Sure, you may be familiar with some basic sensor operation. But as an automotive technician, you may not really understand all of the ins and outs of many common sensors. Do you really understand how they work, how the PCM views them and the possible scan tool and DVOM results when things go wrong? If not, stick around, because I plan to teach you.
Though I will be using Ford vehicles to reference, much of what is discussed here can be applied to other makes. Most other makes’ sensors operate with the same basic principles as what will be brought out here. However, all measurement values will be based on Ford, so be sure to always review different makes’ specifications before applying any of this information to the vehicle on which you are working.
The throttle position sensor (TPS) is a type of variable resistance sensor. Many other sensors such as the fuel level sensor and Intake Manifold Runner Control (IMRC) position sensor inside the IMRC unit function similarly.
In Figure 1, the PCM provides the sensor with a 5-volt supply. The PCM also provides the ground. The current and voltage flow between the ground side of the sensor and 5-volt side of the sensor remain unchanged regardless of the what is present on the signal wire.
The PID displayed on the scan tool reflects whatever value is present on the signal wire. The signal wire is connected to a sweeper inside the sensor. Wherever the sweeper is moved to along the sensor’s internal resistor, the signal’s value will change accordingly. To illustrate, look at Figure 1. At the VREF side of the sensor, we have 5 volts. At the ground side, we have zero volts. All across that sensor, the voltage is being reduced from five down to zero as the voltage travels from one end to the other. All we are doing is tapping into that sensor’s resistor and making voltage measurements at different points across that resistor. If the sweeper were moved to the ground side of that sensor (the red sweeper), what do you think the voltage should read on the scanner’s PID? It should read close to zero. If we moved the sweeper over to the 5-volt end (green sweeper), we should read close to 5 volts. The middle of the resistor (black sweeper), should likewise read close to half of the reference voltage.
OK, so now that you see that, let’s see what happens if things go wrong.
In Figure 2,we have that same TPS, only with a broken VREF wire. What do you think the PID on the scanner will read? The PID will read whatever voltage is present on the signal wire. With no voltage getting into the sensor, there is obviously none that can get to the signal wire.
Therefore, the PID will read zero volts. This also would be the same result if the signal wire was broken.
What would the PID look like here (Figure 3)? The ground wire is broken, so the current can’t get to ground. It is however, going to the signal wire just fine. For the most part, the PID will be stuck at 5 volts. If you were to open the throttle plate, you might find a minor fluctuation in the PID value. After all, the sweeper is still intact and there is some variable voltage drop happening across half of the TPS. However, for the most part, it will be stuck near 5 volts.
The engine coolant temperature (ECT) sensor and intake air temperature sensors are both thermistors. A thermistor is a type of variable resistance sensor. As the sensor is exposed to heat, its internal resistance lowers. As the heat exposure rises, the resistance lowers even more. In Figure 4,the PID circuitry is monitoring the voltage drop occurring across the 249k ohm internal resistor. As the ECT sensor heats up, its internal resistance lowers. The lowering of the ECT’s resistance provides a greater alternate path for the reference voltage to get to ground. Since more of the reference voltage is then traveling through the sensor and not the internal 249k resistor, the voltage drop across the internal resistor is lessened. As a result, the PID circuitry displays this value as a rise in engine temperature.
OK, so let’s break one of the wires going to the ECT; What do you think will happen? With an open sensor circuit, the reference voltage has no other path than through the 249k resistor (see Figure 5). The PID will display a temperature value that the PCM’s software sets as its lowest possible value, which is typically -40°. If you were to view the ECT voltage PID, you will see a value very close to the 5-volt reference. It may not display a perfect 5 volts due to the drop that occurs across the 20k internal resistor. Usually, the voltage shown in this case is around 4.5 volts.
So now we have the opposite problem above as we did before (Figure 6). This time the sensor signal wire has been shorted to the sensor ground lead. In this case, there will be no voltage drop occurring across the 249k internal resistor because all of the reference voltage will travel through the 20k ohm resistor and bypass the 249k resistor straight to ground. The PID will display the highest allowed value by its software. This value is usually somewhere around 260°. The ECT voltage PID will display zero volts, or a value very close to it.
In this scenario, the value of having that 20k resistor in place becomes quite clear. With a shorted sensor circuit, that resistor becomes the only thing protecting the PCM’s 5-volt reference circuitry from literally burning out like a very expensive fuse. Good thing it’s there.
