ADS8168: Unnexpected SPI communication results

Igor Langer

Part Number: ADS8168

Tool/software:

Hello there,

I have a ADS8168 in a custom board, and the device is not responding as expected. I'm trying to write 0x02 to the PD_CNTL register (0x04), and then reading it back again to check whether the write was successful.

What I'm doing? (First image)

- enabling write access, by writing 0xAA to register 0x00 with the read command 0x08;

- writing the 0x02 to the 0x04 register;

- reading the 0x04 register with two 24-bit frames, it returns 0x38

Something weird I'm seeing, for some reason the data bits are launched by the ADS8168 on the rising edge of SCK, this edge is where the bit should be sampled. Even the scope has trouble trying to decide whether a bit is 1 or 0. (second image). According to the datasheet, by default the device is in SPI mode 0, which means the CPOL=0 and CPHA=0.

Important info:
- The device was manually soldered, and was heated twice.

- the IC is reseted before init.

1 month ago

0 Samiha Sharif 1 month ago

TI__Expert 8625 points

Hi Igor,

Thanks for your post. On page 78 of the datasheet, the peak soldering temperature is listed as 260C. Exceeding this may cause damage to the device. Was this temperature exceeded?

Assuming the peak soldering temperature was not exceeded, could you try to read back the SDI_CNTL and SDO_CNTL1 registers? When in SPI-00 mode, the SDO launch edge (for the previous frame's data) is CS falling edge, followed by subsequent SCLK rising edges launching the rest of the data. The SDI data of the current frame is latched on the SCLK rising edge as well.