• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 45 subscribers
  • Views 410 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.

ADS1220: Absolute Value of AIN when PGA is used

Damien Dusha
Prodigy 190 points
Other Parts Discussed in Thread: ADS1220

Dear all,

I am looking to use the ADS1220 in an application that is pseudo-differential with AIN1 = 0.25V - 4.75V and AIN2 = 2.5V.   Both AIN1 and AIN2 are ratiometic to supply (AVDD = 5.0V, AVSS = 0V), with VREF = 2.5V, via a voltage divider.

According to Equations 4-6 (p19), this meets the common-mode voltage requirements for the sensors with the PGA enabled.

The datasheet does not place any constraints on the absolute value of AIN when the PGA is enabled - only when the PGA is disabled.  Are there any constraints that need to be observed, even if specific to the voltage range in this application?

Also, with regards to the reference, the datasheet states the following (with added emphasis):

"The reference inputs are internally buffered to increase input impedance. Therefore, additional reference buffers are **usually** not required when using an external reference and the reference inputs do not load any external circuitry when used in ratiometric applications."

Is there any further guidance for the reference on this device?  Are there any further recommendations for filtering on the reference?

Thank you in advance for your help,

  -- Damien

  • 0
    TI__Guru** 113765 points

    Hi Damien,

    The absolute input voltage will relate to the Absolute Maximum Ratings specified in the table on page 2 of the datasheet.  This is much different than the measurable input range with the PGA enabled.  With the PGA enabled you can only measure as specified by the common mode input range calculation even though you could apply a voltage outside of the common mode input range based on the Absolute Maximums.

    The measurement itself will only be as good as the reference.  If the reference drifts, so will the measurement.  So you have added external components (resistor divider) that will add noise and other error to the measurement. This will be true even though the measurement is ratiometric as the 2.5V analog input (AINN) will drift independent of the AINP input.

    Depending on the sensor output, it may not be necessary to use the PGA enabled.  With this device there is not much to be gained by using the PGA enabled at a gain of 1 other than higher input impedance.  So you could use AVDD (5V) as your reference and measure your input directly relative to AGND.  I realize that you lose 1 bit to extending the reference and one bit to losing dynamic range.  However the noise will be more significant than you might realize and the result may not be worth the effort of using the resistor divider in the end.

    That said, the sensor may indeed require a high impedance input and the circuit you described.  If you use the same voltage divider for both the reference input and the AINN input for the ADS1220, you will have some current flowing into the reference (+/-10nA as stated in the electrical characteristics table on page 3 of the datasheet) and bias/leakage current flowing into the input (see the Figures on page 11 of the datasheet). 

    Filtering with a ratiometric measurement can be a little tricky as the noise/drift canceling effect can be greatly disturbed if the input filtering and the reference filtering are mismatched.  So it is best to match the corner frequeny of both filters so the relationship of response between them remains intact as much as possible.  Generally you would design the required antialiasing filter for the analog inputs, and then match the response with the reference filter.  You can view a sample of this type of calculation by reviewing TI application note SBAA201.

    Best regards,

    Bob B