LIGHT'S OUT? EFFECTIVE NO-CODE DRIVEABILITY DIAGNOSIS

Motor,  May 2004  by Seyfert, Karl

• E-mail
• Print
• Link

Before vehicles could store even a single DTC, every driveability problem was a no-code driveability problem. The methods used to diagnose vehicles then work just as well today.

It didn't take long to get used to working on 1996 and later OBD II-compliant vehicles, each equipped with a PCM that can report hundreds of diagnostic trouble codes (DTCs). There are so many DTCs, it can be hard to imagine a situation where a vehicle could have a driveability problem without a corresponding DTC to help you determine the probable cause.

Of course, it's entirely possible for a late-model car to have a driveability or other running problem before the engine management computer has any inkling that anything is wrong, and long before it gets the urge to illuminate the malfunction indicator lamp (MIL). When this happens, you need a strategy to quickly and efficiently home in on the possible causes and make the repair. That's what this article is about. DTCs can be your friends, but they aren't the only friends you have.

DTC History & Limitations

Before we kiss off DTCs, a little time spent on their history might be instructive. Once you understand what DTCs can and can't do, you'll see why it can be dangerous to place too much trust in them in the first place.

The earliest DTCs were based on very simple concepts. The PCM was primarily interested in spotting sensors that were reporting out-of-range values. It accomplished this by comparing the sensor readings to standards recorded in its permanent memory. If the readings were out of range or missing for a specified length of time, the PCM set a DTC. Most of the time, it took a sensor that was either open or shorted to get the PCM's attention. If the sensor was reporting just about anything else, the PCM was happy.

As programming became more sophisticated, PCMs began to look for situations that would not make sense if the system were working properly. Suppose the coolant temperature sensor continued to indicate that the engine was cold, even though it had already been running for several minutes. The PCM should be equipped to recognize this as a problem in the coolant temperature sensor circuit (or a defective thermostat) and record the appropriate DTC.

The PCM may also compare two or more sensor signals to each other when it conducts what are called rationality tests. For example, if the vehicle speed sensor is reporting that the car is at cruising speed, yet the throttle position sensor is sending a signal indicating idle, a DTC should set. One of the two sensors must be lying.

To get a complete picture, the PCM also watches actuator circuits, such as those that power the automatic idle control motor, the fuel injectors, the canister purge solenoid, etc. If there's a short, an open, high resistance or other trouble present in any of these actuators, a DTC should set.

Sounds pretty good so far, doesn't it? But how many times have you found a problem in a system that was missed completely by the PCM? Or perhaps the PCM registered a DTC that sent you on a "snipe hunt." Maybe it directed you to a sensor failure when an entirely different part of the system was the real culprit. No PCM self-diagnostics yet devised can help you in all cases. As sophisticated as it may be, it still has limitations, and those limitations have a lot to do with basic assumptions. In addition to the data it collects, the PCM must make decisions based on those assumptions, which may not always be correct.

For example, the PCM has to assume that the engine has normal compression in all cylinders. If compression is low in one or more cylinders, this assumption will be incorrect. The PCM may respond by sending inappropriate commands to the fuel injectors, ignition system and other components that would make an engine with good internal components run properly, but not be able to make up for a low-compression condition.

Suppose one fuel injector is clogged. That cylinder will then pump enough air into the exhaust to make the oxygen sensor produce a low-voltage lean signal. The computer, assuming all injectors are operating properly, will increase the pulse-width command to all the injectors to enrich the air/fuel mixture and correct the lean condition. This will cause the other cylinders to run too rich.

The PCM also assumes it has a reliable power supply and good system grounds. When the vehicle left the factory, all ground circuits and connections in the system were free of resistance. Years later, this may no longer be the case. Connections may be corroded or loose, which will cause incorrect information to be fed to the PCM-information it will assume to be correct.

The PCM also assumes the vehicle has a properly functioning cooling system, a free-flowing exhaust system, correct fuel pump pressure, good spark plugs and other secondary ignition system parts and mechanical emissions control systems in working order, to name a few. That's an awful lot of assumptions to place your trust in, don't you think?

Where to Begin