Two basic principles are involved with generating current mechanically. First, when a current flows through a wire, a magnetic field is set up around that wire. The higher the current flow, the larger and stronger the magnetic field. Second, when a wire is passed through a magnetic field, a voltage is "induced" in that wire. Given a place to go, that voltage will cause a current to flow in the wire. The important element here is there must be movement between the wire and the magnetic field. A similar visualization can be made with your hand in a tub of water. Think of the induced voltage as the height of the water next to your hand. As long as you don't move your hand sideways, the water stays level representing no voltage. When you move your hand, the water forms a wave. The high and low spots represent the induced voltage. You can increase the voltage, (height of the water) by sticking your hand deeper into the water, taking longer strokes, or by moving your hand faster. These three variables correlate to the three things that affect how much current a generator can deliver. Those are size of the magnetic field, number of loops of wire the voltage is induced into, and the speed at which the magnetic field is moved.
The magnetic field around a single wire is very weak and of not much use, but when the wire is wrapped up in many loops, the magnetic fields add together to become stronger. Add an iron core to wrap the wire around and the magnetic field will be concentrated and still stronger. Now it is usable. In the generator, this wire is called the "rotor" or the "field winding" because it creates the magnetic field. It is the part that spins.
Similarly, the voltage induced in a single wire will not be very high, but if that wire is formed into many loops, the same voltage will be induced in each loop and all of those voltages will add together. AC generators use three sets of these loops and together they produce a three-phase output voltage. This is the stationary coil of wire attached to the case of the generator. It is called the "stator" winding. Adding just a few inches of wire to form one or two more loops increases the maximum amount of current the generator can produce.
The third component is the movement. As the rotor spins, thanks to the drive belt, the magnetic fields spin with it. The iron core of the rotor has fingers extending from each side and wrapped around the outside of the coil of wire. Every other finger concentrates and directs a north magnetic pole and every other one directs a south magnetic pole close to the stator windings. Just as moving your hand faster in the tub of water produces a higher wave, the multiple fingers cause many changes from the north to the south magnetic poles per each revolution. This has the effect of speeding up the movement and increases efficiency. Generators do not work well at low speeds when the movement between the magnet and coil of wire is slow. The movement is sped up by those multiple fingers, using a smaller drive pulley than the engine's crankshaft pulley, and by increasing engine speed.
While generators produce voltage and current mechanically, only batteries can store that electricity and they do that chemically. When recharging the battery, the voltage must be controlled to prevent overheating it and boiling the water out. Controlling the voltage is also necessary to prevent damage to light bulbs and computers. As the battery becomes fully charged, system voltage will slowly rise. Something must be done to reduce generator output voltage. The number of loops of wire in the rotor and stator can not be changed to meet the constantly changing demands on the system, and it is not practical to slow the engine down when the battery is charged, or to increase engine speed when you want to roll a power window down. The only variable that can easily be changed is the strength of the magnetic field.
Most field windings have close to 4 ohms of resistance so the maximum current flow through it will be 3 amps. That's when the magnetic field will be strongest. To reduce output voltage, it's very easy to reduce the strength of the magnetic field by reducing the current flow through the rotor. Current is reduced by adding resistance to the circuit. That's the job of the voltage regulator. The regulator circuitry is in series with the field winding. It becomes a higher resistance when reduced generator output voltage is desired.
In its most basic form, there are only two parts to the regulator. There's the part that varies the amount of resistance in series with the field winding, and the part that measures system voltage. System voltage is battery voltage.
Since the field winding and the voltage regulator are in series, they can be in either order. Knowing that order is key in determining the steps in troubleshooting the system.
In the most common circuit, starting at the battery positive post, current goes to one terminal of the generator first, through the brushes and field winding, then through the voltage regulator, ground, and back to the negative battery post. The regulator comes after the field winding so it is referred to as an "A" circuit. All regulators need three terminals, the system voltage sensing wire, the reference to compare it to, (ground), and the control circuit that is in series with the field winding. As just mentioned, current flows through the field winding and through the regulator, then goes to ground and back to the battery, so the regulator has a connection to ground already. To sense system voltage, all that is missing is the wire to battery voltage. It was common on older cars to provide this circuit through the ignition switch. Besides sensing system voltage, this circuit also provides the power source to run the electronic circuitry in the regulator. Switching this current through the ignition switch turns that circuitry on when the engine is running and we want the charging system to run, and turns it off to prevent draining the battery when the engine is off.
