• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 328 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.

ADS8661: Discrepancy between the theoretical value and the measured value.

Conor
Guru 11355 points
Part Number: ADS8661
Other Parts Discussed in Thread: ADS8665

Hi,

In "Device Transfer Function", the FSR/2 code is 0x800. From this table, the expected value code at 5.12V is 0x800. 

Also, in Figure 6-19, the output code for input = 5.12 V is Mean = 2045(0x7FD). 

Question
Could you please give us your opinion on the cause of the difference between 0x7FD and 0x800?

Thanks,

Conor

  • 0
    TI__Guru**** 178821 points

    Hi Conor,

    Take a look at Figure 6-28, the difference between Figures 6-19 and 7-15 comes from offset error.

  • 0 in reply to Tom Hendrick
    Guru 11355 points

    Hi Tom,

    The offset error of ADS8665 is approximately 0.2m (@26℃) on the 0-10.24V line.

    Also, the offset error is 0.2mV (typ).

    1LSB in the 0-10.24V range is 2.5mV, but I think that the offset error = 0.2mV is quite small and falls within 1LSB.

    The difference you asked about is 3LSB, so I have some doubts.
    ・Actual value of 5.12V: Figure 6-19 > 0x7FD
    ・Theoretical value of 5.12V: Figure 7-15 > 0x800

    Can you give me any comments regarding the above?

    Thanks,

    Conor

  • 0 in reply to Conor
    TI__Guru**** 178821 points

    Hi Conor,

    I don't have a better answer at the moment - perhaps the device used for the histograms had worst case offset and gain.  Are you having issues with production material or just curious to understand the figure differences?

  • 0 in reply to Tom Hendrick
    Guru 11355 points

    Hi Tom,

    Even if the deviation between the measured value and the theoretical value is large for a worst-case device, the error equivalent to 3LSB is outside the specification of the offset error in the datasheet. Could you please explain what values each parameter is assumed to result in an error equivalent to 3LSB?

    From the datasheet, Eo (Offset error) is equivalent to 0.8LSB at worst 2mV @ 25℃. Background of the question is that the ADC evaluation results show an error between the theoretical value and the actual value, similar to that in the datasheet.

  • 0 in reply to Conor
    TI__Guru**** 178821 points

    Hi Conor,

    I took a closer look and something does see to be a bit off somewhere.  I'll spend some time with the product definer and see if I can't get more detail on the data collection that was used to take the histograms.

  • 0 in reply to Tom Hendrick
    Guru 11355 points

    Hi Tom,

    For your convenience, I will provide additional information about what I investigated.

    Datasheet P.26 7.3.7.1 Internal Reference shows the reference error and offset after reflow.

    The reference voltage is 4.096Vmin/4.096Vtyp/4.097Vmax.
    It is stated that after reflow, the reference voltage will be offset due to thermal stress.

    In Figure 7-9, there is an offset of -6mV to +1mV after reflow.
    When considered in absolute value, there is a possibility of ±6mV offset.

    If the reference voltage is offset by 4mV after reflow, it will be 4.1V.

    FSR is 4.1x2.5=10.25V

    If you input 5.12V in this state, 5.12V/10.25V=0.4997=0.4997/FFF=7FD.

    If the REF voltage is offset by 4mV, the data may look like Figure 6-19.

    The difference between the ADC's theoretical value and the actual value is thought to be greatly influenced by the reference voltage after reflow.

    Thanks,

    Conor

  • 0 in reply to Conor
    TI__Guru**** 178821 points

    Thank you!  We're looking into this.