TMAG5273: B1 Variant Not Responding

Stuart Baker

Part Number: TMAG5273

I'm working with a customer on the TMAG5273. They have designed their application to accept any one of the 8 variants of this device (4 addresses, 2 sensitivity levels). This is done by scanning all four addresses and then reading the device ID.

The customer just built up a bunch of boards with the B1 variant, but it does not seem to be responding to commands on the appropriate address of 0x22. We can ABA swapped it with both an A1 and A2 variant, which is working fine.

Are there any other differences between the B1 variant and the A1 variant other than the default I2C address that we should be aware of?

The devices were purchased through TI.com and I have the purchasing information for lot tracing if it is needed.

Thanks,

Stuart

over 2 years ago

0 Alicia Rosenberger over 2 years ago

TI__Genius 13710 points

Hi Stuart,

Thank you for posting to the Sensors forum!

Would it be possible to see a waveform of the customer trying to communicate with the B1 variant?

Best,

~Alicia

0 Stuart Baker over 2 years ago in reply to Alicia Rosenberger

TI__Genius 10620 points

Yes, we should be able to provide it. It may take a few days to collect.

Thanks,

Stuart

0 Alicia Rosenberger over 2 years ago in reply to Stuart Baker

TI__Genius 13710 points

Hi Stuart,

Thank you for letting me know.

Best,

~Alicia

0 Stuart Baker over 2 years ago in reply to Alicia Rosenberger

TI__Genius 10620 points

Alicia,

Further troubleshooting with the customer has revealed one important difference between the A1/A2 variants they have and the B1 variation that didn't originally work.

It seems the B1 Variant is reporting back a value of 0x5 for the Device ID register. The A1/A2 variants report back a value of 1 or 2 respectively. The datasheet seems to imply that bits 7-2 of the Device ID register should read back 0. Can you comment as to if it is expected that bits 7-2 would not have a value of 0?

The customer software verifies the device ID and manufacturer ID before it will allow to continue. This is because the TMAG5273 is an optional chip in the system, and they use these read back values to positively identify it/what is present. If we mask off the upper bits of the Device ID, and look only at bits 1-0, then the B1 variant is now working as expected.

Thanks,

Stuart

+1 Alicia Rosenberger over 2 years ago in reply to Stuart Baker

TI__Genius 13710 points

Hi Stuart,

We are currently working to update the datasheet for the TMAG5273 as bits 7-2 should be seen as 'don't-cares'. So the B1 (or any x1) variant reading back a 0x5 for the Device ID register would be expected behavior.

Best,

~Alicia