Remember the old carnival game where a dealer hides a pea under one of a half-dozen walnut shells? After the dealer artfully shuffles the six shells, you’re supposed to pick the shell with the pea hidden under it. Good luck with that. As veteran diagnostic technicians know, diagnosing a no-code intermittent stalling complaint can be like playing the old carnival shell game. We know it’s one of maybe six sensors, but why can’t we find the diagnostic trouble code (DTC) that tells us where to find the pea?
SENSORS AND ECMS
Sensor failures normally occur as one of three failure categories: open-circuit, shorted-to-ground, and shorted-to-voltage. In many applications, the ECM feeds a bias voltage into a circuit to detect when the circuit is open, shorted to ground, or shorted to voltage. If, for example, the ECM detects 0.45 bias volts on an inoperative circuit, the ECM senses an open circuit. If bias voltage is pulled down to zero, the ECM senses a shorted-to-ground circuit. If circuit voltage is greater than bias, the ECM senses that the circuit is shorted to another voltage source.
Sensors can also lose their calibration, which usually results in an engine performance complaint. Calibration failures are detected by a rationalization process in which the ECM compares one sensor output with similar sensor outputs, or with an anticipated range of sensor performance. In any case, the ECM doesn’t always store a code when a circuit or calibration failure occurs, which is the topic we’re going to discuss in the following text.
Writing enable criteria for setting codes isn’t an exact science. Enable criteria might include the length of time required for the engine to reach operating temperature. Enable criteria might also include a momentary disruption (glitch) in the voltage return signal. At the other end, a sensor failure might also include a specified amount of time before a DTC is stored in the ECM. If a sensor failure falls outside of designated parameters, the ECM might not set a DTC. In addition, be aware that some enable criteria, as written, can be very misleading.
TEMPERATURE SENSOR DIAGNOSTICS
Temperature sensors contain a thermistor that changes electrical resistance in response to changes in temperature. Temperature sensors are normally two-wire circuits (see Photo 1). A five-volt reference flows from the ECM, through the thermistor, and returns to the ECM as a reduced voltage signal. The scan tool changes the voltage signal into Fahrenheit or Centigrade degrees and displays it as a temperature parameter indicator (PID).
Temperature circuit failures can cause an engine to stall, whereas sensor calibration problems usually cause an engine performance problem. An open circuit normally indicates -40º F, which can cause a major rich-fuel condition. In contrast, a circuit shorted to reference voltage indicates about 300º F, which can cause a major lean-fuel condition. Any calibration failure is caused by a temp sensor supplying an incorrect voltage to the ECM, which usually causes a minor rich or lean condition.
WHEN TEMP SENSORS FAIL
The ECM normally sets a temperature sensor code (ex. P0116) by comparing all of the temperature sensor inputs after an overnight cold-soak. Many intermittent no-code stalls are caused by millisecond-duration glitches in sensor signals that fall outside of the enable criteria. Calibration errors can also fall outside the range of enable criteria. In many cases, it’s more cost-effective to replace a suspect sensor than to spend valuable diagnostic time trying to catch it in a failure mode.
CRANKSHAFT SENSOR DIAGNOSTICS
Misfire diagnostics aside, the most important crankshaft position sensor (CKP) function is indicating top dead center (TDC) on one number-one piston. Since the CKP signal is vital for activating the fuel pump, fuel injectors and ignition systems, it can also be a prime suspect when diagnosing an intermittent, no-code engine stall.
CKP sensors fall into three basic types: magnetic reluctor, Hall Effect and magneto-resistive designs. The two-wire magnetic reluctor type generates a variable alternating current (ac) or sine-wave scope signal that’s greatly affected by the clearance between the crankshaft reluctor and the sensor body. Excessive clearance can cause a weak signal that the ECM can’t read. See Photo 2.
In addition to sensor clearance, crankshaft speed is also critical for creating a readable magnetic signal voltage. Since cranking speed is normally about 350 rpm, the magnetic reluctor signal must stay strong enough to produce a readable signal, otherwise, the engine won’t start. Magnetic reluctor sensor failures can be very temperature-sensitive as well and should be tested both at cold-soak and prolonged engine operating temperatures.
