Ammeter acting up

[Dave WM]

so after all my 70 mile trips and all good, I start out on my 400 mile trip (800 round) about 200 miles it I notice my headlights getting much brighter (more white than yellow) while the ammeter is acting up again....

so we really do have a excess voltage causing excess charge current. Here is the symptoms (I had 200 miles of night time driving to panic about it, lots of observation going on).

there was a pattern, high charge for a couple seconds, between 2-3 seconds I would guess, followed by nearly as much of a discharge for about the same amount of time. Charge then Discharge, then A.OK for a variable interval of seconds to minutes..

My guess is the alternator was at times overcharging causing the VR to kick down the field winding current throwing alt into a discharge as soon as the intermittent overcharge would stop. How would alternator actually do this? again a guess if one or more of the diodes in the alt was failing open, then I presume the overall output would be less, the VR tries to compensate by increasing the field current at which point the intermittent open diode begins conducting again, now with excessive field current, suddenly the over charge, the VR sees this, cuts the current  and then the diode goes open again now with reduced field current therefore reducing alt output going into discharge mode. the actual cycle starting on over charge.

So the diode is mostly open but will momentary conduct throwing the charge cycle into the charge/discharge mode. Just a guess. I was able to get a reman alt on the road, swapped it and the problem ceased. I still get a little jumpiness at idle but its only a fraction (maybe a few amps tops) and only while sitting at idle a completely different situation. will go back and review the my connections as the swap was done in the field with not my best tools.

again the pattern was Normal reading (center) then a sudden charge for seconds of 30-40 amps, followed by discharge close the same amperage and time the return to center on the scale. pause for seconds to minutes to hour...random time between events, but each event lasting approx. 4-6 seconds (2-3 charge, 2-3 discharge).

due to circumstances beyond my control I was NOT able to have them test the alternator that I replaced. so will never have a definitive test to know if that was indeed the issue.

I know typically SS devices fail either open or shorted, but perhaps its not the diode but rather the connection of the diode? Its some kind of top hat looking diode soldered into a mount (multiple diodes that is).

 

 

 

 

 

 

 

 

 

Edited January 12, 2021 by Dave WM

[Dave WM]

another data point, the OE Nissan green fuse link looks like the plastic heat shrink on the female spade terminals cooked itself off. Ordered some new ones. The wire is still intact but I suspect it got hot...

I have searched on intermittent over follow by under charge then return to normal in google, nothing came up. there was something on shorted windings but it was a blocked site for me.

Edited January 13, 2021 by Dave WM

[Captain Obvious]

When I get a chance, I'll take another look at the wiring diagram and see if I can draw any conclusions.

I still don't think you were really pushing 60A into anything. That's welding current. 

[Dave WM]

Those headlights were getting quite bright. I know it was not just a meter twich now.

[SteveJ]

I would suspect that the voltage regulator was messing up if you were seeing voltage spikes.

The symptoms you described would be along the lines of a loss of sensing (maybe from the VR overheating) causes current injection into the stator of the alternator, kicking up the voltage. However, shortly after that, sensing resumes, the system stops injecting current into the stator, dropping the voltage of the alternator (maybe too much).

Is your charging system stock, or do you have an internally regulated alternator?

Edit: I forgot to add that you could get something like this to get a voltmeter in your car: https://www.amazon.com/gp/product/B07VZFSY4SInsert other media

Edited January 13, 2021 by SteveJ

[Dave WM]

I discounted the external VR as an issue (stock system) since I got the exact same behavior after switching to a backup new VR. I could accept a bad VR even a bad second VR but for the  identical pattern, "ok, then high charge then low charge then return to normal" exhibited by both.

So since I know its not a meter malfunction (much brighter head lights during the charge event) and the behavior is the same regardless of a different VR, that leaves me with 3 options

Bad battery (somehow influencing the charge/discharge cycle).

Wiring (I checked for good ground and good connections thinking the VR's may be sensing a low voltage) but the recurring pattern seems odd for that.

