2G Help Me Diagnose my Charging System Issue

[shinzon]

So I recently picked up ECMLink for my '96, and I noticed that the computer is only seeing 13.8v with the engine running at idle. I checked battery positive and negative with a multimeter and saw the same 13.8 volts that the computer is reporting.

I checked for voltage between the positive pole on the alternator and battery negative and found 14v. Lastly, I did a voltage drop test between the positive pole of the alternator, and battery positive and found 0.8v difference, which would roughly make the difference between 13.8v and 14.5v.

The alternator in question is a many years old re-manufactured Canadian spec 90amp alternator. Battery is still in the factory location, and the battery/starter harness is brand new OEM.

I was originally thinking the voltage regulator was getting flaky, but the voltage drop test has me thinking there's something wrong in the wiring between the alternator and the battery.

Any advice?

 

[streetraceboi]

I wish I had that kind of voltage on my car

 

[barza21]

With the car powered off, you could disconnect the alternator cables at the junction box and at the alternator, then measure the resistance. If its something considerable, then I would say start looking at those wires and terminals very hard.

 

[shinzon]

With the car powered off, you could disconnect the alternator cables at the junction box and at the alternator, then measure the resistance. If its something considerable, then I would say start looking at those wires and terminals very hard.

Hmm, I'll give that a shot in the morning.

Would anyone happen to know the part number for those two sets of white wires that run between the junction box and the positive battery terminal? I noticed some corrosion where the wire meets the ring terminal. I can't seem to find them on CAPS.

 

[chrysler kid]

Gross, factory cables. They had to make 2 of them instead of just using a heavier gauge wire

Just use a heavier gauge wire

 

[shinzon]

Took the car out for an extended drive, got it good and hot, and I saw voltage get as low as 12.7v at idle. That would seem to suggest heat is a factor...

Gross, factory cables. They had to make 2 of them instead of just using a heavier gauge wire

Just use a heavier gauge wire

What gauge would you recommend?

 

[shinzon]

Ha, we've got an answer!

Took the car out this morning for a quick run to the store to get some logs on what the alternator would do on a completely cold motor. Stayed at roughly 13.8v for the trip there, but on the way home my brake light, check engine light, battery light, and A/T temp light came on, and I also noticed that my temperature gauge climbed to 3/4.

Computer was reporting that the alternator was putting out nearly 19v. Limped the car home, and confirmed with a multimeter.

Voltage regulator must be bad...

Still, that 0.8v drop between the alternator and battery bothers me. I'll probably replace those wires from the junction box as well.

 

[miguelmcv]

To the OP.
Your drop voltage lies from the alternator to the battery and not from the battery to the starter, The new terminal cable wont fix the problem.
You have to change the 2 white wires coming from the positive of the alternator to the fuse box then from the fuse box to the battery.
Just change to a thicker wire and you should be ok with the drop voltage.

You have a second problem besides the drop voltage, your alternator is going out, it’s not because of the drop but because not charging when hot, it’s a common problem with dsms alternators, just change it.

Quick advice, try to search for an OEM alternator, most of aftermarkets aren’t that good,. A good oem alternator will hold a lot better even when die a repair is way way better then a remanufactured,.

Good luck

 

[shinzon]

Swapped out the alternator for another 90amp and it's putting out 14.5 volts at the terminal on the alternator, and roughly 14.38 at the battery. I'm now measuring a 0.1v drop between the positive terminal on the battery, and the positive terminal on the alternator... better than 0.8v but not perfect.

Turned out to be a simpler fix than I was expecting.

To the OP.
Your drop voltage lies from the alternator to the battery and not from the battery to the starter, The new terminal cable wont fix the problem.
You have to change the 2 white wires coming from the positive of the alternator to the fuse box then from the fuse box to the battery.
Just change to a thicker wire and you should be ok with the drop voltage.

There are two sets of two white wires that run from the junction box to the positive terminal. Without digging out a wiring diagram, I assume one set is for the alternator, and one set is for main power to the junction box. What gauge wire would you recommend I replace these wires with?

Thanks for the help everyone; just about have this one licked.

edit: as an aside, I saw this on the DSMFAQ and it's dumb as hell:

...It has been reported that Mitsu alternators rotate in the opposite direction of most alternators, making aftermarket units problematic. Read about this here. For this reason it is possible that aftermarket parts may not ventilate correctly, leading to premature failure.

https://www.dsmfaq.com/q/340

First, 4G63 spins clockwise like most engines, especially those that can trace their linage back to longitudinally mounted RWD platforms (like the 4G63). Second you can look into the alternator and see that the cooling fan blades are oriented in such a way that it will move air when it's spinning clockwise. No idea how something like that got on the FAQ.

