Galaxie500XL

iTrader: (1)

Well..new problem. The IROC is having trouble starting cold. Playing with it this morning, it’s setting a Code 34....MAF low.

They do this if the reported airflow is below a certain threshold. A few things can cause this:

MAF relay

MAF burn off relay

Wiring issue to MAF

MAF dirty (usually because the burn off relay isn’t keeping the heated wire clean)

Bad MAF

The last one is bad news. Bosch made the OEM MAF, and they’ve been discontinued for many years. Many of the ‘remanufactured” MAF’s are poorly redone, and pretty much anything out there no longer uses the heated wire setup to operate, which can cause all sorts of interesting problems on TPI systems.

The OEM MAF works by heating a wire placed in the airflow of the MAF. Since ambient air temperature is known by the temperature sensor in the plenum, the amount of current required to maintain the heated wire’s temperature is used to infer airflow through the MAF. More current required=more airflow.

I’ll start out by getting some MAF cleaner, and a new MAF relay and burn off relay. Those are the simple and cheap things to start with.

Who makes the best MAF these days? I see Delphi now has AF10320 available...but from what I understand, it’s not a heated wire MAF like the original Bosch MAF.

IROCThe5.7L

I believe they also don't make these anymore. But I've had a Microtech MAF in my car for over 15 years. No heated wire setup. And it's been fine. I wouldn't let that bother you.

Galaxie500XL

iTrader: (1)

That’s very helpful. Thanks!

Kincer

I replaced mine with a Delphi unit, solved my problems, no issues. I've heard you can send the Bosch units for a rebuild but not sure how true it is.

Vader

I've had a Wells TFT sensor in my '86 TA since April 10, 1999. Not a problem at all. Every new vehicle since about '95 has used a TFT sensor instead of the hot wire, and that's a LOT of MAFs.

Since we are only dealing with a few mV DC up to about 4½V, connections are critical. A little resistance from poor connections will degrade the signal enough to make the ECM very unhappy. 35 year old connectors and crimped wiring can have more than enough oxidation, contamination, and corrosion to degrade the signal. Again, remember that a few millivolts is the difference between a good signal level and a level which doesn't satisfy the ECM threshold. My '86 didn't like a 14mV drop in MAF signal at idle.

To translate that to resistance terms, 3 to 6Ω in the wiring between the MAF analog output terminal and the ECM connector will ruin your day. Cleaning and deoxidizing WeatherPak connectors, ECM edge connectors, and splices within the factory harness is not fun, it's not exciting, it's not sexy, and it's plainly tedious and boring, but it needs to be done to properly restore operation.. Later vehicles used a F-M scheme for MAF signal at full voltage to avoid these kind of problems with eventual (and nearly certain) signal degradation, but we're stuck with what we have and need to address it.

Coincidentally, even modern vehicles use a similar analog voltage scheme (although AC voltage) for ABS wheel and axle speed sensors, and also operate at mere millivolts. As a result, a LOT of vehicles with a little age and mileage generate nuisance ABS codes from signal loss, and the industry makes lots of money on replacement hubs, sensors, wire hanesses, and parts to combat that.

Galaxie500XL

iTrader: (1)

This is extremely useful information. Thank you!

Vader

A scanner will never replace a multimeter, and my opinion is that far too many repair people (not "technicians") use the scannner as a crutch, and believe in what it reports without question. A true "technician" knows better because she/he is better trained.

That said, this is one of the instances where scanned data is a valid tool for diagnosis. Reading what the ECM thinks it is seeing can be an indication of signal loss. Any ECM/PCM is actually pretty stupid - It only believes what it is told. If a tech were to measure 750mV at the MAF signal terminal and the ECM reports only 650mV, that is useful information.

The same applies to the 5V regulated reference voltage supplied TO the sensor, and the signal ground everywhere. 4.8V at the MAF sensor reference terminal will ruin your day. All of those circuits need to be at specification or the signal to the stupid ECM will only be a lie, making it much more stupider.

Galaxie500XL

iTrader: (1)

This is critical in our cars...30 year old wiring can’t be assumed to still be in good shape—it’s critical to actually check it.

An example...I replaced the burn off relay, and cleaned my MAF, and seemed to have solved the code, but within a day, the car became hard to start. I decided to get out the fuel pressure gauge, and found that the system wasn’t priming at key on...a little while later, I found the problem.

windsurfer

The manual says in regards to the 5 volt reference signal connector B to ground: "Less than 1 volt indicates a faulty CKT 492." Why do you think 4.8 VDC will be a problem?

windsurfer

The manual says in regards to the 5 volt reference signal connector B to ground: "Less than 1 volt indicates a faulty CKT 492." Why do you think 4.8 VDC will be a problem?

Vader

A weak or low 5V ECM reference may not be as much of a problem for a MAF directly since the MAF electronics are designed to produce their own 0-5V output from the 12V supply voltage, but a weak ECM 5V reference can be a problem for the system in general. That reference voltage is used internally by the ECM for the IAT and CTS inputs, and is also used as a comparison to the MAF output, as well as being used externally for the TPS. If the input analog signals are skewed several mV due to an improper 5V comparator reference, the resultant "data" used by the ECM could be off. As we know, these analog voltage signals are relatively small to the point where a few mV can make a substantial difference.

As you alluded, the ground circuit (452) is also critical so that the 0V reference point for both the ECM 5V supply and MAF 0-5V signal are solidly connected. Again, a few mV difference there can create a substantial problem.

Galaxie500XL

iTrader: (1)

That 5v reference signal is critical. Everything is based on that value being constant, because the computer is making decisions based on the voltage difference between that 5v reference signal, and the signal being received from the sensors.

Vader

Bingo! That includes the grounding of the ECM, its various analog grounds, and grounding for the powertrain sensors and devices (EST, ICM, EGR/EVAP solenoids, etc.).