Tuning Wiki

For a practical workshop checklist, see common MAF sensor faults.

Where this sits

The meter types, the makers and the Bosch curves are in the airflow section, starting at Airflow metering. This page keeps the voltages, frequencies and fault codes used to test the common families.

Why the control unit measures the air

The control unit needs the air being drawn in so it can set injection quantity, ignition timing and, on some cars, the shift point. The meter sits in the intake stream between the air cleaner and the throttle, where it sees the air the engine is actually taking. A speed-density system has no air-mass meter. It estimates the flow from throttle position, manifold pressure, intake-air temperature and engine speed.

A blocked measuring passage often lets the engine start and then run badly or stall, and it may never set a fault code. Look through the passage before condemning the electronics.

Testing a modified intake

Meter readings depend on the sensor and its installation. Compare the output with the correct vehicle-specific reference and retain the normal airbox and ducting during the initial test. Bosch notes that removing the unit from its airbox can change airflow and the signal; air leaks, contamination and wiring can also cause misleading results. Bosch cautions that impact or solvent cleaning can damage the sensor. Bosch: air-mass sensor diagnosis.

Where an aftermarket ECU uses a MAF input, confirm its transfer calibration and whether the signal is voltage or frequency. A sensor or housing change is not automatically covered by the previous calibration. Monitor the sensor signal and calculated airflow over the operating range; an output at its specified upper limit may no longer distinguish further airflow increases. Validate using the sensor and ECU documentation, fuel-pressure and mixture measurements rather than assuming a universal voltage limit. Haltech: MAF sensor calibration.

Hot wire

A hot-wire meter has a thermistor that reads the incoming air, a platinum wire, and a circuit that holds the wire a fixed temperature above the thermistor. More air cools the wire faster, the circuit passes more current to restore the temperature, and that current is turned into a voltage, often called VG, in proportion to the current. A separate intake-air temperature sensor may live in the same housing, with its own ground, distinct from the meter ground. The constructions, including Denso, are on hot-wire mass airflow sensors.

Two descriptions of the temperature offset are in use, and they are not the same number. On one family the platinum wire is held 212 °F (100 °C) above the incoming air, and a hot-film foil is held 167 °F above the surrounding air. On another description the heated wire is held near 200 °C and the current that holds it there is the flow signal. Cooling also depends on the temperature, density and humidity of the air, which is why the current follows air mass rather than air volume. A further current, larger than the measuring current, can be passed through the wire after shutdown to burn off deposits and limit oxidation.

Bosch hot-wire meters, used with LH-Jetronic from the late 1970s and on 1985–89 GM 5.0 and 5.7 litre tuned-port engines, produce an analogue voltage from 0 to 5 V. Idle is usually 0.4 to 0.8 V and wide-open throttle is 4.5 to 5.0 V. After shutdown the wire is heated to 1000 °C for one second. The control unit does that through a relay. If that burn-off never happens, deposits sit on the wire and the meter reads slowly or low. On those GM installations, code 36 is the failed burn-off on the 5.0 and 5.7. Power arrives through two relays, one for running supply and one for the burn-off. An orange-and-black wire is the 12 V feed and a black wire is ground. The output, a brown-and-white wire back-probed at terminal B6 of the PCM, is checked around 2.5 V and must rise as the throttle opens with the engine running. A reading that will not move is a failed meter, a failed relay, or a wiring fault, in that order of checks.

A hot-wire example at idle is about 0.5 V. A failed example of the same style sat at 2.0 V at idle because the wire circuit had gone high-resistance. That pair of numbers is an illustration of one sensor, not a limit for every car. The resistance check uses the diagram and the resistance for the car in front of you. With the engine idling, confirm the feeds and the earths, then the rest of the wiring. For the waveform, use a breakout lead. Piercing the insulation leaves a hole that becomes the next fault. Snap the throttle open and watch the trace move.

Hot film, as a frequency

AC Rochester hot-film meters, introduced in 1984 on the Buick turbo 3.8 litre V6 and also used on the Chevrolet 2.8 and the GM 3.0 and 3.8 litre V6, produce a square wave rather than a rising voltage. Frequency runs from about 32 Hz at idle to about 150 Hz at wide-open throttle. In 1990 most GM engines went back to speed-density. The Buick 3.3 and 3.8 moved to a Hitachi meter instead. The film technologies are on hot-film mass airflow sensors.

