ADM3053
Production
The ADM3053 is an isolated controller area network (CAN) physical layer transceiver with an integrated isolated dc-to-dc converter. The ADM3053 complies...
Datasheet
ADM3053 on Analog.com
Hello, I am encountering an interesting issue when testing my circuit board, which uses the ADM3053 transceiver. The signal being sent to the transceiver (scoped on the isolated GND2 side) appears to be very nice, but the acknowledgement and return signal sent by the ADM3053 appears to be garbage (strange pulses that don't even look like CAN data). Right now it seems like the root cause lies somewhere in the hardware configuration, but I can't seem to figure out anything I did that's outside the recommendations of the datasheet. The speed of the communication is set up to be 500kbps. I have attached images below of my CAN schematic and the oscilloscope CAN waveforms. If anyone has seen something like this before and might have an idea of what could cause weird behavior like this, your expertise is greatly appreciated.


Hello,
Looking at your schematic, I see that you've set Rs to 47k which sets a lower driver slew rate. With a 500kb/s data rate, this potentially could be an issue. You may want to reduce Rs to allow a faster driver edge rate.
Eric
Thanks Eric, it turned out to be some EMI filter capacitors at my connector that were too large for the 500kbps CAN signal.
My dear,
As I look at your schematic, your Rs is 47k, resulting in a lower driver slew rate. Considering the 500kb/s data rate, it may pose a problem. For a faster driver edge rate, you may want to reduce Rs.