Hall-Effect sensors use three wires: a reference voltage wire, a ground wire and a signal wire that transmits a square-wave signal to the ECM (see Photo 3). Hall Effect CKP sensors generally use a crankshaft-mounted trigger wheel to generate a crankshaft position signal to the ECM. Before the ECM can read the digital square-wave signal, the signal must travel vertically from zero to full reference voltage.
The ECM might fail to read the Hall Effect sensor if the square-wave signal doesn’t reach zero volts or is distorted by rounded corners or electrical interference. The ECM might not also accurately read the signal if the Hall Effect sensor isn’t perfectly grounded. Since magneto-resistive CKP sensors are relatively uncommon, it’s most important to know that, unlike conventional Hall-Effect sensors, the square-wave signal “floats” above zero on a bias voltage when displayed on a lab scope.
WHEN THE CKP FAILS
Intermittent magnetic reluctive and Hall Effect sensor failures are often caused by broken wires, bad connectors or trigger-reluctor wheel location errors. In most cases, wiggle-testing the connector wiring can duplicate an intermittent failure. The sensor itself can fail due to changes in operating temperature, so the CKP signal should be monitored on a lab scope until well after the engine warms up.
In many cases, the engine will immediately re-start after a stalling event and will not set a DTC in the diagnostic memory. In other cases, a CKP failure will set a DTC in pending codes, but not in history codes, which is why it’s so important to access all code histories when diagnosing intermittent CKP failures.
MASS AIR FLOW DIAGNOSTICS
A mass airflow (MAF) sensor measures grams per second (gps) airflow into the engine, which is reported to the ECM by a frequency or voltage signal. A calibration error can occur if the MAF sensing wires are dirty, or if some type of debris is blocking or disturbing air flow around the sensor resistor (see Photo 4).
WHEN THE MAF FAILS
Not to mention a loose connector pin on a Nissan MAF sensor, I’ve had everything from carpet yarn to sunflower seed shells cause symptoms ranging from major fuel calculation errors to intermittent no-code engine stalls (see Photo 4).
Don’t forget that cracked or loose ducting between the MAF sensor and throttle plate is the most common cause of intermittent, no-code engine stalls (see Photo 1). This condition occurs most often in response to engine torque. Now you see it, now you don’t. It’s all part of the old shell game we call no-code diagnostics.
The Default Mode: When It Is What It Isn’t
1. When a critical sensor fails, an ECM might substitute a default value to keep the engine running. In this case, the value we might see on the scan tool is substitute data rather than real-time data.
2. Most of us have disconnected a mass airflow (MAF) sensor to diagnose a stalling condition. The reason the engine runs with a disconnected MAF is due to the ECM entering a default mode when it doesn’t sense a MAF return signal.
3. We can also see a default strategy occur when the camshaft position (CMP) sensor is disconnected on some engines.
4. When the ECM doesn’t see a CMP signal, it might follow a default strategy that uses a “roulette” method of locating TDC compression stroke.
5. If the ECM doesn’t see an increase in cranking speed indicating that the engine is firing on compression stroke, it changes the spark timing by 180º to correctly time the fuel injection and ignition. When the CKP/CMP signals are synchronized, the engine starts.
6. In many cases of open and shorted-to-ground circuits on late-model vehicles, the controlling module will disable the circuit or the sensor reference voltage. This strategy is common in exterior lighting and on three-wire wheel speed sensors.
7. The key to diagnosing a reference voltage issue is to locate and repair the circuit failure. To restore reference voltage to the repaired sensor circuit, simply erase the DTC.
8. This summer, I diagnosed an intermittent poor performance complaint on a 1995 OBD I Nissan pickup equipped with the ubiquitous 3.0L V6 engine. In the process, I discovered a rather unusual default operating strategy.
9. The Nissan’s optical distributor generates the CKP and CMP signals. With the distributor disconnected, the engine would continue to start and run in a “limp-in” mode.
10. It took me several weeks to really understand how the ECM was doing that, so, for now, just take my word for it …