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.

TMS320F28375S: GPIO pin inrush current handling capability.

Part Number: TMS320F28375S

Hi Ti E2E.

I will describe my problem as simple as possible.

I need to use a GPIO pin to drive a bus that is capacitive. The bus is capacitive because I need to put in a capacitor for noise mitigation purpose. The capacitance is somewhere from 100pF to 10nF.

When the bus is driven low, there will be an inrush current to discharge the bus, which may be harmful to the GPIO pin.

My question is: Is there any specification for inrush current handling capability of TMS320F28375S's GPIO pins? Such as maximum inrush current, maximum duration, discharge rating...

Thank you for reading.

  • 6576943,

    Please see the following E2E post. 

    https://e2e.ti.com/support/microcontrollers/c2000/f/c2000-microcontrollers-forum/678911/tms320f28377d-tms320f28377d-gpio-inrush-current-issue 

    Please also note that there is a maximum total current that the device an sync as well. This can be found in the Absolute maximum ratings table of the Datasheet.

    Regards,
    Cody 

  • Hi Cody,

    Thank you for your answer.

    According to the absolute maximum rating, I think that it would be necessary to put a series resistor to limit the inrush current to within 20mA.

    However, this lead to another question: Since any digital bus has a parasitic capacitance, does that mean we need to put a series resistor to any digital bus?

    I have thought about this for a while. People don't usually do this. So, I think there must be a limit to the amount of energy that the GPIO can dissipate in a short duration, without having any risk of pin malfunction. 

    I would like to have your opinion on this question. I really want to know if there are some related ratings.

    Thank you very much.

    Regards.

  • 6576943,

    That is actually a good thought. 

    My first thought was "these currents should be very small since the parasitic cap is so small". After doing a quick calculation the current does seem to be higher than expected.

    Give me a chance to check my numbers and look at this bit closer, will respond back soon.

    Regards,
    Cody 

  • I am waiting on confirmation from a colleague of mine.

    Regards,
    Cody 

  • Thuc,

    I have received confirmation that parasitic capacitance is understood and expected during design. Small capacitances like these are OK due to their short duration. These values have an RMS current value well below 4mA even when switching at well over 100MHz.

    If you have larger explicit cap you will still need to worry about sustained charging currents and limit these in some way.

    Regards,
    Cody 

  • Hi Cody,

    Thank you for your answer.

    I also got similar answer from other forum.

    However, is there any way to do some calculation for problems like this?

    I want to understand the mathematics of it. It seems normal calculation does not give a proper answer.

    Regards.

  • Thuc,

    I started with the maximum allowed current and a few assumed conditions for parasitic loading. I used the assumed parasitic load and the known GPIO buffer resistance from the datasheet to calculate the charging and discharging current due to switching high or low.

    I then found the smallest switching period for which the RMS value of this current was below the maximum allowed current. This is the theoretical maximum frequency you can switch a GPIO with a parasitic load without risk of damaging the GPIO buffer. I found that this value is well above the possible switching frequency of the device, which confirms that some GPIO designer(which knows a lot more than I) correctly designed the circuit to be safe for all parasitic loads.(kudos to the designer)

    The formulas for calculating RC current, an RMS value, and realistic parasitic loads of PCBs are all readily available on the web. Best of luck.

    Regards,
    Cody 

  • Dear Cody,

    I think your method is a good way to solve this problem. I totally understand the underlying mathematics of your method.

    Thank you very much for your time and knowledge.

    Regards,

    Nguyen Duy Thuc.

  • Good to hear. Feel free to start another thread if you have more questions.

    Regards,
    Cody