One-, Two-, Three-, or Four-Wire Alternator?

Last week, I talked about troubleshooting a car-died-while-driving condition on my 1969 Lotus Elan +2, which I eventually traced to a broken contact face—on a set of points that didn’t even have 300 miles on them. In the piece, I said that it happened while I was out for something of a victory lap in the car, having just diagnosed and repaired an overvoltage condition. I thought I’d circle back and describe that, as I’d never encountered this kind of charging-system problem before.

As I described in this piece for Hagerty eight years ago, a car’s charging system not only keeps the battery charged, it also feeds the demands of the electrical systems while the car is running. The resting voltage of the battery should be around 12.6 volts, but when the engine is running, the charging system should output about 13.5 to about 14.2 volts. People often think that the alternator is the charging system, but really, the charging system consists of the battery, the alternator, the voltage regulator, and the cables connecting them.

As its name implies, the job of the voltage regulator is to adjust how much voltage the alternator is pumping into the car’s electrical system. Without the regulator to moderate it, the alternator outputs about 17 volts, which is enough to literally boil the acid in the battery, producing that characteristic rotten-eggs smell. The voltage regulator is basically a switch that opens and closes rapidly in order to supply an average voltage in the range of (roughly) 13.5 to 14.2 volts.

On most cars built before 1975, the alternator and voltage regulator are separate units, with the regulator being a box, often mounted on the inner fender wall of the engine compartment, with mechanical contact points inside it that look quite similar to ignition points. In addition to the thick “B+” battery cable post that’s on the back of every alternator, there are typically three wires connecting the alternator and regulator (on European cars they’re labeled D+, D-, and DF). So, if the regulator itself fails, or if any of these three wires or the connectors at both ends break (that’s nine things to go wrong), the alternator won’t charge the battery.

Beginning in about the late 1970s, cars began coming with solid-state voltage regulators that were integrated with the alternator. This design is inherently more reliable than that of the external regulator and its failure-prone wiring. If you’re going to road-trip your car, it’s a good idea to upgrade to what’s often mistakenly called a “single-wire alternator” (we’ll get to that).

Whether internal or external, mechanical or solid-state, voltage regulators usually fail in the open position, where they’re not bringing the alternator into the charging circuit at all. The result is that, when the engine is running, nothing is charging the battery, and the voltage soon drops from 12.6V to low enough that the car won’t start without a jump. Eventually the voltage drops so low that, while the car is running, it will sputter and die because there isn’t enough voltage to fire the coil. On a car with an external regulator, you can try to troubleshoot it and see if the cause is a broken wire or an unseated connector (it often is), but if the car has an internal regulator, there’s no choice but to replace it, and if that doesn’t fix the problem, to replace the alternator.

External mechanical regulators do sometimes fail in the closed position. This causes the alternator to go “full field,” output 17 volts, and, as I said, boil the sulphuric acid in the battery. I described this situation here during a road trip two years ago; I eventually traced a companion’s problem with pitted points to a voltage regulator that was stuck in the closed position, but I kicked myself for not immediately identifying the sulphur smell as boiling battery acid.

Another difference between older and newer charging systems is whether and how well the regulator senses the electrical load and increases the alternator’s on-time to raise the voltage. In the vintage BMWs with original, external mechanical regulators with which I’m most familiar, my experience is that, basically, they don’t—or if they do, they don’t do it particularly well. When running with no electrical accessories switched on, I never, ever see 14.2 volts. I’m lucky if I see 13.5. With lights and wipers on, it may well drop below 13. With lights and air conditioning, it may dip alarmingly close to 12.6. Hey, the car is 50 years old; it’s doing the best it can.

Modern computerized cars go one better: The ECU knows what the voltage demand is, and there’s typically a plug on the back of the alternator where the ECU supplies a signal to tell the alternator/regulator to raise or lower the voltage.

Above, I mentioned the “one-wire” alternator. While there are single-wire alternators with only a fat B+ wire going to the battery and/or fuse box, most 25-to-40-year-old cars that are equipped with internal regulators have two-wire alternators. The second, smaller wire is usually labeled D+ and goes to the alternator/battery warning light on the dashboard. This light is actually part of the charging system, and is used to bootstrap the alternator into producing current. Once it does, the light goes out. When you upgrade an alternator-and-external-regulator configuration to a two-wire alternator with an internal regulator, you typically need to find the D+ wire, splice a new connector onto it, and plug it into the back of the new alternator.

There are folks who turn two-wire alternators into single-wire alternators by making a little jumper and permanently connecting the D+ terminal to B+, but I’ve always been hesitant to do it. It disables the battery warning light, and on a classic car, I’m just so accustomed to cracking the key to ignition, verifying that both the oil and battery warning lights come on, then starting the car and verifying that they both go out.

