So, you want to know about airflow meters?

(Or How I Learned to Stop Worrying and Love the MAF)

What is an airflow meter anyway?

The clue is in the name, airflow meters (AFMs) or mass airflow meters (MAFs) are devices which measure the volume of air being drawn into an engine and convert it to an electrical signal which is passed to the engine management unit (ECU). Additional components and calculations may then be able to deduce the actual mass rather than just volume. The ECU uses this to calculate the correct amount of fuel to inject to give the best proportion for combustion (stoichiometric ratio). Actually the ECU will take into account a lot of other factors when calculating the exact amount of fuel but they are not covered here.

What form of wizardry makes it work?

Well if it's a Bosch HFM5 then a lot of them either don't or won't work in the future hence the reason for this diatribe. The most common method of operation is to pass a known and highly accurate voltage across a thin wire or metal plate (sensor element) to heat it. The current flowing through the circuit will alter in direct relationship to the temerature of the sensor element, thus as more air passes over the element the more the cooling effect and the greater the change in current. Additional circuitry isolates the measuring elements to convert it to a signal more appropriate for connection to external equipment, this commonly takes the form of a voltage output rising and falling in time with airflow. In actual fact the sensor does not measure the entire volume of air passing through the AFM but rather just a sample which can then be calibrated against the bore of the airflow meter's pipe body (venturi). This also means that the sensor assembly (the expensive bit) can be used in different sized plastic venturis (the cheap bit) to give a product range suitable for different capacity engines for minimal production costs. All that is required is an output graph for each diameter of metering tube for the ECU to work to. Many AFMs also measure the temperature of the incoming air to give an even more accurate guide to the best fuel mix. ..and before you ask, no, the heating effect of the sensor on the incoming charge is negligable.

The unit installed in the affected GTVs uses something Bosch calls ‘Hybrid Thin Film Technology' for the main airflow measurement and a simple (and reliable) glass thermistor for the charge temperature reading. The output is a voltage in the range 0VDC rising to 5VDC with an increasing airflow.

What goes wrong then?

Two things; either degradation over a time period measured in months, years or picoseconds, or secondly, sudden complete failure. Gradual wear is typically indicated by a lack of midrange power in the 3000-4000 range (give or take)*. Total failure can happen at any time and is symptomised by the car starting but then refusing to rev over 1000rpm before dying. The engine warning lamp should also illuminate. Several folks have reported sudden failure when ‘giving it some' and indeed in my own case it seems to have been triggered after I floored it, missed a gear and bounced it off the rev limiter.

Many reasons have been put forward for failure with non-standard air filters being the most common excuse. Certainly an over-oiled aftermarket element could be a valid source of contamination as would leaks into the intake system allowing the ingress of dirt particles. However, many failures have been reported on standard, well maintained vehicles.

This technique for measuring airflow has been well used for over 20 years, virtually every injection car, van or truck (petrol or diesel) has one and in general they are as reliable as any other electronic component. It is also worth adding that during the course of researching this subject I found many articles about the nature of this particular unit especially amongst the Volkswagen/Audi TDi fraternity so it's not just an Alfa issue. It was in the discussion forums of these other owner groups that I was introduced to the possibility of an alternative solution.

*Flat spots in the power delivery can also be attributed to other factors which should be investigated. These factors may include cracked or split intake hoses, clogged air filters or fuel delivery problems. Also, the factory exhaust system and intake pipework do a fine job of strangling the car straight from the factory!

Ah ha..you wrote of an alternative?!?

On the aforementioned forums and websites came talk of something that worked, something that didn't keep going faulty and it wasn't a Bosch (wooohh) and they spoke of none other than the name of the device that was to come after and it was.. it was.. (yes).. it was Forty Two Pierburg. Happenstance and kismet found Pierburg marketing a sensor that had an all but identical output curve and an identical electrical connection plug as the VAG and Mercedes iffy (allegedly) Bosch HFM5s AND it's fitment and fixing dimensions allow it to be used in the Bosch venturi – very handy that as this must form an airtight seal. Lucky huh? Hmm. My own theory is that Herr Mercedes who was having terrible trouble with his Boschs, despite the antibiotics, leant over to Herr Pierburg and asked if he could help him out without having to modify either the looms or intake plumbing of his big expensive cars. Just a theory mind you, I have other theories but none of them are about airflow meters. The end result, however it came about, was a component that deserved some closer attention by people such as myself.

Now I did say that the output curve was almost identical. The only real variance is in the idle range and even then, in my opinion, is minimal and probably more than compensated for by the ECU which can dynamically control the throttle (fly-by-wire) to maintain a decent idle speed. One guy did design a compensation circuit to match the responses of the Bosch and Pierburg but he had built an extensive test rig to carry out his experiments on and very impressed I was by it too (see references at foot of page). But I haven't built his design so am unable to give any comparitive tests, perhaps a nice research grant might tempt me? All I can report is that since installing the Pierburg in August 2004 to the time of writing this (March 05) I have covered approximately 10K miles with no problems.

The only real issue as far as I'm concerned is the fact that the electrical plug connection is completely different for GTVs. There are several options here, you can either :

a) Source the appropriate plug and socket and make an adapter lead.

b) Cut the connector off the loom and either fit a new Pierburg socket or just solder the wires directly onto the new metering sensor.

c) Hacksaw the connector off your old unit and solder it onto the Pierburg so that you can still unplug it if necessary (like wot I gone and did).

d) Come up with your own combination of one of the above!

PDF for Bosch HFM5 mass airflow meter (77Kb)

PDF for Pierburg airflow sensor (59Kb)

Comparision of Bosch (LHS both images) & Pierburg (RHS both images)

Ok, my afm is faulty what next?

First unplug the connector from the side of the AFM. The plug can be a little tight so you may need to use a small flat screwdriver to lift the locking tab. This will put the car into what has been called ‘limp home mode'. Bit of a misnomer actually as the car will drive perfectly well without the AFM connected at the expense of a little power and extra fuel consumption until a replacement is organised.

Now go to your local authorised Alfa Romeo dealer and pay 170GBP (2004 prices) plus installation charges and have it repaired ‘correctly'. [/END]

…you still reading this?
You're not happy with option 1?
Dear oh dear. Perhaps a nice Honda or Volvo would suit you sir/madam/other?
No?
Dammit, I'm going to have to write detailed instructions for fitting the Pierburg aren't I ?
( yaaay )
Oh okay, but it'll cost you a pint of your finest ale barkeep….

***COMING SOON - Installation instructions***

Note:
Some owners have tried cleaning their airflow meters using solvents such as ethyl alcohol or carbon tetra chloride with varying levels of success. Upto you on this one, never tried it and not likely to unless I hear of more consistent positive results. There is no point trying this on a dead unit however as once it's gone, it's gone so get over it okay.

References

TDi Club article

Technical comparision of Pierburg and Bosch AFMs

GTV Forum