TMAG5170-Q1: SPI Communication

Arun,

Thanks for reaching out with your concern.  Your SPI write does appear to match the CRC commands you are sending. I also see that your clock polarity appears correct.  TMAG5170 will latch data with a rising clock (data is valid prior to clock edge), and is activated by a low chip select.

 CMD 0 on the EVM is currently programmed to always be '1', but I otherwise attempted to mimic your first command.  You'll notice that MISO is all "0"s up to the CRC bits, and then the correct expected output of 0x9 is transmitted.

So, we see that in your case TMAG5170 is not generating any CRC output, and so we have to assume there is some other issue. On your very first command we should at least receive 0x00000009.  I would start by double checking that there's no issue with the measurement setup.  I would probably attempt swapping scope leads with another channel which is known to be working and verify that there's no issue there. 

It would also be good to ensure that the connection to the PCB is good and that there is no continuity issue between the probe point and the pin of TMAG5170.  Similarly, it would be helpful to verify that the other SPI lines are successfully reaching the device.

It would also be good to ensure that all voltage domains are consistent.  If the clock edges are not sufficient to satisfy the Vih or Vil of the device then it may be possible that the device is unable to properly latch inputs.  Looking at your waveforms, I would guess that this is not the case, but it's never a bad idea to double check.  

Please try these debugging steps out and let us know if you continue to have trouble communicating with the device.

Thanks,

Scott