• State
  • +1 person also asked this people also asked this
  • Locked Locked
  • Replies 9 replies
  • Subscribers 32 subscribers
  • Views 479 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Original question:

PGA450-Q1: reducing measurements blind spot limit

PGA450-Q1: low temperatures causes no response from sensor

Eyal Ron
Intellectual 295 points
Part Number: PGA450-Q1

Hi,

we've encounter a wired issue using PGA450 in our device.

in some cases, when our device is working at a location with low temperatures (less then 10[C] but more then 0[C]), the PGA450 stops responding to UART requests.

for testing purposes, I've placed our unit inside a refrigerator with ~4[C]. 

after a few minutes I could not get a response from the PGA450 (not even for sensor address set in EEPROM Addr 0x31).

after getting the device out to room temperature, the PGA450 started responding again.

normally, this would suspect peripherals but I found that adding a delay of ~50[ms] between writing bytes of the command, reduces the problem significantly.

is there any specific delay I should use between writing bytes? other then baud rate and other regular UART definitions, I cant find any specifications in the data sheet.

thanks!

  • 0
    TI__Expert 7505 points

    Hi Eyal,

    Can you tell me the following:

    • What clock source are you using for the PGA450?
    • Do you observe this issue with multiple units or just one unit?
    • Have you tried this test on our EVM to see if you observed the same issue?

    Thank you,

    Mekre

  • 0 in reply to Mekre Mesganaw
    Prodigy 70 points

    Hi Mekre,

    thanks for your replay!

    We are using Internal oscillator without synchronization with communication (from the code --> CLK_SEL = 0x00;)

    According to our analysis, this issue reproduced on about 70 units already.

    I'm not sure if the EVM is relevant as I tried a few other units and it can't be reproduced on all of them.

    Thanks,

    Guy

  • 0 in reply to GUY WAISMAN
    TI__Expert 7505 points

    Hi Guy,

    Are each of these 70 units using a different microcontroller?  If so, can you take a microcontroller from one of the working units and connect it to the PGA450 on one of the units that are not working?  

    Can you also take a scope capture of the UART TX and RX signals?

    If possible, it might also be a good idea to try using an external crystal on one of the failing units to see if it makes a difference.  

    Regards,

    Mekre

  • 0 in reply to Mekre Mesganaw
    Prodigy 70 points

    Hi Mekre,

    I have no access as the units are installed at our customers sites. we received 3 for debugging.
    each PGA unit is connected to a micro controller, 
    in addition, we cannot connect the PGA to an external crystal as we do not use one on our HW board.

    as requested, please see attached scope capture video files.
    one each cycle we:

    1. send a request to perform a measurement.

    2. read the FIFO buffer to analyze the distance on our own,

    as you can see in the video, in some cases we loop the measurement request till we get a response before asking for the fifo buffer 

    P.S - I also learned that in some cases when I do get a response - I can see very long saturation phase and this also happens only when the temperature is very low. any idea why this happens? 

    thanks,

    guy

  • 0 in reply to GUY WAISMAN
    TI__Expert 7505 points

    Hi Guy,

    If possible, could you replace the microcontroller and any utilized microcontroller crystals on one of the faulty boards?  Could you replace these parts with components from a working board?  I want to see if the MCU clock frequency affects this issue.

    I can’t see the data in your videos.  Do you have a logic analyzer screenshot of the data that shows the actual byte values observed?

    Also, can you elaborate what you mean by the output being saturated?  Are you saying that the output stays high instead of changing values as expected?

    Thank you,

    Mekre

  • 0 in reply to Mekre Mesganaw
    Prodigy 70 points

    Hi Mekre,

    I checked and verified that this issue can be reproduced even when I replace our microcontroller but I don't have any logic analyzer to prove what is the actual bytes value.

    regarding the output: Yes. in some cases, I see a response from the PGA450 but the fifo buffer contains high values only and do not change over time.

    thanks,

    guy

  • 0 in reply to GUY WAISMAN
    Prodigy 70 points

    adding an excel with example of the output I get after reading the fifo buffer BinId - 3538 _(1)_LogFile_30.01.22_15.xlsx

  • 0 in reply to GUY WAISMAN
    TI__Expert 7505 points

    Hi Guy,

    Thank you for this info.  I need to check internally with our team on a few things in regards to this.  I should provide my next response tomorrow.

    Regards,

    Mekre

  • 0 in reply to Mekre Mesganaw
    TI__Expert 7505 points

    Hi Guy,

    Are you able to test if the SPI is functional when you are having communication issues?  Also, can you try reducing the baud rate to see if it helps.

    Thank you,

    Mekre