Martyn wrote:Digital lines can be used for serial decoding although the signals have to adhere to the specified voltage limits.
Dr. Nicolas, one of the great things about the Picoscopes, and being able to see the serial-bus signals in the analog domain, while still decoding them, is that you get to see what's happening... warts and all. Not after it has been 'sanitized' by threshold detection, which may vary between your test device and transceivers on the CAN bus.
In your example, the analog signals are very instructive. The CANL signal looks good. Proper amplitude and offset, with low noise evident. CANH on the other hand, looks problematic. The CAN signal is riding on top of a bias offset, has lots of noise spikes, and has really poor amplitude uniformity... especially toward the end where the CRC field resides.
These factors could make it difficult for devices on your bus to properly decode the data stream, not just PicoScope. This could result in a lot of bus errors, and retries. I'd see if I could confirm those problems on CANH (zooming in would help), and track them down and remedy them, before focusing on trying to decode the data itself.