Connect with us

Basic questions about microphones and mic preamp design...

Discussion in 'Electronic Design' started by Jaye Gallagher, Dec 13, 2004.

Scroll to continue with content
  1. Hi there,

    I'm wanting to be able to connect a microphone to an recording device I'm
    thinking of designing based around the AD1871 24/96kHz ADC. The AD1871 is
    great in that it's very easy to configure it to accept either single ended
    or differential line level input, however, I'd also like to be able to
    attach a microphone to this thing, and that seems to be where things get
    complicated.

    Essentially, I need a microphone pre-amp of some kind, it's just a matter of
    what kind I want, and how on earth to scrounge a design for it. My questions
    are:

    * There seem to be two kinds of microphones in use -- condenser and dynamic.
    Can a single pre-amp design support both of these? What is the difference in
    a electrical sense?

    * There also seem to be two different ways of connecting a microphone to
    stuff -- single ended or differential. Again, can a pre-amp support oth of
    these, or only one?

    * What is phantom power, do all microphones require it, and it is only used
    in differential microphone connections?

    * All the pre-amp designs I've found seem to require at least +9V and -9V
    supplies to operate. I'm hoping to produce a portable device, and so
    operation from, say, 2 AA batteries is desirable. Is this impossible? What's
    going in things like my little Minidisc player that has a microphone input,
    and thus a preamp, but runs from a single AA?

    * Are there any IC based solutions to the above, since, as I'm hoping for a
    portable device, board space is at a premium.

    Any help, opinions or rampant speculation greatly appreciated. Miraculous
    delivery of useful schematics may be rewarded with sexual favours ;).

    Thanks,
    Jaye.
     
  2. You may need to supply phantom power to a condenser mic. (I say 'may'
    because you have the option of using an external phantom supply and
    some condenser mics contain batteries.)

    A single preamp design can support both of these.
    Any serious mic will be differential. A differential mic preamp can
    support a signal ended mic by grounding the other input. This is
    sometimes achieved with a special cable (i.e. the preamp is always run
    as differential).
    Phantom power is a way of supplying power from the preamp to the mic
    using the same wires that are used for the signal (hence 'phantom').

    It is needed if (1) there is a condenser mic that needs a high voltage
    bias (typically 48V) or (2) the mic has a preamp in it (typically 12V)
    e.g. Sennheiser MKH series.

    There are standards that defined the electrical characteristics of
    the "P48" interface: IEC 268-15, EN 61938, etc. It is a three wire
    system, supplying +48V as a common mode signal on the signal wires,
    with the common as return.

    The 12V phantom is a true two wire system. The third connection is
    used solely for shielding.
    I expect it doesn't have very good performance.
    AD, nee PMI SSM2017 (now obsolete!) or SSM2019
    http://www.analog.com/en/prod/0,2877,SSM2019.html

    Thatcorp 1510:
    http://www.thatcorp.com/1510desc.html

    These are very high end devices. Since you say minidisc, you might be
    after something with lower performance.

    The app notes for those parts should give you some ideas though. I
    suggest you read them very closely (particularly the thatcorp ones).
    The biggest mistake I've seen in supposedly 'professional' preamp
    designs is too much gain before the first variable gain stage.
    There's nothing you can do if the first stage in your preamp clips.

    WHAT DO YOU MEAN, THE MICROPHONE IS PUTTING OUT HALF A VOLT?

    Regards,
    Allan
     
  3. Jaye Gallagher wrote:
    (snip)
    There are quite a few variations of microphone standards. you should
    be familiar with them before you decide exactly which you are going to
    support with this design. I suggest you start with a google search
    for [microphone tutorial].
     
  4. I read in sci.electronics.design that Allan Herriman <allan.herriman.hat
    > wrote (in <057qr09dbrcu8pv89utpqadfbigchnco
    >) about 'Basic questions about microphones and mic preamp
    design...', on Mon, 13 Dec 2004:
    IEC 268-15 (aka 60268-15) was replaced by IEC 61938, and EN 61938 is a
    clone of that.
     
