Connect with us

design discrete mic pre?

Discussion in 'Electronic Basics' started by tempus fugit, Mar 24, 2005.

Scroll to continue with content
  1. tempus fugit

    tempus fugit Guest

    Hey all;

    I'm thinking about building yet another mic preamp. I've currently got one
    built around the SSM 2017, and it sounds pretty good. However, I notice that
    many of the high end designs use discrete (i.e. all transistor) designs. How
    feasible is it for a hobbyist to build/design something like this? I've seen
    some schems on the net that use a transistor input stage with an opamp as a
    follower, but I'd like to try an all discrete design if possible. I suppose
    I should also mention that I'm a musician how doesn't have a lot of spare
    time. I'm pretty happy with my current (SSM) micpre, and could easily build
    another (I need another micpre), but if it isn't going to take tons of
    fiddling and redesigning, I wouldn't mind trying another option.

    Thanks
     
  2. Ban

    Ban Guest

    I think you won't get much better performance, unless you invest in a high
    quality transformer (i.e. Jensen). With discrete transistors you might
    achieve a slightly better noise floor however linearity and distortion are
    probably worse unless you invest into special matched parts like MAT04 or
    LM394.
     
  3. Gareth

    Gareth Guest

    It is certainly possible, but I doubt you will notice any improvement
    over a design using an SSM2017.

    There is a fairly simple design here for example:

    http://sound.westhost.com/project13.htm

    Gareth.
    --
     
  4. tempus fugit

    tempus fugit Guest

    This is the one that I was thinking about:

    http://sound.westhost.com/project66.htm

    And that's kind of what I was wondering - will it sound much different from
    what I have now? I've heard many people say that a discrete micpre sounds so
    much better, but I wonder myself.
     
  5. Gareth

    Gareth Guest

    It may be possible to get slightly lower noise with discrete parts as
    Ban said, but the SSM2017 is very low noise and I doubt that you would
    notice any improvement. I'm not really an expert on this though, and I
    have no idea what the pre amp you have now sounds like either :)

    Have a look at this datasheet:

    http://www.analog.com/UploadedFiles/Data_Sheets/23265255MAT03_c.pdf

    Figure 4 is a circuit diagram for a pre-amp with about 1/2 the noise and
    distortion of the SSM2017. it uses the MAT03 low noise transistor pair
    and an op amp to get the advantages of both (low noise and low
    distortion).

    You could reduce the noise still further by putting several MAT03s in
    parallel as in Fig 3a. However that will get very expensive, and I
    doubt you will notice any difference.

    It would be an interesting experiment to get some of these people who
    say that discrete pre amps sound much better and do some blind tests to
    see if they can actually tell the difference.

    --
     
  6. tempus fugit

    tempus fugit Guest

    Thanks Gareth.

    I've got that MAT datasheet already, and this is one of the things that got
    me thinking in the discrete direction. I'd still have to figure out a way to
    get rid of the opamp to be all discrete, od I could use fig. 4, but it is
    only a single ended input; I'd need to figure out how to make it balanced.

    I should note that the SSM I'm using now is REALLY low noise, so it's not
    really the noise level that I want to improve upon. I guess I'm wondering if
    one will actually sound better. I know that I can hear a difference (a big
    one) among different mic pres I've heard, but then some of them may indeed
    be crap in the first place. I often wonder if I would hear much of a
    difference between my SSM design and one of the high end ones.

    Thanks
     
  7. An opamp is in the end nothing but a circuit made from transistors, just
    in a small case. You can also build an OpAmp from discrete transistors
    and use that as a sensitive preamplifier. Elector had a circuit like
    this many years ago, with 5 parallel input transistors to reduce noise.
    IIRC it was meant as preamp for a moving coil pickup (those were the
    days!).

    But I doubt that a good preamp from discrete transistors would be
    audibly better than one made with a good OpAmp, and vice versa. For
    example a noise floor of -100 db is not audibly worse than -105 dB.
    There is a lot of fetishism in this area, and some people are making
    good money from it.

    In addition, what kind of musik is the amp intended for, and in what
    environment will it be listend to? Dance musik with little dynamic in a
    loud bar has lower requirements than recording classical musik in a
    studio.
     
  8. tempus fugit

    tempus fugit Guest

    Hi Dr.
    These are things that I wrestle with - I know that an opamp is what you say,
    so how can it sound different? My 1st mic pre was an all transistor design.
    It never really got off the ground, and after much work and frustration, I
    tried an 072, which sounded infinitely better than my transistor design. I'm
    a little more knowledgeable in electronics now though, so I was
    wondering....
    This is why I wonder what the big deal is about discrete - and why I
    wouldn't mind trying one just to see.
    I record primarily classical guitar, but am looking at recording a rock band
    in the near future.
     
  9. Guest

    Back in '82 I built a transformerless mic preamp to use with my
    Technics SVP-100 digital audio recorder. I was messing with a
    commercial mic preamp with a Triad (not Jensen) input transformers and
    was measuring nasty THD at 30 Hz-- like 10% or more. Looking back now
    it may have been 60 Hz hum raising the noise floor since the HP THD
    meter simply nulled out the fundamental and called the rest distortion.


    I copied the mic preamp out of an EV (Tapco?) board but used matched
    transistors and resistors. With the AKG C451s, it hears better than I
    do (even 23 years ago when I could still hear). The transformerless
    unit measured 0.01% at 30 Hz. BTW, I was recording classical pipe organ
    with true 32' stops (17 Hz) so the low THD seemed needed. It certainly
    made some nice recordings

    I have som SSM2017s to rebuild that thing 'real soon now'.
    Where's my round tuit?

    Good Luck
    GG
     
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

-