Issues with capacitive coupling of 2 opamps

Discussion in 'General Electronics Chat' started by Nano001, May 29, 2013.

  1. Nano001

    Nano001 Thread Starter Active Member

    Jan 12, 2010
    Hi, I built the attached circuit which I am using to detect low levels of optical pulses. I am using the first opamp as a transimpedance amplifier and the second one to amplify the signal from the TIA. The 1uF capacitor and 10k resistor is used to remove the dc signal of the first amplifier.

    The circuit works well, except that there is a background frequency (about 1 khz) when there are no optical pulses which I am trying to remove. I noticed that this background oscillation is only present when I couple the two circuits with the 1uF cap. When I probe the output of the first amplifier both before and after the 1uF capacitor when it is NOT connected to the 2nd opamp, there is no oscillation. As soon as I connect everything together I see oscillations before and after the capacitor (points A and B in the diagram), and these oscillations get amplified at the output of the final opamp (point C). So, my problem lies with capacitive coupling of amplifiers.

    I have tried many combinations of different R and C values, and 1uF and 10k gives me the best output signal with minimal oscillations. However, I am looking to completely remove them. The circuit works without oscillations if I completely remove the 1uF and 10k component, but the DC signal out of the first amplifier is always varying which makes it nearly impossible to set the non-inverting bias point of the second stage always equal to the middle of the first amplifier pulse.

    Please let me know if you can help with this circuit, thanks.
    • TIA.jpg
      File size:
      26.6 KB
  2. MrChips

    MrChips Moderator

    Oct 2, 2009
    With low level signals and sensitive circuits I avoid using multiple amps in an single package. I stick to one amp per package.

    The other thing is two inverting amplifiers make a non-inverting amplifier which produces positive feedback. Any chance you can change the second stage to non-inverting?
    absf likes this.
  3. #12

    #12 AAC Fanatic!

    Nov 30, 2010
    Good idea. 2 inverters makes an oscillator if there is even a tiny bit of coupling through the power supply or radiated energy. Simple solution: Don't invert twice.
  4. Nano001

    Nano001 Thread Starter Active Member

    Jan 12, 2010
    Ok, that is a good suggestion. I will try it out.
  5. ramancini8

    ramancini8 Member

    Jul 18, 2012
    Try terminating the photodiode to lower the impedance of the inverting op amp input.