1.8t alternator / charging issue

[mdk]

My battery is not receiving any power from the alternator. The car starts and runs if I charge the battery separately but I do NOT have continuity between the lug on the alternator and + battery terminal. Concerningly, my multimeter briefly rings when I touch alternator lug to ground, but only for a second after a few minutes delay.

I'm trying to work out the wiring diagrams for the alternator and starter. Does anyone have a good visual or explanation of where the (2) ends (and (2) connectors) of the alternator harness go? It seems that there are (4) ends on the cable but ebahn only seems to show a single (2-sided) connection.





I'm going to try to find where it might be shorted, but any other suggestions would be greatly appreciated.

Thanks in advance

[mdk]

Ok, so I think I understand now. eBahn shows the 16,0 mm^2 (6 AWG) "sw" (black) cable run from starter to alternator. Also shown coming off the starter is the 2,5 mm^2 (14 AWG) red/black cable. This smaller conductor is part of the starter/alternator harness, but leads to the clutch switch.

Since the starter receives 12V, it seems like my problem must be with the 6 AWG connection between starter and alternator. Still open to other thoughts.

[mdk]

Well, now I'm further confused. I did replace my starter -- it was a while ago, but the car hasn't moved much since. Definitely possible that the charging issue began when the starter was replaced.

Does anyone have a picture of a correctly-cabled starter? Again, my starter is working, but googling provides some confusing results in terms of which lug is supposed to be in use. eBahn shows the cable connected at the '30' location, but I'm pretty sure mine was (and is) at the '15a' location.

Old starter for reference - note '15a'


New starter installed

[Turbo510]

That's an awful big cable to not have any continuity. I suppose for the time being, you can just run a new cable direct from the #30 Alternator terminal to the batter +.

[mdk]

Yeah, agreed, I must be missing something.

Is that cable fused in-line at all? (I know some other companies run fused links that just look like a cable ... but no mention of that in the wiring diagram here)

Temporary jumper is a great idea though, thanks!

[EuroxS4]

Put on lead to ground the other on the threaded connection on the alternator with engine running. If you have 13.7-14.5 volts alternator is charging. IIRC alternator is grounded to the block. You have a signal wire at the back of the alternator and a charging wire that runs from alternator to starter and battery IIRC.

[mdk]

I disconnected the battery completely and checked continuity between battery+ cable and starter+; my meter doesn't ring. If I put it all back together and the battery is charged, the car starts (I did this right after checking continuity to confirm). Perhaps CCA is enough to overcome whatever is going on?

Once again I also confirmed that I have "brief" continuity between battery+ cable (disconnected from battery) and ground. The meter rings for a second but then stops as it shows resistance quickly climbing. Obviously this is an issue but I don't know what it points to. Bad starter? I guess I'll have to start disconnecting things until the problem goes away.

[Turbo510]

Once again I also confirmed that I have "brief" continuity between battery+ cable (disconnected from battery) and ground. The meter rings for a second but then stops as it shows resistance quickly climbing.

Are you sure that you are using the meter correctly? The battery + (positive) cable should be checked between the battery + clamp, and the alternator's B+ post, not ground. When you put your meter between the battery + cable and ground, you are charging up capacitors throughout the car, which is why your meter reacts to initial current, which goes away as soon as the caps reach your DMM's test voltage.

[mdk]

checked between the battery + clamp, and the alternator's B+ post

This test shows no continuity, same result when testing between battery + clamp and starter lug (even though I can start the car ... how is that possible then?)

When you put your meter between the battery + cable and ground, you are charging up capacitors throughout the car

Ok - good to know, so not necessarily indicative of a short to ground.

[mdk]

Ok. So I do have continuity between alternator B+ and battery (+) clamp, it's just at a higher resistance than my meter will ring for (0.4 ohm).

I'm now wondering it my voltage regulator could be bad. I replaced my alternator "recently" with a used unit to combat this charging issue, but my original was behaving similarly, which is why I went chasing the wiring. It appears my positive cable is indeed intact. I have voltage on both pins of the small alternator connector when ignition is on (so those seems intact).

An autozone alternator bench test is sufficient to test alternator AND voltage regulator in this car's case, correct?

[Turbo510]

I have to assume that any alternator tester puts a load on the output, and verifies that the Voltage is within acceptable limits.

[EuroxS4]

Your old starter was fine just needed a new ground as it was broken. 0.4ohms is not high on that big of a wire. Where did the used alternator come from and is it a oem unit or a reman?? Also is one a bosch and the other a valeo?? They are different internally in terms of voltage regulator and diodes an they do have a different ripple.

