alternator with some abrasive paper to clean the slip rings on the rotor. They should be bright copper, but I bet yours are fouled with carbon. If that doesn't fix it, then replace the alternator.

fahrvergnugen · #14 09-22-2014, 12:56 AM

Very good fellas, thanks very much to both of you!

I hate removing the alternators on these engines, that top mount is an abortion at best. Once I take it out, I will clean it very carefully. It does not help matters that I have had SLS fluid spraying everywhere, as both the low and high pressure lines began to spew on a trip here after I replaced the T-stat gasket and bypass hose.

The joys of resurrecting a 31-yr-old car that has sat for three years.

whunter · #15 09-22-2014, 01:17 AM

Quote:

Originally Posted by Mxfrank

Wrong on theory, maybe right on conclusion. The alternator diodes in these cars (actually, most cars) are neither zeners nor are they avalanche diodes. They are simple half wave rectifiers. Although they are mounted to a mechanical fixture called a bridge, they are, confusingly, not arranged as a full-wave bridge in the electrical sense. The voltage regulator is a separate piece of electronics, which controls output voltage by regulating the duty cycle of the field. Let me explain what's happening here.

First of, the voltage regulator is a completely separate set of electronics from the diode assembly: it's the little can you mount on the back of the alternator. The diodes are internal to the alternator. There are nine, see the photo, nine diodes inside the alternator. This is because there are three A/C windings, creating a three phase output. Each phase has a negative and a positive diode, which accounts for six diodes. These diodes are probably fine in your case (you wouldn't be able to make these high voltages if even one diode was bad).

Which leaves us with three small diodes, which are appropriately called a trio. These three extra output (+) diodes feed the voltage regulator's internal sense connection (see photo). When the alternator is running, the voltage regulator uses these diodes for both power and sense. By power, I mean that the VR is directly powered by the trio. By sense, I mean that it uses the trio voltage to test the output voltage of the alternator. Based on the sense, it will control the duty cycle of the field coil (rotor) keeping the voltage in at spec.

When the car is first starting, the alternator has to be "bootstrapped" before it will charge. What does this mean? If you understand the previous paragraph, you understand that the voltage regulator is powered by the alternator itself when everything is running. But at start up, the alternator isn't generating power. To get things started, a bootstrap current has to be fed to the VR from outside. This is provided, and I know it's confusing, by the dash lamp. One side of the dash lamp is connected to the battery via the ignition switch. The other side is in series with the VR, actually feeding power to it via the same connection as the trio.

Since the trio isn't providing power at rest, the dash lamp finds ground through the VR. Once the engine starts and the alternator starts generating current, the dash lamp will have 12V+ on both leads, and so won't light.

What's wrong here? Assuming the dash lamp is good, you have 16V at the B+ and at the trio output, otherwise the lamp would light. The most likely explanation is that there's something not right about the VR.

There's one other possibility, and it depends on how the dash lamp is wired (I don't have specific info for your car). In many cases, the engineers will add a resistor in parallel with the lamp. If this is the case, the lamp can be blown, but the alternator will still bootstrap. The trio and B+ could be out of sync due to a bad trio diode, and there's no lamp to light. If the lamp never ever lights, even when the key is in position 2, check your bulb.

Your AC readings are garbage. Because these diodes are half wave rectifiers, the output of the alternator isn't flat DC. It's a ripple wave form, which can't be read correctly by many AC meters. Just worry about DC.

What I would do is take the VR out, and reach inside the alternator with some abrasive paper to clean the slip rings on the rotor. They should be bright copper, but I bet yours are fouled with carbon. If that doesn't fix it, then replace the alternator.

There is no resistor in parallel with the battery lamp on W123 or W126 , and this is true of many other MB chassis.

This battery light is a true diagnostic tool:
#1. Illumination with Key on = alternator excitation circuit continuity.
#2. Failure to illuminate with Key On = NO continuity = NO charging.

.