• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 50 subscribers
  • Views 607 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.

obtaining noise current density from Ibias -OPA365

Jorge Rive
Prodigy 80 points
Other Parts Discussed in Thread: OPA365

it is my understanding that the input referred noise current (density) can be obtained from Ibias via

inoise=sqrt(2*q*Ibias), where q=1.6e-19.

on the datasheet for the OP365, both Ib and the input referred noise current are specified.  When I plug the Ibias from the datasheet into the equation above, I do not get same shown on the datasheet for Inoise.

ie, Ibias=10pA.

inoise (calculated)=sqrt(2*q*Ibias)=1.7fA/sqrt(Hz).


the datasheet shows 4fA/sqrt(Hz).  Why the difference?

Thanks!

  • 0
    TI__Genius 12325 points

    Hi Jorge,

    You are correct that the shot noise calculation of the input bias current is typically used to determine the input current noise spectral density in CMOS and JFET input amplifiers. However, this only holds true at low frequencies.

    As the frequency increases an additional noise current will be introduced by the input voltage noise of the op amp. The input voltage noise produces a changing voltage across the input capacitance of the op amp, causing the input capacitance to draw a current. For this reason the input current noise spectral density of an op amp tends to increase with frequency. Because the input current noise of the OPA365 is specified at 10kHz, it may be including the additional effects of the input voltage noise and input capacitance. 

  • 0 in reply to John Caldwell
    Guru 21975 points

    Jorge,

    John raises a good point but there is an additional issue to consider with CMOS op amps:

    The input bias current of most CMOS op amps is dominated by the leakage current of the input protection circuitry. The input current of the actual transistor gates is virtually zero. In the simplest form of this protection, this comes from the leakage of diode clamps on each input to the positive and negative (or ground) rails. These two currents can be nearly equal and thus cancel one another, leaving a small residual input bias current. See this blog for more detail…

    Input Bias Current of CMOS and JFET Amplifiers

    Each individual leakage current has an associated shot noise current. Even if the two diode leakage currents are identical (leaving zero input bias current) the two shot noise currents do not cancel. They will add randomly, equal to root-two times the noise of one of the currents. This is the same noise you would calculate for the sum of both currents. For this reason, the noise current may be significantly higher than you would calculate from its specified input bias current.

    Also, consider that the leakage of these diodes doubles for each 10'C increase in temperature. So the input current noise will approximately double for a 20'C rise in temperature. See this blog for details…

    Temperature Effects on Input Bias Current…z

    So, noise current of a CMOS op amp is a tricky matter. In the end, it rarely matters as it requires exceedingly high source impedance for the current noise of most CMOS op amp to affect the application. For this same reason, it can be exceedingly difficult to accurately measure.

    By comparison, the shot noise calculation for a bipolar-input op amp will likely be pretty accurate. That is, unless the bipolar op amp has input bias current cancellation. See this blog for details…

    Internal Input Bias Current Cancellation of bipolar (BJT) op amps

    Regards--  Bruce

  • 0 in reply to Bruce Trump
    Expert 1220 points

    I never knew you could calculate the input current noise density using that equation. I learned something new!

    -Ken