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.

PCA9535: Driven LEDs at Output Bank flashed at Startup

Part Number: PCA9535

Hi,

I'm using this I/O-Expander to drive some LEDs on a measuring device. The LEDs are set up in a row to show the measured value on a scale. It works perfectly using a uC to send the data over the I2C bus.

The Issue:

When restarting the device some LEDs sometimes randomely flahed at the very beginning of the bootup. The PCA9535 is getting a 3.3V (Vcc) Power Support immediately after the device is turned on.

Unfortunately, I can't handle that effect in the firmware bc it happend before the firmware even start to initialize the I/O-Expander.

Below is the circuit that I'm using to control the LEDs:

I hope you can help me solving this issue by eliminating the described effect on bootup! Thanks a lot!

 

  • Hey Cam,

    A suggestion from me: for future designs you should look to use our TCA family rather than PCA as they fix some issues our PCA line up has had and in some devices better input voltage ranges. TCA lineup also is USUALLY more cost competitive. (TCA9535 is my recommendation)

    The I/O pins of this device are high impedance after start up however from what you are describing it may be possible the NFET (Q2 in figure 17 of datasheet) is being partially turned on and conducting current during the power up phase (the FET will look like a short to GND for a small period of time). If this is the issue, it's not something we can easily fix (maybe turn on device first then turn on Vcc to LEDs via a switch/FET).....

    Another possibility it could is the node on the "Pn" pin is actually not seeing the exact same voltage as the Vcc line because of the 100k resistor. Essentially there is a time delay between voltage at Pn and Vcc where the voltages are different. Think of parasitic capacitance on the I/O in the degree of 10pF forming an RC circuit with the 100k resistor. The voltage at Pn will be time dependent on the parasitic capacitance and the pull up resistor:
    Vcap=Vcc(1-e^-t/RC)

    If Vcc-Vcap is larger than Vdiode, then the diode will temporarily conduct. So the solution here would be to lower the pull up resistor value and see if this corrects the issue. I would try something along the lines of 10k or 5k ohms. (lower resistance will draw more power when you turn on the diode during normal operation).

    Thanks,
    -Bobby