DZR now with option for "VAGCOM Stopwatch"

Hello,

In collaboration with "Interpolator" = Jan6K , I have implemented a calculation feature in the DZR that allows determining the time spent within the 2000-4000 rpm range, based on VAGCOM logs.
(Download: DZR version from September 29th.)

To verify the accuracy of this measurement method, it would be interesting to learn about practical experiences with measurement series and the observed variation in the results, ideally in comparison to measurements taken with a stopwatch.

So, if anyone ever has the time, the inclination, and a liter of diesel left over -> feel free to join us.

Hello community!


I've been reading for a long time, but this is finally my first post. Post and it's not even a question .


I have calculated the DZR (Direct Shift Recording) time on several occasions using VAG_com logs and subsequent post-processing with interpolation. For each dataset, I always take 4 measurements, with 2 measurements in each direction.

As a comparison for the accuracy that Ulf mentioned, I can offer the differences from a total of 10 measurements, each taken in the same direction (i.e., from 5 different datasets).

Minimum deviation: 0.00 seconds.
Maximum deviation: 0.13 seconds.
Average deviation: 0.06 seconds.

Average deviation (in %): 0.8


By the way, the line numbers in the DZR file for the time calculation formula at 4000 rpm may be off (as of September 30th, 2:00 AM). For example, B9 instead of B10, etc.


Best regards,
Joe.

Hi Ulf,

Joe is right... I only checked the formula for the year 2000 and assumed that the others were just copies.

Joe's error analysis actually aligns very closely with my assumptions/calculations.

Assuming that the MSG (presumably a device or system) accurately determines the 1/100th position and controls the speed in increments of 21 RPM, the following is true (copied from an email I sent to Ulf):

Quote:

By the way, I immediately tried to figure out what kind of mistake... fits into the 21st-century framework by incorporating "real" data once. (from one of my 2002 DZR experiments), I interpolated it once. reduce the speed by 10, and then increase it by 10, because that covers the range. That's exactly the right size band. With that, I'm achieving a shift of 0.03 seconds. at the moment of impact, so in the case of acceleration, the... achieved a score of 2000 (the error depends on how quickly...). (as the speed increases), in the worst-case scenario, an error of 0.06. Seconds (this is exactly when, at 2000, both values indicate an error). (with one direction having a value of 1000, and the other direction having a value of 4000). IMO is fine. that still... always, provided that the MSG is accurate. It provides timestamps for the measurement data. I assume that... This is because for all the calculations related to fuel injection, there MUST be a... It requires a very precise measurement, where even fractions of a hundredth are considered relatively coarse.

While this is only half of what Joe determined, I think it's also worth considering that two attempts of this nature can never be exactly the same, which could explain these discrepancies.

In my attempts in 2002 (there's a thread with all the data and analysis), the deviations were actually a bit larger, but the track wasn't ideal either, because the wind conditions wouldn't be exactly the same between the sections due to the trees and the two curves.

Best regards,

Jan.

joe* wrote:

By the way, in the DZR file, the line numbers in the formula for time calculation at 4000 rpm might be off by one (as of September 30th, 2:00 AM). For example, B9 might be incorrectly labeled as B10, etc.

Oh, how embarrassing .
That, of course, is the time calculation for the tonne .

Thank you for paying attention!

EDIT:
The corrected version is available online (the download name has been changed to "..." for better distinction). October...").