The Differential Pressure Feedback Exhaust (DPFE) sensor is a capacitive type differential sensor. It is monitoring the pressure difference between the two different sides of a restrictor orifice located inside the exhaust gas recirculation (EGR) supply tube. The PCM uses this sensor to monitor the amount of EGR flow.
The sensor uses three wires: one 5-volt reference, one ground wire, and one signal wire that will also have 5 measurable volts found on it if the sensor is disconnected. The voltage found on the signal wire comes from the 5-volt reference through a 4.7K ohm resistor that links the signal lead to the reference voltage lead inside the PCM. The DPFE works by creating a varying voltage drop on the signal wire (see Figure 7). At rest, the DPFE pulls the signal lead to around 1 volt for plastic-bodied DPFE sensors and as low as 0.5 volts on the older metal-bodied DPFE sensors. As the EGR flows, and the pressure differential becomes greater across the restrictor orifice inside the EGR supply tube, the DPFE allows the voltage on the signal wire to raise. This signal is linear. Being so, if you were to manually apply a vacuum to the upstream hose of the DPFE, you would find that the voltage on the signal wire will change evenly per equal amounts of pressure change.
Being that the PCM reads this sensor in a voltage drop fashion, an open circuit in any of the three wires or internally to the DPFE will result in 5 volts being displayed on the scan tool as well as found at the sensor with a DVOM.
Crank Sensor and Other Speed Sensors
Crank sensors, cam sensors, wheel speed sensors and other “speed” sensors work on similar principles. Crank position sensors and cam position sensors are really just speed sensors that are monitoring a reluctor that contains missing teeth.
In the case of a VRT-style sensor, inside the sensor is a wire wound coil surrounding a permanent magnet. As the teeth of the reluctor wheel pass close to the sensor’s tip, the permanent magnet’s field is distorted in the rhythm with the teeth, see Figure 8. That magnetic distortion induces an AC signal into the coil windings and is feedback to the module. This sensor is a “signal generator” and produces voltage all on its own. This signal is very sensitive to outside interference. The wires to the sensors are usually shielded and/or twisted around each other to resist radio frequency interference being induced into the wiring.
In the case of a Hall effect-style sensor, the sensor is simply turning “on” and “off” a 5-volt reference signal that the module is sending to it, much like a switch. On a scope, this signal will appear as a square wave.
In the case of a wheel speed sensor, each tooth is evenly spaced 360° around the reluctor. The signal produced at any one point in the rotation is identical at any other point in the rotation. The ABS module is simply looking at the frequency of the signal. The faster the signal cycles, the faster the wheels are turning. This also applies to vehicle speed sensors and turbine speed sensors.
In the case of a crank position sensor, one tooth is missing. The PCM is looking at the frequency of the signal to determine how fast the engine is turning. The time frame between the humps aids in misfire detection. Also, the PCM is looking for the missing tooth. The missing tooth causes a missing “hump” in the sine wave. That tells the PCM when #1 cylinder is at TDC. It does not, however, tell the PCM if it is on TDC of exhaust stroke or intake stroke. The number of humps after that missing tooth tells the PCM the exact location of crankshaft in its rotation.
In the case of a cam position sensor, the majority of the reluctor is void of teeth. There will only be one tooth. Every time that one tooth comes around, one sine wave signal is produced. When coupled with the missing tooth signal from the crank sensor, the PCM “knows” that cylinder #1 is at TDC compression stroke.
If either wire were to break or short to ground to any of these speed sensors, the sensor would be completely inactive. If the harness becomes unshielded, then interference from such things as secondary ignition wires can induce false “humps” into the signal.
The knock sensor is a tuned accelerometer (see Figure 9). It converts engine vibrations into an AC signal. Inside the sensor is a crystal. When the sensor is vibrated at just the right frequency, an AC signal is generated out of it. This type of sensor is a signal generator. It literally produces a voltage signal, rather than just manipulating one that is sent to it.
Doesn’t the engine make a lot of normal vibrations already? How does it know the difference between normal vibrations and those from “spark knock”? It is the balance between the sensor’s tuning (by construction) and filtering software in the PCM that determines the difference between a “YES” or “NO” on the KS PID. The PCM uses this input to retard the ignition timing if excessive spark knock is detected. If either wire is broken or shorted to ground, the sensor will still generate a signal, but the PCM’s PID will not be able to detect it.