[Kevin C]

The 15A location on the starter is where you want the battery and the alternator connected to. The other large lug is the side of the solenoid that is energized when the starter is operating, if you connected either terminal to that you would have a problem. If it was the battery the starter motor would be spinning all the time. If it was the alternator it would only be connected when the starter was active. Neither is a good situation. That said I'm pretty sure that's not your issue.

What it looks like to me:

Since your meter only rings if its showing less than .4 Ω its reasonable to assume that the resistance from the alternator terminal to the battery is a value that is higher than that. So what would happen if you have exactly .4 Ω's of resistance in your alternator's power connection? Assuming that V=IR is still valid and your system load is 25 amps when the car is running: .4Ω*25A= 10V. That wire would have a 10 volt drop across it!!!

.4Ω is way, way too much resistance. To maintain a charge on your battery your alternator will need to put out an extra 10 volts. This is a problem for a couple of reasons.

1: The alternator would need to produce 22.6 volts just to maintain a charge at a very low load (25 amps).
2: The regulator on our cars is not externally referenced. It only looks at the voltage is the output terminal. If there is extra resistance it has no way of knowing its not charging the battery.

What should the resistance be? 6 AWG wire is listed at .3951 Ω per 1000 feet @ 20°C. I'm going to guess that your alternator wire is shorter than that so it should be less. For discussion lets say it's six feet. That should put your alternator to battery wire resistance at .0024Ω's. Your resistance reading is about 160 times too high.

I would measure the resistance with an ohm meter, something in that circuit has too much resistance. In my experience diagnosing voltage drops in these circuits is not intuitive what appears to be a small resistance number is actually very large when your dealing with highs currents. When in doubt do the calculations.

The normal drop on that wire at the alternators full output would be 150A * .0024Ω's = .36V.

A problem is that many meters don't measure low resistances very well. You can also measure voltage drops if the wire has a load across it.

[EuroxS4]

If your using a FLUKE they are accurate. Its the cheap meters that arent worth jack. I looked at the repair manual this morning.As per repair manual anything under 0.3ohms is considered in the norm anything over is no bueno. So correction on my part. Long story short a good meter is key. Also make sure to measure the leads first and subtract that readings from the total resistance.

[Kevin C]

If your using a FLUKE they are accurate. Its the cheap meters that arent worth jack. I looked at the repair manual this morning. As per repair manual anything under 0.3ohms is considered in the norm anything over is no bueno. So correction on my part. Long story short a good meter is key. Also make sure to measure the leads first and subtract that readings from the total resistance.

If you do the math on .3Ω at max current it doesn't make sense (it's a 150 amp alternator) . .3Ω x 150 A = 45 volts. To get the rated current output of the alternator it needs to put out ~57 volts. I don't think the people that came up with that figure thought this through. My suggestion is look a the circuit and do the math.

For my job to validate circuit performance I have I have had to deal with making accurate measurements of low resistance circuits. This is to certify that grounding systems will work in the event of a short and is a product certification requirement. There are a number of factors that come into play and while meter accuracy is one of them its not the only one. Typically a dedicated meter is used that is designed for this type of application. There are a number of complications including that the meters are a lot more expensive than your typical hand held. Without going into a ton of details its not practical in a shop or hobbyist environment.

What does work very well is to apply a load to the circuit and measure the voltage drops. A hand held meter can find the obvious large problems. Accurately measuring resistances under .1Ω requires a bit more work.

The resistance needs to be around .003 Ω not .3Ω

Some background: https://www.testequipmentdepot.com/m...ce-testing.pdf

A simple hack: https://www.instructables.com/Simple...Milliohmmeter/

Some Good Discussion: https://www2.theiet.org/forums/forum...threadid=22130

https://www.dataloggerinc.com/resour...nt-techniques/

[mdk]

Where did the used alternator come from and is it a oem unit or a reman?? Also is one a bosch and the other a valeo??

It is a Bosch off of a junkyard A4.

The 15A location on the starter is where you want the battery and the alternator connected to.

Since your meter only rings if its showing less than .4 Ω

If your using a FLUKE they are accurate.

Ok - all good to know. I am measuring resistance with a Fluke 117

.4Ω is way, way too much resistance.

I checked it again. I had removed the 2-pin connector from the back of the alternator, and having done so, the resistance from alternator (+) to battery (+) is 0Ω. When I re-connect that 2-pin connector, resistance on the positive cable then fluctuated between 0Ω and 0.4Ω -- usually at 0.2Ω. Not sure what to make of that.

Since it appears I do indeed have continuity on that cable, I'm going to try replacing my voltage regulator. Thanks so much for the comments so far -- feel free to keep them coming.

[john_gonzo]

My battery is not receiving any power from the alternator.

I skimmed the thread, didn't see slip rings mentioned. Alternator-new-slip-rings

[Kevin C]

It is a Bosch off of a junkyard A4.



