- Joined
- Apr 6, 2021
- Messages
- 57
- Reaction score
- 37
- Location
- Taufkirchen, Germany
- VCDS Serial number
- C?ID=418771
When logging with HEX-NET (CB 0.4555, FW 1.64), once in a while a row in the resulting CSV file is shorter than it is supposed to be:
In this specific case, the engine speed time stamp and value are missing from the line starting with ",192.67". Other values from the same group (engine torque and fuel consumption) are there. The problem is that this breaks the entire line. If the value obtained from the ECU was invalid, I would prefer that it maintains the CSV column structure.
I stumbled over this when making an application that creates a gnuplot data file from a VCDS recording (plotting with LibreOffice is way too much manual work). In one file with 8690 lines, there are 6 lines showing this problem, and it's not always the engine speed that was dropped. As a workaround, I'm only using rows where the number of columns is the same as in most other rows.
I've also looked through older recordings I've made, and it's been there in CB 0.4543 as well. VCDS on Windows doesn't seem to show this behaviour.
Code:
Fri,Dec,10,2021,08:37:40 -VCID:28135BA3CDC20B2BF482A78-807D,VCDS Mobile: 0.4555
03G 906 013 K,ADVMB,R4 1,9L EDC G000SG 2940,
,,G002,F3,G006,F0,G015,F0,G015,F1,G015,F2,G041,F1,G041,F2,G062,F2,G073,F1,G074,F1,G000,F0,G000,F0,
Marker,TIME,Group 2 - Field 3,TIME,Group 6 - Field 0,TIME,Group 15 - Field 0,TIME,Group 15 - Field 1,TIME,Group 15 - Field 2,TIME,Group 41 - Field 1,TIME,Group 41 - Field 2,TIME,Group 62 - Field 2,TIME,Group 73 - Field 1,TIME,Group 74 - Field 1,,,,,
,STAMP,Coolant Temperature (G62),STAMP,Vehicle Speed (actual),STAMP,Engine Speed (G28),STAMP,Engine Torque ,STAMP,Fuel Consumption,STAMP,Position (spec.) EGR-Actuator,STAMP,Position (act.) EGR-Actuator,STAMP,Ambient Temperature,STAMP,Distance since Regeneration,STAMP,Temperature bef. Turbo Charger,,,,,
,, °C,, km/h,, /min,, Nm,, l/h,, %,, %,, °C,, km,, °C,,,,,
,0.38,5.4,0.58,0.0,0.79,903,0.79,51.0,0.79,0.84,0.99,13.8,0.99,10.7,1.20,3.6,1.40,100,0.17,168.0,,,,,
,1.81,5.4,2.02,0.0,2.22,903,2.22,51.0,2.22,0.84,2.43,12.2,2.43,10.7,2.63,3.6,2.84,100,1.61,162.0,,,,,
,3.25,5.4,3.45,0.0,3.66,1491,3.66,138.0,3.66,5.04,3.86,0.0,3.86,0.0,4.07,3.6,4.27,100,3.04,156.0,,,,,
...
,189.80,39.6,190.01,82.0,190.21,1491,190.21,240.0,190.21,9.03,190.42,0.0,190.42,0.0,190.62,0.0,190.83,100,189.60,354.0,,,,,
,191.24,40.5,191.44,85.0,191.65,1554,191.65,243.0,191.65,9.45,191.85,0.0,191.85,0.0,192.06,0.0,192.26,100,191.03,420.0,,,,,
,192.67,40.5,192.88,88.0,193.08,243.0,193.08,9.66,193.29,0.0,193.29,0.0,193.49,0.9,193.70,100,192.47,444.0,,,,,
,194.11,40.5,194.31,92.0,194.52,1680,194.52,240.0,194.52,10.08,194.72,0.0,194.72,0.0,194.93,0.9,195.13,100,193.92,444.0,,,,,
,195.54,40.5,195.75,95.0,195.95,1722,195.95,126.0,195.95,5.25,196.16,12.2,196.16,9.7,196.36,1.8,196.57,100,195.34,450.0,,,,,
...
I stumbled over this when making an application that creates a gnuplot data file from a VCDS recording (plotting with LibreOffice is way too much manual work). In one file with 8690 lines, there are 6 lines showing this problem, and it's not always the engine speed that was dropped. As a workaround, I'm only using rows where the number of columns is the same as in most other rows.
I've also looked through older recordings I've made, and it's been there in CB 0.4543 as well. VCDS on Windows doesn't seem to show this behaviour.