I have recently got 2 product returns from a few hundred PSoC-containing devices which were sent into the field (industrial devices). In one case, an LED connected to the PSoC wasn't working, and in the other, two cap sense buttons connected to the PSoC were not working.
I've checked all the connections and determined that the microcontroller is at fault. Then I discovered that all these pins happened to be from the same port, port 12, which is SIO only. This got me thinking, because SIO supports a larger voltage range than normal GPIO, do they not have the same ESD diodes clamping the voltage is the positive direction? Thus giving the microcontroller port 12 less ESD immunity and perhaps why these pins are not working anymore.
Note that in both cases, other pins on port 12 were still working, so only the pins had fried, not the entire port.
SIO pins have a similar ESD protection as any other pin.
Since you are using CapSense you are bringing out the traces of the port to the outside and that can be the same for the LED. Here you may have experienced a strong static discharge that destroyed the pin's electronic.
I would suggest to view at your board-design to check if a (very) strong discharge could reach the pin's traces.
I would conclude from the below that SIO circuit structure has some
differences. You might post a "CASE" and let the forum know the
“Create a MyCase”
If you suspect specific pins/external interface consider reving board
and adding protection arrays to the pins. ON Semi, Fairchild Semi,
Vishay have smt arrays of zeners/diodes for that purpose.
Bob, how do SIO pins have the same ESD protection if they support higher voltages than the maximum voltage of the microcontroller?
Normally ESD protection is implemented with a diode from the pin input to the voltage rail (one is formed by the inherent body-diode in the top totem-pole MOSFET also).
These SIO pins cannot have a diode here, since it would conduct and clamp the voltage on the SIO pin to the supply voltage + 0.7V.