Connect with us

Question on audio amplifiers for a change...

Discussion in 'Electronic Design' started by Paul Burridge, Jul 9, 2004.

Scroll to continue with content
  1. Hi all,

    Why does the first stage of all commercial audio amps seem to consist
    of a differential (long-tailed) pair? AFAIA, the chief characteristic
    of this configuration is just to eliminate 'mush' on an input signal,
    like if it's come from a long line input and has a high noise floor,
    which clearly isn't the case between your CD/tape-deck/phonograph or
    whatever. So why?

    p.
     

  2. Because a differential input stage is a useful way to apply global negative
    feedback, which is in turn a good way to improve the response
    characteristics of the amplifier.

    And because a long-tailed pair permits DC coupling of the input.
     
  3. Ben Bradley

    Ben Bradley Guest

    My short answer: it makes it like an op-amp, one base is the signal
    input, the other base is feedback from a resistive divider on the
    output to set gain, with the lower resistor having big cap in series
    to ground so gain at DC is low, keeping the output near zero.
    It's much easier to design this way, using the usual transistors
    that have widely varying beta, yet the amplier's gain from unit to
    unit varies very little because it's mostly determined by the feedback
    resistors.
    You mean like a balanced input to cancel interference induced in
    the cable?
     
  4. Jim Thompson

    Jim Thompson Guest

    Differential input stages afford some cancellation of non-linearities,
    thus reducing distortion more than in a single-ended input stage.

    ...Jim Thompson
     
  5. Genome

    Genome Guest

    | Hi all,
    |
    | Why does the first stage of all commercial audio amps seem to consist
    | of a differential (long-tailed) pair? AFAIA, the chief characteristic
    | of this configuration is just to eliminate 'mush' on an input signal,
    | like if it's come from a long line input and has a high noise floor,
    | which clearly isn't the case between your CD/tape-deck/phonograph or
    | whatever. So why?
    |
    | p.
    | --
    |

    Bugger off Burridge

    DNA
     
  6. Ah! I see. Thanks Ben (& Walter). A clear and concise explanation
    indeed.
    Precisely. As in half a mile of 600 ohm ribbon, for example.
     
  7. Thanks, Walter. By "global" I assume you mean f/back from the ultimate
    stage to the first? I've seen some designs that incorporate an input
    from the supply rail(s) for the same reason. Why not regular
    decoupling? And this might be an appropriate time to ask for a
    definition of what exactly is meant by the terms "open loop gain" and
    "closed loop gain"?
    Thereby saving a thumping great and rather expensive capacitor, I
    guess? Plus the low frequency loss it would introduce?
     
  8. Ken Smith

    Ken Smith Guest

    The other answers plus they are hard to break. With most such circuits,
    you can swing the input above the positive rail and below the negative
    rail without any harm.
     
  9. At this point it would be appropriate for me to recommend Art of
    Electronics, 2ed., by Horowitz and Hill. Any answer I might give you is
    there, more articulately and better supported.
     
  10. Don't do yourself down, Walter.
    The terms I was seeking clarification on came from Douglas Self's
    magnum opus on audio amplifier design, which (with all due regard for
    AoE) is very much more detailed on this particular subject.
     
  11. Fred Bloggs

    Fred Bloggs Guest

    How come school buses are yellow? What is the standard width of a
    sidewalk? How come some trees drop their leaves in fall?
    Where...what...how....why...
     
  12. Paul Burridge wrote...
    Walter was no doubt referring to our discussion of Ebers-Moll, which
    you will find is completely missing from Douglas Self's book. He
    starts off with differential amplifiers, and continues in that vein.
    He only modestly discusses the reason, on page 62, where rather than
    present any mathematical detail, he simply references the reader to
    Gray and Meyer's book (which does have much more detail than AoE).

    But it's a significant question, and therefore deserves coverage.
    If you look at figure 2.36 in AoE, and read the discussion, you'll
    see the rather strong reason for using a differential, rather than
    a single-ended amplifier for the input stage.

    Thanks,
    - Win

    (email: use hill_at_rowland-dot-org for now)
     
  13. So many things to learn. So little time.
     
  14. Thanks, Win. But I don't see the relevance here to differential amp
    stages. The section you refer to deals with the shortcomings of
    grounded-emitter amps (p. 83) and you don't begin to deal with
    differential amps until page 98. The query here on this thread relates
    to aspects of DA amp stages which typically do use emitter resistors.
     
  15. Paul Burridge wrote...
    Really? You asked, "Why does the first stage of all commercial audio
    amps seem to consist of a differential (long-tailed) pair?" The only
    alternate to differential is a non-differential (single-ended grounded
    emitter) amplifier. We answer the question in AoE, with equations and
    plots showing the distortion (with a non-differential choice). Douglas
    Self does not, except a hint in his brief comment that I referenced.

    Thanks,
    - Win

    (email: use hill_at_rowland-dot-org for now)
     
  16. normanstrong

    normanstrong Guest

    A differential input allows you to design with at least one less high
    value capacitor. Since a capacitor is larger and more expensive than
    a transistor, the differential input is desirable at the outset.
    There are many more advantages, but this one alone would be
    sufficient.

    Norm Strong
     
  17. Okay, that's fine. I just wanted to be sure that you did really mean
    to refer to fig. 2.36, that's all.
     
  18. Oh?

    Why?

    Seems like darkness to me.
    Not for small caps.
    This cap one is pretty much irrelevant. No ones uses a diffstage because
    of this somewhat dubious issue.

    Kevin Aylward

    http://www.anasoft.co.uk
    SuperSpice, a very affordable Mixed-Mode
    Windows Simulator with Schematic Capture,
    Waveform Display, FFT's and Filter Design.
     
  19. The Doyen of audio amp design, Douglas Self, disagrees with you, Kev.
     
  20. Tim Shoppa

    Tim Shoppa Guest

    With stereo equipment being so common these days, it's important that
    the gain of the left channel match the gain of the right channel to
    a fraction of a dB. This is easy to do with negative feedback... which
    is easily done with differential pairs.

    Note that if you open up a mono transistor radio and trace the circuitry
    you will not find any long-tailed pairs, most likely. (And if you do
    it's likely to be part of an IC...)

    I'm not so sure what "mush" you're talking about. Real differential inputs
    are Good Things, but outside of the connection between a phono pickup
    and the preamp you are unlikely to see true differential interconnects
    in consumer electronics.

    Tim.
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-