Key--off current draw from 12SI alternator

I have a 78 amp 12SI alternator, out of a mid-80s Riviera, in my '78 Trans Am.

My new battery went completely dead in a week. A meter showed a current draw of about 300mA with the key off. Removing the two-pin connection from the alternator dropped the current draw to about 9mA. This is a completely repeatable result.

The alternator appears to work fine, except maybe for being a bit weak at idle with the backglass defrost on. However, something is obviously wrong inside it.

The '78 Pontiac service manual refers to diagnosis using the "ALT" light, but my T/A has rally gauges and no warning lights, so I can't use the manual's procedure. Nevertheless, the manual says that "if the ALT light is on with key off, remove the two-pin connector; if the light goes out, replace the rectifier bridge." So I'm guessing that my ALT light would be on with key off, assuming I had an ALT light?

The rest of the troubleshooting procedures involve disassembling the alternator, which is more work in winter than leaving the two-pin connector off when the car is parked!

So my four questions are:

1) Do the symptoms as described point to a bad rectifier bridge in my alterntor?

2) For the short term, can I simply disconnect the two-pin connector while not driving, and connect it when I am driving? (It's winter in Ontoario!)

3) I have an overhauled/rebuilt 10SI with unknown components in it. Can the diode trio and rectifier bridge be swapped between a fairly high amperage 10SI and my 12SI?

4) Okay, I've studied the factory manual charging system pages and the wiring diagrams. How on earth can a "ALT" light be on with the key off? The light circuit is interrupted! Yet apparently it can happen?

Thanks Ed

Reply to
Ed Treijs
Loading thread data ...

You have a leaky diode.

Yes.

I suppose so, but it is just going to get worse.

I don't know, but you can look at it and see. Honestly high current diodes should not be a difficult thing to find. Ask your local TV repair shop.

It wouldn't be the first time I have seen the factory manual wiring diagram at odds with the actual vehicle. Just because the manuals say it's wired in a particular way doesn't mean that it is.

--scott

Reply to
Scott Dorsey

A third of an amp drain doesn't seem like much although I suppose if you didn't drive the car it could discharge the battery in a week or so. The most basic test you could do is to measure the voltage across the battery terminals with the engine running and with the engine off. A low voltage with the engine running would probably mean your alternator was the problem or that it's output wasn't enough for what you're running. A low voltage with the engine off would probably mean a problem with your battery or excessive drain. 300mA seems fine to me.

Reply to
dsi1

Okay, it's time to start doing a bit of alternator dissassembly.

May be easier to find a 12SI alternator in the junkyards than a TV repair shop! I do have a decent AC/Delco parts place in my end of the city.

....Ed

Reply to
Ed Treijs

Not necessarily.

A problem is indicated, I'd want to know what it is before I assume that just disconnecting the alternator connector is going to make life better.

Doubt it. There are many different sizes and amp ratings of rectifier bridges.

Because the circuit isn't interrupted?

Here's what I'd do; cull the tan wire from the 2 way connector, leave the two way connector plugged into the alternator so that the red wire is connected to where it normally is. Use your 12 volt test light to see whether there is voltage at the tan wire coming from the ignition switch or whether there is voltage coming out of the terminal on the alternator where the tan wire was connected. Which ever one has voltage dictates which direction you are going.

IOWs, you either have a short that is feeding voltage to the voltage regulator (alternator) or you have a problem inside the alternator which is feeding voltage back out to the ignition switch and (according to the wiring diagram), the heater blower switch.

Reply to
aarcuda69062

Alas, the advice was a little late, as I decided to "do something" and the project just grew. But I do have more questions now.

I took both the 12SI and the 10SI apart. Once I got going, I decided to treat it as an educational experience. If I break something, I will learn from that. If I put something together that works, I learn from that too.

I got puzzling readings from the 12SI bridge: it seemed to have no conductivity either direction. I'd put it down to error or tool misuse (I'm using a digital multimeter that has a diode check setting) but the 10SI bridge checked out as I'd expect it (conductivity in one direction). Additionally, the 10SI diode trio checked out fine, but the 12SI trio again seemed to have no continuity either way!

Anyway, I didn't understand how the rectifier bridge was assembled, and in learning how it's assembled I destroyed the 12SI bridge. The

10SI bridge fits, although the heat sink cooling fins aren't ideally shaped for the 12SI end frame vents.

I presume that the SI-series rectifier bridges come preassembled on the heat sinks, because the diodes appear to be bonded to the sinks (I broke most of the diodes from the 12SI heat sink).

Aside: contrary to the 1978 factory service manual, my car does have a blower off position. My '79 Firebird did indeed have a blower that ran all the time, but my non-A/C '78 definitely has an off position.

Obviously, I should check if there's unwanted voltage on the tan/black wire when the key is off. Would be annoying if I did all this teardown stuff when it really was a wiring problem!

I put the 12SI back together with the 10SI bridge and diode trio. The

10SI did not have a resistor between the +12 brush connector and ground--this is labelled as "Resistor (some models)" in my factory manual, probably R5 in the circuit diagram GM provides. The 12SI does have this resistor: it's about 48 ohms. I wonder why "some models"-- what causes it to be or not to be present? I might guess that the resistor's presence is matched either with the regulator or the diode trio, but which? Should I keep it?

Anyway, if the alternator works at all, I will have to get a proper rectifier bridge, and also get the bearings replaced. 12SI stator is noticeably heftier than 10SI stator; couldn't see significant differences in the rotor.

