I can not receive the data from the peripherical slave device(MAX6675)

[kaochin]

Hi, I write a code for the max6657 temperture via modifying the SPI MASTER Example. The function for reading max6675 via SPI is shown as follow

uint16_t readCelsius(void){
uint16_t flag;
uint16_t temp;
uint16_t tempc;
uint32_t spi_rw_err_code;

APP_ERROR_CHECK(nrf_drv_spi_transfer(&spi, NULL, 0, m_rx_buf, m_length));

temp=m_rx_buf[0]<<8;

flag=temp&0x04;
	if (flag){
	  return 8888;}

    temp>>=3;
    tempc=temp*.25;  	
    SEGGER_RTT_printf(0,"TEMP*4=%6.2f\n",tempc);
nrf_gpio_pin_set(SPI_CS_PIN);		
return tempc; }

but i can the the data 0x0000。

In order to makesure that the SPI is working correct, I use the Logic Analyzer to measure the pattern of the SPI sigal which is shown as follow:

I think that the spi is work well，but it read the data 0x0000!

I review the API function nrf_drv_spi_transfer() and spi_xfer() in the file "nrf_drv_psi.c" I find that only the function "nrf_spi_txd_set" which is a function for writing data to the SPI transmitter register is used. And the function " nrf_spi_rxd_get() " for reading data from the SPI receiver register is never been used. I guess that the function "nrf_drv_spi_transfer()" is "send" only, and I can get the data 0x0000 from the temperature sensor. Is my guess correctly?!

If my guess is wrong, how can I read it correctly? Thank for your suggestion。

By the way The sensor is work well when it is connected to the STM32 board.
The signal read via Logic Analyzer is Except the data length per transfer(8bit in STM32, 16bit in nRF51822), the other configureation of SPI in nRF51288 board is the same as tho one in STM32 board.

Thanks for your review and suggestion. Best Wishes

[Ole Bauck]

From the logic analyzer capture it looks like you actually are actually receiving 0x0000, which indicates that the problem is with the other chip or the connection with the other chip. Have you tested that it works with another chip as master, for example an Arduino?

Do you use blocking SPI (no event handler passed in the init function)? I assume you are using blocking SPI since it works with the STM chip, but If you are not using blocking SPI then you need to wait for the NRF_DRV_SPI_EVENT_DONE event before the data is valid.

[kaochin]

The max6675 is work correctly in both of the STM32 and arduino。

I have assigned a even handler function in the initial function just as the SPI MASTER Example:

    void spi_event_handler(nrf_drv_spi_evt_t const * p_event){
spi_xfer_done = true;
NRF_LOG_PRINTF(" Transfer completed.\r\n");}

I think that the blocking SPI is not used in my code. So,Shall I need to wait for the NRF_DRV_SPI_EVENT_DONE event? I will check it in my code! But I think that it is done in the handle function, is'nt it?

Thank for your suggest!

[Ole Bauck]

Yes, you should wait for the event handler to be executed with the NRF_DRV_SPI_EVENT_DONE event. You should also make the rx buffer global or static such that it is placed in RAM rather than on the stack.

But if you still don't see anything on the data lines (but on the clock line) with the logic analyzer, then the problem is with the connection to the chip, not with the code.

[kaochin]

yes, you are right. This problem is caused by the power of the borad. I have solved the problem via providing a external power the sesor. Thank you very much.