Ok - all good to know. I am measuring resistance with a Fluke 117



I checked it again. I had removed the 2-pin connector from the back of the alternator, and having done so, the resistance from alternator (+) to battery (+) is 0Ω. When I re-connect that 2-pin connector, resistance on the positive cable then fluctuated between 0Ω and 0.4Ω -- usually at 0.2Ω. Not sure what to make of that.

Since it appears I do indeed have continuity on that cable, I'm going to try replacing my voltage regulator. Thanks so much for the comments so far -- feel free to keep them coming.

I tried to explain why normal meters have an issue measuring low resistances. If you look at your maters data sheet is gives an accuracy percentage and a number of bits of uncertainty. Your meter had something like .09% accuracy and two bits uncertainty. The last two digits are +2 bits unknown. If the reading is .4Ω It could be from .2 to .4, the meter does not resolve any better than that, that not including the percentage of accuracy. That's straight from its data sheet.

The next question is are you sure the alternators is good? John comment about the slip rings is valid.

If you have a known good alternator then its worth spending some time on understanding the connections, how to measure them and what resistances to expect.

[mdk]

I tried to explain why normal meters have an issue measuring low resistances. If you look at your maters data sheet is gives an accuracy percentage and a number of bits of uncertainty. Your meter had something like .09% accuracy and two bits uncertainty. The last two digits are +2 bits unknown. If the reading is .4Ω It could be from .2 to .4, the meter does not resolve any better than that, that not including the percentage of accuracy. That's straight from its data sheet.

The next question is are you sure the alternators is good? John comment about the slip rings is valid.

If you have a known good alternator then its worth spending some time on understanding the connections, how to measure them and what resistances to expect.

Ok thank you. Am I wrong to presume then that, given my resistance measurements sometimes showing 0Ω, my 6 AWG cable from alternator to starter to battery is good?

I replaced the voltage regulator with new (tried it on both of the alternators I now have); no change in behavior -- battery still discharging (slowly) while idling/driving.

I had the younger alternator (with brand new regulator) tested again, and it "passed" with 13.3 VDC output. The slip rings are visually in very good condition compared to the image in the linked thread, where they had worn through.

I guess the only other thing I could confirm is the state of the 2-pin alternator harness. I have some voltage between each pin and ground, which made me think they are ok, but maybe I need to dig into that further. Thanks in advance for any other info or confirmation.

[EuroxS4]

IIRC alternator connector should only have 1 signal wire going to it not 2. If it was apart make sure its in the correct designed position.

[Kevin C]

One question is how do you know its discharging at idle? Also the alternator bench test needs to tell you the amperage and the voltage to be useful. Another consideration is the regulator varies the voltage setting based on temperature. The colder the alternator the higher the voltage. A dropping voltage at idle can be the regulator adjusting for temperature. In any case at the battery you should be seeing something above 13.1 volts.

An amp meter on the battery can tell you a lot (charge / vs discharge).

At idle with the headlight on what are you seeing for a voltage across the battery terminals? The next test should be what is the voltage drop from the alternators positive terminal to the batteries positive terminal (same test, some accessories on to draw a load).

There are two terminals on the regulator. One signals the ECU to tell it how hard the alternator is working (allows the engine to adapt to the load) the other is the charge light that should provide a warning if the alternator is not charging. Neither is needed for the alternator to function. This is normal for the newer regulators, they retain some of the original regulator functions (charge light) but have an IC controlling it and are self starting.

The same regulator was used on the B7 A4. On the B8 A4 they went to the single wire system using a LIN bus and added much more sophisticated charge controls.

[mdk]

I’ve been trickle charging up to around 12.5 V, and then starting the car. Voltage drops 0.1 V every several seconds; I usually cut the engine by around 11 V. With the headlights on it of course drains faster.

Therefore the result of testing between terminals with headlights on depends on how long the car has been running for, but it will read between 10-12V.

I’m a little unclear on your second test, could you confirm what you mean by that?

[Kevin C]

Second Test: With the car running put one lead on the output terminal of the alternator and the other lead on the positive terminal of the battery. The voltage should be quite small.

[mdk]

Between alternator lug and battery positive terminal, while running, I see about 1.8-2V. Which I suppose, coming back around, makes sense given the resistance I sometimes saw.

So, how should I resolve? Run a new jumper between alternator and battery, or replace the oem cable?

[tiggerholley]

Between alternator lug and battery positive terminal, while running, I see about 1.8-2V. Which I suppose, coming back around, makes sense given the resistance I sometimes saw.

So, how should I resolve? Run a new jumper between alternator and battery, or replace the oem cable?

Did you ever figure it out? I'm having a similar issue with my 03 a4 1.8 with a valeo junkyard alternator that AutoZone benched good 6 times and a battery that's less than a year old and will not charge the battery and the battery light is on in the dash.