Alternator somehow confusing the VR's, this was the most likely although hard to figure the exact fault. My assumption is the alt was having an intermittent fail, be it loss of internal ground of the diodes, intermittent open of one or more of the diodes, winding faults of the stator or field. Again not exactly how but it seemed the next logical thing to replace.

After replacing the alt I made the 400 mile return leg with out a recurrence of the "charge/discharge/back to ok" cycle. there remains some very small events I noticed at idle but it was a very different event, no pattern and a much smaller scale (amps not 10's of amps flux), and more of a ammeter needle vibration than a sustained 2/3 seconds long event. this event is not constant either, just every now and then when I am sitting at  stop light the ammeter would vibrate a bit. I suspect something is still not 100% correct, I should have just rebuilt the OE part with some bearings and a new diode pack (I kept the OE alt from years ago when it shorted the diode, and I replace the alt back then, did not give it back as core). So the presumed bad alternator that I replaced on the trip was the autozone one with about 5+ years and about 25k miles on it. Maybe that's about what to expect with a reman...

I did order some new OE Nissan green fuse links, since the one in there looks like it got hot.

and after going out to take a ride around the block to test again, I have a FLAT TIRE, thanks to a well place nail, argh...

 

Edited January 13, 2021 by Dave WM

[SteveJ]

Ah, since you seem to have eliminated the VR from potential causes, I would think it's the stator or something related to the stator. 

[Dave WM]

clamp on ammeter to test with

 

[Captain Obvious]

Cool. So are you going to drive around for a while with that attached?

I still haven't looked closely at the wiring diagram to see how things are really connected. I will do that when I get a chance.

[SteveJ]

Since you kept talking about it, @Captain Obvious, I decided to take the plunge. 😉 While @Dave WM is trying to see if it's just the meter, the description of the behavior during the trip tells me it is the charging system itself acting up. 

By the way, Dave, I like the current monitoring setup. The ammeter I linked would work the same way. As long as the place you clamp the meter feeds the white wire in the wiring harness, you should be good. You may want to get a voltmeter that plugs into your cigarette lighter, too. The lights getting brighter tells me your voltage is going up during the issue. I'm betting you'll see the voltage go up when the ammeter says you have current flowing to the battery. The ammeter behavior seen in your video looked typical to me. You have the voltage drop from starting the car, so the alternator is replacing the battery's loss. You didn't have to crank long, so it doesn't take long for the battery to get back to a sufficient level, dropping the alternator output.

Let's consider the operation of the VR as laid out in the wiring diagram.

As the neutral of the alternator builds sufficient voltage, it energizes its coil in the VR.
The contact that closes provides a ground for the second coil that is powered by the battery. Note that this second coil is wired in series with a resistor, probably to drop off the voltage and cause the coil to de-energize should battery voltage drop from discharging.
When coil 2 is de-energized, the switched voltage is going straight to the field wire on the alternator (full field voltage).
When coil 2 is energized, the switched voltage goes to two resistors (voltage divider) to reduce the voltage going to the alternator field, dropping the alternator output.
From reviewing http://www.secondchancegarage.com/public/83.cfm it's obvious for the mechanical VR the contacts would constantly open and close. With a solid state VR, you have discrete components handling the same function in a similar manner.

Here's the thing. If the VR is not dropping the voltage going to the field wire, you're going to see increased output from the alternator. Keep in mind that a flaky ground at the VR could present itself in a similar fashion. I am assuming you're running a solid-state VR. Also, a flaky neutral could also cause the problem. It could also be a winding issue in the alternator if the grounded leg has a winding shorting out close to the neutral. (I don't think you would have an intermittent short on a winding, though. I don't have enough practical experience to say one way or the other.)

Ideally, it would be great to monitor the voltage on the white/black wire with an oscilloscope. You would expect it to fluctuate too much for a typical voltmeter. Monitoring the voltage to ground on the neutral might reveal good data, too.

