 | |||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||
|
Generators and Charging SystemsPosted 10/13/2000By Vince Fischelli
How Not to Check The Charging System Remember the term “alternator” is now GENERATOR. Yes, it will take time to mend our ways but we try, don't we? In any testing procedure, a thorough understanding of how the circuit or system works is essential in designing a test procedure that meets all conditions. So, in our first article on generators, we discussed how generators (alternators) work and produce electricity. We also discussed the role all the mysterious internal components play in the overall scheme of generator operation. Then, in the second article, we discussed the correct way to test a charging system on a car and provided a step-by-step procedure to follow that is easy and quick, yet conclusive. We trust you have been using our method of dynamic charging system testing on the car and found it quick and effective in determining charging system problems. As we have already discussed operation and testing of charging systems, we conclude this series with a discussion of other testing procedures often employed that do not always bring about the desired result of determining if a generator has a problem or not. That brings up another point that needs emphasis. Should we check a generator by itself and assume it works fine in the complete system or should we check a generator as an integrated part of a complete and functioning charging system? The answer should be obvious. It would be better to check the complete charging system as a whole to see if it is performing properly. If performance of the charging system is below specs, then investigate to see if the generator or voltage regulator is at fault, or if a bad connection exists in the charging system. Too many times a tech will load test a generator and find that under load the generator can put out 95 percent of its rated output in amps. Since the generator meets specs, the tech erroneously concludes the complete charging system is functioning properly in the vehicle when it may not be. Another common mistake in testing charging systems is to check for AC ripple using a DMM set to AC volts. A reading below 0.50 V AC is accepted as a charging system with acceptable ripple voltage. But is something being forgotten here? Yes, it is. What about electrical noise that a computer can sense and react to but a DMM cannot see? Let's discuss these two topics that cause a lot of confusion and misunderstanding in our industry. Load Testing a Generator
Figure 1 shows this typical test setup. A variable carbon pile battery load tester is connected to the battery terminals. The engine is started and allowed to warm up. Then the variable carbon pile knob is adjusted to crank in an additional electrical load on the charging system. The additional load of the carbon pile demands current from the generator, which dutifully begins to do what it is being paid to do - put out up to 90 percent of its rated current output. The current absorbed by the carbon pile is indicated on the AMPS meter on the load tester - 40 amps, then 50 amps, 60 amps, 70 amps ... Boy, this is fun! Listen to that old generator hum under the load - 80 amps, 90 amps - look, I'm getting over 90 amps outta this thing! I think I'll hold it there for a minute or two just to make sure the generator doesn't fail. Hmmm, what happened? The amps dropped to almost zero. Well, I guess the generator just blew up. Oh well, it's under warranty. I'll pull it and get another one. There are at least two problems with this test scenario. First of all, when the engine is running, normal vehicle circuits are already drawing current from the generator. A typical PCM and ignition module pull about 6-8 amps with the engine running. The fuel pump is pulling about 6-8 amps. The engine cooling fan is pulling about 4-5 amps. The passenger compartment blower on low speed is pulling about 3-5 amps. That amounts to about 20-25 amps without lights, A/C or heater, just to run the vehicle. Was this considered when the alternator (oops!) ... generator, rated at 105 amps, was load tested to see if it would put out 90 amps into the carbon pile? If the carbon pile is cranked up till the AMP meter reads 90 amps, and the additional load of the vehicle is 20 amps, the generator is actually producing 110 amps (90 + 20 = 110 amps). Will the generator get hot? You bet! Will the generator last very long under this extreme test condition? Not likely! There is a good chance the generator will be fried like a crispy critter because today's generator is not designed to operate at maximum output for very long. The second problem with this test scenario is that the generator could be loaded down so much that the charging voltage drops below battery voltage (12.66 V). Then what happens? Think about this! If the charging system voltage drops below battery voltage, the battery becomes the voltage provider to the vehicle because it has more voltage than the loaded-down generator. At this point the carbon pile is showing a high