Some manufacturers use the "B" circuit where, beginning at the battery positive post, the voltage regulator comes before the field winding. In many systems where the regulator is built into the generator, it is easy to connect the system voltage sensing terminal directly to the generator's output terminal since it is connected directly to the battery already. To prevent a drain on the battery when the engine and ignition switch are off, the regulator is typically switched off electronically. When the ignition switch is turned on, current flows through the warning light circuit on the dash to the regulator to turn it on. It is important to understand that the system voltage sensing wire, (and power supply for the regulator's circuitry), the reference, (ground), and the control circuit are all inside the generator's case and often are not accessible. The addition of the fourth circuit, the turn-on circuit from the dash light, may be the only test point provided.
Knowing which circuit is used is necessary when performing a full-field test. This test is done to determine which parts of the charging circuit are working. The voltage regulator is bypassed, removing it from the circuit. If the generator begins charging at a very high rate, the regulator has been identified as defective. When there is little or no charging during this test, further diagnosis is required, but there is a very good chance the voltage regulator is not the problem.
To full-field any AC generator, one field terminal must be grounded and the other one must have full battery voltage applied. The manufacturer always provides one of those in the circuit; you must provide the other one to bypass the regulator. In the "A" circuit, battery voltage is applied directly to one field terminal either all the time, or through the ignition switch or a relay. You must ground the other terminal. Some older generators such as GM's "SI" series from the 1970s and 80s, provide a test point specifically for that purpose. Older Chrysler alternators have a terminal right on the back of the case, or you can find the same wire in the plug for the voltage regulator.
When the manufacturer uses the "B" circuit, one field winding terminal is already grounded. You must provide the battery voltage to full-field this system. Doing so involves running a jumper wire between that field terminal and battery voltage. This was fairly easy to do right at the voltage regulator on older Fords.
See the "Testing" section for your charging system for more details. The guidelines here apply to any AC generator but many systems do not have test points provided. When testing is possible, the procedure will be found in the manufacturer's service manual.
Every generator has a low-current input circuit and a high-current output circuit. Up to now, the entire discussion has been about the input circuit. The voltage regulator is always a part of that circuit because it is very easy to control the very small current found there. Even though the current is very low, the input circuit is where you will find the majority of charging system problems.
The output circuit consists of the stator, (stationary) winding and the diodes. Three-phase windings are used to reduce the amount of ripple voltage in the output voltage. Excessive ripple voltage may be heard as a high-pitched whine on an AM radio and can affect computer sensor readings. A good battery will dampen most ripple voltage to an acceptable level. For a more detailed explanation, see "Basics of Batteries".
Each of the three stator windings connects to a pair of diodes capable of passing the very high output current. A diode is a one-way valve for electrical current flow. AC generators develop alternating current which is constantly changing direction. The diodes direct the current from the various windings to all go the same direction to the battery. Batteries can only store direct current. There is no practical way to store alternating current. During the half cycle when current flow is blocked, the diode "sees" the voltage that is trying to cause current to flow. The voltage they are capable of withstanding without damage is designed in and is typically around 30 volts, well above normal system voltage. The more the voltage goes above that designed-in reverse voltage, the more likely the diode will become shorted. Even if the regulator were to short or you were to bypass it, the battery will resist allowing the system voltage from going excessively high. This is important to understand when warning against performing a trick used by uneducated mechanics many years ago.
It is so easy to determine whether the charging system is working by simply using a voltmeter to measure battery voltage while the engine is running, but before mechanics understood these systems, a favorite trick was to remove one battery cable while the engine was running. If the engine continued running, current for the ignition system had to be coming from the generator so it was assumed the system was working. There are a number of problems with that procedure. First of all, it is common on many cars with computers and electronic controls for the engine to stall when a cable is removed, even when the charging system is working correctly, so the test is meaningless. More importantly, without the battery in the system, there is nothing to limit system voltage. Even though the voltage regulator tries to lower the amount of current flowing through the field winding, it can't cut it to 0 amps. There will be some output voltage developed and with no battery to absorb the ripple voltage and spikes, that voltage will go right back to power the field winding. As the output voltage increases, it causes increased current flow through the field. The resulting increased magnetic field causes a higher output voltage which goes right back to the field. A viscous circle is formed and output voltage keeps on increasing. The voltage regulator doesn't have enough control to prevent this from happening. The only thing going in your favor up to this point is low engine speed. The third problem is this makeshift test has no way of showing if there is a shorted diode or how much current the generator can produce.
