Hi guys. Aarnold1, you don't realize it but you're suspecting the battery and that's a very good possibility. There's no way to sugar-coat it, beginning with the '87 model year, GM went from probably the second best generator design to by far the world's worst pile ever. Due to the "switch-mode" voltage regulator, it turns the field winding on and off about 400 times per second. Average voltage is controlled by varying the "duty cycle"; that's the on-time vs. Off-time percentage. When it turns off, being a coil of wire, the field winding produces a huge voltage spike when the magnetic field collapses, just like an ignition coil does. This is such a problem that they even use special "zener" diodes to short out those spikes.
The rest of the job of dampening and absorbing those spikes is done by the battery. As they age and the lead flakes off, their "internal resistance" goes up and they lose their ability to dampen those spikes. The spikes wreak havoc with computer sensor signals so computers do weird things, and those spikes can easily destroy the internal voltage regulator and diodes. It is REAL common to have repeat generator failures. Many people go through four to six in the life of the vehicle. What many professionals are finding out is to prevent those repeat failures, replace the perfectly good battery at the same time the generator is replaced. That old battery might work okay in an '86 or older model.
I can't say all the symptoms, (nice job of documenting them, by the way), are related to the battery, but especially since it is that old, get a new one in there, then continue with the troubleshooting. At least we can eliminate that potential variable. Everything you mentioned is in some way connected to a computer on the truck so multiple symptoms can be expected. If the battery does solve some of the problems, you would have been chasing your tail trying to figure out some other cause that wasn't there.
As for the starter voltage, the industry standard, unless the manufacturer specifies differently, is 9.6 volts or higher during cranking. This is where that internal resistance comes into play. A good, fully charged battery will measure close to 12.6 volts. Where did that three volts go during cranking? Whenever current flows through anything with resistance there is a voltage drop, just like there's a pressure drop at a kink in a garden hose or compressed air line, but only when water or air are flowing. Battery cables have a little resistance which is undesirable, but that's why a tiny amount of voltage is dropped across them during cranking. We would like ALL the battery's voltage to be dropped across the starter, but a good chunk of it is dropped right across the plates in the battery, ... Two or three volts worth. We can't measure that resistance with an ohm meter, but we can see the results of it during cranking or a load test. Current flow through the battery's resistance creates that two or three volts voltage drop, and you find, in your case, 10.5 volts where you expected to find 12 volts. 10.5 volts is very respectable and means the battery can supply enough current for the demands of the starter motor.
There's two ways to test a battery, and I'm guessing your mechanic either didn't fully understand what he was doing with that piece of equipment or he wasn't good at describing what he did. On higher-class testers you dial in half of the battery's cold cranking amp rating, then watch to see if the voltage stays above 9.6 volts for 15 seconds. If it drops below 9.6 volts in 15 seconds or less, the battery doesn't pass the test, but you have to make a judgement call on how bad it failed by. If it hit 9.5 volts right at the end of the 15 seconds, yeah, technically it failed, but it's still doing a fairly good job if it's pretty old. A one or two-year-old battery that just squeaks by isn't going to get better with age, just worse, so that would be more cause for concern.
Some less-expensive battery testers have just one resistor to draw the same load on any size battery, then you have to look at a chart to see how much voltage it should be able to maintain with that load. A tiny riding lawn mower battery will have the same load put on it as a huge diesel truck battery. The little battery might pass if it can struggle to maintain 6.0 volts while the big battery might fail if it can't maintain 11.5 volts. That's like comparing the weight lifting ability of two men. You would be impressed if a 90 pound weakling could lift 100 pounds, but if the 300 pound giant couldn't even lift 200 pounds, you'd be disappointed. I never used that kind of tester so the voltages I used in my story are just examples. Same with the weight lifters!
The bottom line is your battery passed a load test with flying colors, ... I think. They said they found 12.47 volts but only 382 cold cranking amps. That sounds like the first type of tester I described. They have a variable resistor with a big **** that lets you dial in one half of the 700 amp cca rating. It's impossible to dial it in exactly but 382 is more than the desired load of 350 amps and it still maintained at least 9.6 volts. Assuming they did the test correctly and for the full 15 seconds, cranking-wise, the battery is fine. It still can have high internal resistance though which will allow the generator's voltage spikes to be an issue.
To my knowledge, only GM has the voltage spike problem. Their older "SI" generators from the early '70s through '86 also had an internal voltage regulator and it also switched on and off 400 times per second but they never had this problem. I don't have an answer for that. Chrysler switches theirs the same way but the regulator is in the Engine Computer where it can adjust the target voltage for many other factors such as wide-open-throttle, hot or cold ambient air temperatures, and things like that. They also use a switching regulator without any problems. VCR power supplies, which I'm very familiar with, use the same system. If you understand electrical theory, volts times amps equals power, (watts). Watts equals heat, the deadly enemy of electronic components. When the switching transistor is turned off, 0 amps flows, so there is 0 watts dissipated and no heat. When it is turned on and current flows, there is no voltage dropped across the transistor, so again, no watts, no heat. That means they can control the rate of charge with a tiny inexpensive transistor. Think of letting the water drain out of your bathtub. It takes no effort on your part to let 'er flow as fast as it wants to, and it takes very little effort to totally block the drain with your hand. Full on or full off, with very little effort. If you wanted to slow the rate of draining, it would again take very little effort to let 'er rip for a few seconds, then completely block it for a few seconds, and keep switching that way. If you tried to block exactly half of the drain opening, you would struggle and use a lot of effort to hold your hand in the right spot. THAT'S where a regular transistor used to control the average generator field current would have some voltage across it and some current through it. Multiply those two numbers and you get some watts which equals lots of heat, a big power transistor, and it's mounted to a huge aluminum heat sink to keep it cool.
Way more technical than you need to worry about, but in spite of those voltage spikes, they did have a reason for designing the voltage regulator the way they did. I'm not an engine performance specialist but I can share that the oxygen sensors on your truck develop a small voltage that should switch between around 0.2 and 0.8 volts a couple of times per second. When the generator develops a voltage spike, the higher voltage, (electrical pressure), causes a corresponding spike in current flow to the battery. Current flowing in a wire sets up a magnetic field around that wire. (That's how kids make electromagnets to pick up paper clips with a small battery). The opposite is also true. When another wire is nearby, that magnetic field from the adjacent wire "induces" a voltage in the second wire. That's the way the secondary winding in an ignition coil works. Same for any power transformer in your house. If a big voltage spike occurs in the wire going from the generator to the battery, it can easily induce a voltage spike of a few volts in an adjacent wire in that same harness, and that wire could be for one of the oxygen sensors. The Engine Computer is purposely turning the injectors on and off in such percentages to make the mixture switch rapidly from too rich to too lean, and it expects to see corresponding readings from the oxygen sensors. Now add in these extra induced voltage signals and the computer sees different signals that it may try to correct by making changes in the amount of fuel it commands from the injectors. The result could be too much fuel and poor fuel mileage, or not enough fuel and hesitation and stumbling. Driveabilty experts get involved at that point. A simple trick to identify that cause of a running problem is to unplug the small electrical connector on the side of the generator so it produces no output and no voltage spikes. If the running problem clears up, it's a good bet the generator / battery is at fault.
Thursday, July 21st, 2011 AT 9:27 AM