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 opto-isolation at analog layer

Other Parts Discussed in Thread: ISO1050

I've used the TI ISO1550 in a few designs now - works great!

The ISO1550 provides interface between CAN_l/CAN_H <=isolator=> TX/RX;
ie analog CAN levels on one side and digital levels on the other.

For a new application, I need to provide isolation on analog side only;
that is CAN_L/CAN_H <=isolator=> CAN_L/CAN_H

Does TI have a part that does this ?
Or can someone recommend an isolation circuit for this?

Thanks!
Best Regards, Dave

  • Hi Dave,

    Unfortunately, the isolators we have are digital isolators and therefore only have push / pull outputs. That means the outputs are either pulled high to VCC or pulled low to GND. They have no way of creating the 2.5 volts common mode reference.

    But for my understanding, can you explain why you need to isolate the analog signals of the bus? Can you share a block diagram of what you are trying to do? I’d like to see if there is an alternate way we can achieve what you are looking for.

    Thanks,

    John 

  • Also, I am guessing that you mean the ISO1050, and not the ISO1550. Am I correct?

    Thanks,

    John

  • Apologies, ISO1050 (not ISO1550). Need more coffee...

    The application is a CAN-repeater with isolation.
    A CAN repeater requires anti-feedback logic to avoid locking up the bus.
    The repeater part I'm using has one logic-level port with anti-feedback,
    to which I've attached a ISO1050. This works but....

    Unlike the old SI9200 transceiver, the CAN ports on the repeater
    cannot exceed Vcc. This creates a hazard during power-up, as the
    if the local Vcc powers up slower than some other node, the CAN
    bus voltage can destroy the chip. I discovered this when carelessly
    hot-plugging my repeater...

    So, to prevent this hazard, ideally I should isolate the other side of
    the repeater chip, which has CAN (not logic-level) IO...

    Hope that all makes sense!
    Any suggestions ?

    Thanks again!
    Best Regards, Dave

  • Whoops, my problem is actually with the ISO1050 !

    I erroneously convinced myself that I destroyed the CAN
    repeater, but actually it is fine and my interface should be
    bullet-proof...

    I will post a new question with a more appropriate subject
    line...

    Sorry!
    Thanks,
    Best Regards, Dave