This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TMP107: Getting different device ID with the TMP107B

Part Number: TMP107

Hi Team,

I would like to get your inputs on the customer situation below with their 1000pcs of TMP107B.

The customer has a sensor that uses TMP107B on it. The PCB shop delivered 1000pcs to them and none of them are working

When they checked the device ID on the chip, they are getting 0x9107 when they believe it should be 0x1107

At least the first 80pcs are the same situation.
Their old sensors report address 0x1107 when they read them.

One other thing they noticed was the sensors seem to have a large degree of difference in temperature readings
the old sensors are very accurate whereas these new ones appear to be +/- 5 degrees different from each other.

Please let me know if you have any questions for the customer.

Thanks,

Jonathan

  • Jonathan - 

    Do you have good vs bad UART captures to share?

    With regards to the temp offset, this device is supposed to be +/-0.4C accurate - (we don't even make a temp sensor that is +/-5C) - so something must be very wrong there - did they inspect for any damage or bad soldering, etc. yet? If you can get any images of the board area where the IC is, that might help, too.

  • Hi Josh,

    Upon discussing your answer with our customer. Here are more details they have provided:

    In a deeper dive into the problem, they discovered that about 167 out of 1000pcs are okay. The 833 parts that appear to be defective are the chip itself. They removed several bad chips and placed them on brand new boards that were working (removed working chips) and the same issue moves with the chip itself.

    They also noticed that in the data it appears to be the 9th bit is always a '1' in the data. So this directly affects the temperature on the device when read, the ID of the device, etc...

    Here are some photos of the boards and the logs. They have been able to isolate the issue to the chip but would like to understand how this could happen. ESD? Over heated? anything like that.



    Other information:
    - we are reading at a baud rate of 9600
    - cable length is less than 12 inches for testing
    - PCBs are very clean and where assembled by pick and place with stencil so soldering is well controlled

    Please let me know if this can be better to discuss offline.

    Regards,

    Jonathan

  • Jonathan - 

    thanks for the extra info - the soldering pins on the right side of the image look a little odd, and i would have expected the top side marking to be more visible - at any rate, the board looks correct as you said - if you could get a better image, that may be helpful - but I think we may need to involve CQE since there are so many - I'll send you an email on this. - do you by chance have or can you get the date codes and/or topside marking on these? 

  • moved to email - closing the post