Alternator overcharging

Tiny
WILDMTBILL
  • MEMBER
  • 1995 DODGE DAKOTA
  • 5.2L
  • V8
  • 4WD
  • AUTOMATIC
  • 150,000 MILES
Alternator charging at 18 volts, replaced alt and didn't fix. Have read some other answers on similar problems, blue wire reads 18 volts, green wire reads.23 volts?
Do you
have the same problem?
Yes
No
Thursday, March 28th, 2019 AT 5:06 PM

12 Replies

Tiny
CARADIODOC
  • EXPERT
You are correct, 18 volts is much too high, but an alternator can not cause over-charging unless it is assembled incorrectly with one of the brushes grounded through its mounting screw. Over-charging is caused by the voltage regulator driving the alternator too hard.

There's two possibilities for this condition. The voltage regulator is shorted, which is fairly uncommon, or the control wire going to it is grounded. There is a third possibility, but it doesn't apply here. That is a break in the system voltage sensing wire used by the voltage regulator. In this case, the voltage regulator is inside the Engine Computer. Since Chrysler invented the electronic voltage regulator for 1970 models, they have always had temperature compensation built in to help batteries charge in cold weather. By putting them in the Engine Computers now, they can also modify charging voltage based on other variables such as wide-open-throttle, engine overheating, and things like that. That voltage sensing wire is the feed circuit for other things inside the computer, so if that wire has a break in it, there would be other symptoms besides the over-charging problem.

The voltage regulator can not draw the voltage on the green control wire down much below 4.0 volts when it is working properly. The lower it takes that voltage, the greater will be the voltage difference between the two field wires, and the bigger will be the electromagnetic field created in the rotor, or "field" winding. That results in the highest current output.

The clue here should be the extremely low voltage you found on the green control wire. Even when the voltage regulator circuit becomes shorted, I've never seen that voltage drop below around 2.0 volts due to all of the other circuitry inside the computer. To get down to.23 volts, the better suspect is that green wire is grounded. In fact, grounding that wire is part of a diagnostic test we can do in this system.

To verify this, with the ignition switch off, connect your ohm meter with small jumper wires so you don't have to hold onto the probes. Connect one to a paint-free point on the engine, and the other one to the green wire. Actually, you can use either small terminal on the back of the alternator. The resistance between those two terminals is only four ohms, but be aware, it is common for them to read open circuit when the rotor is not spinning. That's due to the less-than-perfect contact the brushes make when they're standing still. If you get an open circuit reading, move the probe from one small terminal on the back of the alternator to the other small terminal.

I should mention too, the reason for the apparent confusion is those two wires go through a black plastic block, and it's that block that has the two terminals. It's impossible to tell which wire goes to which terminal.

You should find an open circuit during this test. If the green wire is grounded, you'll find a very low resistance reading. Carefully move and wiggle the wire harness to see what makes that reading go higher or open circuit. Look for a harness that's draped over the sharp edge of a metal bracket. Another common place to find grounded wires is under the battery tray. On some models, that wire harness slides back and forth a little as the engine rocks between accelerating and coasting. The paint will wear away and wire insulation will wear down until a wire becomes grounded.
Was this
answer
helpful?
Yes
No
+1
Thursday, March 28th, 2019 AT 5:37 PM
Tiny
WILDMTBILL
  • MEMBER
Will test tomorrow evening and get back with results. Thanks for the timely response.
Was this
answer
helpful?
Yes
No
Thursday, March 28th, 2019 AT 8:15 PM
Tiny
WILDMTBILL
  • MEMBER
Checked green wire to ECM plug and got no resistance, no grounding, also checked blue wire to ECM plug, was also good, checked blue wire to and relay and found a plug that had no resistance. I can do that because we tried an external regulator before I found your info so I have male and female plugs on both the green and blue wires about 4" behind the black plastic at the alternator. Thoughts?
Was this
answer
helpful?
Yes
No
+1
Friday, March 29th, 2019 AT 6:09 PM
Tiny
WILDMTBILL
  • MEMBER
Should read ASD relay.
Was this
answer
helpful?
Yes
No
Friday, March 29th, 2019 AT 6:10 PM
Tiny
CARADIODOC
  • EXPERT
Forget the blue wire. The automatic shutdown, (ASD) relay turns on for one second when the ignition switch is turned on, then again during engine rotation, (cranking or running). That relay feeds the ignition coil, injectors, fuel pump or pump relay, oxygen sensor heaters, and the alternator field winding. That's the blue wire. It will always have full battery voltage on it when that relay is on. That blue wire goes one more place, that is back to the Engine Computer, but it has nothing to do with the charging system there. The computer thinks it turned the ASD relay on. 12 volts on that wire simply verifies that occurred.

