Miles: How hard is it to do a diagnosis on an alternator that isn't charging?  Very difficult if you don't know what you are doing.  This involves a 1997 Chevrolet Cavalier with a 2.2L engine that came in with an alternator that wasn't charging.   This vehicle came in with a rebuilt alternator that another shop installed for the same symptoms.  Be warned all of the following information is boring until you encounter this particular problem and really need this information.. 

That last sentence I found very interesting. It is so true! So many times, this stuff is boring. Until we suddenly need to make a diagnosis on an unfamiliar system. At that point, we then go scrambling for information and answers.

Miles: I must say that the key to this diagnosis was to follow the trouble tree.  My thinking on a trouble tree is it is the last thing you look at when trying to do a diagnosis.  I treat trouble trees  like the assembly instructions that came with my computer.  I look at the instructions only after I have assembled all the components and when the computer doesn't work.  I think this mentality has something to do  with me being a mechanic.... I could put anything together without reading the instructions.

Well heck, I have problems putting things together with or without instructions.... (bg). Normally, I would be against trouble trees, but in this case, it would have indeed helped if one was unfamiliar with the system. Lets take a quick look at the description of the charging system (on this vehicle) found in the GM service manual

The generator provides voltage to operate the vehicles electrical system and to charge the vehicles battery. A magnetic field is created when current flows through the rotor. This field rotates as the rotor is driven by the engine, creating an AC voltage in the stator windings. The AC voltage is converted to DC voltage by the rectifier bridge and is supplied to the electrical system at the battery terminal.

The operation of the charging system is monitored by the PCM. The PCM monitors the charging system loads and adjusts the idle speed accordingly. The PCM also sends information to the instrument panel cluster (IPC) when the charging system is inoperative and tells the IPC to illuminate the volts telltale lamp.

The PCM supplies 5 volts through the L circuit to the generator. If the generator becomes inoperative, the PCM can detect the fault through the L circuit and tell the IPC to illuminate the volts telltale lamp.

The PCM monitors the duty cycle of the generator through the F circuit. As the generator load increases, the PCM will adjust the idle speed accordingly. If the IPC does not see any activity from the PCM on the F circuit, the IPC will illuminate the volts telltale lamp.

Ok, here I want to clarify something. The PCM enables the generator to charge, it does not control the rotor fielding. Simply put, if the L terminal does not receive the 5 volt enable signal from the PCM, the alternators internal voltage regulator will not apply field current to the rotor. Consider the L terminal (being PCM controlled) an ON-OFF switch of sorts.

Once enabled, the alternators internal voltage regulator will control rotor field current. Lets read a tiny bit more from the GM Service Manual:

The regulator voltage setting varies with temperature, and limits system voltage by controlling rotor field current. It switches rotor field current on and off at a fixed frequency of about 400 cycles per second. By varying the on-off time, correct average field current for proper system voltage control is obtained. At high speeds, the on-time may be 10 percent and the off-time 90 percent. At low speeds, with high electrical loads, on-off time may be 90 percent and 10 percent respectively.

Now on this particular vehicle, the F terminal of the regulator is used. This signal is an OUTPUT that originates with the internal voltage regulator. As mentioned above, the PCM uses this "information" to help anticipate and control idle speed.

Now with this background, how would we go about diagnosing the problem vehicle that Miles had? The first step, of course, would be to verify that there is either no or low alternator output. Now knowing that there is no output, we need to verify that the PCM is enabling the alternator on the L circuit.

This is where things can get confusing or go wrong....
Assuming  NO alternator output and B+ at the battery (output) post...

MOST of the time, if you simply unplug the regulator connector and (with the key on) find 5v (with DMM) at the L terminal (harness end), you have a bad alternator. So why only most of the time? One weakness of this test is that there is no load on the circuit other than your DMM. Excessive voltage drops can occur if you have bad connections etc... On the plus side though, the regulator L circuit draws very little power. Let's take a look at the L and F circuits on a properly operating vehicle.

Now if we plug the connector back into the alternator (engine still not running), what do you suppose will happen to the L circuit voltage? Remember this from the manual: If the generator becomes inoperative, the PCM can detect the fault through the L circuit and tell the IPC to illuminate the volts telltale lamp. So lets plug it back in (engine still not running) and see what happens.

The alternator is not charging (of course) and, seeing the no charge condition on the L circuit (low voltage), the PCM informs the IPC (instrument panel cluster) to illuminate the volts telltale lamp. Now, lets start the engine up and see what happens to our L and F circuits!

The L circuit has gone high again and the volts telltale lamp goes out. 

Do we now know enough to determine what was wrong the vehicle Miles was working on?

Although scan data is available for the L and F terminals (as well as ignition voltage), I do not use it for charging system diagnostics unless I find the alternator is not being enabled. In part 3 of this article, I will go over the available scan data as well as answer the above question!

Stay tuned for part 3!