i've just been playing around a bit with EG33 data logs from evoscan, and i've found a few discrepancies with the 7225xx definitions file posted here.

- O2 conversion factor: the conversion factor should be x*5/256 (like the O2 min/max).

the data.xml file here has x*5/512, which gives values in the range 0-0.5V. they should be 0-1V. changing the O2 sensor conversion factor to /256 gives the correct range, and plotting them with the min and max shows them all working together. i verified this against Phil Skuse's original conversions at the vwrx site, and an SSMI eavesdropping report at TomsSVX site.

- MAF conversion factor: this data.xml uses x*100/255, which gives values expressed in % of max. using x*5/256 gives values in volts, which imho is more useful for diagnosis (corresponding to the service manuals), and is also the standard byte to voltage conversion factor.

- TPS conversion factor: the conversion factor for TPS should be x*5/256, again to read in volts.

several other request ID's differ from Phil's:

- battery voltage - here: 102F, phil's: 12A2. both show nearly the same readings, but in my test runs 12A2 showed the voltage dip at starting time.

- speed - here: 1071, phil's: 12A4. both are within sampling error of each other.

- rpm - here: 106F, phil's: 1290. both withing sampling error, but 106F correctly shows zero at pre-start; 1290 shows an arbitrary large rpm.

- air/fuel correction - here there are 2 for the two banks: 1097/109F, phil's had only one: 12AD.

also, in the logs i've seen, the canister purge parameter and the barometric pressure don't produce sensible values....