• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 2157 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.

ADS1115 accuracy

NAOKI AOYAMA
Guru 11215 points
Other Parts Discussed in Thread: ADS1115

Hello,

My customer is evaluating the ADS1115.
It is configured 128sps data rate, single-shot mode.
He evaluates the ADS1115 two conditions as follows.

Case1
start conversion
wait 10ms
read result
start conversion
wait 10ms
read result

repeat ten times

Case2
start conversion
wait 150ms
read result
start conversion
wait 150ms
read result

repeat 10 times

He check the variation of the data of above two Cases.
As a result, the accuracy of the Case2 result better than Case1 result.  (but only 1LSB difference)

Do you think the length of wait time impact the accuracy?

Could you give me any idea about the cause of above result?

Regards,

Naoki Aoyama

  • 0
    TI__Guru* 97745 points
    Naoki-san,


    Theoretically, the two results should be identical. Does the input change at all between any of the readings?

    The ADS1115 is single cycle settling and each of the readings are settled as long as the inputs are settled. The noise level is low compared to the LSB size that small changes in the input may alter the statistics of the noise. You can verify this by using the EVM and making small voltage changes to the input of less than 1 LSB.

    Can you share their data and their setup? It might help if we know what they were measuring.


    Joseph Wu
  • 0 in reply to Joseph Wu
    Guru 11215 points
    Hi Joseph-san,

    Thank you for your fast reply.
    I will get their data and their setup, will reply again.

    Thanks,

    Naoki Aoyama
  • 0 in reply to Joseph Wu
    Guru 11215 points
    Hi Joseph-san,

    I got their data and setup.
    They want to observe DC voltage, single-ended input.
    Their setup and data as follows.

    Case1

    Configure [D5 83] ;AIN1,FS=2.048V,Single-shot,DR=128sps

    wait 10ms

    read result

    Configure [E5 83] ;AIN2,FS=2.048V,Single-shot,DR=128sps

    wait 10ms

    read result

    Configure [E5 83] .......repeat ten times

    Case2
    as same configure as Case1 except wait time is 150ms

    The deviation of read result is below.
    (These values are three sigma.)
    Case1 10ms : 2.1 LSB
    Case2 150ms: 1.2 LSB

    Regards,

    Naoki Aoyama
  • 0 in reply to NAOKI AOYAMA
    TI__Guru* 97745 points
    Naoki-san,


    Are they able to provide a schematic? I was also hoping they would be able to send us the raw data from the ADC as well. The raw data may show noise and the exact settling behavior of the measurement.

    Are they switching anything else in the measurement? I'd like to know what it is they are measuring (is this just a DC voltage source?) and if there are any other external configurations for this system. I'd also like to know what the voltages are for each of the measurements.

    I'm not aware of anything internal to the device that could cause this behavior and I'd like to know as much as possible to try to duplicate this setup.


    Joseph Wu
  • 0 in reply to Joseph Wu
    Guru 11215 points
    Hi Joseph-san,

    I will ask the customer to provide a schematic and raw data.
    I will reply again.

    Thanks,

    Naoki Aoyama
  • 0 in reply to Joseph Wu
    Guru 11215 points
    Hi Joseph-san,

    I asked the customer to provide a schematic and raw data. But they cannot provide it.
    I got the block diagram instead.

    photo sensor - amp - active filter - RC filter - ADS1115
    There is no external switch.

    I think it does not progress of this discussion without the raw data.
    So, I cancel this question. Sorry.

    But I have a another question about the ADS1115. I need your help.
    Which is the best configuration for observe DC voltage accuracy?
    DR= 8sps or 128sps, FS=?, single-shot mode or continuous mode.

    Regards,
    Naoki Aoyama
  • 0 in reply to NAOKI AOYAMA
    TI__Guru* 97745 points
    Naoki-san,


    I don't know if there is a best configuration for observing the DC voltage accuracy.

    Certainly the DR=8sps would be better because of the low noise, but for each of the ranges, the offset error is going to be typically 1LSB. I don't think any range has better performance. It should not be different from single-shot mode to continuous mode.

    However, for measuring the DC accuracy, just remember that if you are measuring a DC source, the source should be very low noise as well or the measurement will appear noisy.


    Joseph Wu
  • 0 in reply to Joseph Wu
    Guru 11215 points
    Hi Joseph-san,

    Thank you for your suggestion.
    I will feedback to the customer this suggestion and evaluate more.

    Thanks,

    Naoki Aoyama