AutoFocus: Nissan Sentra RPM comes back from the dead

Sept. 22, 2010

I recently called a friend at a local Nissan dealership. As soon as I said hello, he said, “Have I got a car for you!” He had been working on a 2001 Nissan Sentra for two days. The original customer complaint was that the engine would intermittently misfire and run terrible. The vehicle came into the shop running fine and became a no-start. Now two days later, nothing had changed.

Camshaft Position (CMP) sensor DTC P0340 would set right away and Crankshaft Position (CKP) sensor DTC P0335 would set after cranking the engine for a while. Scan tool data showed zero engine RPM while cranking. Unfortunately the dealership did not have a scope to check the CMP and CKP sensor signals. The CMP and the CKP sensors were tested with a voltmeter and then replaced. Both sensors had good battery voltage and ground and their signal wires had good continuity to the ECM. There was no change in the problem.

The ECM was then replaced and was programmed so the immobilizer system would allow it to start; however, it still had the same problem. The Nissan tech line recommended replacing the timing chain due to chains stretching and causing CMP to CKP signal phasing problems. At the end of Day Three, a new timing chain had been installed and the same problem existed.

I was fortunate to have a training day that week and thought this would be a great opportunity to help a friend at the same time. I arrived at the dealership on the morning of Day Four with my scope in hand. The first thing I did was scope the CKP sensor signal at the ECM; it had a perfect square-wave pattern while cranking. Scoping the CMP sensor signal at the ECM showed a square wave pattern as well. In the beginning I thought the CKP sensor was the main input for engine RPM on the scan tool so I didn’t look that closely at the CMP sensor signal. With all components in the system replaced and having CKP and CMP sensor signals at the ECM, it seemed the new ECM might be defective but that was very unlikely.

All the ECM powers and grounds were tested for voltage drop. The 5-volt reference was verified coming from the ECM to components such as the TPS as well, so we knew a 5-volt reference wasn’t shorted affecting the ECM. Fortunately, my friend also owns a 2001 Sentra but his did not have the immobilizer system so the ECMs were not interchangeable. I wanted to find out if a missing CMP sensor signal could cause no engine RPM reading on a scan tool while cranking so we brought his vehicle into the shop and unplugged the sensor. To my surprise, the scan tool showed zero RPM while cranking.

It was obvious now that I needed to look at the CMP sensor signal closer on the problem Sentra. I increased the millisecond setting on my scope to get a better look at the CMP sensor signal groupings and the pattern. When I rechecked the CMP sensor signal at the ECM, there was no scope pattern at all, which seemed odd because there had been a pattern before. The CMP sensor signal was checked at the CMP sensor -- still no pattern.

When I back-probed the CMP sensor signal at the sensor with the key “on,” I noticed the voltage line on the scope would sometimes jump to around 10 volts and sometimes stay at zero volts. As I wiggled the sensor connection and harness, the voltage would come and go. I verified the battery voltage to the sensor and it never went away.

I put a voltmeter on the CMP sensor signal on my friend’s car and the voltage read 10.08 volts from the sensor with the key “on.” With the sensor unplugged, the signal voltage read zero volts so the voltage was coming from the sensor.

Nissan has a new terminal testing tool that is the size of the male terminals used in their connectors. The tool is spring-loaded so when it is pushed into the female terminal in the harness connector, it will compress the spring. If the spring does not compress, then the female terminal is spread out too far, causing a poor or loose connection. The terminal testing tool had been used at the ECM pin terminal connections earlier in the diagnosis and everything tested fine. Using the terminal testing tool on the CMP sensor harness connector, the spring did not compress on any of the three CMP sensor female terminals.

The connector was taken apart and the terminals were tightened. The connector was reassembled and plugged back into the CMP sensor. Now the CMP sensor signal voltage with the key “on” showed 10.08 volts. After I connected the scope back to the CMP sensor signal and cranked the engine, there was a perfect CMP sensor signal. The scan tool showed normal engine RPM cranking and the spark tester was showing spark. The spark plugs and ignition coils were reinstalled and the engine started right up -- problem solved.

Lessons learned: The CMP sensor signal is necessary for engine RPM input on this vehicle, and the CMP sensor and CKP sensor should have about 10 volts coming back out of the sensors, checking the voltage while back-probing with the key “on.” The 10-volt signal may be affected by the engine crankshaft position. In that case, it may be necessary to remove the sensors from the engine during testing.

Note: The sensor(s) should always be plugged in during testing. Thank goodness for training days!

(For information on Identifix, visit