During a proper full-field test, it is very possible to reach a system voltage of 18 volts when raising engine speed. The battery is helping to hold the voltage down, but he is not happy about it. If a battery cable is removed, there is nothing to hold the voltage down. If engine speed is raised, the reverse blocking voltage of the diodes will be exceeded and the diodes will be permanently shorted. If the right two or more diodes short, there will be a direct current path from the battery positive post to ground. Many diodes will burn open once they've become shorted. That will eliminate the short, but if that doesn't happen, a number of things could occur. With just one shorted diode, the generator will overheat but will still provide a little output current. A second diode will still block current flow from the battery when the engine is not running. With one open or shorted diode, the symptom during a load test will be the generator can only provide one third of its rated capacity. A 90 amp generator would still provide close to 30 amps which can be enough to run the car with only very subtle symptoms.
Even if a diode doesn't short, when a battery cable is removed, system voltage can go high enough to destroy computers and burn out any light bulbs that are turned on. The engineers have hung insane, unnecessary computers on every part of today's cars from power windows to heater controls. Some GM vehicles can have up to 47 computer modules, many of which could be damaged from excessive voltage. There is a much faster, simpler, and more reliable way to determine if the charging system is working. Use a digital voltmeter to measure battery voltage while the engine is running. Expect to find between 13.75 and 14.75 volts. You'll still need to perform a load test to verify the generator is delivering full current.
One common question has to do with determining the capacity of the generator. This only becomes an issue when ordering the correct replacement. For testing purposes, an AC generator is incapable of delivering much more than its design current. They are self-regulating in that respect. If your load test reaches 80 amps, you likely have an 85 amp generator. About three amps go right back to run the field. If your generator has a built-in voltage regulator, that three amps will never be seen by the current probe at the output terminal. If the load test shows only 40 amps, you either have a 120 amp generator with one defective diode, or you have a good 40 amp generator. Forty amps is very small and would only be found on older cars with few electrical accessories.
Many manufacturers used two or three different generators with different capacities depending on the electrical options on the car. Using one with a higher capacity will not affect anything else on the vehicle since it will only deliver the needed amount of current. Where the problem could arise is when performing a load test. Output circuits used to have a fuse link wire leading to the battery. Many newer cars have a bolted-in fuse in an under-hood fuse box. The size of the fuse device was dependent on the capacity of the generator installed at the factory. If a larger capacity generator is installed later, its output could exceed the rating of the fuse during a load test.
Alternating Current Output
We've discussed the three things required to generate a voltage mechanically, a magnet, a coil of wire, and movement between them. What we haven't covered is which direction the resulting current will flow. That is dependent on the orientation of the north and south magnetic poles of the magnet; electromagnet in this case. Because the iron core has fingers extending from both ends, and every other one is connected to the north pole and every other one is connected to the south pole, the polarity of the voltage, and therefore the direction of current flow in the stator changes direction each time the next finger passes by that coil of wire. That is why AC generators produce alternating current.
Figure 1. Voltage starts to rise.
As one of the magnetized fingers on the rotor begins to pass by one of the coils in the stator, the induced voltage begins to increase. The red arrow in Figure 1 shows the voltage rising over time.
Figure 2. One half cycle is complete.
In Figure 2, the first half cycle is complete. At the maximum voltage the rotor's magnetized finger is at its closest point to the stator coil. Then, as the north pole finger moves away and the south pole moves closer, the induced voltage goes to 0 volts.
Figure 3. One cycle is complete.
As the south magnetic field gets closer the induced voltage changes polarity and begins to increase. Current will flow in the opposite direction as in the first half cycle.
Figure 4. Two cycles.
In Figure 4 two cycles are complete. Two north pole fingers and two south pole fingers have moved past one of the stator coils. Because there are many north and south pole fingers on the rotor, many cycles will be repeated for every rotation. That increases the frequency and therefore the efficiency of the induced voltage and current.
Figure 5. Voltage from two stator windings.
As the rotor turns, the magnetized fingers move to the next group of coils in the stator and induce a voltage there. Here the voltage from the first stator winding is shown with the lighter line, and the voltage from the second winding is shown in black.
Figure 6. Three phase output.
The output of all three stator windings is shown here. This is three phase output and is very efficient.
Figure 7. One phase rectified by the diodes.
