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.

ISO1500: baud rate

Part Number: ISO1500

We are using the ISO1500DBQ isolated RS485 transceiver IC in one design to give us an RS485 MODBUS port.  On another design we are using a Linear Tech component.  Same code is being used on both boards to communicate on the bus.  On the board with the ISO1500DBQ we are unable to run the bus any faster than 33,600 baud rate.  On the board with the Linear Tech component we can go all the way up to 115200.  I see a big difference between these two components is that the IS01500DBQ has a max speed of 1Mbps whereas the Linear Tech component is 20Mbps.  Is there any reason the ISO1500DBQ wouldn’t be able to run 115200?

Thanks

Viktorija

  • Viktorija,

    I would expect the ISO1500 to be able to pass but I have several questions.

    Can you share the part number of the Linear Tech device? 

    Is the bus terminated or unterminated?

    What protection devices if any are on the bus?

    How long are the RS485 bus line cables?

    What are the physical differences between the two designs mentioned? 

    Can the customer provide waveforms of the bus when the ISO1500 is not operating correctly?

    Respectfully,

    Lucas

    1. The Linear Tech Device is the LTC2862AIS8-2#PBF
    2. There are two boards:
    1. On Board 1 (the one with the ISO1500 on it) Bus is Terminated on both ends  R1 on the PCB and the USB to RS485 cable we are using also has a termination resistor on it.
    2. On Board 2 (the one with the LTC2862 on it) Bus is not terminated on the PCB, though I can populate one if necessary.  We are using the same USB to RS485 cable here and it is terminated there.
    • The protection devices are the same on both boards, see schematic sent via email.
    • RS485 to USB cable we are using is the USB-RS485-WE-1800-BT from FTDI.  Standard length, trace lengths are maybe an inch long.
    • Differences in Design:

      1. Biggest difference between the two designs is that the one using TI needed to be isolated from the outside world, which is why we used your ISO1500 instead of the LTC2862.
      2. The ISO1500 provides an RS485 connection to an STM32F070 MCU from STMicro.  The LTC2862 is connected to an STM32F091.  These processors are very similar, the 091 just has more peripherals, Flash and RAM.  They are both Cortex M0 MCUs
      3. RS485 side of the design layout is very similar
    1. Sorry, but I am unable to send waveforms.

  • As per there schematic there could be 2 different protection devices on the bus along with a comment that says "Do not populate unless needed". I'm assuming both are present based on the previous answer. The ESDONCAN1lt1g is meant for CAN and should be removed. 

    Providing waveforms can help us see if there is excessive capacitance on the bus lines or otherwise diagnose the issue. Unfortunately this will be hard to solve without waveforms. 

    Are both of these boards communicating on the same bus to each other? If so, then both RS485 devices should be terminated.

  • No, both devices are not on the same bus at the same time…we are communicating with them “point to point”

    Is there a particular reason that this TVS Diode won’t work for RS485?

  • The clamping voltages of the ESDONCAN1 only make sense for the CAN devices that can handle bus voltages that high, The ISO1500 can only handle +/-18V volts on the bus lines with respect to ground. This means that the clamping voltages of the ESDONCAN1 at about 33V won't actually do any good to protect the RS485 device. Generally, the SM712 protection diodes are better suited for RS485. Several manufacturers make the SM712 with varying performances. 

  • Lucas, 

    Thanks for catching the TVS diode issue.  

    Turns out it was a combination of a software corruption problem during programming and a broken USB to RS485 interface cable. 

    Thanks for your help!