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.

INA116 non-linearity

Other Parts Discussed in Thread: INA116

I am trying to understand the nature of the non-linearity of the INA116 amplifier.

I have a voltage measurement instruments which uses the INA116 as the front end sensor at unity gain due to its high input impedance and ultra low input bias current. The data sheet indicates +- 0.0005 of FSR (which I assume is the supply +-15V) giving ~7.5mV. What I am seeing is a 20 mV positive to negative jump in the residuals within 30mV of zero Volts. Is this a known feature of this device? All of the non-linearly occurs near the zero difference of the inputs?

The testing is done with one input grounded and the input signal on the other and then the test is switched.


I have seven different builds of the same instrument and the residuals from a linear voltage sweep on each build are within about 1mV of each other. The instruments and testing are essentially identical. They all show a very sharp jump from +10mV to -10mV when within +-30mV of zero.


Thanks,

Charles

  • Hello Charles,

    Please correct me if I'm wrong, but this is what I gather of your system:

    Dual supply (+/-15V)
    Vref=(Vin-)=GND
    (Vin+)=input voltage
    G=1V/V (i.e. RG disconnected)

    If so, the output should be within +/-4mV (or +/-10mV, depending on device grade) of (Vin+). However, as you decrease (Vin+) toward 30mV, the output suddenly jumps -20mV given a much smaller change in input signal. Correct?

    Also, what lab equipment are you using to provide the input signal and measure the output? Finally, are you also measuring the input voltage at the pins of the device? Just want to make sure that it's not the input that's experiencing the voltage jump.
  • Pete,

    I think you have it all correct. The testing was done with either a Keithley source meter, 2400, or a waveform generator, 3390. I also did another test on one of the circuits using a completely different setup and set of equipment. I was quite careful with signal grounds. I was not measuring the input at the output but was instead looking at the signal through the whole chain and examining ADC output.

    To Aid in the discussion the INI116 is being used as a high impedance electrometer for measuring electric fields by determining the potential between two spaced probes connected to the inputs of the INA116.  The schematic of the circuit is included here as a PDF

    EFP12.pdf

    The tag EF_V1N1 (5-A1) goes off to another INA116 input. In other words two INA116's share the same input point. The residuals of a linear fit are attached here for this channel. The suspected reason for the non symmetry is due to the second INI116, not shown on this figure, tied to EF_V1N1 (5-A1).


    Residuals EFP12.pdf

    for a different channel that does not share a test point with another INI116 the residuals are more symmetric and look like:

    Residuals EFP34.pdf

    Probably more information than you wanted.

    Charles

  • Hello Charles,

    Thanks for the verification and additional information. Have you tried observing the 'jump' at the pins of the INA116? I want to ensure that the anomaly comes from the INA and nothing else in the signal chain (specifically the op amps).