[wizenhymer]

This happened last night, while making my weekly 5 hour commute. About 2 1/2 hours into the drive, there is a hilly region where the cruise ordinarily will downshift and RPM's go up to about 3500 for a while. "A while" being 30 seconds to 2 minutes. Duration of downshift depends on load and setpoint, and I used to be very nervous about this, but eventually I figured "I'm burning out carbon deposits", and it's nowhere near the redline.

Anyway, I hit the first uphill segment, and when I reached the top of the hill I touched the coast button (as I normally do) to get it to shift back to normal RPM's. Everything normal. Then I hit the next uphill segment, and when it downshifted, RPM's went to 3500 as always, BUT, the headlights went very dim and the battery indicator light (for the charging system) was flashing rapidly on and off. As I began to realize that my alternator was dying in the middle of the night, in the middle of nowhere, I hit the coast/decel button to drop the RPM's down. When the RPM's dropped down, the trouble vanished, and did not recur during the rest of the trip. 2-1/2 hours later, as I pulled into my destination, I briefly revved the engine to 3500 RPM to see if it would cause the headlights to dim and the battery light to come on. It did not. So RPM's alone are not the cause. Car ran fine today. Later in the day, in an attempt to re-create the problem, I accelerated uphill with 3500 RPM and headlights on, everything the same as last night, but did not get the light show.

If it were the alternator, I'd think it would stay bad. I replaced the brushes last spring, so I don't think it's brushes. Also checked bearings for play at that time; there was none. I have a hard time imagining the diodes / regulator in there could go bad just for a 30 seconds and then go good again.

Could the tilt from going uphill cause the electrolyte sloshing in the battery to create a momentary short? But then how would the timing coincide exactly with the RPM's dropping back down to normal? Seems possible, but not plausible.

Could the solenoid that kicks in for downshifting be momentarily shorting out the electrical system? It seems to fit the symptoms perfectly, but seems unlikely from an electrical standpoint. Solenoids typically fail by failing to actuate; a mechanical phenomenon. Electrically, I have only ever seen them go open, not short. Once open, they tend to stay that way. I cannot imagine how a solenoid could momentarily short, unless there is a bad connector.

Could there be a problem with the heavy ground or power connections between battery, alternator, and frame that would act this way? This seems most believable as a failure mode but doesn't seem to fit behavior too well.

Car has 230,000 miles and gets regular oil changes every 3-4,000 miles.

Possibly unrelated, it has had a few rough cold starts in the last 4 months. At one time set a P0301. Examined cylinder 1 plug and found nothing. Sprayed plug wires with WD-40 and juggled wires 1 and 5 at coil pack to see if code follows coil or wire, but code has not set again since then. I'm wondering if somehow a coil pack that is on the edge of failing could produce this bizzare and scary behavior.

I'm hoping to reproduce the problem sometime in the next week so I don't have to experience another knuckle-biting 5 hour drive. Making it reproducible makes it easier to confirm a good fix.

Any ideas?

 

[sadistic]

have you checked serpentine belt tension and tension pulley as in how much flex room is still left in it. so next time you hear a lot of squeaking at a light on a cold morning or rainy day it most likely a dodge or chrysler product serpentine belt squeaking. so look to see what type of car it is if you think of it. had a old 1988 reliant k car every time it would rain the belt would squeal loudly till it down shifted. some times when i hit a puddle it would slip so much the head lights would dim all the way home. i wondering if the increase cause the belt to slip and the alt to not charge temporarily. but all i can say is check the tensioner pulley spring for enough adjust ment in it and belt tension is tight enough.

 

[360alaska]

http://www.taurusclub.com/forum/82-...-95-sable-3-0-intermittent-battery-light.html

 

[wizenhymer]

have you checked serpentine belt tension and tension pulley as in how much flex room is still left in it. so next time you hear a lot of squeaking at a light on a cold morning or rainy day it most likely a dodge or chrysler product serpentine belt squeaking. so look to see what type of car it is if you think of it. had a old 1988 reliant k car every time it would rain the belt would squeal loudly till it down shifted. some times when i hit a puddle it would slip so much the head lights would dim all the way home. i wondering if the increase cause the belt to slip and the alt to not charge temporarily. but all i can say is check the tensioner pulley spring for enough adjust ment in it and belt tension is tight enough.

