• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 47 subscribers
  • Views 251 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

ADS8686SEVM-PDK: Strange behaviour on 1-wire SPI when DVDD is at 1.8 V (systemic IC design fault)

Rune Warhuus
Prodigy 60 points
Part Number: ADS8686SEVM-PDK
Other Parts Discussed in Thread: ADS8686S

I'm experiencing strange behaviour when the ADS8686 is run at 1.8 V DVDD (within spec).

Some register values are read wrong (repeatable, not random) and the ADC readings are very unstable and erroneous (brownout).

A basic step by step to validate the unexpected behaviour:

  1. Set up the ADS8686EVM-PDK for 1-wire SPI
  2. Feed DVDD from a 1.8 V source
  3. Set register 0x04 - 0x07 to 0b10101010 (±5 V)
  4. Read back registers 0x04 - 0x07

The result will be:

0b11110000
0b11110000
0b11110000
0b11110000

The expected result should be:

0b10101010
0b10101010
0b10101010
0b10101010

Also observe the erroneous channel readings.

Feed DVDD with 3.3 - 5 V and observe correct behaviour.

  • 0
    TI__Mastermind 18900 points

    Hi Rune,

    I suggest taking a look at these FAQs first:

    In summary, we'll need an oscilloscope (not just a logic analyzer) and verify what is sent to the device is what you're expecting. I'd start with a read, see if you can read device ID and capture the data for that. Then we can do a write and see how the waveform looks as well.

    Speaking from experience, most of the time, these kind of issue are the result of bus contention. Essentially, the logic rail at the IC is set to some voltage (e.g. 1.8V) and the MCU or some other IC on the SPI line is expecting a different voltage (3.3V). The result is waveforms that don't look like square waves but instead look more like a square wave stacked on top of a square wave.

    As for "systematic IC design fault" we validated the device over process and temperature with DVDD at 1.71V, we test the device at 1.8V DVDD before it leaving TI, and I haven't heard of any of our other customer talk about this so I doubt its a systematic issue. But we'll keep that in our back pocket. 

    Best,

    -Cole

  • 0
    TI__Mastermind 18900 points

    Hey Rune,

    I was talking with the team a bit, I made the assumption that you are using your own microcontroller and sending commands to the EVM without the PHI.

    If you are using the PHI, it doesn't work down to 1.8V.

    You'll need to make sure you remove R28 and supply DVDD externally.

    Best,

    -Cole

  • 0 in reply to Cole Macias
    Prodigy 60 points

    Your assumption is right, I'm not using the PHI. There are conflicting specifications on the minimum DVDD voltage of the ADS8686. The datasheet for the IC states that it will work down to 1.8 V, but a implementation note for the evaluation board states the lower limit for DVDD to 2.3 V.

    Why would removing a 0 ohm shunt fix the issue?

    Have you tested the EVM to work down to 1.8 V like I describe above?

  • 0 in reply to Rune Warhuus
    TI__Mastermind 18900 points

    Hi Rune,

    Rune Warhuus said:
    There are conflicting specifications on the minimum DVDD voltage of the ADS8686. The datasheet for the IC states that it will work down to 1.8 V, but a implementation note for the evaluation board states the lower limit for DVDD to 2.3 V.

    The datasheet is the actual specification. As I mentioned earlier, the PHI doesn't work at 1.8V (or less than 2.3V) so I believe this is why the note exists in the user guide. To be more specific, if the DVDD pin = 1.8V, then the PHI will continue to work at 3.3V (on the DIN pin) and the device absolute maximum for digital inputs will be violated. I agree it causes a lot of confusion, especially because the note talks about the absolute maximum specification in the datasheet in the next sentence, so its possible. 

    Rune Warhuus said:
    Why would removing a 0 ohm shunt fix the issue?

    It won't in your case. The 0ohm connects ADS8686S DVDD pin to the PHI, so if you disconnect it, you can use an external supply instead of conflicting with the PHI. Since you're not using the PHI, its not helpful.

    Also helps with measuring DVDD supply current.

    Rune Warhuus said:
    Have you tested the EVM to work down to 1.8 V like I describe above?

    No I have not. I only have the PHI to communicate with the device and I will damage the device if I set DVDD to 1.8V separately.

    Looking forward to your oscilloscope and logic analyzer pictures.

    Best,

    -Cole