CAN bus in infiniti qx80

If you are not a PicoScope user and want advice with a diagnostic problem then post here. Forum members may be able to help.
Post Reply
sleepwalker
Newbie
Posts: 2
Joined: Wed Jan 08, 2020 11:45 am

CAN bus in infiniti qx80

Post by sleepwalker » Wed Jan 08, 2020 2:40 pm

Hi,
recently I got air bag light blinking, radio not working (no volume change from 0) and a few more errors about around view camera U1300 and can bus problem U1000.
I checked all fuses, grounds, and start to play with can bus signals.
I always beveled that idle voltage for can H and L must be close to 2,5V but
On ACC this level is around 2V (H and L) but on ignition voltage jump to 2,5V like use to be. I didn't decoded those because looks bad.
Tried to trace module by module but no luck, resistance is ok
May someone of you have any idea or seen something similar?
Thank you for help
Dan
Attachments
CAN oscil H _L.jpg

liviu2004
TwoWaves
TwoWaves
Posts: 206
Joined: Sat Mar 19, 2016 8:23 am
Location: Netherlands

Re: CAN bus in infiniti qx80

Post by liviu2004 » Wed Jan 08, 2020 7:16 pm

post psdata file.

sleepwalker
Newbie
Posts: 2
Joined: Wed Jan 08, 2020 11:45 am

Re: CAN bus in infiniti qx80

Post by sleepwalker » Wed Jan 08, 2020 8:19 pm

Hello,
Please see this file, recently I measured line H and L to ground and show around 1Mohm, between H and L is exactly 60 Ohms, and still this different voltage on ACC - idle around 2V, on IGN 2,5V.
Thank you for help
Attachments
20200107-0001.psdata
(3.72 MiB) Downloaded 30 times

ben.martins
Pico Staff Member
Pico Staff Member
Posts: 204
Joined: Tue May 16, 2017 1:02 pm

Re: CAN bus in infiniti qx80

Post by ben.martins » Fri Jan 10, 2020 11:13 am

Hello Sleepwalker,

Thank you for your post. Taking a look at the capture you have posted it appears to be similar to a problem I had on a Mazda with a loss of audio. A link to the case study can be found here - https://www.picoauto.com/library/case-studies/no-audio

You'll see that I use a couple of math channels to help with the diagnosis on this vehicle being A+B and A-B. Typically when looking at CAN we have a differential network where the voltage levels will sit at 2.5V when idle and then pulled high and low, 3.5V and 2.5V, when data is being transmitted. Using the math channel A+B, we should see that the voltage level, regardless if data is being transmitted or not, will sit at 5V.

A-B will give you what the ECU's will be seeing, the physical layer. Here any interface present in both H and L will be dismissed and no longer be present in the signal. This is what makes CAN extremely fault tolerant. The following image shows your uploaded capture with these math channels.
CAN with MATHS.png
You'll notice that the A-B reveal an extremely clean signal despite how it appears in the raw signal captured. What gives us some direction is the A+B where we were expecting 5V but only seeing 4V. If we look back to the case study, the fault here was an ECU corrupting the 2.5V idle voltage for the whole network and when removed we saw the 2.5V return and the signal for H and L restored.

There is of course a subtle difference here and that is you are seeing these voltage issues with the ignition set to ACC but with the IGN on and engine running, the signal are as expected. Is this correct? I have seen this before where it is in fact normal to have an odd CAN signal which is then resolved when the IGN is switched ON. Personally I would like to verify this before moving on with any diagnosis. If you have access to another vehicle to try then great, if not hopefully a member on the forum will be watching and could help.

Hopefully something in there helps with the diagnosis but please update us with anything you find.

Kind regards

Ben

Post Reply