Every dataset output is only as good as its input. We have created maintenance procedures to keep our cars for 300,000+ km with heavy-duty use. But what if your sensors have false readings?
The Maintenance trap
While I keep my car in good shape with proper maintenance – who could imagine there is something like sensor wear. So much time wasted by getting optimal engine calibration based on input data containing errors.
How bad it can get?
I discovered all by accident like usually. We have been working on optimal AFR, and there were still large discrepancies between reality and theory.
I washed MAF & MAP sensors because they tend to get dirty. Wow, it was like new car. No placebo, confirmed by logging.
Nowadays, I am changing all important sensors automatically with every used-car purchase.
MAF sensor had ~8% lower air mass readings. You have just lost 8% of engine power!
Bosch MAF sensor
Bosch’s lifespan recommendation:
- 4 years
- 150,000 km
Wash it carefully first, or simply replace it (it costs €100 for 2.0 TDI)
Russian Roulette
Still having weird readings? Rinse and repeat – as you can see from the picture in the beginning, it can be any of the sensors in the diagram.
I ended up replacing MAF, MAP, and Lamba sensors. Best chiptuning ever.
Leave a Reply