• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 233 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.

ADS1118: Input filtering for common AIN3 configuration

Georgi Beloev
Prodigy 30 points
Part Number: ADS1118
Other Parts Discussed in Thread: ADS1115

Hi,

The ADS1118 data sheet (p.30) states that AIN0, AIN1, and AIN2 can be measured with respect to AIN3. What is the recommended input filtering option in this case?

Also, the data sheet mentions that this configuration doesn't offer common-mode noise attenuation. However, the selected inputs (for example, AIN0/AIN3 or AIN1/AIN3) are stil subtracted, i.e., the ADC processes the differential signal. It seems that common-mode signal disturbance will be canceled out. Could you clarify?

Finally, comparing ADS1118 with ADS1115, it seems they use the same ADC core with different digital interface (SPI vs. I2C). Fig 2 in the ADS1118 data sheet and Fig 19 in the ADS1115 data sheet both show total error vs input signal -- but although they look similar in shape, one has error scale in millivolts while the other one in microvolts. Is this a typo? I imagine the mV version is correct?

Thanks!

  • 0
    TI__Guru* 92655 points

    Georgi,


    Thanks for the interest in the ADS1118, I'll address your questions below.

    With single-ended measurements, I would just use a simple RC filter. Generally, I would select a bandwidth of about 20x of whatever data rate you intend to use.

    Also, with single-ended measurements with the negative input tied to ground, you already remove the common-mode component. With a differential signal, you may still have a common-mode component, which could be removed by common-mode filtering even if the common-mode rejection for the ADC is high (higher than 90dB with different FSR and input signal types as listed on page 7 of the data sheet).

    You are correct that the ADS1118 and the ADS1115 are very similar (the ADC is the same with some small changes to the configuration. You are also correct about the total error vs input signal plot. The ADS1115 incorrectly labels the y axis in uV, and it should be mV. This error is dominated by the gain error.


    Joseph Wu

  • 0 in reply to Joseph Wu
    Prodigy 30 points

    Thanks Joseph!

    Let me clarify a bit to make sure we are on the same page. I don't have negative input tied to ground. Instead, I have a common offset voltage (VDD/2, buffered) applied to AIN3. I want to measure AIN0-AIN3, AIN1-AIN3, and AIN2-AIN3 to make use of the entire (positive and negative) input range of the ADC. So it looks like a differential measurement setup even though AIN3 is common to all three measurement.

    Normally a single-ended RC filter has the C connected to ground -- but in this use case should it be connected to AIN3 instead? This would be somewhat similar to a differential filter that has 2xR on each signal line, and C between them.

    Cheers,
    -- Georgi

  • 0 in reply to Georgi Beloev
    TI__Guru* 92655 points

    Georgi,


    I would still do the same type of connections to AIN3 as I would to ground (basically treating AIN3 as my ground). Because AIN3 is a buffered connection to VDD/2, it would be a the similar to a low impedance AC path to ground.

    This might be something you probably should simulate with TI-TINA. Some op-amps acting as a buffer might not be stable when there's a capacitance tied to the output. This may depend on the source impedance of the sensors that you are measuring.


    Joseph Wu

  • 0 in reply to Joseph Wu
    Prodigy 30 points

    Got it, thanks!