• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 50 subscribers
  • Views 884 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.

OPA2188: Input current noise

paolo carniti
Prodigy 30 points
Part Number: OPA2188
Other Parts Discussed in Thread: OPA192, OPA140

Dear Texas Instruments,

I am using the OPA2188 in an active filter (SKT) with digitally programmable cut frequency using a trimmer. The frequencies of interest are pretty low (10 Hz - 1 kHz), and I am forced to use quite high resistances (100k) in order to use NP0 capacitors.

I am experiencing unexpected high noise and I am suspecting that input current noise is ~100x higher than the 7 fA/Hz^0.5 quoted in the datasheet.

I built a small test circuit, measured it and also simulated it with TINA.

TINA seems to behave as expected from the datasheet values, while measurement shows 800 fA/Hz^0.5.

Do you have any ideas or suggestions?

Best regards,

Paolo Carniti

  • 0
    TI__Guru 60610 points

    Paolo,

    The Tina spice macro-model of OPA2188 does NOT model transient effect of OPA2188 chopping action and its input current noise spectral density of 7fA/rt-Hz is solely based on the average integrated value (+/-160pA) of the input bias current that is dominated by the current spikes (up to 850nA!) caused by closing/opening of the front-end switches. Thus placing a large input resistor will convert the current spikes into voltage spikes adding to the overall input voltage noise. Please see the attached presentation describing in greater details the input bias current spikes in chopper amplifiers like OPA2188, their commutation into the voltage noise, and the ways to minimize their effects.7635.2188 Input Bias Current Commutation.ppt

  • 0
    TI__Guru* 93105 points

    Hi Paolo,

    Based upon the explanation Marek provided for the input current noise you may want to consider a non-chopper op amp; one with a more conventional FET input design. Take a look at the OPA192 datasheet as an example. The current noise is very low, 1.5 fA/√Hz, and there's no chopping spikes with this type of input design. It offers precision performance:

    Regards, Thomas

    Precision Amplifiers Applications Engineering

  • 0 in reply to Marek Lis
    Prodigy 30 points
    What I am experiencing is a white (1 <> 2000 Hz) noise increase, not spikes at the switching frequency.

    Switching frequency (which I understand should be ~600 kHz) is far beyond the cut-off frequency of the filter (~2 kHz), so even a 850nA*100kOhm = 85mV spike should be greatly attenuated by the 6-pole transfer function, almost to the level of white noise. Unless the spectral components of such pulses can go down to very low frequencies, but this conflicts with the 7 fA/Hz^0.5 at 1 kHz quoted in the datasheet.
  • 0 in reply to Thomas Kuehl
    Prodigy 30 points
    Unfortunately OPA192 doesn't match our specifications due to the high low frequency noise.
    The only choice would be a JFET input opamp or a chopped/AZed CMOS.
    OPA140 would be suitable, except for the input common mode range and power consumption which are a bit too high for our needs..