O2 Sensor and Catalyst Monitor
Currently, the most popular type of oxygen sensor in use is the zirconia oxygen sensor(see Figure 10). It is a signal generator that is fully functioning after 600° F. The zirconia oxygen sensor is a type of chemical battery. It contains two dissimilar materials that when heated and exposed to correct elements, produce a voltage. Those dissimilar materials are zirconia and platinum. The zirconia is formed into a hollow bullet-shaped porcelain cone located inside of a protective cage. The platinum is “painted” onto the zirconia’s surface.
The “elements” needed to generate the electrical energy are oxygen from the ambient air and ionic gases in the exhaust stream, such as hydrogen and carbon monoxide. The O2 molecules on the ambient side of the sensor are “wanting” to give up their electrons to the ionic gases in the exhaust stream. There is some controversy over exactly which ionic gas it is in the exhaust stream that is attracting the O2’s electrons. High levels of oxygen in the exhaust block the ionic gas’ ability to attract the O2’s electrons by oxidizing those gases. When the oxygen content in the exhaust is low, the ionic gases are free to attract electrons from the O2 gases outside.
Simply put, the oxygen sensor is not sensing oxygen in the exhaust. But rather, it is sensing an oxygen differential between what is inside the pipe as to what is outside the pipe. It generates voltage when the O2 is now in the exhaust.
The internal heater is simply used to get the sensor’s tip up to operating temperature more quickly. It is powered by 12 volts. The ground side of the heater circuit is monitored by the PCM to determine the health of the O2 heater.
Looking at the O2 diagram (Figure 11), you can see that the sensing circuitry has a ground supplied by the PCM. If this ground wire, or the positive feed directly to the input PID circuitry, were to break, the PID would read as though the sensor was inactive. In reality, the sensor would still be working, but the PCM would not “see” it. So anytime a scan tool shows an O2 or cat monitor as completely inactive, a voltage measurement should be made again directly at the wires coming out of the sensor to compare.
What if the 12-volt heater circuit shorted to the O2 signal wire? Will the PCM display 12 volts? No, at least not on a Ford PCM. Since the normal maximum output of any automotive zirconium O2 sensor is about 1.0-1.2 volts, the PCM software isn’t programmed to recognize O2 voltages over 1.67. So, if the O2 heater power feed were to short over to the O2 signal lead, the PCM PID will display a constant 1.67 volts. However, you would measure battery/charging system voltage on that lead if you measured it with a DVOM directly.
What if the 12-volt feed or the ground feed leading to the O2 heater PID should fail open or shorted to ground? Would the O2 sensor still operate? Yes, it would. It would just take longer for the sensor to “wake up” since it would not have the benefit of the internal heater to warm up the sensor as rapidly.
Analog Mass Air Flow Sensor
The analog mass air flow (MAF) sensor is a little module all to itself. It can also be called a “signal generator.” Figure 12is an illustration of a MAF sensor without a built-in Intake Air Temperature (IAT) sensor.
The hot wire MAF uses two wires exposed to the intake air inside the sensor. One is a reference wire and the other is a hot wire. There is a Wheatstone Bridge that connects the reference wire and the hot wire. The reference wire is used to determine ambient air temperature as a reference point for the hot wire. The Wheatstone Bridge increases or decreases amperage, in the range of 500 ma to 1,200 ma, to maintain the sensing wire’s temperature 100° C above that of the ambient reference wire. The amount of amperage needed to maintain the sensing wire at that temperature determines the output voltage of the MAF to the PCM. The air mass passing the hot wire cools it. As the wire cools, its resistance climbs, requiring more amperage to maintain the wire’s temperature, and vice versa. The more amperage needed, the greater the voltage signal sent to the PCM.
If you’ll notice in my drawing of the approximate output voltage versus the approximate air flow volume, the output is not linear. Most of the MAF’s resolution is concentrated in the first 40% of the total output. That is because most of its usage is spent in that region and must be the most accurate in that area for emissions, driveability and fuel economy reasons.
The PCM infers BARO at large throttle angles, 75% to 100%, from the MAF sensor for most of Ford’s applications. The MAF is the sensor that “wears the pants” in the sensor family. The MAF’s reading determines the load percentage, which determines the base injector pulse width and ignition timing across each rpm.
This sensor’s most common enemy is dirt. Any dirt particles, especially those soaked in oil, are damaging to this sensor’s accuracy. The dirt builds up on the wire and insulates it at moments of high CFM. Inversely, dirt increases heat transfer properties by increased overall mass at moments of low CFM. So, with a contaminated MAF sensor, you can have it over-report air flow at idle. At the same time, it can under-report air flow at large throttle angles.
If the signal wire or ground wire between the PCM and the MAF were to break, the MAF would still function. However, the PCM’s PID would show it as inactive. If the 12-volt power feed to the MAF or the ground lead from it were to break, the MAF would truly become inactive.
