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.

ISO1541: Can the Master be on port 2?

Part Number: ISO1541
Other Parts Discussed in Thread: ISO1540

Section 8.4 suggests that Port 1 has precedence but that isn't stated specifically.  Can the master be on Port 2 or will that possibly allow Port 1 to load the SDA2 line? 

Is the ISO1541 designed to be placed between an I2C Master and an Isolated bus, or can it be placed between an I2C bus and an isolated I2C slave?  I ask because Port 2 can handle much more parasitic capacitance (400 pF) than Port 1 (40 pF), so a bus on Port 1 doesn't make sense. 

Thanks,

  • The 1541 has a unidirectional clock line, which implies that the master (which drives the clock) must be connected to port 1, and that any slave connected to port 2 must not use clock stretching.

    In theory, it is possible to have other slaves on port 1 (together with the master(s)), as long as the maximum load is not exceeded.
  • Clemens Ladisch,

    Thanks for answering the query.

    John,

    I Agree with Clemens Ladisch, The master must be connected to port 1 and slave to port 2 to ensure the clock direction. And the load on port 1 side should be less than 40pF and load on port 2 should be below 400pF

    Regards
    Tejas
  • Thanks. I see now on p. 5 that the image shows the unidirectional clock signal. If I switch to an ISO1540 can I have the master on Port 2? Or are there other restrictions?

    If not, what is the recommended way to have an I2C bus (C > 40 pF) connect to an isolated slave (C < 40 pF)? I'm thinking that if I use the ISO1541 I'll need an I2C bus splitter (with <40 pF) upstream to drive Port1 while my slave is connected to Port2. That's more parts and PCB area so at that point it seems I might be better switching to a different isolator.

    On a side note, I originally started this thread because my I2C bus was loaded and non-functional. I've found the ISO1541 was not the problem; another device on slave side (currently Port 1) was dragging SDA1 low and thus dragging SDA2 low.
  • On the 1540, both lines are fully bidirectional, so the isolator does not care which side contains master(s) or slave(s).