• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 45 subscribers
  • Views 632 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.

ADS1262: input offset current and DC precision

Tillmann Stübler
Prodigy 20 points
Part Number: ADS1262

Hello,

We are using the ADS1262 for high precision measurement equipment. We need exceptional DC precision. A constant ADC offset voltage is not a big problem, but we have to keep offset drift very low over hours and days.

Our current design includes the following input filter:

The main purpose of the 1k resistors is current limitation when for whatever reason the input signals exceed supply voltage. Now, my question is how those resistors affect measurement performance. Both absolute and differential input current (PGA disabled) is specified as typ. 150nA, resulting in 150µV voltage drop. I assume the input current is temperature-dependent and contributes to the overall temperature drift of the circuit. Would the elimination of those resistors bring a significant benefit in DC precision? For example, I could bypass them using relays which are fast enough to open under all realistic overvoltage scenarios in our application.

I know that I could use the PGA to reduce input currents, but one of the input signals is always close to GND, so I can not meet the PGA's absolute input voltage requirements without greatly complicating the whole design.

Is there any other suitable way to improve DC performance? Would the chop mode eliminate the voltage drop caused by input currents as well? Does the chop mode alter input impedance or input currents in any way?

  • 0
    TI__Mastermind 45135 points

    Hi Tillmann,

    The series resistors can be used to limit the current and protect the ADC input, also can filter noise with the capacitors together. I'm sure you will use low tolerance and low temperature drift resistors to achieve higher accuracy because any mismatching between them will create an additional offset error at room and over temperature. Another concern is the absolute and differential input current are not constant and they are related to the applied voltage and temperature, thus these current can create additional error over temperature. There are a couple of ways to eliminate or minimize these errors:

    1. Measure the temperature in the system and implement a calibration. This will require an additional temperature measurement circuit and may not be easy to be accepted from hardware and software perspective.

    2. A calibration can be done periodically without additional hardware circuit, the calibration interval depends on your software and temperature change.

    3. The Chop mode is a good way to cancel the input offset including the error caused by the mismatching from external series resistors. Also, the Chop mode can reduce the ADC noise by a factor of 1.4 when it's enabled. This Chop can be done periodically based on the design requirement.

    4. Other ways to impact DC performance include: 

    • A good pcb layout design, for instance, a solid and dedicated ground plan is recommended.
    • A clean power supply is recommended.
    • Reference and PGA output capacitors should be placed to the ADC as close as possible.

    By the way, the common-mode capacitor in the RC filter should be chosen to be 10 times smaller than the differential capacitor, thus the resulting cutoff frequency for the common-mode filter is roughly twenty times larger than the differential filter, this makes the matching error of the cutoff frequency less critical and can ensure that mismatching of the common-mode filtering capacitors will not be translated to a differential voltage error. However, all three same value 22nF capacitors are used in your circuit. 

     

    Best regards,

    Dale

  • 0 in reply to Dale Li
    Prodigy 20 points

    Hi Dale,

    Thanks for your answer. So I think using the chop mode is a better and less complicated solution than using relays to bypass the 1k resistors.

    I was wondering whether the chop mode might introduce additional input currents due to a "switched capacitor" effect at high sample rates.

    Thanks for your remarks on the input filter, I will change that.

    Best

    Tillmann