During each half cycle current flows from the stator windings through a pair of diodes and flows out of the output terminal, then to the battery. Here, one phase has been rectified. Current changes direction repeatedly in the stator coils but is directed in one direction to the output terminal by the diodes. The voltage from this one part of the stator goes from 0 volts to around 14.5 volts, back to 0 volts and repeats. Even though it is pulsing, current is only flowing in one direction. This rectified single phase is very inefficient because the voltage drops to 0 volts at times and is only atits maximum voltage for a very short time.
Figure 8. Three phases rectified.
Figure 8 shows all three phases rectified. Compared to the single phase in Figure 7, this is very efficient because voltage and current are always present, if not from one stator winding, then from a different one.
Figure 9. Ripple voltage.
The result of rectifying three phase current is very low "ripple voltage". The voltage leaving the output terminal is shown in red, and only varies about half of a volt. That ripple voltage, or variation in voltage is very easy for the battery to smooth out. Excessive ripple voltage won't hurt light bulbs but it is a common cause of a whining noise from an AM radio. See the more in-depth discussion on how ripple voltage can damage a battery in the "How Batteries Can Be Damaged" near the bottom of the "Basics of Batteries" page.
Excessive Ripple Voltage
Figure 10. Excessive ripple voltage due to one defective diode.
A common failure in many AC generators is one defective diode. Each of the six main diodes is used by multiple stator windings at different times, so the loss of that diode will result in a loss of very close to two thirds of the generator's capacity. A 90 amp generator will produce only 30 amps under a load test. To help identify the cause of low output, most professional load testers display ripple voltage, not as a number or voltage, but as a relative amount on a bar chart.
Figure 10 shows the output when one phase is missing due to that defective diode. Now the voltage will drop very low at times. Here the ripple voltage is 5.2 volts. Whenever the voltage is less than 12.6 volts, current will leave the battery to power the vehicle's electrical system. When output voltage increases above 13.75 volts, some current will go back into the battery to recharge it. That rapid change in direction mimics alternating current and leads to overheating of the plates in the battery. This problem often goes unnoticed on older vehicles without computers and electric fuel pumps. As long as the generator can meet the electrical demands of the vehicle, voltage will be within the acceptable range. In some cases it may even be a little high because the voltage regulator is reacting to the intermittent drop in voltage. On vehicles with voltage gauges in the instrument cluster, you'll see the voltage drop considerably at low engine speeds and when a number of loads are turned on. At night you'll notice the head lights dim when you turn the heater fan on the higher speeds, and turn signals may flash slower than normal. AM radios will pick up this excessive ripple voltage too. You'll hear it as a high-pitched whine that changes frequency with changes in engine speed.
For vehicles with computers, they are susceptible to unstable supply voltage and can operate improperly. This will get worse if the battery is less than fully charged. Engine hesitation and stumbling on acceleration are some of the more common symptoms. Digital dashboards are very intolerant of unstable supply voltage. Parts of the display may flicker, be dim, or intermittently go out completely. To identify if excessive ripple voltage is the cause of these symptoms, start with a fully charged battery, disable the generator by unplugging the field wire(s), then drive a short distance to see if the problem clears up.
NEVER DISCONNECT A BATTERY CABLE WHILE THE ENGINE IS RUNNING! One of the tricks mechanics performed many years ago was to disconnect one battery cable while the engine was running. They didn't understand how these simple circuits work and it was believed if the engine stalled, it was because the charging system was not working, and if the engine continued running, the charging system must be okay. There are a number of problems with doing this test and it can very easily result in a costly repair. The safety issue has to do with the sparks that will be created, the expensive components that can be damaged, and is explained in the "Do-It-Yourselfers" safety page. Here we are concerned with why that is an unprofessional procedure and an invalid test.
By now you understand ripple voltage and how it is affected when one diode is defective. A proper test is a load test with a professional load tester. If a generator has one bad diode, the most output current you'll get is one third of the unit's rated capacity. With the old trick of disconnecting a battery cable, the engine would typically still run with that reduced output so what did that procedure prove? It would appear the generator is working but if tested properly, it would not pass the professional load or ripple voltage tests.
Another problem is on some cars the high ripple voltage that results from a good generator when the battery is disconnected will cause the engine computer to shut down and the engine will stall. That makes it look like the generator is not working when in reality, it's just the procedure that is improper.
For the specifics of your charging system, click the appropriate link: