ADS1261: Where to get the firmware source code of ADS1261EVM?

Hi Yue,

I think we should try to collect more information from the customer before trying to replicate it, otherwise it will be very unlikely that we'll succeed in doing so.

For starters, what is (command) pattern they are sending, and how often are they sending these commands?...it looks like they are performing repeated RREG and WREG commands.

Can you also have the customer look at the SPI communication on a logic analyzer or oscilloscope? It would be most helpful if they could capture an SPI frame with a CRC error. If the error is infrequent enough, they might need to configure their micro to toggle a GPIO pin when the CRC error is detected.

What stands out initially to me is that the "echo byte 2" recieved does not always match the "arbitrary" byte they transmitted, as shown below:

Echo byte 1 appears to be okay most of the time except when DataRx = 0xFF, which would indicate that the device is not responding at all. In this case, it is import to check the /CS, /RESET, and /PWDN pins to ensure that the device is active and selected on the SPI bus. If there are other devices on the SPI bus, then they may need to check if other devices are active and holding DOUT high.

Echo byte 2 seems to be incorrect more often. Since this data is getting printed to a terminal, I cannot tell if the SPI communication is the issue or if the code that prints this data to the terminal is causing the issue. Therefore, it is important to determine if this the data printed to the terminal is correct and if the same mismatch is occurring on the DIN and DOUT pins.

If the data is getting corrupted on DIN/DOUT, then I would pay attention to whether the SCLK/DIN/DOUT signals are clean or if there are large noise spikes that bit be causing transmission errors. Also, check if any other device's /CS pin (sharing same SPI bus) is getting enabling during this time.

Tom,

I on-site supported them, turns out there is a little bug in TI offered reference code. after this modification, so far, no further issue reported. 

/* (OPTIONAL) Write to all registers */

uint8_t startIndex = REG_ADDR_STATUS;

uint8_t count = NUM_REGISTERS - startIndex;

writeMultipleRegisters(startIndex, count, &initRegisterMap[startindex]);

void writeMultipleRegisters(uint8_t addr, uint8_t count, const uint8_t data[])

{

/* Check that register map address range is not exceeded */

assert( (addr + count) <= NUM_REGISTERS );

uint8_t i;

for (i = addr; i < (addr + count); i++)

{

   writeSingleRegister(i+addr, data[i]);

}

}

Changed to :----------------------------------------------------------------------------------------

/* (OPTIONAL) Write to all registers */

uint8_t startIndex = REG_ADDR_STATUS;

uint8_t count = NUM_REGISTERS - startIndex;

writeMultipleRegisters(startIndex, count, initRegisterMap);

/* (OPTIONAL) Read back all registers */

readMultipleRegisters(startIndex, count, ADC_RegisterMap);

void writeMultipleRegisters(uint8_t addr, uint8_t count, const uint8_t data[])

{

/* Check that register map address range is not exceeded */

assert( (addr + count) <= NUM_REGISTERS );

uint8_t i;

for (i = addr; i < (addr + count); i++)

{

   writeSingleRegister(i, data[i]);

}

}