As a side note, when you have a crank / no-start condition on any Chrysler product, and don't know where to start, check for 12 volts on that blue wire during cranking. On later models that is almost always a dark green / orange wire. It will be the wire that is the same color at every injector, as well as the ignition coil pack and alternator. If you see 12 volts for one second when you turn on the ignition switch, the ASD relay and its circuit are working. What's important is if that voltage comes back during cranking. When it does not, the crankshaft position sensor and the camshaft position sensor are the two best suspects. Their signals are what tells the Engine Computer the engine is rotating. Those sensors commonly do not set diagnostic fault codes right away, so don't let the absence of a related fault code fool you.

The engine runs, so we know the ASD relay is turning on, and voltage is feeding the alternator's field winding. If it was missing due to a corroded splice in that blue circuit, you'd have a no-charge condition. To have such a serious over-charge condition, either the green wire going to the voltage regulator is grounded, as I suspected, the voltage regulator circuit is shorted inside the computer, which I'll admit is a possibility, or the system voltage sensing circuit has a break in it. The sensing circuit is usually also the feed circuit for other systems, so when that has a problem, there's usually other symptoms in addition to the over-charging.

I had a lot of "bugs" built into donated cars for my students to diagnose. Chrysler was very good to my school. Their instructors would drive a vehicle for a few months, and they'd be poked and prodded by their students, so they didn't want to sell them to the public. We had a '97 Dakota, quite a few Shadows, and a '95 Intrepid. Some of my bugs that could be switched in were that green wire was grounded in multiple places. Each one could be narrowed down by disconnecting plugs to isolate different parts of the circuits from each other. I did have bugs built into Engine Computers too, and those included a defective voltage regulator. Even that couldn't pull the green wire down to much less than four volts.

If you are right about the regulator being shorted, a replacement computer will solve the problem. If my best guess is right, that green wire has to be grounded somewhere.

Adding an external voltage regulator that you mentioned is a way around this problem if the one in the computer is shorted, but it comes with a whole new set of problems. The obvious one is when it's built into the computer, it can turn the alternator off completely during wide-open-throttle when you might need that extra five to ten horsepower. It can reduce alternator output when the engine is running hot, to reduce load, and it can anticipate the switching on of the AC compressor to eliminate the momentary dimming of head lights, dash lights, and slowing of the wipers and heater fan, by bumping up output a fraction of a second before it switches the compressor relay on. Those features are lost with the use of the external regulator.

The bigger problem is the Engine Computer watches the field current going through the regulator circuit. That actually pulses on fully, to around three amps, then off fully, to 0.0 amps, about 400 times per second. The ratio of on-time to off-time is varied to adjust the average field current according to the needs of the electrical system. When you cut the green wire to run it to the external regulator, that leaves the computer's circuit open, and there's no field current to switch on and off. That lack of current will be detected, and it will set a diagnostic fault code for "Field circuit not switching properly".

Two problems result from this fault code. First, the fuel injectors will not open at the right time or will not stay open long enough when supply voltage is wrong. That can result in increased emissions. Any fault code that is related to something that could adversely affect emissions must turn on the Check Engine light. The "field circuit not switching properly" code is one of them that turns the light on. Once that occurs, how will you ever know if a second, unrelated fault code sets? That second code could result in an expensive repair, such as a melted catalytic converter. You'd never know that second problem developed.

The other problem is for any fault code to set, there is always a long list of conditions that must be met, and one of those is that certain other codes can't already be set. As an example, with low system voltage, computers do weird things, and they read sensors incorrectly. You could develop a problem related to just one injector circuit, but the computer has stopped running tests on that circuit because it knows the results can't be trusted since a code is set for the charging system. A problem could develop, but with the tests suspended, you'd never know it. This happens a lot with engine performance problems and there's no related fault code to tell you where to start looking.

