TM4C123GH6PM: GPIO ports not reading input correctly

Hi,

  Sorry, I'm currently traveling. I will reply you next Monday. 

Hi,

  Please refer to this FAQ. https://e2e.ti.com/support/microcontrollers/arm-based-microcontrollers-group/arm-based-microcontrollers/f/arm-based-microcontrollers-forum/1020817/faq-ek-tm4c123gxl-how-to-resolve-unexpected-behavior-on-pins-pd0-pd1-pb6-and-or-pb7/3772683#3772683

 PD0 is connected to PB6 for boosterpack compatibility reason. If you want to use PD0 for other functionality please remove the R9 0ohm resistor. Same for PD1. 

This is not the problem as I already have taken R9 off. The resistance went up by .1 ohms but it is still too low. Any other suggestions?

Hi,

  Can you repeat the problem after a power cycle? Please also refer to the below errata and if you increase the rise and fall time on the input to the GPIO will make a difference. 

The problem does happen after power cycle. Some more research suggests that an ESD could be damaging the pins (including debugging pins) disabling programming in the future. Some pins are stuck high/low (possibly all of them) do you think that this is the case?

Hi,

  Do you know if your students subject the device to out of spec operating condition that led to the damage? If some pins are stuck high/low then I agree it may be damaged. You can do some resistance check on the pins to confirm. I think you already did some in your opening thread where you said the resistance is only about 130ohm. Please do not subject the MCU to out of spec operating condition. To see 7 boards in a row with the same failure is a strong indication that it is operating out of spec.

Probably, we're looking to get better protections against esd since we moved labs & now there is carpet. This in conjunction with a higher rate of boards becoming damaged seems to be the case. I believe the protections are coming in in a couple of weeks. I can let you know status then. 

As for the resistance check, what should the resistance be approximately with a good board? I have checked others & I see around 47k but others I see 1M or even higher that it overloads the multimeter. Clearly the 130ohm is too low but other pins also seems to have a low resistance, (PD1 with 2k), but was still resistive enough for the application so it doesn't seem broken. By the way, I am comparing resistances with the usb shield, but a coworker pointed out that there may be an analog ground that could be referenced. Is this a case I should worry about when checking resistances? 

While the datasheet does not specify input impedance on the I/Os, I expect them to be very high in Mohm range. 

Okay that's what a board board seems to produce. Thanks for the info.

While we're on the subject, there is no way for a GPIO pin to be similarly "stuck" in a high or low state through modifying any of the gpio registers correct? We initially thought that the students were causing the issue 2 years ago by inadvertently setting the wrong bits when debugging their code. I do know that the debugger JTAG pins are on port c but we only use ports a (uart) b, d & f for this specific class. Now that it's totally possible for a student to accidently set port c instead of b but the flash programmer seems to work fine for these cases (based on the last 2 years; i still need to actually test this myself) 

If pins can get stuck through a program, that may also be a possibility. Thanks for the help so far appreciate it!