On a GM Delco meter the idle output, read as a voltage on the signal terminal, should sit steadily near 2.5 V. A light tap that makes the voltage jump, or that makes the engine misfire for a moment, is a failed meter. The same test after heating the meter with a hair dryer catches a heat-sensitive element. Older AC Delco units, such as a 2.8 litre V6, read about 32 Hz at idle and about 75 Hz at 3500 rpm. Later Hitachi units, such as a 3800 V6, read about 2.9 kHz at idle and about 5.0 kHz at 3500 rpm. A sudden frequency change when the meter is tapped is a failed meter. A scan tool may show the same signal as grams per second: about 4 to 8 g/s at idle and 100 to 240 g/s at wide-open throttle.

Fault codes that point at these frequency meters, on GM multiport engines, are code 33 for a frequency that is too high and code 34 for a frequency that is too low. Ford codes are 26 for a reading out of range, 56 for an output that is too high, 66 for an output that is too low, and 76 for no change during a throttle snap. A code is not guaranteed. Low compression, low vacuum, low fuel pressure, leaking or dirty injectors, an ignition misfire and a blocked catalyst all produce the same driving symptoms: hard starting, stalling after start, hesitation under load, surge, a rough idle, and a mixture that is far too rich or far too lean. A sudden throttle movement can also make the engine hiccup when the meter is slow.

The trace on a scope is a square wave whose frequency rises smoothly as speed and load rise. Steps, dropouts or heavy noise are a failed meter. Injector pulse width is a second view of the same fault. If the meter is producing no signal, injector time at idle or during cranking stretches to about four times the normal width and the engine can be too rich to start. Under a light cruise the same measurement is about 2.5 to 2.8 ms, richer as the throttle opens, leaner on a light cruise, and cut during deceleration.

A micromechanical film with a fixed analogue curve and return-flow detection is the Bosch HFM 5. A frequency output that falls as flow rises, used on the 1.6 HDi and TDCi, is the Pierburg frequency meter. Those two must not be diagnosed with the GM frequency numbers on this page.

Vane air-flow meter

A vane meter measures volume, which the control unit uses for basic injection time and basic ignition advance. Intake air pushes a measuring plate against a return spring. Plate angle follows the volume. A compensation plate in a damping chamber is fixed to the measuring plate and stops the plate fluttering. The shaft turns a wiper on a potentiometer, and the voltage at the VS terminal follows the plate. A resistor in parallel with the main potentiometer track keeps a VS signal available if the main track goes open. A fuel-pump switch inside the meter closes once air is flowing, and holds the pump on after the engine has started. An intake-air temperature sensor is in the housing. An idle-mixture screw is factory set and covered by a tamper plug. There is no published procedure for resetting that screw once the plug has been disturbed. The mechanism is on vane airflow meters.

The older vane meters run from battery voltage, and the signal voltage rises as the plate opens. Later meters run from a regulated reference. Some rise from 0 V toward 5 V or toward about 9 V as the plate opens. Some fall from 5 V toward 0 V. The direction has to be measured on the car: with the duct off, ignition on and the engine stopped, only one wire changes as the plate is pushed by hand, and the closed-plate and open-plate voltages identify which kind of meter it is. Worn tracks cut a groove in the resistive element. The voltage then drops out at particular plate angles, which feels like a flat spot, a stutter, a loss of power, or a failed emissions test.

Replacing a vane meter with a hot-wire or hot-film meter means teaching the control unit a new relationship between voltage and flow. That job is covered under AFM to MAF conversion.

Karman vortex

A Karman meter also reports volume. Air passing a vortex generator sheds a street of vortices, the same pattern as the wake of a boat. The rate of shedding follows the air speed, and therefore the engine load. The vortices reach a metal-foil mirror through a pressure port. The mirror oscillates, and a photo-coupler, an LED and a phototransistor, sees that motion as an interruption of the light. The phototransistor grounds and releases a 5 V KS signal, so the control unit receives a 5 V square wave whose frequency rises with flow. The frequency is high enough that a frequency-capable meter or an oscilloscope is what shows it. A DC voltmeter averages the wave and hides the flow. Mitsubishi’s ultrasonic version of the same street is on Karman vortex airflow meters.