ADM3056E CAN_H and CAN_L waveform

Hi Pat, 

Looks common-mode. What's the applications termination circuit look like? Split termination with a filter cap? Are the network cables very long or transceivers being supplied from independent sources? 

Regards,
Jason

Hi Jason,

yeah, definitely common mode. The bus is on both ends terminated with 120R across CANH and CANL. Cables are unshielded but quite short : the longest one is just about 1m (about 3ft).

The bus sides of the trancievers share a common supply of 5V. Power and ground are also distributed on the same cable as the signals but are clean. Logic sides are of course supplied by individual supplies, therefore the isolator.

[EDIT]

Well, sometimes it's the simple(-ish) things. I just tried split termination and now everything looks fine.

Do you have any input on what I was seeing? It did not register as high-frequency noise or anything reflection-related to me, so I still don't see why the low-pass of the split termination is so obviously beneficial.

Thank you for your time and kind regards,

Patrick

Hi Patrick, 

Its not high frequency noise or reflections. During the dominant to recessive transition there are two different devices turning off quickly. Any difference in the turn off characteristics will translate into common-mode voltage. There may also be some difference between CANH and CANL parasitics created by the bus (layout, protection components, construction…). Those may contribute to common-mode as well. 

Regards,

Jason

Hi Patrick, 

Its not high frequency noise or reflections. During the dominant to recessive transition there are two different devices turning off quickly. Any difference in the turn off characteristics will translate into common-mode voltage. There may also be some difference between CANH and CANL parasitics created by the bus (layout, protection components, construction…). Those may contribute to common-mode as well. 

Regards,

Jason