nRF52 failed to read from max30102 through I2C

RE: nRF52 failed to read from max30102 through I2C

Petter Myhre — Mon, 23 Jan 2017 13:54:16 GMT

Yes, you should wait for m_xfer_done to be true before you start a new transfer. I wouldn't use a while, I would check the flag in the main loop and start the next transfer from there. Or maybe you can just use the driver in blocking mode?

RE: nRF52 failed to read from max30102 through I2C

djqtsg — Mon, 16 Jan 2017 00:07:06 GMT

Now I know where is the problem is:

After calling nrf_drv_twi_tx(), I should wait for twi_handler() being called and m_xfer_done changed to true.

err_code = nrf_drv_twi_tx(&twi_instance, device_address, &reg_addr, 1, true);

while(!m_xfer_done);

err_code = nrf_drv_twi_rx(&twi_instance, device_address, part_id, 1);

Now my question would be:

Is it safe to add while(!m_xfer_done)? What if in twi_handler() events other than TWI_EVT_DONE happens, and m_xfer_done is not changed to true, that while statement loops forever, isn't it?

How about I use nrf_delay_us(10) instead of while(!m_xfer_done);

RE: nRF52 failed to read from max30102 through I2C

Petter Myhre — Fri, 13 Jan 2017 10:03:10 GMT

Is nrf_drv_twi_rx() always returning 0x11? This means NRF_ERROR_BUSY - If the driver is not ready for a new transfer.