You might have to experiment with adding about a five to six ohm resistor between the blue wire and the green wire that goes to the computer, after you cut it off. That would give the voltage regulator a small current to switch on and off, and that might prevent the computer from detecting a problem, setting a fault code, and turning on the Check Engine light. Chrysler's old external voltage regulators from the '70s will run this alternator just fine, but I never tried using a resistor to trick the computer.
Was this
answer
helpful?
Yes
No
Friday, March 29th, 2019 AT 7:46 PM
Tiny
WILDMTBILL
  • MEMBER
We are going to try to find a ECM at a salvage yard. How important is the ECM number being the same? We've looked for one at part stores, but the number on ours is a " remove and rebuild" and that I'm not doing. I'm really curios about that resistor between the green and blue wires. I will let you know if I try. Thanks again for your expertise.
Was this
answer
helpful?
Yes
No
Friday, March 29th, 2019 AT 8:15 PM
Tiny
CARADIODOC
  • EXPERT
That resistor is only what I'm guessing would be needed if you go with the external voltage regulator modification.

As for Engine Computers, head to any pick-your-own-parts salvage yard and just go by the model year and engine size, regardless of the part number on it. The Transmission Computer on my '93 Dynasty was not designed to be flashed with updated software, so Chrysler gave me a new one under warranty. That one had a giant heat sink on the side, it could be flashed with updates, and it had a totally different part number. Both part numbers will work in that application.

Chrysler always wants old computers returned to them when they're replaced under warranty, then they autopsy them to see what failed. When they're repaired, they get improved parts to enhance reliability for all of the common problems they've seen, even if some of those don't apply to each computer they're rebuilding. Software is modified too to address any known emissions or customer satisfaction issues. Each time a change is made, that computer gets a new part number.

Often the first revision gets the part number changed from ending with an "AA" to an "AB" or "AC". More significant changes might get it a new sticker with a new part number. That's why we just go by the application, not the exact part number.

There's one major clinker you should be aware of, but I don't know if it applies to your truck. I do know that with '92 and '93 Dynastys, and '94 and '95 Intrepids, you can create a crank / no-start by installing the wrong used Engine Computer or Body Computer. The easiest to remember is if your truck has the factory-installed optional anti-theft system, you can use any Engine or Body Computer for that application, but you absolutely must not take those out and install them into a vehicle that doesn't have the factory anti-theft system.

If your vehicle doesn't not have factory anti-theft, you have to install a used computer from another vehicle that also did not have it. That can be impossible to know with salvage yard vehicles.

What happens is the anti-theft programming lives in those two computers. When you get a new or rebuilt computer from the dealer, they always come without that anti-theft programming. If it is needed, it will learn it from the other computer the first time you turn on the ignition switch. After that, the anti-theft programming can not be undone, and that computer will only work in a vehicle with anti-theft.

If your vehicle does not have anti-theft from the factory, and you plug in a computer that has it, it will teach that programming to the other computer when you turn the ignition switch on. At that point both of them are waiting for the disarm signal, but it's never coming, so both computers will cause a crank / no-start. The only fix then is to remove both of them at the same time, install two new computers at the same time, then start the engine.

People get in trouble when they throw random parts at a problem, and this one is a doozy. Crank / no-starts are fairly common, but Chrysler has had extremely little trouble with their Engine Computers. Those should be the last thing on the list of suspects. If the vehicle doesn't have anti-theft, and you pop in a used Engine Computer to try, and it has anti-theft software activated, it is going to teach that to the Body Computer. Engine still doesn't start, so you put the old Engine Computer back in. Now, the next time you turn on the ignition switch, the Body Computer that just activated the anti-theft software is going to teach it to your original Engine Computer. Now you have two computers that must be replaced, plus the original cause of the crank / no-start.

Again, if your truck has anti-theft, you can put in any computer for that application.