 

[shinzon]

Well, hell, it's not solved after all. Took the car out for an extended drive and and it's back to outputting 13.5 volts. I had the old alternator tested and it failed, so it would seem I got a bad alternator out of the box, likely murdered by the first heat-cycle since it was outputting 14.5v after I installed it. I can't think of anything else that would cause 13.5v at the alternator's output terminal--it's clearly getting signal from the ECU to turn on (or rather, not getting signal to turn off is how I think it works on DSMs) if it's outputting 13.5v, it's grounded internally to the case so it likely has a good ground, and I inspected the condition of all the wires and they all seemed fine. Am I missing anything?

It's Easter and I'm over messing with it for the day. I'll have to pick a new alternator up tomorrow.

 

[pauleyman]

To the OP.
Your drop voltage lies from the alternator to the battery and not from the battery to the starter, The new terminal cable wont fix the problem.
You have to change the 2 white wires coming from the positive of the alternator to the fuse box then from the fuse box to the battery.
Just change to a thicker wire and you should be ok with the drop voltage.

You have a second problem besides the drop voltage, your alternator is going out, it’s not because of the drop but because not charging when hot, it’s a common problem with dsms alternators, just change it.

Quick advice, try to search for an OEM alternator, most of aftermarkets aren’t that good,. A good oem alternator will hold a lot better even when die a repair is way way better then a remanufactured,.

Good luck

Oem alternator FTW. I never replaced my alternator all the way back from 1990. When it went out I had the original oem rebuilt. My guess is the rebuild companies use cheap voltage regulators as that was the most common problem

 

[shinzon]

Oem alternator FTW. I never replaced my alternator all the way back from 1990. When it went out I had the original oem rebuilt. My guess is the rebuild companies use cheap voltage regulators as that was the most common problem

Yeah, my '97 still has its factory alternator, 21 years and 130k later; speaks volumes for the quality considering its proximity to the turbo. I really wish I would have had my OEM alternator rebuilt when it died on my '96 about 7 years ago.

 

[miguelmcv]

Yea OEM are the best, rebuilding takes a risk if cheap parts are used but at least you will have the rest (not replaced) of the parts OEM, but if you get a reman alternator all 100% of the parts are cheap, obviously there are some rare ocations, some companies use oem quality but to find those companies is fricking hard.

To the op
A 3 gauge wire would be a good choice to swap with.

 

[chrysler kid]

Make sure the belt is tight, then make sure it's tight again, then tighten it again

And yes part store alternators for cars this age are hit and miss, I went through 2 of them, the third one works but it's got a stripped bolt hole. So yeah


3 gauge wire, haha. I've literally never seen a gauge of wire sold in anything other than even numbers

Just use 4 gauge wire. You can buy a 56inch strand for 10.99 at the auto parts store

 

[jim95redgsx]

Well, hell, it's not solved after all. Took the car out for an extended drive and and it's back to outputting 13.5 volts. I had the old alternator tested and it failed, so it would seem I got a bad alternator out of the box, likely murdered by the first heat-cycle since it was outputting 14.5v after I installed it. I can't think of anything else that would cause 13.5v at the alternator's output terminal--it's clearly getting signal from the ECU to turn on (or rather, not getting signal to turn off is how I think it works on DSMs) if it's outputting 13.5v, it's grounded internally to the case so it likely has a good ground, and I inspected the condition of all the wires and they all seemed fine. Am I missing anything?

It's Easter and I'm over messing with it for the day. I'll have to pick a new alternator up tomorrow.

Actually, this is fairly normal for a 2G. The 2G ECU will adjust voltage as needed to keep the battery charged. When the car is first started, the alternator will charge at 14.3 - 14.5 volts to replace the current lost during the engine start. As the battery comes up to full charge, the voltage will drop to between 13.2 and 13.8 volts (depending on load - fans, lights, radio, heater, etc).

The ECU gets its reference voltage from the headlight circuit to adjust voltage. It uses a duty cycle signal to turn the voltage regulator in the alternator on and off.

Jim

 

[miguelmcv]

3 gauge wire, haha. I've literally never seen a gauge of wire sold in anything other than even numbers

Just use 4 gauge wire. You can buy a 56inch strand for 10.99 at the auto parts store

Sorry my bad, 4 gauge it is.