The first victim… err… I mean, uhhh… “patient” is a 2002 Explorer with a 4.0L SOHC engine and automatic transmission. This vehicle was towed in for a “cranks, but no-start” condition. A quick parking lot check found ignition fire and enough fuel pressure present on the rail to provide a healthy spit when the Schrader was depressed. The odor of raw fuel at the tail pipe suggested that the cylinders were flooded out. Pressing the gas pedal to the floor with a little extended cranking time brought the engine to life slowly, one cylinder at a time. It’s a good thing too, because I don’t like to push anything I don’t absolutely have to. Once it started, it actually ran surprising well, with only the heavy exhaust odor of a rich running engine. The check engine light immediately illuminated on the drive into the bay.
When the scanner was connected, a very long string of fault codes came up. See Figure 13. A list this long can be startling to see. What can set this many codes at once? Wiring? PCM? Well, yes, could be. But it’s important to look at what all of the “failing” sensors have in common versus the ones that worked fine. If you’ll notice, none of the “signal generators” set codes (such as MAF and O2s). Looking at the data for some of the effected sensors (in Figure 14), what do we see?
We see a lot of sensors stuck at or around reference voltage. We could have a failing PCM, but more than likely the 5-volt reference is being shorted to ground. But where? It could be in any length of 5-volt reference wire, or in any sensor that operates off of 5 volts.
The fastest and easiest solution to diagnosing this problem was simply to start disconnecting sensors one at a time and seeing if the other sensors came alive. I started with the easiest to access sensor of the bunch, the DPFE. The DPFE and the Manual Lever Position (MLP) sensor both have a track record for shorting to ground. Upon disconnecting the DPFE… viola! (Figure 15).
The other sensors started displaying believable values. The DPFE, even though it was the tiny little plastic tube-mounted version, and not the old metal one that was traditionally known for this problem, had shorted its 5-volt reference to ground. The fix of course was a new DPFE and an oil change due to the earlier engine flooding.
The next vehicle is a 2003 Taurus DOHC 3.0L with automatic transmission. The problem is a steady misfire on the #6 cylinder, and setting a code P0306. Is it lack of fuel to that cylinder or lack of ignition fire maybe? You techs with a gas analyzer can use the HC and CO readings to narrow down the search quickly. Is there anything we can look at on the scanner to help narrow it down before you can even warm up the gas analyzer? How about O2 sensor readings and fuel trim?
This first capture (Figure 16), of course, is normal O2 operation and trim for bank 2.
An ignition misfire was induced into this second capture(Figure 17) about halfway across. I unplugged the secondary ignition wire to the #6 cylinder. During an ignition misfire, unburned oxygen is “pumped” into the exhaust system and fuel is still being introduced as well (depending on type, duration of misfire and PCM strats). Since there is still fuel being added to that cylinder, then that cylinder is still contributing some of the ionic gases to the exhaust that we discussed earlier. Those ionic gases are still being detected by the O2 sensor, so the PCM only had to add a small amount of extra fuel to the remaining cylinders to get the O2 sensor back to cycling again. Notice the trim percentage to the right of the screen is on a 15% range. The trim percentage here is bouncing between a positive 10 to 13%. Notice also the O2 sensor’s new “switching” point is lower than normal. It is important to note that if an ignition coil primary circuit were to fail, the PCM will shut off the injector to that cylinder to protect the cats. In that case, the O2s and trims will appear as the ones in the next capture (Figure 18).
In Figure 18, I unplugged the #6 injector about halfway across the screen. The fuel trim immediately shot to double that of what the ignition misfire did. It is important to notice the new trim percentage scale to the right. This trim went to a positive 26% and completely left viewable range to the top of the screen when it was set at the 15% scale used for the ignition misfire. The total lack of fuel to that cylinder left an ionic gas deficiency in the exhaust. Therefore, the O2 sensor triggered the PCM to add an even higher amount of fuel to the remaining cylinders in order to get the O2 sensor active again. Notice also the higher O2 “switch point.”
When using the O2s to aid in misfire diagnosis just remember this: A slightly elevated fuel trim with a low O2 switch point suggests ignition misfire, and a highly elevated fuel trim with a high O2 switch point suggests injector misfire.
Also comparing your cat monitor results to the upstream O2s can clue you in. An ignition misfire is often accompanied with a very high cat monitor voltage, and an injector misfire is often accompanied by a very low cat monitor voltage.