Alternator not charging; known good alternator and battery. Any Ideas?

[DiscoClay]

Some time in the last couple days my charging system stopped...charging. Everything else is working, starter, ignition, lights. etc.. Indeed, per Lucas POD SOP, NO warning lights have illuminated to indicate a battery/charge problem at all The only tell when the tach and speedo started doing their low-voltage dance this morning when I tapped the brakes. Alfred left me high and dry at the local coffeeshop. I got it jumped and made the short trip home.

Here is the lay of the land:

• The battery was replaced less than a month ago.
• The battery dropped down to 9.75v.
• I pulled the positive cable while the engine was running.. to do the old fashioned alt-check... the engine did something odd:
  • attempt 1 (battery @9.75v): the engine immediately died: Making me immediately think the alternator was bad (it *IS* the original)
  • attempt 2 (battery externally charged, via trickle-charger, @11.6v): engine started and ran, when I pulled the positive cable.. the engine continued to run for about four (4) seconds and then sputtered (and almost died).. i quickly put the lug back onto the + post and the engine recovered.
    • I tried this a few times; the engine continues to run for a few seconds (high idle speed only) and then sputters and dies.
• I swapped the alternator with a known-good unit but still: no charging.
• I installed new headlight lamps a couple (remarkably coincidental) days ago.

*note: thanks to a few years up in America's Hat (where batteries die of frostbite and oil freezes back into tar for the winter) I have an external 120v plug and a hard-mounted trickle charger, so the battery will hopefully be topped up by tomorrow).


My questions, beyond "any else already solve this problem" are:

1. Is there a fusible link that directly stops the alternator from charging but does not impact battery voltage from energizing the rest of the system?
2. Is there something else I need to check? e.g.: engine ground strap, some obscure (or not so obscure) relay, fuse, fusible link, or ground cable that would cause the alternator to stop working but leave everything else (engine, lights, etc) alone?
3. Can anyone explain how the engine would continue running for a few seconds before dying, when the battery terminal is disconnected?

The brown fuse-link cable mentioned in this thread seems possible, but it would more likely impact all systems generally, not just the charging?

Related to this maybe:
https://discoweb.org/index.php?threads/1996-discovery-new-battery-wont-start.95464/


Please let me know.

 

[Greg_M]

Is there something else I need to check? e.g.: engine ground strap,

I'd look at the ground connections.

 

[p m]

Is there a fusible link that directly stops the alternator from charging but does not impact battery voltage from energizing the rest of the system?

No, it's a fat wire that goes from the alternator to the fuse box and from there - to the battery.

Is there something else I need to check? e.g.: engine ground strap, some obscure (or not so obscure) relay, fuse, fusible link, or ground cable that would cause the alternator to stop working but leave everything else (engine, lights, etc) alone?

Have you got a copy of Rave PDF? You definitely need one.

Can anyone explain how the engine would continue running for a few seconds before dying, when the battery terminal is disconnected?

What's so odd about it?

There's a number of ways alternators can fail.
- Brushes wear out, causing first intermittent then complete failure (most frequent and easiest to diagnose).
- Voltage regulator can die, resulting in higher-than-normal, or lower-than-normal, or none at all, output voltage (rare but easy to diagnose).
- Diodes in the rectifier can burn out or short out. Shorted diodes usually cause quick battery death. Burned-out diodes create large pulsations of the output voltage; usually a good battery absorbs these and this condition can remain undiagnosed for a long time. An oscilloscope is often needed to detect this.
- The rotor or stator winding can burn out or short out (see above).
Many of these failures don't manifest as a "dead" alternator, and these are often not diagnosed by a parts-store check.

 

[DiscoClay]

From the official Discovery Electrical Troubleshooting Manual:

Discovery - ETM - 1995 - 2nd Edition - Eng

Charging System

When the Ignition Switch (X134) is in position II, battery voltage is applied to the charging system fault light. When the Generator (Z106) is being turned by the engine, its stator windings are excited by voltage applied to the Generator (Z106) via the charging system fault light. The Generator (Z106) begins to produce electricity in order to charge the vehicle Battery (P104). If the Generator (Z106) fails to produce power, the Generator (Z106) grounds the fault light control wire, causing the charging system fault light to glow.

The tachometer displays engine speed in rpm. Voltage pulses are taken from the Generator (Z106) and are generated when the engine drive belt turns the Generator pulley. The tachometer responds to the frequency of the voltage pulses, which increases proportionally to that of the engine speed.
​

I guess I'll check:

• That little fault light.
• The [maybe] fusible link

Some potentially related troubleshooting from the manual:

Discovery - ETM - 1995 - 2nd Edition - Eng




SYSTEM DIAGNOSIS

1. If the Starter Solenoid (K136) does not click and the engine does not crank, do Test A. [not applicable in my case]
2. If the Starter Solenoid (K136) clicks but the engine does not crank or cranks slowly, do Test B. [applicable but... I already know the battery is low due to lack of charging]
3. If the charge warning light does not light with the engine off and the Ignition Switch (X134) in position II, do Test C. [I need to check this.. I can confirm though, that the Charge Warning Light *does not light up* with the engine running.. even though the system is not charging.

