OBDLink LX CAN-bus not sending some ID

jeby

Hi,

I'm using an OBDLink LX on RaceChrono Pro 7.1 (Android) as raw CAN-bus reader.

The app can read some ID such as 0x201 (rpm, accelerator pedal position, car speed) but not IDs 0x81 (steering) and 0x85 (brake)

For 0x81 it seems to read the first value and then get stuck at 00 00 00 00

Is this something to be fixed to RC side or a limitation on OBDLink side?

I know those IDs are available on the port because I logged them with a PiCan2.

aol_of_RaceChrono

Turn on the "Save device output" in expert settings, and it will save everything it sees. The logs are saved in the .rcz files when you share the session. You could also use serial terminal and connect to the OBDLink directly. With ATMA command it displays the messages on the bus.

jeby

device_output_200.obd.txt

serial_20210206_135726.txt

@aol

serial_20210206_150313.txt

I used the ATMA command with Serial Terminal Bluetooth for Android (I'm attaching both "text" format and "HEX" format) and then I went out and exported the device_output. It is not really clear to me how to interpret the results, except that I see in the device_output a lot of "data error" associated to a bunch of IDs including the 0x81 and 0x85 that I need.

PS: sorry for the double quote, something went wrong editing the post from android

@aol

aol_of_RaceChrono

@jeby I can see immediately there's some problems with RaceChrono. Can you export the Vehicle profile and send it to me tracks(at)racechrono.com please? I'm sure I can fix it.

jeby

https://racechrono.com/forum/discussion/comment/11346#Comment_11346

Done, thanks!

zixuw

I got the same problem! I believe there's a bug in RaceChrono that does not send CAN pass filters for 1 byte IDs correctly. Guess I should have read more forum threads before doing my own investigation...
Anyway, I think I've found the problem, see https://racechrono.com/forum/discussion/1854/can-bus-data-logging-was-custom-obd-ii-channnels-for-miata-nd2#Comment_11392

aol_of_RaceChrono

@zixuw your analysis is spot on, even if I figured it already out couple weeks ago This was a simple printing bug ("%3x" vs. "%03x"). Fixed in upcoming v7.2.0.

timurrrr

That's awesome, can't wait to try it!
Would be happy to test the fix with OBDLink MX+ on my Subaru BRZ as soon as the beta is available.