So, with that background, here’s what happened with the Elan +2.

When I bought the car last fall and began sorting it out, I put a voltmeter on the battery and verified that, with the engine running, it read about 14 volts. At that point, the car had many needs, so I was happy that the charging system didn’t appear to be among them. This spring, I solved a hot-running problem by replacing the unbranded cooling fans (which turned out to be puller fans in a pusher configuration, wired with their polarity reversed) with a proper set of Spal pusher fans. Anxious to check this problem off as “solved,” I took the car for a drive to watch the engine temperature as the fans turned on. I also plugged in a cigarette-lighter voltmeter so I could see what effect the fans had on the charging voltage. To my stunned surprise, I saw that, as soon as I began driving the car and the revs came up, the reading was over 15 volts. Yeesh! I turned on the fans and headlights, and the voltage fell to about 14, but as soon as I shut them back off, it immediately climbed back over 15.

Now, these cig-lighter meters aren’t terribly accurate; I mainly use them to be certain that the running voltage is simply higher than the resting voltage. But a reading over 15 volts was cause for concern. I came straight home and tried another plug-in voltmeter. It read even higher, about 15.6 volts. So I took out my trusty Fluke multimeter and checked the voltage at the battery. It confirmed that what I was seeing was basically accurate: The charging system was outputting about 15 ½ volts. Any higher and I would’ve likely smelled sulphur.

I’ve had my ’74 Lotus Europa Twin Cam Special for 12 years, but I’ve only owned the Elan +2 since last fall. Although they share the same engine, not all their accessories are the same. The Europa came with its original alternator, but it took some research to learn that the Elan +2 is old enough that it originally came with a generator.

I looked closely at what is currently installed in the car, and it appeared to be a two-wire alternator. It looked very similar to a Denso alternator that was sold as part of a conversion package by a known Lotus parts house. On closer examination, I found that it is actually a Mitsubishi A430. I even found 13-year-old receipt for it in the folder from the previous owner.

I did some reading on common vintage Lotus alternator conversions, and it initially appeared that there was supposed to be a little voltage adjustment on the back of the alternator, but mine didn’t have it. I looked into replacing the internal regulator, but then learned that it’s not the two-screws-and-out-comes-the-regulator-and-the-integral-brush-pack design that it is on my BMWs.

Then I saw a reference to the “voltage sensing line” on the back of the alternator. I looked at my car, and sure, enough, next to the terminal labeled “L” for “Lamp” was a second terminal labeled “S” for “Sensing.” This was something I wasn’t familiar with—a three-wire alternator. For these, the idea is to run a sensing wire from battery positive to the “S” terminal so it can sense the electrical load that the car is actually trying to draw from the battery. On the Elan +2, the battery is in the trunk, so I simply made a short jumper wire and connected it to “B+,” which is directly connected to the battery anyway.

The jumper wire solved the problem. The over-voltage condition immediately went away, and the charging voltage was capped at about 13.5 volts.

In truth, all alternators are supposed to be “voltage-sensing,” but those without a sensing wire read the voltage from the same line they’re outputting it on. The use of a sensing wire makes, well, sense: In theory, the voltage coming out of the alternator should be the same as what it is at the battery. In practice, due to voltage drop, it isn’t.

Here’s the full alternator-wire taxonomy, at least as it applies to the alternators I’ve dealt with:

• A single-wire alternator has an internal regulator and uses the fat B+ wire as the voltage output to the battery/fuse box, the sensing wire, and the bootstrap wire.
• A two-wire alternator has an internal regulator and uses the fat B+ wire as the voltage output to the battery/fuse box and as the sensing wire, but has a separate D+ wire to the battery indicator light that acts as the source of bootstrap current.
• A three-wire alternator has an internal regulator and uses the fat B+ wire as the voltage output to the battery/fuse box, but has a separate D+ wire to the battery indicator light that acts as the source of bootstrap current, and a separate voltage-sensing wire.
• A four-wire alternator from the 1970s uses the fat B+ wire as the voltage output to the battery/fuse box, and three wires to an external regulator.
• An alternator from the post-2000-ish era likely has the fat B+ wire along with a multi-pin connector carrying signals from the ECU that perform the voltage regulation.

Now that both the over-voltage problem and the no-spark problem are solved, I hope I can give the Elan +2 the sudden-death-free pleasure drive both it and I richly deserve :^)

***

Rob’s latest book, The Best Of The Hack Mechanic™: 35 years of hacks, kluges, and assorted automotive mayhem, is available on Amazon here. His other seven books are available here on Amazon, or you can order personally inscribed copies from Rob’s website, www.robsiegel.com.