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.

PCA9306: Questions about hot plug design

Part Number: PCA9306


dear teams,

System scheme diagram is shown in the figure above.

9306 plays the role of bridge between the main board and child module. The child module can be inserted and unplugged.

Our  customer's design is that after the child module is inserted and stabilized, the child module is powered on by the control board. And during the whole process they keeps EN pin high.

So the question is that does a design like this have an bad effect on hot-plugging?

 

Regards,

Lijia

 

 

  • Hey Lija,

    "So the question is that does a design like this have an bad effect on hot-plugging?"
    Because enable is held high the entire time, the 1.8V I2C bus will see any glitches that the hot plug event causes. (The child module has parasitic capacitance which will temporarily drive the I2C bus low during the insertion event.)

    One question is, what else is on the I2C bus? Is it only the FPGA? If so, I don't think a hot plug event will cause the FPGA a problem. If there are other slaves on the bus, then we may be concerned.

    At the moment if you just leave the PCA9306 enabled then it will only provide level translation, it will not help with a hot plug event.

    Do you have any control of the child module? We do have a device which could help with this.

    Thanks,
    -Bobby
  • Hi, Bobby

    Here is another question, could you please evaluate whether such an approach will do harm to the chip itself.

    In other words, will hot-plug damage the 9306 ?

    There is no need to worry about possible interference signals in the process of plugging. Customers will filter it out by software.

    Thanks,
    Lijia
  • Hello Lijia,

    "In other words, will hot-plug damage the 9306 ?"

    The short answer here is I do not think it would (unlikely) but if it did there are discrete solutions to resolve this.

    Typically hot plug events will drive a signal low to ground for a short duration of time. If there is a lot of inductance between the PCA9306 and the daughter card then I supposed the hot plug event could damage the device is the swing to ground cause a large undershoot to violate the absolute minimum of the device. It may be a good idea to place a clamping diode on the SDA/SCL lines to provide this. Voltage suppression diodes could also be useful if we think the SDA/SCL rails get pulled up above the allowed voltage range during the hot plug event. (I don't think it would but I am not certain).

    -Bobby