• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 49 subscribers
  • Views 1055 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.

INA197: Unexpected output of negative going current sensor

Steve Schmidt
Prodigy 80 points
Part Number: INA197
Other Parts Discussed in Thread: INA152, INA282, INA213

Hi,

I am trying to build a current sensor for a +/- 2mA sinusoidal current source injected into a 5k to 10k Ohm load. I am using the circuit shown in figure 29 of the datasheet for the INA197 (with a 33 Ohm shunt resistor) but getting some weird output from the INA197 which senses the negative-going phases of the sinusoid. The sensing of the positive-side by the other INA197 works great. The issue with the negative going output only happens for higher current values. Images below:

This is the output from the INA197 which senses the negative phases of the sinusoid.

Below are (yellow) the output of the INA152 and (cyan) the voltage waveform at the load.

  • 0
    TI__Genius 14825 points
    Steve,
    Just to confirm, the INA152 has a 2.5V reference attached to the REF pin, correct? That could be the issue. Please let us know if that's not the case.
  • 0 in reply to Jason Bridgmon
    Prodigy 80 points
    Jason,

    Thanks for your response. For the traces above I am actually using V+ = 5V, V- = -5V and Vref = GND. I have now tried it with 5, GND, 2.5 (respectively) with similar results. The issue with the INA197 occurs in isolation; even when it is not connected to the INA152.

    Thanks again,
    Steve
  • 0 in reply to Steve Schmidt
    TI__Genius 14825 points
    Steve,
    From the looks of the waveform, I'm concerned that you are effectively changing the common mode of the signal across the supply level, which is causing a change in input bias current. Please refer to figures 9 and 10 in the data sheet. There is a kind of hand-off internal to the device that happens near the supply level wherein the inputs start to draw more current to power the front end of the amplifier. This is why the part inputs can exceed the voltage rail. Since you only see this happen when you have larger signals, I think that perhaps you were not crossing that threshold, and when you do have larger signals, you start to see this.

    Another reason I think it is this is because your current is very small, just a few mA. The input bias currents are 10s of uA, which to an application that is running 10A is almost nothing, but in your application it is very relevant. It's a little odd that we're not seeing it on the positive-going device unless that part is always above supply perhaps? How many boards have you tried and observed this behavior?
  • 0 in reply to Jason Bridgmon
    Prodigy 80 points

    Hi Jason,

    I see what you mean. The unexpected output on the negative side does start when the common mode is roughly 4V in the image above. This is probably the 5V VDD minus a 1V turn on voltage? The positive-going device gets the same input, with reverse polarity, and I haven't observed it there. I've observed this behavior on both boards that I have. I'm using breakout boards into a breadboard for testing and have been able to observe this for both when negative-going and neither for positive-going.

    Is there a better chip for low amplitude applications?

    Thanks,

    Steve

  • 0 in reply to Steve Schmidt
    TI__Genius 14825 points

    Most of the current shunt monitors have a little handoff circuitry around where the common mode crosses supply.  I would recommend trying the INA213 (gain 50V/V) for common modes less than 26V and the INA282 (gain 50V/V) for bidirectional operation.  These are parts that have low offsets, great performance and an external reference pin so you only need one part to do bidirectional measurements.  These both have a jump in Ib like the INA197, see figure 11 of the INA213 data sheet and figure 7 of the INA282 data sheet.

    I was wondering though, are you just simulating a load with the resistor and common mode change and the eventual application will just have a steady voltage with varying load current?  If so, then you may not need to worry too much about the handoff region as long as the common mode is fairly static.  I would still recommend switching to the INA213 or INA282 though as they work very well with low sense voltages.

    Hope that helps,

  • 0 in reply to Jason Bridgmon
    Prodigy 80 points

    Thanks! Our common mode ranges over about 75V so I will look into INA282.