Okay, now I need to find all of my O-scope probes so I can see if I can configure something for an educational video.

 

[Dave WM]

13 hours ago, Captain Obvious said:

Cool. So are you going to drive around for a while with that attached?

I still haven't looked closely at the wiring diagram to see how things are really connected. I will do that when I get a chance.

yes that is the plan I just have to make a better hookup to the meter. I was also going to try and simulate the intermittent by poking around the wiring harness, tapping the VR etc to see if I can locate. This setup has a better resolution and I can see it while poking under the hood.

[Dave WM]

12 hours ago, SteveJ said:

Since you kept talking about it, @Captain Obvious, I decided to take the plunge. 😉 While @Dave WM is trying to see if it's just the meter, the description of the behavior during the trip tells me it is the charging system itself acting up. 

By the way, Dave, I like the current monitoring setup. The ammeter I linked would work the same way. As long as the place you clamp the meter feeds the white wire in the wiring harness, you should be good. You may want to get a voltmeter that plugs into your cigarette lighter, too. The lights getting brighter tells me your voltage is going up during the issue. I'm betting you'll see the voltage go up when the ammeter says you have current flowing to the battery. The ammeter behavior seen in your video looked typical to me. You have the voltage drop from starting the car, so the alternator is replacing the battery's loss. You didn't have to crank long, so it doesn't take long for the battery to get back to a sufficient level, dropping the alternator output.

Let's consider the operation of the VR as laid out in the wiring diagram.

As the neutral of the alternator builds sufficient voltage, it energizes its coil in the VR.
The contact that closes provides a ground for the second coil that is powered by the battery. Note that this second coil is wired in series with a resistor, probably to drop off the voltage and cause the coil to de-energize should battery voltage drop from discharging.
When coil 2 is de-energized, the switched voltage is going straight to the field wire on the alternator (full field voltage).
When coil 2 is energized, the switched voltage goes to two resistors (voltage divider) to reduce the voltage going to the alternator field, dropping the alternator output.
From reviewing http://www.secondchancegarage.com/public/83.cfm it's obvious for the mechanical VR the contacts would constantly open and close. With a solid state VR, you have discrete components handling the same function in a similar manner.

Here's the thing. If the VR is not dropping the voltage going to the field wire, you're going to see increased output from the alternator. Keep in mind that a flaky ground at the VR could present itself in a similar fashion. I am assuming you're running a solid-state VR. Also, a flaky neutral could also cause the problem. It could also be a winding issue in the alternator if the grounded leg has a winding shorting out close to the neutral. (I don't think you would have an intermittent short on a winding, though. I don't have enough practical experience to say one way or the other.)

Ideally, it would be great to monitor the voltage on the white/black wire with an oscilloscope. You would expect it to fluctuate too much for a typical voltmeter. Monitoring the voltage to ground on the neutral might reveal good data, too.

Okay, now I need to find all of my O-scope probes so I can see if I can configure something for an educational video.

 

its a mechanical VR, I have not had a recurrence of the orig problem since replacing the alt. I wish I had kept the core to evaluate. Now the only issue is an occasional flux, happens at idle and after the car has warmed up some and there is a load beyond just the EFI, like the fan or light. No more wild swings that last or flare up of voltages (head lights flare up) since the change.

My SWAG is after warm up my idle is less stable, prob due to leaner mix, I have it set up pretty darn lean (pulling the dip stick can effect the idle, remove the oil filler and it dies in a second). I may richen up the idle circuit just so its a bit smoother sitting at a traffic light. But anyway the theory is, unstable idle means unstable alt speed, add a bit of a load and the VR will have a time switching the field current quickly enough. The ammeter needle does seem to go with the idle changes. I am slowly noticing these things as I generally don't sit and stare at the ammeter, but of course since the orig problem (flaring up lights etc...) I have become much more aware of it.

I suspect the alt was the whole problem, I will be reinstalling the OE hitachi VR now just to confirm it was never the issue.

Edited January 23, 2021 by Dave WM