Foreign Service

Motor,  Aug 2004  by Marinucci, Dan

• E-mail
• Print
• Link

'Saving a customer some money' by recommending the replacement of only the most obviously defective parts could result in a misfire of a most unwelcome variety.

The older I get, the more I realize that successful diagnosis is part science, part art. The science part is using all kinds of test equipment to gather relevant data. But the art side of the business is the instinct or sixth sense top technicians have that prompts them to go beyond the raw data. Ultimately, going beyond the raw data prevents comebacks and creates happier customers.

This month's story concerns a 1997 Honda Civic that came back because it received unequal parts science and art. When the car first rolled in, DTC P0302 (cylinder 2 misfire) was set. After questioning the customer carefully, the technician assigned to the car learned that the engine hesitated during acceleration. The problem usually occurred on damp mornings and would go away within a few minutes of driving.

This tech habitually performs a careful visual inspection of the ignition system whenever he diagnoses a misfire problem. If you aren't familiar with this Civic, its coil is located inside the distributor. In this case, the coil looked powdery white and all of its terminals were corroded. Internally mounted ignition coils tend to get hot anyway, but the burned look of this coil convinced the tech that it had been running hotter than normal. High firing voltages can overheat a coil, and worn spark plugs are a common cause of this condition. Sure enough, the tech found extremely wide spark plug gaps when he removed the plugs. It appeared that the plugs hadn't been replaced for a long time.

The tech replaced the coil and spark plugs and scope-checked the ignition system. The engine ran smoothly and the misfire code didn't come back. He cautioned the customer about the need for regular maintenance and scheduled the car for a timing belt job several months later. The good news is that the customer showed up for that appointment. The bad news is that the car was hesitating again and the ECM had flagged misfires in both cylinders 2 and 4 (P0302, P0304).

Now the customer claimed that the engine hesitated only on cold mornings, but the symptom was more prevalent in damp or rainy weather. He complicated matters by demanding that he get his car back as soon as the timing belt was replaced. He claimed he could live with the hesitation for the time being but didn't want to risk a broken timing belt.

The major maintenance services went smoothly and the tech had a little time for testing before the customer showed up. He grabbed a lab scope for some quick ignition and injector tests. His reasoning was a practical combination of art and science: All the wires were easy to reach and an ignition or injection problem could very well cause the misfire codes.

Fig. 1 on page 14 displays primary ignition voltage (top) and current (bottom). Both patterns show clean, consistent primary ON/OFF signals as well as a normal primary ON time (dwell period). This suggests that the ignition module is doing its job. But on the upper pattern, both the firing and spark lines are erratic and unstable. (The firing line is vertical, the spark line is horizontal.) In this screen capture, the spark line has disappeared altogether.

An old ignition axiom states that the primary circuit reflects the secondary circuit. Of course, the secondary is the higher-voltage of the two circuits. A corollary to that axiom is that the primary ignition spark line is a miniature version of the secondary one. Simply put, the shenanigans in Fig. 1's voltage pattern strongly suggest serious secondary ignition trouble brewing somewhere.

The injector wires were easy to reach and the tech already had his inductive, low-current pickup hooked up for the primary ignition test. Maybe a quick injector check would uncover the cause of the misfire codes. The injector current pattern in Fig. 2 shows a normal slope and clean ON/OFF signals. But the straight vertical line dropping down to 0 suggests that the injectors were shutting off momentarily for some reason. Hmm...a car with multiple problems is not uncommon, but by now all the tech could do was plan more-detailed tests whenever the car returned.

Our intrepid tech got a big break when the customer dropped off his Civic while he went on vacation and it began raining the same day. By now, the engine would almost stall during acceleration. Distracted by the unusual injector current patterns, the tech swapped injectors from cylinders 2 and 4 with those in cylinders 1 and 3, respectively. However, the unusual patterns didn't follow the injectors to their new locations.

Another big break occurred when he listened more closely to the distributor cap. By now he could hear popping and snapping sounds inside the cap! Instinctively, he shut off the lights in his bay and watched the distributor cap and wires. Occasionally he could see voltage arcing from a distributor cap terminal right down the side of the cap. Little lightning bolts were also visible between the cap and the distributor housing.