• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 2 answers
  • Subscribers 49 subscribers
  • Views 2157 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.

INA219: ina219 low pass filter

Philipp Overath
Prodigy 140 points
Part Number: INA219

Good day,

so I wanted to implement an analog low pass filter as mentioned in the datasheet.

I wanted to implement a cut off frequency of around 10hz. But when I use resistor over 1k the signal has a bigger drift than 320mV. Why is it limited to 10 Ohms for the resistor? And why do you need two restistors instead of a "normal" low pass filter consisting of only one resistor?

Thanks

  • 0
    TI__Mastermind 26640 points
    Hello Phillip,

    We recommend limiting the input resistors to 10 Ohms because of the difference in input bias currents or Input offset current (Ios), where Ios = (IB+) - (IB-).

    Usually, as the shunt voltage increases, IB+ will become more positive and IB- will become more negative, or vice versa. Thus, as the shunt voltage increases so does Ios. Placing input resistors at the IN+ and IN- inputs causes voltage drops across Rfilter. Since IB+ and IB-become different values, these voltage drops also are different and thus more voltage offset is introduced at the input, thus causing more error.

    Having two input resistors for the filter means that this is a differential input filter configuration as opposed to having a common input filter ("normal") where you would have shunt capacitors going from IN- to GND and IN+ to GND. We recommend using the differential input filter for several reasons:

    1. The differential filter has less components.
    2. The common input filter will suffer greater error because of capacitance/tolerance mismatch between two same capacitor components.
    3. The differential filter is really all that is necesary since INA219 is measuring a differential voltage.

    Note the INA219 also has an internal input resistor (necessary to measure bus voltage) that contributes to input bias current. You can read about error analysis with input filters and the INA219 here in our FAQ articles.
    e2e.ti.com/.../3576.ina219-ina3221-input-impedance


    Hope this helps,
    Peter Iliya
    Current Sensing Applications
  • 0 in reply to Peter Iliya
    Prodigy 140 points
    Thank you very much for the long and informing response.
    But increasing the filter resistors lowers the current flowing, right? So I thought that this decreases the IB+ and IB-. Anyway, why do the bias currents increase with the shunt voltage? Could I use a higher Capacitor to increase the cutoff frequency? (100uF?).
    thanks.
  • 0 in reply to Philipp Overath
    Prodigy 140 points
    Oh and would you recommend a different current measuring chip which would be compatible with a analog low pass with such high filter resistors?
  • 0 in reply to Philipp Overath
    TI__Mastermind 26640 points

    Hey Phillipp,

    Increasing the filter resistors does not have an effect on the input bias current, at least significantly. I believe input bias current rises with shunt voltage because of device internal resistance. Using a bigger capacitor (100uF) is valid if the higher cutoff frequency is needed.

    I list our lowest specified IB current shunt monitors in the article below, but what really dictates the IB is the value of the common-mode voltage so keep this in mind and additionally what dictates the filter error is the Ios. Overall, you won't see a tremendous reduction in IB, but bringing IB from tens of uA to single uA is your best shot.
    e2e.ti.com/.../3593.how-vos-and-ib-affect-small-current-measurements

    Sincerely,
    Peter Iliya
    Current Sense Amplifiers

  • 0 in reply to Peter Iliya
    Prodigy 140 points
    Thanks,
    So I used a potentiometer to see the effect of changing the resistor value. Even without any big current running through the shunt, I could change the measured voltage by changing the potentiometer. And for using a bigger capacitor I thought it would bring down the cutoff frequency?

    And sorry for asking: what exactly do u mean by common mode voltage?
    Thanks for the article

    Greetings
  • 0 in reply to Philipp Overath
    TI__Mastermind 26640 points
    I am not sure what you are implying with the measured voltage changing. Is the circuit working as you expect? Is your potentiometer the filter resistor or shunt resistor?

    Yes, your are right. Using a bigger capacitor will bring down cutoff frequency. I miswrote that in previous post.

    Common-mode voltage is the voltage at the input pins (IN- and IN+) with respect to ground. What the device is actually measuring is the differential voltage (the voltage at IN+ with respect to IN-). The differential voltage floats with the common-mode voltage.

    Sincerely,
    Peter Iliya
    Current Sensing Applications
  • 0 in reply to Peter Iliya
    Prodigy 140 points

    Hello,

    The circuit is working. But not with high resistors. So to see the effect of the filterresistor, I used a potentiometer to change the value. In the datasheet it says that it has a bias current of 20uA. So with higher filterresistor the voltage drop increases and so does the measured value of the ina219.

  • 0 in reply to Philipp Overath
    TI__Mastermind 26640 points

    Phillipp,

    This make sense. At some point the resistance will get too high and begin to dominate the input offset of the part. This gets gained up and shifts the output voltage readings, thus introducing error.

    Best,
    Peter