Search code examples
can-buscanopen

CANOpen protocol overhead

Can someone help me understand how much overhead CANOpen protocol adds on a CAN bus. In other words, if a node is generating a CANOpen PDO message of 8 bytes of application data, how much header/overhead does CANOpen protocol itself add? I'm not talking about the overhead of CAN bus itself which are the CRC, DLC, RTR, EOF and other bits. What is the extra load on the bus for using CANOpen vs using CAN directly. Thanks.

Solution

  • There is no "CANopen protocol" as such, it's the name for the whole application layer. Simplified, the actual protocols used by CANopen boil down to:

    • NMT frames, network management. Controlling the state of nodes and the network as whole.
    • SDO frames, service data objects. Configuration of individual nodes.
    • PDO frames, process data objects. The actual data frames.

    There's various other special-purpose ones such as EMGY, LSS etc as well, but the above are the main ones.

    In other words, if a node is generating a CANOpen PDO message of 8 bytes of application data, how much header/overhead does CANOpen protocol itself add?

    PDO have zero overhead. You get 8 bytes payload in an 8 byte PDO frame.

    SDO and the various NMT have some overhead though, typically 1 byte.