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.

ADS1018: ADS1018 SPI communication issue

Part Number: ADS1018
Other Parts Discussed in Thread: ADS1220

Tool/software:

Hello,

I have a communication SPI issue with ADS1018. The symptom is the DOUT data is not the same with DIN data on CONFIG MSB& CONFIG LSB when using 32-bit data transmission cycle. Here is DIN data: 0x48,0x6A,0x48,0x6A, but DOUT data: 0x22,0x92,0x29,0x02

Hardware setup:

I am using our controller board to interface with ADS1x18EVM board. We have 5v signal level available on Spi connector, where ADS1x18EVM is connected. 

So I did below modifications on ADS1x18EVM board.

1. Removed the SPI level shifter (U2) and short below pins

    • Short U2.2  with U2.13
    • Short U2.3  with U2.12
    • Short U2.4  with U2.11
    • Short U2.5  with U2.10

 I have also attached the scope waveform for reference. 

SPI configuration : SPI clock 500Khz, SpiDataShiftEdge = Leading, SpiShiftClockIdleLevel = LOW

Re->Chip select,  Yellow->Clock;  Blue-> DIN;  Green ->DOUT 

Could you pls let me know what went wrong?

Regards

Shyam 

  • Hi Shyam,

    I'll take a closer look and get back to you by tomorrow.

    Best Regards,

    Angel

  • Hi Angel,

    I will wait for your input !!

    Thanks

    Shyam

  • Hi Shyam,

    Are you cycling through the MUX to set different channels while using continuous conversion mode?

    When changing the configuration register during an ongoing conversion in continuous-conversion modes, the new configuration settings do not take effect immediately. Unlike many other delta-sigma ADC devices, writing a new configuration does not reset the digital filter and restart conversions.

    This will require that the next conversion after the mux change needs to be thrown away so that the conversion result is valid for the desired mux input.  This differs from the ADS1220, for example, where the conversion will automatically restart when writing a new configuration to the registers.

    In single-shot mode the mux change and read of conversion can take place at the same time.  Once it is determined that the conversion has completed, the new configuration can be written and the conversion started at the same time as reading the previous conversion result as the SPI communication is full-duplex.  For this use case, there is no need to throw out an invalid conversion and the only delay time is the length of time from DOUT/DRDY going from high to low until the conversion data are read while starting the next conversion.

    The following diagrams help explain this behavior:

    Device behavior in continuous-conversion mode when writing new configuration register settings:

    Device behavior in single-shot conversion mode when writing new configuration register settings

    Also, have you tried probing the digital communications lines using our controller board (PAMB) that the EVM comes with and our online GUI?

    There seems to be some additional strange behavior, such as data transitioning on the falling edges of SCLK, when data should only be transitioning on the rising edges of SCLK, and clocked in on the falling edges of SCLK since the device operated in SPI mode 1 (CPOL = 0 and CPHA = 1):

    There is also this lingering discharge of DOUT (green) at the end of the frame. 

    Could there be some unwanted capacitance in the digital communication lines since the EVM has been altered that causes this behavior?

    I would suggest ensuring all the connections are ok according to our data sheet since we don't typically recommend altering our EVM boards.

    Best Regards,

    Angel

  • Hi Angel,

    Thanks for your reply,

    Answer to your questions,

    I was sending  above configuration periodically rate of 1 ms.

    I could not tried with TI PAMB board with with GUI as getting below errors.

    As you have mentioned data transition There seems to be some additional strange behavior, such as data transitioning on the falling edges of SCLK, when data should only be transitioning on the rising edges of SCLK, and clocked in on the falling edges of SCLK since the device operated in SPI mode 1 (CPOL = 0 and CPHA = 1):

    I have checked with different modes. (All possible modes available in SPI) Please see the results below.

    Also I have change the configuration from continuous to Power down and one shot mode.

    so DIN now 0x496A 496A (total 32bits config data sending periodically 1 ms)

    Red->Chip select,  Yellow->Clock;  Blue-> DIN;  Green ->DOUT 

    1. SpiDataShiftEdge = Leading, SpiShiftClockIdleLevel = LOW

    2. SpiDataShiftEdge = Leading, SpiShiftClockIdleLevel = HIGH

    3. SpiDataShiftEdge = Trailing , SpiShiftClockIdleLevel = LOW

    4. SpiDataShiftEdge = Trailing , SpiShiftClockIdleLevel = HIGH

    In all above cases didn't get the desired data ( read back Config data) from ADS1018. 

    Just FYI, Along with ADS1018 there are two more ASCIS are connected (with separate CS) on the same SPI channel and working fine. 

    Pleased let me know if your input on the above results. 

    Waiting for your replay.

    Shyam 

  • Hi 

    I could not tried with TI PAMB board with with GUI as getting below errors.

    This is an issue when installing the drivers. The following threads give a step-by-step on how to fix this issue:

    ADS1115EVM-PDK: Connection failure to the GUI - Data converters forum - Data converters - TI E2E support forums

    [FAQ] How do I resolve PAMB EVM connectivity and/or driver installation issues - Data converters forum - Data converters - TI E2E support forums

    SPI mode 1 communication to ADS1018 example:

    (from meeting: additional testing will be done to verify communication to the ADC)

    Best Regards,

    Angel