current reading. Can a battery put out 100-150 amps into a load tester? Of course it can. The problem is the current is coming from the battery and not the generator, but the generator gets the credit. Does the tech understand this fact? Probably not. He probably thinks the generator is producing the current when it's really the battery. To prevent misinterpreting the current reading on the load tester, the charging voltage should be monitored with a DMM to see if the charging voltage gets close to or below 12.66 V, at which point the battery is discharging into the load tester instead of the generator. Now two meters have to be monitored at the same time, increasing the probability of confusion and misinterpretation while the generator is on the brink of destruction. It is better to use the test procedure from the last article where we loaded down the charging system with normal vehicle loads at 1500 rpm to see how low the charging voltage dipped. If the voltage stayed above 13.10 V we know the charging system is working under load. If the charging system stays above 13.50 V it means the generator is in excellent shape to handle normal electrical loads of the vehicle without dipping too low. There is only one meter to watch, the DMM, and the generator is not placed in any extreme jeopardy. The generator is safely tested under realistic operating conditions without danger of being damaged. Testing the Charging System For Noise
A DMM set to measure AC volts will indeed check for AC ripple, as shown in Figure 2. This is a generally accepted test procedure in the industry. The reading should be less than 0.5 V AC. More than 0.5 V AC is considered excessive ripple due to a defective diode or stator winding in the generator. When ripple voltage is too high, the DC charging voltage is often fairly normal until the generator is loaded down. Then the charging voltage drops below an acceptable level and battery or driveability problems result. But (there's that word “But” again!) there is something else to check in a charging system. It is electrical noise. It usually wasn't a problem, except for creating weird noises in the car radio, until computers began appearing in cars. Now, electrical noise is a big issue that is not getting enough attention in the industry. Electrical noise can be detrimental to computer operation.
This is very confusing to a PCM. It causes the PCM to lose its place in the engine run program and “reset,” which restarts the computer program at the beginning. The result is erratic spark timing, missing fuel injector pulses and significant driveability problems that are not corrected until the generator is repaired or replaced. Voltage dropout occurs when an internal problem in the generator, such as an intermittent diode or connection, a dirty or loose brush contact, etc., causes momentary loss of output voltage. There could also be a loose or corroded connection in the B+ supply between the generator and the destination of the B+, such as the B+ pin on the PCM. Negative spikes can only be detected with a lab scope. A DMM cannot react fast enough to indicate the presence of noise spikes. If a tech checks a generator for ripple with a DMM on AC volts and gets a good reading, will he “assume” the generator is good, no spikes are present and the generator is not the reason for the driveability problem? Will he assume there are no voltage dropouts? Obviously, a lab scope is the only way to detect ripple and those pesky negative voltage spikes at the same time. Don't rely solely on a DMM to check generator ripple and ignore the possibility of negative noise spikes. It's lab scope time! We'll have to talk about positive voltage spikes at some other time because that gets more involved. By considering all the possibilities, you are more likely to find the problem the first time without missing the real cause of the problem. Don't ever leave home without your lab scope.
|
||||||||||||||||||||||||||||||||||
Copyright (c) 1996-2008. Automotive Service Association. All rights reserved.
 | |||||||||||||||||||||||||||||||||||
What about putting a load tester on the battery terminals with the engine running to see if the generator can meet current specifications — is that OK? Is that the best way or is there a better way to test a generator?
What is the proper way to test a charging system for noise? What about using a DMM to check for AC ripple? Is that enough? Is there the possibility of missing another problem that should be found? Yes, there is.
Electrical noise “rides” on the generator voltage in the form of negative “noise” or voltage spikes and at times, even positive voltage spikes. Figure 3 shows negative electrical noise as short duration spikes tracing down from the B+, which is riding at 14.2 volts. The spikes momentarily decrease the B+ and are sometimes called “voltage drop-out.” During the short-time duration of the negative spike, the PCM gets less B+ and reacts as if it were turning OFF. Once the spike passes, the PCM is abruptly turned “back ON” as the charging voltage rises to a normal 14.2 volts.