 

[barza21]

Like I mentioned before, if you wanna completely rule out the wiring, you are going to have to disconnect at the fuse box and at the alternator. Then "ohm out" the wires. One of the other members mentioned something about the ECU controlling the voltage. I cannot speak to this with a 2g 4g63 (you're gonna have to research), but with a 420a ECU, the voltage regulator is inside the ECU.
http://www.dsmtuners.com/threads/where-is-the-voltage-regulator.391464/

When I upgraded and relocated my alternator, I used 4awg from the alternator to the fuse box. 14.3+ at idle on a 1g 4g63t.

As far as to why your alternator died so quickly, I don't know, but I have read several instances that find the proximity to the downpipe (heat) as the culprit (one of the reasons why I relocated the alternator to the back of the block). Or like you said, it could've been bad out of the box.

 

[shinzon]

Make sure the belt is tight, then make sure it's tight again, then tighten it again

Just use 4 gauge wire. You can buy a 56inch strand for 10.99 at the auto parts store

Might be over tight, actually. It's squealed a bit on cold start since I reassembled everything after I did some work inside the timing cover a couple of months back. I generally eyeball about a quarter inch of deflection, but I took it a bit tighter than that this time.

Thanks for the recommendation on the wire. I'll look into replacing it. Why did Mitsu use two wires per lead; just to save money by not having to use a heavier gauge wire?

Actually, this is fairly normal for a 2G. The 2G ECU will adjust voltage as needed to keep the battery charged. When the car is first started, the alternator will charge at 14.3 - 14.5 volts to replace the current lost during the engine start. As the battery comes up to full charge, the voltage will drop to between 13.2 and 13.8 volts (depending on load - fans, lights, radio, heater, etc).

The ECU gets its reference voltage from the headlight circuit to adjust voltage. It uses a duty cycle signal to turn the voltage regulator in the alternator on and off.

Jim

Very interesting, I didn't know this. To test this, with the engine hot, I turned on all of the electrics I could: heater, rear defroster, wipers, highbeams, etc. Watching the live data in ECMLink, I saw the voltage drop from 13.5v to 12.7v, so this would seem to suggest that the alternator is not able to keep up with the increased load. Thoughts?

When I started the car this morning, I logged 14.1v and it held steady regardless of load, perhaps suggesting that heat is playing a factor in its inability to meet demand.

Like I mentioned before, if you wanna completely rule out the wiring, you are going to have to disconnect at the fuse box and at the alternator. Then "ohm out" the wires. One of the other members mentioned something about the ECU controlling the voltage. I cannot speak to this with a 2g 4g63 (you're gonna have to research), but with a 420a ECU, the voltage regulator is inside the ECU.

From what I was taught, for anything that carries more than an amp of current flow, ohming a wire is inferior to doing a voltage drop test on the circuit while it's fully loaded. Since a meter outputs a tiny amount of current when conducting an ohm test (the meter is acting like the battery), and since that amount of current flow is orders of magnitude less than what the wire would see at full load, it is possible for there to be a few stands of good copper left in a corroded wire through which your meter will find a good path, but when loaded that wire cant flow the full amount of current.

In the service manual, Mitsu actually suggests doing a simple voltage drop test (only time I've ever seen one recommended in the manual) between the output terminal of the alternator, and battery positive with the headlights on high beam and the engine held at 2500rpm. Specs for this test are <=0.5v drop; I read 0.30v.

There's a more complicated output voltage test where you add an amp meter inline and test voltage drop at a certain amp load, but I don't have another multimeter to attempt it.

 

[chrysler kid]

Might be over tight, actually. It's squealed a bit on cold start since I reassembled everything after I did some work inside the timing cover a couple of months back. I generally eyeball about a quarter inch of deflection, but I took it a bit tighter than that this time.

Thanks for the recommendation on the wire. I'll look into replacing it. Why did Mitsu use two wires per lead; just to save money by not having to use a heavier gauge wire?



Very interesting, I didn't know this. To test this, with the engine hot, I turned on all of the electrics I could: heater, rear defroster, wipers, highbeams, etc. Watching the live data in ECMLink, I saw the voltage drop from 13.5v to 12.7v, so this would seem to suggest that the alternator is not able to keep up with the increased load. Thoughts?

When I started the car this morning, I logged 14.1v and it held steady regardless of load, perhaps suggesting that heat is playing a factor in its inability to meet demand.



From what I was taught, for anything that carries more than an amp of current flow, ohming a wire is inferior to doing a voltage drop test on the circuit while it's fully loaded. Since a meter outputs a tiny amount of current when conducting an ohm test (the meter is acting like the battery), and since that amount of current flow is orders of magnitude less than what the wire would see at full load, it is possible for there to be a few stands of good copper left in a corroded wire through which your meter will find a good path, but when loaded that wire cant flow the full amount of current.

