I’m curious if other calibrators have seen slow degradation of AEM map sensors. I installed an AEM 5.2 bar absolute pressure sensor in my 08 STI manifold about 2.5 years ago as part of a build up (documented here: http://forums.nasioc.com/forums/showthread.php?t=2472813 ).

Over the last year I started to notice that my boost was running a bit higher than what I targeted. At first I thought the boost gauge might be off a little bit, but when I added pre and post intercooler sensors it was clear that was not the problem. I noticed before starting the car that the barometric sensor was saying 14.4psi while the MAP sensor was saying 12.9psi. That is a ton of error. I dug around in my logs, and it appears the sensor has been slower degrading and reading lower values. I had a spare sensor that was on Perrin’s 06 STI laying around, and sure enough it also read more then 1 psi low. (similar age). I got a brand new sensor and it was spot on (within about .1 psi of the baro sensor with the engine not running).

Since the MAP sensor is the core sensor for not just boost control but also most of the ECU calculations used for fuel and timing when in speed density the accuracy is important. Since the sensor tends to drift to a lower reading your AFRs would go leaner and timing would likely increase. Not a great combination in a highly tuned car.

I believe the AEM sensor is made by Kavilco, and this particular type of sensor is direct media interacting, so whatever is pushing in the sensor is acting directly on a small film. Over time that film can get damaged, and I would guess that a leak across the film would result in lower readings.

Something to keep in mind if you have one of these sensors on a car for more than a year or two. It is easy to check if you have a newer car that has an independent baro sensor.