If you do not have anti-theft, what I would do is get the used Engine Computer and the Body Computer from the same truck in the salvage yard. Plug both of them into your truck at the same time, then see if the engine starts and runs. If it does not, those used computers have anti-theft software activated, and neither can be left in. I'd mark them to that effect for future reference. Your original Body Computer is sitting on the bench and is still safe. If the engine does start and run, neither computer has anti-theft software activated, and you can save the Body Computer in case you need it in the future.

There's one more thing to be aware of that can avoid lots of possible frustration. Very often the engine will not start and run when there is actually nothing wrong. Some people get all "wrapped around the axle" replacing parts and tearing things apart, ultimately causing a lot more problems. This applies to every Chrysler product when memory power is lost to the Engine Computer. That can be from nothing more than replacing the battery or disconnecting one cable to do other repairs, and it will occur when unplugging the Engine Computer. All the learned data will be relearned as soon as you start driving again, except for "minimum throttle". The Engine Computer needs to see a very specific set of conditions to know when your foot is off the accelerator pedal, then it puts the throttle position sensor's signal voltage in memory. From then on, any time it sees that voltage, it knows it has to be in control of idle seed. Before that is learned, idle speed will be too low. You may need to hold the accelerator pedal down 1/4" for the engine to start and stay running. You won't get the nice idle flare-up to 1500 rpm at start-up, and it will tend to stall at stop signs.

To meet the conditions for minimum throttle to be learned, drive the truck at highway speed with the engine warmed up, then coast for at least seven seconds without touching the pedals.
Was this
answer
helpful?
Yes
No
+1
Saturday, March 30th, 2019 AT 11:14 PM
Tiny
WILDMTBILL
  • MEMBER
Your a wealth of information, thanks for sharing. I'm not able to find any resistors locally in Helena Mt yet, but will let you know if it tricks the computer when I do
Was this
answer
helpful?
Yes
No
-1
Sunday, March 31st, 2019 AT 7:06 AM
Tiny
WILDMTBILL
  • MEMBER
So we installed the external regulator and left the two wires coming from the computer hanging lose, still charging at high 17's and the voltage meter in the dash is also working, maybe because there no resistance between the wires yet?
Was this
answer
helpful?
Yes
No
Sunday, March 31st, 2019 AT 5:33 PM
Tiny
CARADIODOC
  • EXPERT
Do you have the regulator bolted to a paint-free point on the body? The case is the third terminal. The regulator senses system voltage between the blue wire and ground, which is the housing.

If that isn't the issue, use a digital voltmeter with the negative probe clipped to the battery's negative post, then measure the voltages on the blue wire, green wire, and case of the regulator, and tell me what you find.

Also try unplugging the regulator, then see what you get for charging voltage.
Was this
answer
helpful?
Yes
No
Sunday, March 31st, 2019 AT 8:14 PM
Tiny
WILDMTBILL
  • MEMBER
Sorry took so long to get back, everything was grounded and good, with the external regulator installed was still overcharging, the only wire I hadn't done anything with was the positive cable. Started checking it and found that the positive battery clamp was corroded on the inside where the cables went into it. $5.00 and a new clamp everything works fine.
Thanks again for your help and sharing your knowledge.
Was this
answer
helpful?
Yes
No
Sunday, April 7th, 2019 AT 10:56 AM
Tiny
CARADIODOC
  • EXPERT
Dandy. Thank you for posting the solution. Where were you measuring that 17 volts, at the battery or on the large output stud on the alternator? If you found that at the alternator, your discovery would make sense. The voltage regulator is seeing only battery voltage of 12.6 bolts, so it runs the alternator harder to try to get that voltage back up. With a corroded cable going back to the battery, the alternator's output voltage and current never got to the battery and it never got to the Engine Computer / voltage regulator to be seen by them. That's why the regulator saw 12.6 volts and not 17 volts. It kept driving the alternator harder and harder until it was developing 17 volts.

Think of this as a municipal water pump developing 100 psi, but in the output pipe, the valve was closed. The pressure gauge at the water tower shows only 50 psi, so the pump has to run harder to build more pressure. As it runs harder and harder, it builds more and more pressure, but with the closed valve, that higher pressure is never seen at the water tower or pressure gauge.

Happy to hear you solved it.
Was this
answer
helpful?
Yes
No
+1
Sunday, April 7th, 2019 AT 7:31 PM

Please login or register to post a reply.

Sponsored links