In the service manual, Mitsu actually suggests doing a simple voltage drop test (only time I've ever seen one recommended in the manual) between the output terminal of the alternator, and battery positive with the headlights on high beam and the engine held at 2500rpm. Specs for this test are <=0.5v drop; I read 0.30v.

There's a more complicated output voltage test where you add an amp meter inline and test voltage drop at a certain amp load, but I don't have another multimeter to attempt it.

You're over thinking it. Dc current has a natural decay.

In my 98 fbody the body control module just adds in more wiring, then the drivers side door runs the power to the passengers side window.

At any given time being routed through the body control module with the windows being rolled down I can view up to 1~2 volt drop across the entire interior harness when using my power windows


You have to remember link is getting it's power signal from the OBD2 port which is already wound into the interior engine harness after being fed from the battery. By the time you factor in the amount of drawn power by each accessory you can view a large standard fluctuation in the interior wire harness compared to what the battery is seeing directly from the alternator

What about the injectors and engine pieces you say, well that's why we use relays to send power further distances with stronger current. Which is the reason so many people add in a fuel pump relay directly to the battery, voltage drop and voltage decay

It is far more efficient to monitor voltage at the battery

You must be logged in to view this image or video.

 

[miguelmcv]

I assume you did confirm the voltage at the ecu by checking with a multimeter, because the ecu ( ecmlink ) sometimes doesn’t report exactly the right voltage, that’s why ecmlink has the ability to adjust the voltage shown on the log.

In my case, ecmlink was showing higher voltage then what the actual voltage was at the battery and I had to adjust link to the designated voltage I was getting , so in a fuw words you have to configure ecmlink to the right voltage you have in the car.

 

[barza21]

Heat will directly affect the resistance of winding inside the alternator. If it's getting too hot, then that is the reason why the output may be lower as temperature increase. I still believe that you should investigate, replace or upgrade the alternator to fuse box wiring. A .8v drop is considerable on a 12v system.

 

[shinzon]

Swapped out for another alternator, and it seems to be outputting a steady 14.1-14.5 at the alternator output terminal, even after an extended drive to get it good and hot. Might have it fixed this time. Thanks for the help everyone.

You're over thinking it. Dc current has a natural decay.

Ah, I get you. Appreciate it, man, you're always informative.

I assume you did confirm the voltage at the ecu by checking with a multimeter, because the ecu ( ecmlink ) sometimes doesn’t report exactly the right voltage, that’s why ecmlink has the ability to adjust the voltage shown on the log.

In my case, ecmlink was showing higher voltage then what the actual voltage was at the battery and I had to adjust link to the designated voltage I was getting , so in a fuw words you have to configure ecmlink to the right voltage you have in the car.

I did not, actually. I read voltage from battery positive to negative with a multimeter, however there was only a minute variation between what Link was reporting and what I was measuring, so I guess I got away with it. I also checked it directly off the alternator output terminal. I'll keep it in mind to check at the ECU next time this happens... and there will be a next time, since it's an aftermarket alternator.

Heat will directly affect the resistance of winding inside the alternator. If it's getting too hot, then that is the reason why the output may be lower as temperature increase. I still believe that you should investigate, replace or upgrade the alternator to fuse box wiring. A .8v drop is considerable on a 12v system.

Oh absolutely; I noticed a bit of the same green corrosion that murdered my battery harness and it appears to extend under the insulation--these wires need to be replaced.

At this point, I'm only reading a .30v drop between the alternator output terminal and battery positive, and the service manual says less than .50v is fine, but I'm just going to replace the wires for peace of mind.

 

[shinzon]

Oh, and for some bonus content. For at least 10 years I've been looking for an intake leak. It sounded like it was coming from under the intake manifold, but I could never locate it. I rebuilt the throttle body, replaced the intake manifold gasket, all the vacuum lines, injectors seals, and confirmed the brake booster line wasn't leaking--it still sounded like air was rushing out from under the throttle body side of the manifold. I just simply could not find that leak and it frustrated me to the point where I gave up on it.

Since my flex section has been leaking for ages, I was thinking that it was the 30 year old, never rebuilt turbo leaking air passed the seals and into the exhaust. Well, I was wrong.

I did a BLT yesterday and I finally found it. It wasn't coming from under the manifold; it was coming from the charcoal canister. Plugged up the emissions nipples at the throttle body and the car is running better than it has since time out of memory. Unbelievably pleased. Should have deleted that stuff years ago...