Excellent idea! I will check that!

 

[wizenhymer]

http://www.taurusclub.com/forum/82-...-95-sable-3-0-intermittent-battery-light.html

I don't think it's my brushes because I recently replaced them. I will certainly look though if I can't find anything else.

 

[DLC360]

I had a simular problem on my '99 vulcan. At 3000 / 3500 RPM, under a load only, the battery light would flicker. over time that RPM limit slowly creaped down until it was comming on around 2000 / 2500 RPM. I changed the laternator and all was well. Those diodes are getting old and you may be on the verge of a thermal fault in them as well, all things leading to alternator failure.

 

[wizenhymer]

I have since been able to reproduce the problem 3 times. It is related simply to RPM's; happens at about 4,000 RPM.

Also took it to the local FLAPS and they checked the charging system with a midtronics EXP-1200.

We went through a series of tests with engine off, starting, idling with no loads, 2,000 RPM, idling with all loads, 2,000 RPM with all loads.

According to this device, my battery CCA's are down to 491. When I had the battery tested in April on an EXP-800 it had 639 CCA's.

Starter test says cranking normal, though I note battery dips to 8v.

Charging system test says "no problem" with ripple of 62 mV. Delivers 13.76V @ 5A with no loads; 13.61V @ 2.9A with full load.

So I'm thinking I'll check tensioner / belt slippage, and double check battery.

If the problem grows worse with time, it becomes easier to pinpoint. I'm thinking if I have a cell going out on the battery it might load down the alternator and exhibit this behavior... maybe.

Does anyone know this: is the battery indicator an LED, or is it a lamp?

 

[wizenhymer]

Thanks for the info dlc360. I also have a vulcan engine. I will keep an eye on the RPM limit and see what happens.

When mine does it, the headlights also dim and you can feel that the car has less voltage. Is that what happened with yours?

 

[wizenhymer]

Today at work a colleague gave a plausible explanation:

Windings on the rotor have lost varnish and loosened with age, and when RPM's are high enough, centrifugal force causes them to fly outward far enough to graze against the inside of the stator housing, which is grounded.

Tomorrow night I will remove and disassemble the alternator. He's probably right. If so, I should see some shiny copper on the outermost windings of the rotor.

In the meantime, I will avoid the 3,000+ RPM range.

 

[wizenhymer]

Took the alternator out. First step in disassembly was to remove the brushes. I didn't expect to see what I saw. I had replaced them in April, I think.

One brush so long, there was 1/8 inch of brush beyond the retainer hole.

The other brush has obviously broken off and was just barely sticking out.

I did not disassemble any further as I feel I have the answer staring me in the face.

xmanglc, you win the kewpie doll! :wub:

$9.66 for a new brush set at the FLAPS tomorrow and I will be back in business!

 

[wizenhymer]

Postscript

Problem persisted at high RPM's even with new brushes.

I subbed in a new battery (July '10) from a different vehicle to eliminate suspected dead cell. No joy.

Noticing the belt was outside spec on tensioner indicator, replaced belt to no avail.

The dénouement was to replace the alternator.

I had the alternator apart to inspect for slung windings, but had failed to loosen the pulley bolt in situ, so I ended the post mortem with slung windings as the presumptive cause.

Replacement alternator was a reman with a stripped bolt hole. Sick of trekking back and forth to the FLAPS; used my friend JB weld to re-form threads. This is a lightly loaded bolt with virtually no tensile stress (7-1/2 ft lbs) ; the formed metal bracket and 1/2" nut endures strong upthrust.

Electrical problem solved.

 

[wjc]

FYI it's been my experience that replacing internal alternator parts isn't worth the time. They have a nasty habit of disintegrating a number of key components almost simultaneously.