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

Periodic oscillations in the output signal

  1. Jul 28, 2010 #1
    Dear All,

    We have an voltage amplifier. The input is given in small square wave steps to check rise time and noise level of an output signal. The rise time looks good, but we see periodic (noise)oscillations after reaching the plateau. We measure signal voltage across capacitor (10 micro farad) and 1k in series. Can anyone suggest how to get ride of this noise of periodic oscillations in the plateau. Thank you.

  2. jcsd
  3. Jul 28, 2010 #2
    We need some more information in order to help.
    1. Is the amplifier home built or a commercial one?
    2. What is the bandwidth of the amplifier?
    3. What is the input impedance of the amplifier and is the input ac or dc coupled?
    4. What is the rise time of the pulses and their duration?
    5. How are the series 10 uF cap and 1 K resistor connected to the amp? Are they in series with a load or are they the load (connected between the output and ground)?
    6. If they aren't the load, what is the load?
    7. About what is the frequency of the oscillations?
    8. At the beginning of the flat part of the pulse, does the waveform of the oscillation rise above the pulse first or fall below it first?
    9. What is the output impedance of the amplifier?
    10. If you have tried any fixes, what were they?
    11. A circuit diagram or schematic always helps.

    Oh, and welcome to Physics Forums.
  4. Jul 28, 2010 #3


    User Avatar
    Science Advisor
    Gold Member

    Just from reading the OP it sounds like ringing.
  5. Jul 29, 2010 #4
    Hi here are the answers to the questions, hoping to see reply.

    1. Amplifier: Commercial
    2. Bandwidth: -3 dB
    3. Input impedance: 10 KΩ and DC coupled
    4. Time: the time to reach plateau is 200ms
    5. 1K resistor is associated with circuit and with load, the load connected to the output socket (BNC). The capacitor is connected at BNC and GND.
    6. The output is 75V.
    7. The frequency of the oscillation is around 60-65 Hz
    8. At the beginning of the flat part of the pulse- the waveform of the oscillation is raise above the pulse first.
    9. Output impedance: 150Ω.
    10. We are trying to change the value of capacitor and resistor and to see the output.
    11. No circuit diagram, since it is commercial one

    Thank you
  6. Jul 29, 2010 #5
    ye it sounds like ringing with some frequency
  7. Jul 29, 2010 #6
    I agree it sounds like ringing however there is the possibility it may be some powerline ripple if your powerline frequency is 60 Hz. Is your input circuit shielded? Do the oscillations decay or is the amplitude constant?

    For bandwidth I was interested in frequency bandwidth at the -3 dB points.

    Do I understand correctly that your risetime is 200 milliseconds? What happens if you disconnect the 10 uF cap?
  8. Jul 29, 2010 #7
    I too agree that the frequency(60Hz) may come from powerline, but am not very much sure.

    Circuit is shielded.

    The oscillations decay is constant.

    Yes the raise time is 200ms.

    Removal of capacitor causes high frequency noise with spikes.
  9. Jul 29, 2010 #8
    I do not understand "The oscillations decay is constant." Does this mean the amplitude is constant or that the amplitude decays (diminishes).

    Is the input risetime also 200 ms. I'm wondering if the risetime is bandwidth limited.

    How much is the voltage gain of the amp? (What are the input and output amplitudes of the signal?)

    What is the amplitude of the oscillation? Is it minor compared to the pulses or does it seem significant?

    If you decease the resistance that the amplifier output sees, does that increase or decrease the amplitude of the oscillation?
  10. Jul 29, 2010 #9
    More questions:

    Where do you think the high frequency noise and spikes are coming from?

    Can you reduce the value of the capacitor an reduce the oscillations without increasing the noise?
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook