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.

TLC5917: Bus Contention

Part Number: TLC5917

Hello,

We believe we are seeing a bus contention issue using the TLC5917. Through testing, we were able to determine that the TLC5917 causes SPI bus contention and prevents the DRV9 MCU from reading data back from the CP2400 via SPI. When I scope the signal, I can see the CP2400 is trying to transmit data, but it can’t fully pull the MISO line up to 3.3V or down to ground so I either read back 0xFF or 0x00.

If we remove the IC from the board, this issue goes away. If we just disable all software control of the TLC5917, we still run into the bus contention issue.  

Attached is a scope capture showing the issue. The SPI transaction with the issue is the last one in the waveform (yellow = MOSI, blue = CS of CP2400, green = MISO, pink = SCLK).

Have you run into an issue like this before?       Thanks very much for your help, Keith

  

  • Hi keith,

    how to connect the SDO? SPI is a series peripheral interface, which means the first SDO should connect to the next SDI when cascaded.

    Regards,

    Shawn.

  • Hi Shawn,

    There are 3 devices using the same SPI bus so the SDO of the TLC5917 is connected to the SDO's of the other 2 devices since they are all slaves to the DRV9 (master). The same is true for the SDI line. The SDO line of the TLC5917 is then connected to the MISO pin of the DRV9 and the SDI line is connected to the MOSI pin of the DRV9. If the design had multiple TLC5917 devices then I would agree with you, but that is not the case here. Does that answer your question?

    Thanks,

    Michael

  • Hi Michael,

    It may be not the right connection way for the series peripheral no matter the device is TLC5917 or not. the data on SDI will go to SDO with a shift register length delay unless the device can be shut-downed. the different data flowing to the same SDO will cause a big problem.

    Regards,
    Shawn.
  • Hi Shawn,

    How do I shutdown the TLC5917? I believe that I have already tested every combination of setting/clearing the LE and OE signals and none of the combinations fix this issue.

    Thanks,

    Michael

  • Hi Shawn,

    Here is another scope capture but with the SDO signal (pin #14) of the TLC5917 disconnected from the bus. Channel 1 (yellow) is the SDO bus signal, Channel 3 (blue) is SCLK, and Channel 4 (pink) is the SDO pin of the TLC5917, which is lifted off of the PCB. As shown in the capture, the TLC5917 is clearly "blocking" some of the data that comes in on the SDO bus, which is preventing another one of my devices from operating correctly.

    Thanks,

    Michael

  • HI Michael,

    The device cannot be shut down with the logic configuration. I have no idea how to fix with the parallel connection. I might suggest you to change the parallel to series.

    Regards,

    Shawn.

  • Yes, you can disconnect the SDO if the error flag is no need to read.

  • Hi Shawn,

    I cannot change from parallel to series since this would not work for my application. I was also hoping to use the SDO signal to read the error status in the TLC5917, but that hasn't been working correctly either. I think it is a bit ridiculous that I have now found 3 ways in which this device violates its own datasheet and does not operate as expected. But since the TLC5917 is causing other parts in my design to work incorrectly, I will likely have to disconnect the SDO signal from the main SDO bus.

    Michael
  • OK,please let me know if you need further help.

    Regards,

    Shawn.