Help needed with BJT differential amp

Discussion in 'The Projects Forum' started by Athelionas, May 7, 2013.

  1. Athelionas

    Thread Starter New Member

    Mar 12, 2012

    I'm in the midst of designing my first audio power amplifier and I ran into a problem with the input stage.

    I'd like to utilise a long tail pair similar to the one attached (I made a few changes to make matters simpler) and it looks like it works fine except for a huge positive DC shift at the output and a noticeable amount of distortion of the signal's negative half.

    I've never used a long tail pair before so I don't really know what goes on in the circuit (at least not yet) but I noticed that the offset becomes more positive and distortion worsens as I raise gain.

    So questions are the following:

    1. Why is there a DC offset and distortion in the first place?
    2. Are these two phenomenons related?
    3. How could these issues be eliminated?
    Any help is highly appreciated!
    • LTP.png
      File size:
      19.7 KB
  2. Brownout

    Well-Known Member

    Jan 10, 2012
    The base of Q8 is presumably biased at 0VCD, while the base of Q20 is biased at some point between ground and +V. That would account for your large DC offset and other issues.
  3. #12


    Nov 30, 2010
    Current mirrors don't work very well when made of discrete components. That's an IC trick that works because the transistors are so well matched and thermally equal in a tiny chip.

    You have to think of this as an operational amplifier to get the feedback loop right.