In my February column, I talked about not understanding why the weather service in Tucson reports the barometric pressure to be between 29 in/Hg (inches of mercury) and 30 in/Hg. At 2500 feet above sea level and deducting 1 in/Hg for each 1,000 feet, Tucson's base barometric pressure should be 30 in/Hg - 2.5 in/Hg = 27.5 in/Hg. Looking at the MAP voltage charts for key-on-engine-off, the MAP BARO readings indicate Tucson's average BARO at 27.5 in/Hg. I checked Denver's BARO reading on the Internet and it had the same reading of near 30 in/Hg at 5,000 feet. How could that be?

Well, reader Tom Finn called with the answer. The weather service reports a "corrected" barometric reading. Everybody knows that if the reading is lower (less than 29 in/Hg), a low pressure front is moving in and a storm is on the way. Your local barometric pressure is corrected to sea level by adding on the inches missing from altitude. This gives everyone a standard point of reference even though it is really incorrect. Thanks, Tom, I knew it had to be something like that, but all of my reference materials failed to point this out.

OK, we've been talking about how the computer determines fuel delivery. We've covered the basics, now let's talk about those "wildcards." This subject is huge and getting larger every day with more powerful computers installed in cars. I'm going to try to cover many of them briefly and help you to identify when one is in command.

We know that when a feedback system is working correctly, the O2 sensor takes final trim command for fuel delivery. Usually running the engine at 2,500 RPM is good enough to test this function; however, some cars used the vehicle speed sensor (VSS) and timers to go into a special strategy. This strategy was designed to keep the catalytic converter cool when stopped. So to truly test these cars, you must test drive or drive on rollers to get the VSS to input the needed trigger for closed loop status.

The test drive method will also eliminate any special strategies relative to the PRNDL (park, reverse, neutral, drive, low) switch, assuming that it is working correctly. The PRNDL switch will often eliminate the function of the EGR valve, knock responses, ignore the O2 and use open loop strategies, and invoke rev limiters, etc. Also on the test drive, you should be able to test the response of the O2 sensor by accelerating hard to drive the O2 to the rich limit and decel hard to achieve the lean limit.

Use your scope to run this test. The scanner is not the best tool for this job because the diagnostic mode of the scanner may invoke its own special strategies. Many techs have worked long and hard to find the cause of a high idle with a scanner connected, only to discover this "problem" was the scanner's diagnostic mode. During your test drive, make sure all loads are off that may affect the fuel strategy, air conditioning, lights, etc.

Let's assume your test drive showed the O2 sensor to respond well to rich and lean conditions, so it's OK. However, the system failed to achieve closed loop when it should have. After we eliminate the possibility of basic sensors causing this problem (MAP, ECT, TP, IAT, etc.), we check for special sensor inputs. These are the other input sensors such as a power steering switch, an air-conditioning input (either AC request or a pressure switch), a turbo overboost switch or any other special input used on that vehicle. How do you determine these inputs? Print out the computer wiring diagram for the vehicle you are working on and use a highlighter to mark all inputs. I never work on a driveability problem without a printout of the wiring diagram that I can write on.

Now test all "special" sensor inputs. If the vehicle has scan data, this is a great place to test these sensor inputs. Watch the pulse width and the idle command, and turn the wheels back and forth to test a power steering switch. Turn on the air conditioning and look for a response. Watch the PRNDL switch operation on the scanner and look for a response in the idle command and pulse width when going into gear. These are important inputs to test when a car is stuck at a higher-than-normal idle.

Now let's cover some not-so-obvious internal strategies. We know that some cars use rich idle strategies and some use lean strategies. Some cars use idle counts to determine fuel delivery at idle. Why? Well, the O2 sensor has a tendency to cool down, and the idle counts are a perfect indication of air delivery at idle. OK, perfect when the car was brand new, problems occur when the throttle sludges up and the idle air counts go up. Now the same amount of air is going in, but more fuel is being delivered, causing a rich condition. How about a vacuum leak? Idle counts go down to achieve the desired idle speed; however, now less fuel is delivered and you have a lean misfire condition you can't fix unless you understand how the fuel is calculated. If you have scan data, check those idle counts! When tuning an engine, clean those throttle bores! Before starting a diagnostic procedure, always clean the throttle body.

Most cars will relearn idle speed pretty quickly after disconnecting a battery. Mitsubishis seemed to take up to three days to relearn idle after a battery service. This was discussed in the Tech section of CompuServe about a year back. The solution was to clean the throttle body when doing a battery service to get back near the base strategy so that the learn procedure didn't take so long. This is the wise move to keep the customer from coming back with that "it didn't do that before" comment. Also it is a legitimate service to charge for and may prevent a dangerous stall condition. I have since been told that Chrysler products can only relearn five idle counts between key cycles. So rather than taking three days, if you do a number of restarts it should relearn idle. I have not confirmed this, but it makes sense to me.

Some cars are now using lean cruise strategies. After you have driven at a minimum speed, RPM, MAP voltage and a timer times out, then the engine may enter a lean cruise mode. A typical example would be 15 minutes at 45+ MPH, a steady MAP reading under 2 volts and more than 1,500 RPM. The computer will start to lean the engine until the O2 reads about .2 volts. The main problem with running lean is the production of nitrogen oxides (NOX); however, at low loads this isn't a problem. So don't be surprised to see lean conditions under these conditions.

Failure strategies can also create some real diagnostic grief. Assume you have a car with no data stream and the engine coolant temperature (ECT) sensor circuit delivers a false high temp signal to the computer. Now the sensor for the dash gauge is separate and no overheat is indicated to the driver because the engine temp is really OK. The computer will invoke a special strategy to prevent or minimize overheat. Usually this strategy will include a richer mixture and retarded timing. On a Cadillac Northstar, this strategy will include switching off injectors so that the engine pumps air on every third power stroke, effectively air cooling the engine even with no coolant in the system. Often, no code is set for this condition.

Some new active test strategies for OBD II compliance will be felt by the driver. Knowledge of these problems will save time chasing ghosts. Just remember, one of these demons may be causing you grief. Eliminate what you can and then expand your search. Use every resource available to you. Happy hunting.

#INCLUDE FILE = "/autoinc/includes/warren.inc"

RATE THIS ARTICLE What do you think of this article? Your input will help AutoInc. develop additional articles on this subject. Share your thoughts! Your name Your e-mail address

MOST ACCESSED ARTICLES Fuel Injection Service, Not Just Cleaning The Art of Extraction EGR Systems: Operation and Diagnosis Proactive Target Marketing:_Rethinking Your Business Strategy Engine Performance: HO2S Diagnostics MOST E-MAILED ARTICLES Developing Employee Potential How Critical Thinking Can Help Your Business How to Diagnose the Ford Glow Plug What to Look for When Shopping for the Right Shop Management Software Putting a Price Tag on Complaints

AutoInc. Web Site | ASA Web Site | NHTSA Air-bag Proposals Being Reviewed, Final Rules Near | Value-Added Programs: Helping Auto Body Repair Shops Prosper | Thinking About Expanding Your Shop? | Automotive Service Network Programs: Components and Requirements | Guest Editorial | Tech to Tech | Tech Tips | Shop Profile | Net Worth | Chairman's Message