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.

Issue with excess current shunted into TCA6408A

Other Parts Discussed in Thread: TCA6408A

I'm having some issues with the TCA6408A regarding uncontrolled current consumption during startup. Please see schematic for connections. On startup we sometimes have a current to the part above 100 mA. This lasts until the microprocessor starts communicating with the part and then the current drops to the expected level. If we force the reset low the part draws the huge current continuely which i find very puzzling. The part is supplyied with +5VDC on both Vcc's. and the output ports are coupled to high impedant loads.


Hope you can help me because the startup current drawn is unacceptable in our product.

  • Moving to I2C forum for support.
  • Hi Kenneth,
    Are you seeing this issue on multiple boards?
    Are any of the IO ports configured as inputs? What are the pull-up resistance on those IO ports? What is the SCL, SDA pull-up resistors?

    Rajan
  • We are seeing the issue on maybe 1/10 parts. The I/O ports are both configured as inputs on startup according to the datasheet. Whether or not the IO ports are input or output shouldn't matter because i have no external pull up attached to them and they are terminated in huge impedances.


    The pull up on SDA, SCL is ~5 kOhm which so the current drawn from these can never get to the level of 100 mA. I have also tried to remove both data lines from a part and it still had the high current consumption.

    We have resently bought new parts from Farnell and tried to mount one of the PCB's where a huge current consumption was measured, and we got the same huge current drawn in the new part.

    Is it possible that the TCA6408A needs to be controlled during startup from a microprocessor(interrupt, reset) ?

  • Ken,
    Are the IO's connected to LED's? How much current are IO's sinking? Do you know what voltages are the IO ports at?
    If the IO's the left floating it could cause current increase as the IO's could be at any floating voltage and the open-drain ports start conducting current.

    I recommend connecting the IO's configured as inputs to pull-up resistors.

    Thanks,
    Rajan
  • No the ports that are used are all connected to high inpedant digital inputs on analog switches. No current in the mA range is able to go that way. There is one floating port but all the rest are connected.

    I tried to pull all the inputs up to supply, and this seemed to remove the startup current consumption. We would need 8 pull-up resistors for this solution but there is simply not enough room on the PCB for it.

    I still don't understand why the part acts as it does when all ports are input at startup. According to the datasheet input means that the two FET's are off. So how does the part consume all that current?

  • Ken,

    I did not see the schematic in the earlier post. Can you please post it again?

    All the IO configured as inputs need to be pulled up. This is because the IO expander ports have input buffers. So in case the input is not pulled up or down and it is floating at an intermediate voltage, there could be shoot-through current through the inverter.

    This explains why when you pull all the inputs up to supply it removes the startup current.

    So please use pull-up resistors on all IOs configured as inputs.

    The datasheet diagram does not include all the intricate details of the design.

    Please let me know if you have further questions.

    Thanks,

    Rajan