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.

CAN Transceiver Understanding

Hello,

I'm a junior EE trying to get into the CAN world. I've done some extensive research into the CAN system and am familiar with the general dynamics of the system (CANH, CANL, up to 1Mbps, extended-ID...

I've hit a road block in trying to grasp why many microprocessors have "CAN" ports, but yet from what I've read, a transceiver is still needed to interface with a CAN system. With my agenda, I'm wanting to sniff a CAN bus and then replay the system when I've taken a node out of the network so that the node still believes it is in the network.

I just don't see what the purpose of a CAN transceiver, such as the Philips PCA82C250, is needed for when I have pins on the development board specifically for CAN. My gut tells me it has something to do with the fact that the micro will work with TTL, whereas the CANH/CANL works by differential voltage which doesn't necessarily follow the 0/5V digital standard.

Am I anywhere close/on track? Thanks for any help you may be able to provide.

  • Hi Kevin,

    just to make you aware this is a TI forum. Unfortunately we don't have Philips or NXP development boards with PCA82C250 as we prefer selling our own CAN transceivers.

    However generally speaking a CAN controller (and its memory) contain the CAN protocol intelligence and operate on a high-speed, low-volt CMOS process using single-ended or unbalanced signaling.

    A CAN transceivers however (often required driving long lines) uses differential signalling with dominant and recessive states and usually comes in a more robust process technology.

    - Intergrating both function has proven uneconmoically, that's why SC vendors keep the transceiver portion as stand-alone.

    - Differential signalling offers high robustness against noise.

    - Dominant and recessive states allow for arbitration during multi-bus access attempts.

    Our CAN specialist, Scott Monroe, has decades of experience in automotive CAN applications and CAN help you should you have further questions.

    Thomas Kugelstadt,

     Applications Manager, Industrial Interfaces

  • Thanks, Tom.

    Yes, I'm aware of this being a TI forum, and just used the PCA82C250 as an example that I'm familiar with, but I still understand your comment.

    Are there any documented ways to use any TI products for SWCAN (Single Wire CAN) transceivers?

    Thanks,

    Kevin

  • Kevin,

    unfortunately I never got the chance to work with SWCAN, but I will be checking the TI website and also will ask Scott Monroe at  s-monroe1@ti.com , whether we have this type of documents available somewhere - once I'm returning from vacation on Janury 9th.

    Happy Holidays, Thomas

  • Sound good, Thomas.

    For the record SWCAN is a variation created by GM as a "lower speed, cheaper" alternative to the traditional CAN. Their version runs at 33.3kb/s.

    Merry Xmas,

    Kevin

  • Kevin, I saw an apps-note at NXP. It showed that the SW-CAN uses the CANH signal but instead of going to 2.5V in recessive mode, it is going to ground.

    I have simulated a cheap SW-CAN transceiver circuit. It does not have any wave-shaping but it's dirt cheap. Here DI stands for Driver-Input and RO for Receiver output.

    By the way, Scott is not only the application person but also defines new products. Perhaps he has some new circuits on his roadmap that will support SW-CAN.

    Merry christmas to you too, Thomas

  • That looks really good. This should help with my application.

    Thanks for you time & help.

    Kevin