 

[StangGT5]

Dead batteries can kill alternators for sure. How known good is your spare? I can't say I've had one die on the shelf, but I've had some funky spare deaths from lots of humidity.

Probably not too helpful of a response, but I agree with Gregg; I'd start with ground connections. I always found on D1s and RRCs that a solid, welder cable grade ground from frame to engine does wonders for slow starting issues. Could help with charging.

 

[special ed]

Just something to consider, I have found alternator ac bracket loose contact with being grounded to the block through oxidation of disimilar metals. Check for grounds at the metal. it was the oddest thing but aluminum and steel dont mix, you can get oxidation of the aluminum to steel bolts to the block and the whole bracket will technically not be toughing the engine. Sent me for a spin one evening on a customers car.

Not saying its the problem but worth a test.

 

[p m]

In any case, the troubleshooting is pretty easy. Start with measuring the resistance between negative battery terminal and alternator body with engine off.

 

[DiscoClay]

Dead batteries can kill alternators for sure. How known good is your spare? I can't say I've had one die on the shelf, but I've had some funky spare deaths from lots of humidity.

Probably not too helpful of a response, but I agree with Gregg; I'd start with ground connections. I always found on D1s and RRCs that a solid, welder cable grade ground from frame to engine does wonders for slow starting issues. Could help with charging.

It's a brand new Odyssey, not a spare

I am checking grounds.. but its certainly not a "primary" circuit. Everything else works...and: the volt/charge light is not illuminated... if the stator energize voltage were dead then the volts would drop across the dummy light. That said, I am gonna dive into that particular circuit tomorrow!

 

[DiscoClay]

Just something to consider, I have found alternator ac bracket loose contact with being grounded to the block through oxidation of disimilar metals. Check for grounds at the metal. it was the oddest thing but aluminum and steel dont mix, you can get oxidation of the aluminum to steel bolts to the block and the whole bracket will technically not be toughing the engine. Sent me for a spin one evening on a customers car.

Not saying its the problem but worth a test.

Yeah, i've ran into the aluminum oxide (ceramic insulator, btw) issue before. I will check again and update here **when** [pure, bold faced optimism in the face of Lucas POD!!]

 

[DiscoClay]

No, it's a fat wire that goes from the alternator to the fuse box and from there - to the battery.

Have you got a copy of Rave PDF? You definitely need one.

What's so odd about it?

There's a number of ways alternators can fail.
- Brushes wear out, causing first intermittent then complete failure (most frequent and easiest to diagnose).
- Voltage regulator can die, resulting in higher-than-normal, or lower-than-normal, or none at all, output voltage (rare but easy to diagnose).
- Diodes in the rectifier can burn out or short out. Shorted diodes usually cause quick battery death. Burned-out diodes create large pulsations of the output voltage; usually a good battery absorbs these and this condition can remain undiagnosed for a long time. An oscilloscope is often needed to detect this.
- The rotor or stator winding can burn out or short out (see above).
Many of these failures don't manifest as a "dead" alternator, and these are often not diagnosed by a parts-store check.

These are great bulletpoints, so I wanna address them individually!

1. it's a fat wire that goes from the alternator to the fuse box and from there - to the battery.
   1. So... the part that goes from alt-->fusebox could be bad? I will check
2. Have you got a copy of Rave PDF? You definitely need one.
   1. No.. I have all those manuals though... If you know where I can download the RAV please let me know I will go digging for it, but if you have a shortcut
3. What's so odd about it?
   1. if the alternator is not charging the system then... when i disconnect the battery [whilst the engine is running] the engine should immediately die (no volts).. but it continued to run for a few seconds and then started to sputter and die... and when i quickly reconnected the battery.. it sputtered and then smoothed out.
   2. It seems to me that if the alternator is not providing any voltage ..and if the battery is not providing any of the voltage.. then the engine *will die*.
   3. In writing this response, I am thinking: okay, we are all making arguments that its a ground/connection/fuseable-link problem.. in that *some* voltage (with very LITTLE current) is getting through, at least for a few seconds. Which also makes me wonder if there is some relay with a hold-down doing some evil here?
   4. I have replaced the alternator.. and the same issue is occurring. I wire-brushed the aluminum alternator bracket before I installed the new alternator.
      1. No voltage spikes appear to be occurring (no charge, lots of drain, brand new battery)... but I do not have an O-scope anymore (why'd I get rid of that old Heathkit O-scope???)
      2. I tested the diode on the old alternator (using diode-mode on my multimeter) and I did not see an obvious indication of a failed diode (if one diode in a pair is open or shorted it will typically show up with this test)
      3. I did not check the regulator. But this assumes both the old regulator (and diodes, for that matter) and the new one are bad... possible but unlikely.

 

[DiscoClay]

Guys, I just gotta say: this is the best, most friendly, and helpful forum.. period. So many forums (especially other rover forums) are just giant troll-fests.

Thanks for all the advice; it's meaningful and helps to ground me (no pun intended) when I am at my wit's end.

 

[DiscoClay]

In any case, the troubleshooting is pretty easy. Start with measuring the resistance between negative battery terminal and alternator body with engine off.

