GPIO P0.19 not working on nRF52840 Custom board

[KenT3]

We just received our latest custom nRF52840 board and have found that there seems to be some issue with PO.19.   We have verified that the P0.19 is not working on multiple boards so I do not think that it is a manufacturing defect.

All of the other GPIOs that we are using work correctly.  Everything else on the board functions correctly.

We are not linking with nrfx_qspi.c and do not have NRFX_QSPI_ENABLED, or QSPI_ENABLED defined in our sdk_config.h.

P0.19 is connected to a pin on a header and has nothing else connected to it.  The Module we are using is a Minew MS88SF3.

P0.19 does not function as an input or on output.

I have also verifed that if I force the pin high or low it does not show up on any of the other port bits so it is not a simple case to the wires being swaped on the module.

We reviewed the Gerber and board files and they seem to be correct.

I looked at system_nrf52840.c and noticed there was a QSPI errata but the code for that errata is never being called since FICR x10000134 contains a 3.  So it has nothing to do with that.

What am I missing?

Thanks for your help,

Ken

[Mehmet Nuri]

Check config.h file. A peripheral might be using it. For example GFX driver sets pin 19 as DC pin..

[KenT3]

Thanks for the quick response:

I am not sure what you mean by GFX driver.  The only Nordic driver that uses P0.19 seems to be the QSPI and I am not using that.  None of the drivers configured in config.h use P0.19, and none for the drivers configured in code use P0.19.  We are not using any drivers.

We are using 15.3 and S140.

I am doing my port tests before any of the drivers are initialized.  After SystemInit() at the beginning of main().  I think this should rule out any configuration issues in config.h.

With everything disconnected and powered down, if I do a continuity test between P0.19 and and it measures open.  All of the other GPIOs measure 3.5-4.5 M Ohms.

Module manufacture says P0.19 is connected.  Board looks like it is correct.  Port does not work.

[Jørgen Holmefjord]

Hi,

Are you able to control P0.19 with the same firmware on a nRF52840 DK (given that you have disconnected the QSPI flash chip and connected it to the pin headers)?

If it does not work on the DK, then there is most likely a firmware issue, but if it works on DK it is most likely a HW issue.

Best regards,
Jørgen

[KenT3]

I do not have a nRF52840 DK.  All of the other nRF52840 boards that I have I do not have access to P0.19.  I can get one but It will take a few days.

When you Disconnect the QSPI flash on your DK and do a continuity test to Gnd what do you measure?

Thanks

[Jørgen Holmefjord]

I measure OL between P0.19 and GND on the DK. I also verified with a logic analyzer that the pin is toggling correctly with this simple program:

#include "nrf_delay.h"
#include "nrf_gpio.h"

#define gpio19 NRF_GPIO_PIN_MAP(0,19)               // LED0 P0.19

int main(void)
{
    nrf_gpio_cfg_output(gpio19); // config the as output pin
    while(1)
    {
          nrf_gpio_pin_toggle(gpio19);
          nrf_delay_ms(1000);
    }
}