Mass Air Flow (MAF) Sensor (G70) – Introduction
The air mass sensor, more correctly known as the mass air flow sensor or MAF. It is located in the airbox housing and sits between the air filter and the compressor side of the turbocharger as shown below. It's sole function in life is to measure the amount of air (by mass) entering the engine at any point in time. The MAF device used on the S2 and RS2 is similar to those used on the Bosch LH and Motronic fuel injection system found on many other cars then and today.
The function of the MAF is to provide a voltage signal into the ECU which determines how much air MASS (or weight if you like) is entering the engine. It does this, by passing an electrical current through a thin platinum wire to heat it 100 degrees celsius above the temperature of the air passing over it. This current needs to be larger with increasing airflow in order to overcome the cooling effect of the passing air and maintain a constant temperature on the platinum wire.
SideNote - On account of the heat involved, this type of device is called a 'Hotwire' MAF. I believe this gives the 'H' notation used in the Bosch LH systems where it was first employed in the late 1980s by Audi and other German manufacturers. It was subsequently used on Bosch Motronic systems, and the concept is still used today in a wide variety of petrol and diesel engined cars. It is an accurate, cost-effective device that can be used by the ECU to both optimise performance and emissions.
The varying control current flowing thru the hotwire is passed thru a calibrated reference resistor network in order to generate a varying voltage at the output pin on the MAF. This voltage signal is therefore a direct measure of the VOLUME of air flowing thru the MAF, but we need to know the air MASS. In order to do this, the density of the air must be known. This is obtained by a thermistor device (which changes resistance with temperature) that forms part of the resistor network such that the MAF output voltage is augmented by temperature effects.
A given volume of cold air is more dense then the same volume of air at high temperature as long as the pressure is maintained constant. Its also useful to remember that a constant volume of air at high altitudes (low pressure) is less dense than the same volume at sea level. These two statements are linked together by a series of important physical equations known as the ideal gas laws. This wiki tells you more on the science fundamentals. Understanding that stuff will help you understand why turbochargers and intercoolers make engines more powerful. Of course the ECU has its own means of measuring ambient air pressure and ambient air temperature by means of F96 and G42 respectively. It uses these values to make vital adjustments to ignition, fuel and boost calculations in real-time for the prevailing conditions.
More MAF Information
A honeycomb type screen is placed in at both ends of the MAF housing so as to protect the hotwire filament from erroneous readings caused by air turbulence. This is a small restriction to airflow thru the MAF - you can reduce the pressure drop across it by removing these screens. It is worth noting that the RS2 MAF, whilst electrically identical to the S2 MAF, has much less restrictive baffles fitted.
Contamination of the hotwire will quickly cause inaccurate readings that can result in loss of performance or foul running. Given the nature of the positive crankcase ventilation (PCV) system on the S2/RS2 which recirculates oil vapour and blow-by products thru to the intake system there is potential for this to collect on the hotwire... Despite the design by Audi to ensure this is minimised by the fact that these vapours are introduced somewhat 'south' of the MAF. To ensure it is kept as clean as possible, the hotwire is heated to a temperature of 1000 degrees C for a one second period each time the engine is shutdown. This should burn off any potential contaminants.
This is all great with the exception that the burnoff sequence actually erodes the platinum hotwire, reducing its accuracy and lifetime. This burn-off signal is triggered by the ECU after the engine is turned off. It is therefore recommended that you wait at least twenty seconds after shutdown before attempting to remove the connector plug from the ECU. Details on how to test the burn-off signal are down here !
Should the MAF fail entirely (or the electrical connector be disconnected), then the engine will run in limp home mode such that when the idle throttle switch (F60) is closed the ECU uses a static ignition map that assumes a certain amount of air flowing at idle. At part load, when F60 is open, the ignition map is fixed for 20 degrees BTDC and the mixture is leaned off to get you home.
technical information on the Bosch hotwire MAF, there is useful FAQ at this
location provided by JDS
A common practice in DIY cleaning of the MAF filament across VW, Audi and other similarly afflicted owners is to submerge the hotwire filament in a tub containing a clear liquid chemical known as iso-propyl-alcohol (often shortened to IPA). It can be obtained at your local pharmacy or from some electronics / components suppliers. Some people claim wonderful results with this cleaning as it allegedly dissolves any remnants of the invisible grime which the burn-off sequence has failed to remove. Others claim no improvements whatsoever. I believe that the benefits observed from the MAF Bath are as much to do with the fact that the MAF connector and wiring plug generally gets a good clean along the way.
If you are having some running troubles, it is worth trying this as it costs almost nothing, but the test sequences described below are rather more objective.
IMPORTANT – Be sure to NOT submerge the part of the MAF with the 6-pin electrical connector in the cleaning liquid… Just the filament !
