Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Low noise, precision amp?

  1. Dec 9, 2005 #1
    with building a circuit. i need a gain of 300 (not exact, but somewhere around there, 200-500) for a small signal (typ. 0.1mV to 5mV) AC input with a bandwidth from 0.5Hz to 120Hz. I need it low-noise, with high precision and reliability..... has anyone got a reference for any sort of help to design with schematic examples etc? :confused:

    I was recommended the AD820, AD822 (dual), and AD824 (quad) op amps as they are apparently precision, very low noise amplifiers. Can someone help with a circuit diagram for connecting one or several of these for the best results?

    err.... and low power too... forgot about that, its going to be battery operated... if i cant get one
    Last edited: Dec 9, 2005
  2. jcsd
  3. Dec 9, 2005 #2


    User Avatar
    Science Advisor
    Gold Member

    You could check out Analog Devices data sheet and applications notes. A quick Google search always helps.
    http://www.analog.com/en/prod/0,,759_786_AD822,00.html" [Broken]
    Last edited by a moderator: May 2, 2017
  4. Dec 10, 2005 #3


    Staff: Mentor

    You will probably want a multi-stage amplifier. Usually you get a very low-noise low-power preamp and then follow it with a beefier power amplifier. At least that is what we do in MRI. The total SNR is pretty sensitive to the noise figure of the preamp, and not very sensitive to the noise figure of the main amp.

  5. Dec 14, 2005 #4


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    The choice of the amplifier will be determined also by the source of the signal to be amplified. Apparently, bandwidth is no issue. But what's the impedance of the source ? You are actually only limited in noise performance if the impedance of the source is rather high. Otherwise, there is a cheap trick to lower the noise in the output: connect several amplifiers in parallel to the same source (this can only be done if the amplifier impedance / number of amplifiers is still higher than the impedance of the source of course), and take the average of the outputs of all the amplifiers. The average signal will be equal to the signal of one amplifier (them being all identical), but the average of the different (independent) noises will of course go down (by about sqrt(number of amplifiers)).

    A few good operational amplifiers to consider:
    OP-07, OP-27, OP-627, ...
    but it really depends upon the application.

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook