Air Metering Systems


Flapper Door - This style of air metering uses a spring-loaded, hinged door to estimate the amount of air entering the engine.  When air flows through the meter body, it has to force the door open, and the amount of travel corresponds to the volume of air flowing towards the motor.

Once the system has determined the volume of the air entering the motor, it needs to solve for the mass flow, which can be done with two additional sensors: temperature and pressure.  Both of these sensors measure right next to the flapper door, as it is crucial that they match the properties of the air at the point where the flapper measured the volume flow.

This system works, but is probably the least effective of all of the ones in this article.  It is fairly restrictive due to the small cross-sectional area of the door, and it also has a limited measurement range, so modified vehicles can run into problems with it.


Karman Hz MAF - When air (or any fluid) flows through a pipe, it has a tendency to form vortices behind any uniform object it comes into contact with.  These vortices are like mini "whirlpools" of air, and the rate at which the form corresponds directly to the speed of the air (along with the pressure and temperature).

A Karman MAF has several pipes passing through it, and one of those pipes contains an ultrasonic sensor which measures the frequency of the ongoing vortices.  The other pipes are simply bypass pipes of known area, such that the ratio of the air volume going through the main section to the total air volume passing through the MAF is known.

The MAF also takes pressure and temperature readings, and sends all three (Hz, Temp, Press) to the ECU.  The ECU then does several calculations, and comes up with the following information:

-Velocity: The velocity can be calculated, using known Karman vortex formulae.

-Area: The total cross-sectional area of the MAF is known, as well as the ratio of unmetered area to metered area.  With these two facts in mind, the ECU can find the volume flow (speed times cross-section area equals volume flow).

-Temperature, Pressure: With the temperature and pressure inputs, the ECU can calculate the mass flow of the air, since it knows the volume already, using basic laws of physics (PV=nRT).

The Karman MAF is fairly accurate, but it has a few shortcomings.  First of all, because of the separate passages and screening, it can be a bit restrictive, although not nearly to the extent of a flapper-door MAF.  Second, there is a limit where the airflow through the MAF becomes turbulent, and at that point the ultrasonic sensor loses its capability to discern between Karman vortices and turbulence, and metering become erratic enough to make the vehicle almost undrivable.


Hotwire MAF - The principles behind a hotwire MAF are very simple, and also very consistent, which makes it one of the better ways to count air.  Basically, there is a single wire that passes through the middle of the MAF, which is nothing more than a pipe with an airfoil in the middle of it to support the wire.  That wire is heated to a certain temperature, which is verified by some thermistors (temperature sensitive resistors) attached to auxiliary wires.

An analog circuit measures how much current is required to sustain the desired temperature, and as the number of air molecules coming through the meter increases, so does heat transfer, and the necessary current to keep the wire warm increases.  This change is directly proportional to the number of molecules passing through the MAF, or the mass-flow of the intake charge.

Because of the way the meter works, there is no need for secondary pressure or temperature sensors, since pressure and temperature changes re compensated for by the primary measurement system.

Hotwire MAF's are very free-flowing, allowing more air with less restriction than anything besides speed density systems.  They automatically compensate for ANY and all changes to the vehicle and any changes to the air coming into the car.


Speed Density - Speed density, also known (incorrectly, perhaps) as "MAP" uses the state of the air entering the motor to figure out how much of it there is.

Using principles similar to those used in the Karman Hz MAF, if you know a bunch of information about the air going into the cylinders, you can figure out how much of it you have. You need the following information:

Pressure - Pressure, measured by the MAP (manifold absolute pressure) sensor, shows what the presure of the air is within the intake manifold, which is the closest you can come to representing the pressure in the cylinder without measuring it directly (which opens another can of worms, I won't go there). Pressure is one of the key components to knowing the air mass, and it's also a fairly good indicator of load. It is not, however, the only information you need.

Temperature - Temperature, like pressure, is measured in the intake manifold (or right before it). You need to know the temerature of air, as its density changes as the temp changes.

Volume - Volume measurement is one of the most difficult parts of a speed-density system. While the cylinders have a constant volume (and tthat is where we want to find the mass), the volume of the air entering the cylinders is not a constant. This phenomenon is described by the term "volumetic efficieny" (or "VE"). What happens is, as the piston speeds through its motion, the volume of the chamber may not necessarily be filled totally by fresh charge air. This can be because of exhaust gas staying in the chamber, the air losing density as it flows past the valves, or other things.

What you have to do, in order to run a speed density system, is you need to create a table of VE compared to engine load (MAP) and engine speed (RPM). Then, the ECU can look up the value for wherever it is at any given time.

Engine Speed - This is pretty simple. If the engine is moving faster, each volume (each cylinder) is filling more per time unit. That means more airflow (for a constant VE).

With this information, the ECU runs some calculations, and you get air mass.



Kyle Tarry 2004