MAF Part Numbers
information for the MAF on the S2, and other vehicles, is shown below. The
Porsche 968 application for the RS2 MAF is proof (it it be needed) of the
fastidious component choices by the Engineers from
MAF Pinout & Circuit Diagram
The table below provides the electrical pinout of the MAF device, along with a description of each pin and where it connects to the ECU on the 3B and ABY engines. Note this information is also applicable to the RR and AAN/ADU engines respectively. Strangely, Audi used the same pin numbers on both of the ECU used for these engines.
The diagram below depicts the electrical schematic of the MAF and how it connects to the ECU. The notation of points A, B, X and Y on the diagram below align with the photographs shown further below and the information on how to quickly test the MAF quickly and easily.
Testing the MAF – Part 1, External Checks
Using a digital multimeter (DMM) the MAF and its wiring can be quickly checked in four steps after removing the connector at the MAF. Pin #1 is the RIGHT-most pin when viewing the MAF from the top and the LEFT-most pin when viewing the harness connector face on.
3. With the MAF connector removed, and the DMM still set for low resistance values, verify that 3 ohms is measured between pins #1 (one) and #3 (three). This is a measure of the internal reference resistor, Ra, shown in the above schematic.
4. Connecting the harness plug back onto the MAF, again teasing back the insulation hood enough to get the DMM probes onto the pins entering the connector, measure DC VOLTAGE across pins #1 (one) and #3 (three).
NOTE – When refitting the MAF connector for the final time, it is important to check the cleanliness of the connector pins and that the insulation hood is firmly in place to ensure that moisture and dirt do not have an opportunity to get to the electrical connections. Such defects can give rise to erroneous MAF readings and some running issues.
If I ever track down a voltage to mass-air-flow graph for the S2 MAF I shall post it here. An internet search might find you a similar device from a Porsche to depict the shape of the curve. The maximum output voltage is ~5V of course... The ECU is aware of the calibration between this voltage value in terms of 'how much air' that corresponds to at any given time - it uses that for fuel, boost and timing calculations.
Testing the MAF – Part 2, The Burn-Off Signal
If the burn-off signal is not visible within the MAF, then test for the trigger signal from the ECU at pin #4 (four) on the MAF connector. If the ECU is providing the burn-off signal and the MAF isn't obeying this command, the MAF should be replaced as it is not a serviceable component.
Testing the MAF – Part 3, Internal Checks – Careful Now
Further tests to the MAF can be performed if there are potential problems identified by the Part 1 and Part 2 tests.
If you have got this far, the MAF is already detached from the airbox and it is now a simple matter of removing it entirely and placing it on the bench for some careful tests with that multimeter. The image above shows the internals of the MAF, viewed from the airbox side of the housing. I have highlighted the hotwire filament in this image for clarity. In order to get this far of course, one has to remove the retaining ring and metal honeycomb mesh at the airbox end of the MAF housing.
In order to align with the schematic shown above, I have added the labels A and B for the hotwire filament and labels X and Y for the thermistor.
With great care, using the DMM set to low resistance values, it is possible to check the following –
… Resistance across the hotwire filament (Rf) should be approx 1 ohm (pins A and B)
… Resistance across the thermistor (Rt) should be approx 370 ohm (pins X and Y) at 20 celsius
… Pins A and X should be a short circuit (0 ohm)
… Resistance between MAF connector Pin #3 and pin B should also be 0 ohm
… Resistance between MAF connector Pin #1 and Pin Y should be Rf + Rt + 3 ohms
Please note that these resistance measurements should be considered as a reasonable guideline for what to expect when testing the MAF – they are not absolute and the value of Rt shall alter noticeably with temperature. This thermistor is a PTC device and will yield lower resistance values at lower temperatures. The DMM I have used is not the most accurate thing in the world.
As a final behavioural check on the MAF, with it connected to the wiring loom, any screens refitted and the ignition on - you can use one of those hair dryers that has the ability to blow at multiple speeds with a couple of heat settings. You should see lowest MAF output voltage with the slowest/hottest setting on the hair dryer and the highest voltage output with the fastest/coldest setting. Use some common sense here so as not to blast hot air directly onto the filament.
Testing the MAF – Part 4, More Tests & Calibration Information
to the MAF, to verify its calibration, require an accurate flowbench and
industrial test & measurement equipment against a known fresh and healthy
reference MAF. One specialist company in the
Bosch are notorious for being protective about component information. As such, it is impossible to find a graph of output voltage against input air mass for the S2 / RS2 MAF. However, I am in the process of working out a way to reverse engineer this information with the help of an entirely different MAF (that does have published data) and the assistance of the flowbench at JDS. Watch this space for an update on that.
It may not be deadly accurate, but by the wonders of applied mathematics, that work shall provide a means to estimate engine power by simply measuring the output voltage of the MAF. If I find a field on the diagnostic channels within the ECU that directly reports MAF voltage, then it will be possible to log this information with VAG-COM. No promises here – it is work in progress shall we say.
Last Updated 23rd November 2007