So to summarize: Didn't check tan (field) wire--need to check

12SI bridge and diode trio check out open; bridge is wrecked anyway 10SI bridge and diode trio check out fine--in use 12SI had resistor, and I kept it Haven't reinstalled alternator to test Certainly have learned a whole bunch about alternator innards, for what use it is

....Ed

Reply to
Ed Treijs

There should be 3 wires involved. There is a hot to the Battery and a ground and a wire from the stator which is where the current being rectified comes from. From Battery to ground.

But anyway the rectifier bridge shouldn't have anything to do with your problem. If I understood your original posting correctly your current leak is in the control circuit not the heavy gauge output to the battery.

You also said you don't have a dash light. Normally the dash light is part of what makes the alternator work. Usually when you disconnect the wire to the dash light and it won't charge because that is where the juice for the field comes from.

If I remember correctly the diodes are pressed in. Here is a wiring diagram and some trouble shooting tips:

formatting link

-jim

Reply to
jim

Sorry, I was replacing stepper motors in a 2003 Silverado instrument cluster. (gotta eat)

That alternator shouldn't have worked at all if the entire rectifier bridge was open.

You do realize that the diode test function on your DVM tests bias voltage correct? You should be reading something like .7 at one polarity and 0.0 when the test lead polarity is reversed.

The big old honkin Fox valley meters that were sold 30 years ago were great for testing alternator internals, the ohmeter was powered by nine volts which was more than enough to bias the PN junction to test the diodes. I think the old Simpson 'tombstone" meters did as well.

They don't come apart, I used to bend the straps up from the studs to totally isolate each individual diode from the rest of the assembly. IIRC, I got bit in the ass by a bridge frame that was shorted between positive and negative sides, time to re-invent the test.

Yes, that is how they come.

Yup.

That resistor was needed on some applications with a warning light, its purpose was to stop the warning light from glowing dimly at idle speeds. With a no warning light application it wasn't needed, its presence or lack there-of makes absolutely no difference in a vehicle with gauges and no light.

Stator is bigger because the output rating is higher. At low amperage, the 10SI bridge will probably work okay, at higher amperage it's gonna smoke.

Check the tan field wire for voltage with the key off BEFORE you re-assemble with the new rectifier bridge, if there is no voltage there, I'd recommend replacing the voltage regulator unless you don't mind taking it all apart again.

Reply to
aarcuda69062

That isn't even it; I have no internet at home and I was sitting around yesterday afternoon, thinking "I gotta do something about that alternator". I appreciate help, especially *accurate* help. Usenet is getting a poor place to expect that, so sometimes I just go ahead on my own.

Yeah, shouldn't, but it did, except big loads at idle would pull the voltage way down after the car, and presumably alternator, warmed up. I can't explain why the one diode trio didn't check out either.

What I get is something like "568" in one direction and off-scale flashing "3000" in the other direction. I think it reads resistance. Alas, my DVM instruction book is.....somewhere, and I've never had to check diodes in my 15 years of ownership. It may actually be .567; kinda hard to see the decimal place on the meter.

To check the key-off draw, my father pulled out a real big (car- battery-sized, and almost as heavy) made-in-England analogue meter that was used by our local power company (Ontario Hydro) technicians. Just in case my key-off draw was in the tens of amps. (My father was an EE for Ontario Hydro, go figure.)

I drove out the pins/wedges staking/wedging each side of the plastic middle piece. With those removed, the plastic piece slides out from between the bridges (or vice-versa). At that point you wind up with four major pieces: upper bridge, lower bridge, plastic holder/ insulator, and a three-post terminal strip to which the stator windings attach. At that point, attacking each of the diodes individually is no problem. Reassembly is inverse of disassembly.

I find it odd that the positive side of the bridge goes via a big nut to the thick B+ stud, while the negative side of the bridge seems to be grounded entirely by the wimpy hold-down screw. I think I want to look at it again.

Hmm. The one thing it does do is to change my meter readings; on continuity setting, suddenly the meter beeps for continuity in both directions! It evidently assumes that a resistance of under 50 ohms is as good as continuous. (Continuity beep works with ohm setting only.) So removing the resistor makes it easier to check for proper diode polarity and working.

Okay, I need to find a new 12SI bridge.

Hmm, 10SI and 12SI regulators look identical, but I don't have field current specs for a 12SI. 10SI would be 2.5A, but this was a "higher performance" rebuild by a reputable electrical shop. What new/upgraded items they threw into this old 10SI (case part # makes it sound like an early '70s build) I don't know.

Here's another question I'm pondering: why is the sense (#2 conector) an external connection? The sensing lead is simply spliced into the heavy main line coming from B+, and runs back to regulator. Surely it could be jumped internally, say from the B+ stud, similar to the diode trio. I'm guessing it's got something to do with troubleshooting options, as otherwise why bother?

Anyway, thanks for all your info. I will soon be the alternator guru on my block! (This summer I took apart my HEI distributor because the mechanical advance had seized solid; yet another learning experience, and the car started right up when I dropped the cleaned/greased unit back in.)

....Ed

Reply to
Ed Treijs

This is correct. You have a .6V drop in one direction, and no current flowing in the other direction. That's what a diode is supposed to show.

However, you can have diodes that measure OK at low voltages but break down in actual service.

--scott

Reply to
Scott Dorsey

The voltage sense wire is extended so that the regulator is sampling voltage at the battery not at the alternator. There *is* resistance even in a 8AWG wire from the alternator through the ammeter and then back to the battery (assuming you have an ammeter, that is.) For troubleshooting purposes on a bench just connect it to the big stud.

I have been watching this thread; I have a 10SI with a Speedway diode harness in my '55 Stude but have not had any issues with it yet (knock on wood) save for low charge at idle, but I thin kthat is just because a Stude 289 has a smaller crank pulley than a SBC

nate

Reply to
Nate Nagel

MotorsForum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.