New MAF, now runs retarded and kicks at 5k

[Pietercc]

I've already tested the maf according to greenbooks, spent a night looking threw mine to find anything there. over all I'm done with this issue, I will be switching to 2.4 soon

 

[Otto Mattik]

Next I would move on to checking the connection and possibly trying another power stage. They can get corrosion in the connector being near the battery. I'd also check the resistance of the coil. Original spark plug wires are around 500ohms per foot.

Does a LH 2.2 760T have a "2.4" type power stage(mounted on fender near battery) ?

 

[apollokid]

Check out this old thread (post #13), it has good MAF measurement information and helpful links:

AMM testing procedure

OK, I need a new AMM and I would like to know the resistance values across the terminals so that I can test junkyard AMMs before I buy one. There has to be a procedure. I think I have even read it before. I have had AMMs that allow the engine to run, but the car idles like crap and gets bad MPG.

turbobricks.com

...
Because of this it makes it next to impossible for the technician to accurately diagnose the air mass in the field. We use a calibrated flow bench that measures the exact CFM air flow to then compare the voltage to. This is not practical in the field because temperature, altitude, humidity and the mechanical condition of the engine will affect how much air the engine is drawing in. So just to say that the Air mass should have xx.xx volts at idle would be completely false since all these parameters must be accounted for. Also even if a range is given just a small amount of deviation in the output causes poor performance.


I can confirm that what he says is true, I've invested a lot of time on various methods to test a MAF and also if when the engine at idle on AMM 016 the voltage should be approx 2.2V this value cannot be used to assume if the MAF is good or not. It can just indicatethat it is not totally broken.
A MAF is not a normal sensor, it must be considered as a thermometer: it have to measure the exact value of the temperature, not just changes the output values between hot and cold.

Last summer I build a simple DIY flow bench project using a shop vaccum cleaner and also if I was not able to drive the MAF to work at it full output I was able to compare its signal with the output signal of a known good MAF connected in series to it (so that they have to measure the same amount of air).
By modulating the sucked air with a DIY throttle plate I was able to analyze the output signal response on different MAFs and I found that most of the MAFs with problems can be grouped into two main classes:
- MAFs with attenuated output signal (found on some original Bosch hot wire ones)
- MAFs with a delay in output response (found on some aftermarket clones)

For the former problem I also tried to recalibrate the output signal of a bad Bosch MAF by making a simple DIY external module and it returned to work very well.

 

[Khrrck]

Last summer I build a simple DIY flow bench project using a shop vaccum cleaner and also if I was not able to drive the MAF to work at it full output I was able to compare its signal with the output signal of a known good MAF connected in series to it (so that they have to measure the same amount of air).

In theory... If you mapped the signal at all flow rates from some good LH MAF sensors, and also made a similar map for a more modern, easy to source MAF from a different car, you could create a conversion table and set up a microcontroller based system to run LH with a modern MAF.

In practice I suspect it's not worth the effort at the moment.

 

[apollokid]

It was the same idea I had but to be able to test also the MAFs of other cars by using the same known well working MAF that I use as a reference.
The picoscope software that I use to plot and compare the two output signal could make this job pretty easy.

An aftermarket MAF today is relatively cheap but what is missing to me (and I believe not only to me) was a valid tecnique to check if a MAF is working well or not that it wasn't to simply try to install it on the car and drive it for a while.

 

[Pietercc]

It was the same idea I had but to be able to test also the MAFs of other cars by using the same known well working MAF that I use as a reference.
The picoscope software that I use to plot and compare the two output signal could make this job pretty easy.

An aftermarket MAF today is relatively cheap but what is missing to me (and I believe not only to me) was a valid tecnique to check if a MAF is working well or not that it wasn't to simply try to install it on the car and drive it for a while.

We would need to get a database started with people stating their own values, we just need to make standard measuring parameters.