I2C driver problem?

[djqtsg]

I am using PCA10040 evaluation board to test I2C connection.

I tried twi_scanner sample code in SDK 12.

I wired SCL and SDA pins to D0 and D1 to a PicoScope device.

When the code was running, I captured signals from those two pins.

The clock and data lines are completely wrong, right?

1. The clock cycles are far below expected: at least 8 for for one write, isn't it?
2. Data line should start before clock line; but in my capture, they started to fall at almost the same time.

Can anyone explain why?

I also did another test with the following code:

int main(void) { nrf_gpio_cfg_output(ARDUINO_SCL_PIN); nrf_gpio_cfg_output(ARDUINO_SDA_PIN);

  for (address = 1; address <= TWI_ADDRESSES; address++)
  {
	nrf_gpio_pin_set(ARDUINO_SDA_PIN);
	nrf_delay_ms(200);
	nrf_gpio_pin_set(ARDUINO_SCL_PIN);
	nrf_delay_ms(200);
	nrf_gpio_pin_clear(ARDUINO_SDA_PIN);
	nrf_delay_ms(200);
	nrf_gpio_pin_clear(ARDUINO_SCL_PIN);
	nrf_delay_ms(1000);
 }

}

The captured two pins signals are as good as expected, which means my way of the signal monitoring is correct, unless for I2C wiring there are special settings I missed.

[djqtsg]

Hi,

forget this question.

The strange signal most likely is caused by PicoScope or my not properly using it; because even when I2C was working, I also captured funny clock and data signal.