So.. I just checked, with the engine off (no key in the ignition switch):

• Negative Terminal<--> Alternator Housing: 3.3 Ω (ohms)
  • From the battery terminal (literally the top of the terminal sticking out of the battery) to the alternator case.
• Positive Terminal<++>Fuseblock-Unfused-side: 40 Ω (ohms)
  • This was to check the brown heavy gauge wire that runs from the battery (pos) to the input lug of the underhood fuse block.
  • The drop across the totality of those fuses is ~4 Ω (ohms)... or ~44 Ω (ohms) from the positive terminal.

40 ohms seems pretty high for that little cable.. it's only a foot long (at best). That said, 40 ohms will get it hot but shouldn't be dropping all the volts(?)
Do any of you know what that should be? Or.. can one of you with a "working" disco pop a multimeter on yours and report in?

 

[mbrummal]

Both of those cables should be way less than 1 ohm.

40 ohms is a LOT of resistance and would explain the battery not charging.

Measure the voltage drop from the alternator output to the fuse box terminal, fuse box terminal to battery positive, and alternator case to battery negative with the engine running. You shouldn't even have 1 volt across those combined.

You probably need to replace those cables because of internal corrosion. You can strip the insulation back a few inches to see.

 

[special ed]

Guys, I just gotta say: this is the best, most friendly, and helpful forum.. period. So many forums (especially other rover forums) are just giant troll-fests.

Thanks for all the advice; it's meaningful and helps to ground me (no pun intended) when I am at my wit's end.

Should have seen it a few years ago. It was a the worst abusive forum ive ever seen.

 

[p m]

Should have seen it a few years ago. It was a the worst abusive forum i've ever seen.

But it was fun!

40 ohms seems pretty high for that little cable.. it's only a foot long (at best). That said, 40 ohms will get it hot but shouldn't be dropping all the volts(?)

I would definitely try to re-measure it just to be sure - 40 Ohms for these circuits is a deal killer.
An ohmmeter should read less than 0.1 Ohms for both circuits. It actually should read less than 0.005 Ohms, but you need a professional tool for that.
Even slight corrosion on these surfaces and/or not very sharp DMM probe tips could affect the reading.

 

[DiscoClay]

But it was fun!


I would definitely try to re-measure it just to be sure - 40 Ohms for these circuits is a deal killer.
An ohmmeter should read less than 0.1 Ohms for both circuits. It actually should read less than 0.005 Ohms, but you need a professional tool for that.
Even slight corrosion on these surfaces and/or not very sharp DMM probe tips could affect the reading.

Confirmed x4

Lowest was 39 ohms, highest was 43.

I have a pretty good multimeter (an HP from back when that meant something). Its very accurate. In a past life I was an electronics tech.

 

[DiscoClay]

Should have seen it a few years ago. It was a the worst abusive forum ive ever seen.

I am just not interested in trashtalkingtrolls. It's a waste of time. This forum is an active user repair guide. Sarcasm is fine so long as its directly coupled to finding the fix. imo.

 

[rover rob]

But it was fun!


I would definitely try to re-measure it just to be sure - 40 Ohms for these circuits is a deal killer.
An ohmmeter should read less than 0.1 Ohms for both circuits. It actually should read less than 0.005 Ohms, but you need a professional tool for that.
Even slight corrosion on these surfaces and/or not very sharp DMM probe tips could affect the reading.

where is dan and kennith when you need them .

 

[kris812]

With that much resistance on a large wire, I really suspect the ends have corroded or it should look smoked to hell. I would take the ends off and clean them up well with a wire brush. It's also possible your meter is on the wrong scale and you're reading it wrong.

Secondly, I wouldn't be pulling and applying the cable while the engine is running. Normally makes for some crazy voltage spikes out of the regulator on other cars and I've personally seen almost every light bulb in an old car blow up from this. One time on a Cessna the external regulator was bad and also blew up every bulb in the aircraft.

The basic and normal test for an alternator I've always done has been a volt meter on the battery 11-12 volts battery is good, running should be 13.8v charging.

On a couple of vehicles I've ran relays to the alternator and connected to my headlights for extra power, that extra 1.8v does a LOT to a headlight!

On other vehicles there's a 12v feed into the alternator that triggers it to charge. I'm guessing with multiple alternators having an issue, this is missing. Checking grounds is a good start here. And seeing as your gauges were jumping around, I suspect a missing ground wire somewhere.

 

[discostew]

It's a brand new Odyssey, not a spare

I am checking grounds.. but its certainly not a "primary" circuit. Everything else works...and: the volt/charge light is not illuminated... if the stator energize voltage were dead then the volts would drop across the dummy light. That said, I am gonna dive into that particular circuit tomorrow!

If you have a bad body ground and a good chassis ground the thing could be charging but the difference in ground will affect the voltage the cluster is seeing. I would check your grounds by measuring voltage from battery pos to a body bolt or ground. Then compare that to the reading you get from battery pos to the alternator case. Another thing I do is run the engine and turn on all the loads. Like blower brake lights heated screen heated seats. Lights on high beam. The whole shooting match. Then make those measurements.