• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 35 subscribers
  • Views 456 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Original question:

TCAN332: CAN BUS dropped packets troubleshooting

TCAN332: Inconsistent Noise on CAN Bus and Ground Reference voltage difference

Gideon Kanikevich
Prodigy 40 points

Part Number: TCAN332

Tool/software:

I'm currently seeing some very inconsistent noise on my CAN Bus.

One packet 'burst' is relatively clean:

While another packet 'burst' seems to have a lot more noise:

I don't understand why these specific packets are either picking up/inducing extra noise when the other ones are not if they are on the same bus. Would the noise not be consistent throughout all boards, regardless of who is sending/receiving data?

Please let me know if more info on the setup is needed.

Thanks in advance!

  • 0
    Prodigy 40 points

    After doing some further testing I realised a stupid mistake I made.
    The packets which seem clean are those which are being sent out of the device in which I am probing at.
    If I move my CRO probes to the other node, the noise seen on the two separate set of bursts is inverted.

    This tells me the design of the PCB is inherently 'okay' and that there is probably a source of noise inside the system.

  • 0
    Prodigy 40 points

    I found the source of the noise to be an on-board power supply injecting lots of noise on the bus. When removing this rail, the supply is quite clean.

    Now that the waveform is a lot cleaner, I am seeing the reference voltage slightly change during communication:

    As can be seen there's roughly a 200mV difference in the reference voltage.

  • +1 in reply to Gideon Kanikevich
    Guru 322780 points

    The reference voltage is only weakly driven.

    Anyway, what matters is the differential voltage. Please tell the oscilloscope to compute CANH − CANL. It looks as if even the noisy waveforms would have a nice differential voltage. (If you have common-mode noise and cannot remove it as its source, common-mode chokes would help.)

  • +1 in reply to Gideon Kanikevich
    TI__Mastermind 31080 points

    Hi Gideon,

    As Clemens said, the differential voltage is more important and it looks find from your waveforms. The system is expected to continue operating with the ground shift present.  

    For noise filtering, you can use common mode choke and that's what we usually recommend for most CAN applications.

    Regards,

    Sean