Issue link: http://undercardigest.epubxp.com/i/532830
38 Undercar Digest gine speed" (rpm), manifold abso- lute pressure (MAP) for load, throttle-position sensor (TPS) to verify its operation. The first thing we wanted to know was if the PCM was acting normally. With the key on but the engine not run- ning, we monitored the TPS signal and confirmed it read 0.5 volts with the pedal released and 4.5 volts with the pedal pressed to the floor. Knowing now that at least the PCM was processing the TPS signal, we then cranked the engine to try to get it to start. The rpm reading was zero most of the time, but we could hear the engine cranking over and the sound of compression so we knew the crankshaft was actually turning. Every once in awhile, the rpm reading would jump to more than 4,500 and back to zero (Figure 4). It would continue to bounce around with numbers less than 100 and more than sever- al thousand. This was totally irra- tional. The engine during cranking just cannot act that way. Sometimes the engine would actu- ally start and idle. When that oc- curred, the rpm reading on the scan tool would read around 750 at idle and would move up evenly as we pressed the accelerator pedal. We then had to find out why the rpm gauge was responding like that. It was time to use our lab scope and actually look at the live sensor. But, which one? Remember critical thinking? Both the crank and cam sensors are inputs for the rpm gauge. The crank sensor is at the back of the engine under the trans- mission bell housing. It is very difficult to get to. But the cam sensor is mounted right on top by the firewall and is very easy to get to. Since they are both inputs to the rpm gauge, we checked the easy one first. Reviewing the service manual again, we found that the cam sensor is a three-wire magnetic pickup that responds to 5 volts supplied by the pcm. That signal will be interrupted as the conductive sprocket passes through the sensor, so the signal will be a square wave reading be- tween zero and 5 volts. Looking at the connector diagram, we found that the middle pin was the cam signal and that was where we needed to back probe (Figure 5). With the "T" pin inserted, we im- mediately read 5 volts on our scope. This confirmed that the PCM was sending a good 5-volt signal to the sensor. We started cranking the engine over, but it would not start. We knew we should be seeing a nice, clean square wave, but during that time our waveform was very erratic and irregular (Figure 6). As you can see, the signal was not clean. It was erratic and irregular. Taking this evidence into consideration, we could understand why the RPM gauge was reading erratical- ly. Remembering that sometimes the engine would start and run OK, we waited for that to happen and monitored the cam sensor waveform and found it to main- tain a nice, clean square wave. Knowing that the sensor has three wires with one being system volt- age and one being ground, we didn't want to condemn the sensor until those were confirmed. We used our power probe and proved both the power and ground were good when the problem was pre- sent. We now had evidence that the cam sensor was bad. So were we done? Did we stop here? Again, critical thinking reminds us that the crank sensor still needs to be tested. But it is very difficult to get to. Again, looking at the service manual tells us that the crank sen- sor is a two-wire magnetic pickup. This means it's going to be a sine wave instead of a square wave. Since access to the sensor itself was so difficult, we wanted to find an Figure 4 Figure 5 Figure 6 Figure 7 Critical Thinking continued from previous page