 | |||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||
|
Picking the Fruit Off the Trouble TreePosted 11/7/2001By Jeff Bach
It was common practice by many experienced technicians throughout the '80s and early '90s to use a combination of high-tech equipment, factory recommended technical trouble shooting diagnostic procedures, previous model familiarity, basic automotive system technical knowledge and some speculation to formulate a hypothesis concerning the cause of an MIL or driveability complaint. To properly diagnose and repair a problem of this type requires years of training and experience coupled with high-tech equipment including electronic meters, scan tools, scopes and such. Equally as common was a methodology that involved a bit more conjecture and a lot less training, tools and experience. This method probably worked as well or better than the aforementioned “proper” method for some technicians. This is what I call picking the fruit off the trouble tree branches. It involves the use of a “secret code reader” and a “decoder card.” These tools, which could be purchased at local chain parts stores, were also used by most “Weekend Home Auto Technicians” (WHATs) with minimal experience to aid in diagnosing their own - and their friends' - MILs, CELs and driveability problems, saving them countless dollars and time they would have spent at an “auto repair facility.” In the early '80s it was simple to cross over those two terminals and blink out a code, then look it up on the card to see which part was bad. A real good code reader (the $29 one) would read out the name of the sensor right on it. If the part didn't fix it, then the computer got replaced for lying. The computer usually took care of the problem if the part didn't, at least for a while. I used to get to see the cars after they had already “eaten” several computers, usually with the presumption that it needed a new wiring harness. I finally realized why I would get so many cars from DIYers that had had new computers and needed wiring harnesses. These guys have somehow gained access to real live trouble trees and figured out the secret ... the fruit is on the ends of the branches. You almost always start at the very end branch with a new computer, then go up to the next branch to see what part made the computer go out. Usually the branch on the right is it if the one on the left says to check the CKT. If you don't find a box that says “replace ####,” and all it talks about is checking for CKT this, or CKT that, you need a new wiring harness. The top of the tree is just there to hold up the branches. Did you ever notice that when you get one of those cars in that has been all over the place trying to get repaired, the list of new parts it's had tried on it is the fruit at the ends of the trouble tree branches? With some of the trouble trees I've seen, this logic pattern makes more sense than it should. Now that the mid- to late-'90s cars have parts that are reaching retirement stage in their designed obsolescence, we are starting to see more and more that DIYers and WHATs are doing less and less to their cars before they bring them in. With the advent of OBD II, redesigned factory troubleshooting guides, trouble codes that won't clear when you disconnect the battery, PCMs that won't run the car until they are flashed, etc., these guys are all but locked out of the weekend repair business. But then, there's always the salvage yard to get your preflashed PCMs from. Forget about axle ratios and optional accessories, it's the right part number and it runs. Then there are the “total builders.” These guys buy a total at the salvage pool, then get a spare from their favorite salvage yard towed to their place of business to use as a donor. Once the bodywork is completed and every conceivable part has been substituted (sometimes including the complete engine harness) in an attempt to turn out the appropriate dash lights - all to no avail - we are given our chance at solving the problem. It gets to be like a carnival game sometimes. It would be nice if a bell rang when the MIL went off, then, a “you win” light flashed like 10 times and the service manager handed you a cigar. There are two kinds of total builders. There are the ones with experience repairing vehicles in professional auto shops who decided to strike out on their own, and then there are the ones who wear ties through the week and watch “Tim the tool guy.” They figured out a way to make some extra vacation money by building and selling totals. These guys are the most likely to commit “sins of omission.” You can usually spot one of their “units” by the glue seam on the air bag cover or the absence of one or two warning lights on the instrument panel (usually ABS, Air Bag or Service Soon). The sins of omission may also carry over to the description of the problem and which parts have been previously replaced. These guys are likely to say something when picking up their vehicles like, “So all it was was a wire shorted, huh?” This is to let you know they know you don't have any tangible assets to bill them for ... just time. These are the kind of customers that are OK with a laundry list of parts on their bill that didn't fix the problem 'cause they got a lot of miles on them and probably needed replacing anyway. They have a hard time relating, though, to diagnostic charges or testing fees. They know that you know exactly which sensor to put on their car to fix it when you hook up your “machine&3148; and that the rest is just going through the motions so you can charge more. I love it! I recently had a customer with a Cadillac that had a “service soon” light on and an error message on the driver information center. This one turned out to be a rather expensive repair as he was getting a code EO49 “Air System Fault.” The car had a rather new looking AIR management system valve on it already. The trouble tree chart led to the valve being the problem again. After checking the circuit, I pulled the valve off and bench tested it with a vacuum pump and test leads while blowing through it with compressed air. As I suspected, chunks of what used to be AIR pump vanes came out of the stuck-open AIR switching valve. The owner just happened to be in the neighborhood and stopped by to check on our progress. I informed him that he would need a new AIR pump and another AIR management valve, and he asked if it was a code 49 again. When I informed him that it was indeed a code EO49, he wanted to know how the computer could tell that the valve was bad. Fortunately, I still had the test and failure conditions fresh in my RAM. I was able to explain to him that the ECM tested the AIR system after about 10 minutes of closed loop operation on a warm engine while the vehicle was stopped with the rpm above 400, by energizing both the AIR switching and AIR divert solenoids, causing the AIR management valve to divert air to the exhaust manifold. The extra air causes the O2 sensor voltage to go low, which the ECM interprets as a lean condition. It then compensates by increasing fuel trim. If the ECM doesn't see an increase of 15 or more counts on the fuel integrator after four tries, it sets code EO49. After hearing my explanation of the test and failure conditions, the customer informed me that the valve, ECM and O2 sensor had all been replaced previously by him. He wanted to know how I had arrived at the conclusion that he now needed both the expensive AIR management valve and the AIR pump. It was then that I produced the waveforms in Figures 1 and 2. I explained that I verified the AIR switching and divert solenoids circuit integrity by connecting a current probe and lab scope to them and through the on-board diagnostic system, I command the ECM to cycle the solenoids while I collected the waveforms. After determining that they were functioning electrically, I ran the engine and physically checked the airflow after having the ECM turn them on through its override function. Once I determined the airflow was not switching, I pulled the valve and operated it on the bench while blowing air through it. I showed him the pieces of the pump and explained that once the pieces are wedged in the valve they won't come out because the valve is a molded piece and doesn't disassemble. Now a “confident skeptic,” he gave the OK to repair the vehicle. He said surprisingly little else before asking for a copy of his waveforms. He has since sent other customers to us, a sign that this weekender has succumbed to the technology battle and surrendered his would-be clientele. This car is more than 10 years old. Today I'm looking at a '97 Cadillac STS with a code for low efficiency three-way catalyst. The circuit description takes a whole page, then there is a list of 42 codes that can't be set, along with 15 test conditions that need to be met. The failure condition is simple “Stage 1 average difference between actual O2 sensor deviation and max allowable deviation greater than 93 millivolts. Stage 2 average of the average differences calculated in stage 1 is greater than 9 millivolts.” If you thought a code 42 was fun, try a P0420. OBD II will serve to weed out the less serious auto technicians, at least the ones that want to pick fruit from the low branches. The fruit is actually better farther up in the tree; you just need a ladder to get to it. I say, “Bring on the high-tech failures. OBD is the best thing to hit our industry since the Montreal Protocol.”
|
||||||||||||||||||||||||||||||||
Copyright (c) 1996-2008. Automotive Service Association. All rights reserved.
 | |||||||||||||||||||||||||||||||||
;){kind=link}
;){kind=link}