  5. Robert Baer

    Robert Baer Guest

    I suppose that few know that "SSM" stands for "Solid State Music", the
    same name given to a small surplus (and new goods) electronic store in
    Santa Clara, run by John Bergoun(sp?).
    John's interest was in organs, more explicitly solid state
    implimentations.
    As a result, he designed numerous solid state audio ICs, which he got
    Exar to make and sell.
    Other companies eventually copied those designs, using the same SSM
    designation that Exar used.
    **
    I agree; a gain of about 10 is sufficent for the first stage; tends
    to prevent clipping and decreases the NF added by the next stage by a
    factor of 10.
    **
    Any "micropnone" that has an audio output signal of 500mV already has,
    by definition, an amplifier stage built in, and that level is
    essentially LINE LEVEL.
     
  6. I read in sci.electronics.design that Robert Baer
    about 'Basic questions about microphones and mic preamp design...', on
    Mon, 13 Dec 2004:
    The source follower or common-source stage in an electret microphone can
    only just be described as an 'amplifier'; it converts the capacitive
    source of the capsule, 2.5 to 10 pF, to a manageable lowish impedance (1
    kohm to 5 kohm). But with a suitable supply voltage (e.g. 12 V), such a
    microphone can produce several volts if placed close to a brass
    instrument.
     
  7. I was actually thinking about a dynamic mic when I wrote that.

    Yes, I have done the measurement (Sennheiser MD421 + very loud noise)
    and seen the results on an oscilloscope.

    Regards,
    Allan
     
  8. Ban

    Ban Guest

    A dynamic mike works like a loudspeaker, a diaphragm moves a voicecoil in a
    gap with a magnetic field created by a permanent magnet. There are several
    different condensor mikes
    1.) Electret, which requires only a lowish(1.5-5V) voltage to operate a
    built-in FET stage which is supplied by batteries or +5V on the center pole
    of the connector(Sony). The output voltage of 1/4" diameter size is around 3
    to 10mV/Pa. Output is unbalanced jack or RCA. Price is very low. There are
    also musicians grade electrets with a built in transformer for balancing and
    phantom power or 2 AA cells.
    2.) real Condensor, needs a higher operating voltage, normally +48V phantom
    power. Has balanced output with XLR-connector. Price from 150$+.
    3.) RF-Condensor, made by Sennheiser, modulates a high frequen.cy. Balanced
    XLR, phantom powered. Price is high 300$+

    Can do both if done properly.
    It is only used with balanced lines. +48V, negative to Pin1(gnd), positive
    through 2 resistors 6k8 into both of the balanced conductors (Pin2 and 3).
    Depends on the dynamic range you need. You can make a preamp with a +2.5V
    only supply, but then you are more limited with the max. Sound Pressure
    Level vs. gain setting. If the mike puts out 20mV/Pa (1Pascal equals 94dB
    SPL) and your pre can output 0.31Veff without limiting, you can have +26dB
    gain, if your input stays below 94dB. A singer can easily output 120dB SPL,
    which limits your gain to 0dB, generally you want some additional
    "headroom", so the gain must be even lower. This raises OTOH the noise
    floor. Depending on the sensitivity of the mike and the required max.SPL,
    you have to find the best compromise.
    You can use any low noise opamp depending on the supply voltage range. You
    will need to decide for an electret with headphone jack only, because the
    XLR connectors are big and heavy.
    If you are Jaye Foucher, you are very welcome. :-])
     
  9. <Vgevd.286921$>) about 'Basic questions about
    microphones and mic preamp design...', on Mon, 13 Dec 2004:

    This is correct, but electret mics are often 'line-powered'. One
    arrangement is:
    ASCII art; use Courier font

    +----o-------o-+-4.7kohm------o +V
    mic | |
    |] +----100 nF----o audio
    +--------+------| FET cable
    | | |] pre-amp/mixer
    mic |O 470 Mohm |
    capsule | | |
    +--------+-------+---o-------o----------------o 0 V
     
  10. well, you can get up to 0dBu out of a microphone,

    http://www.gprime.com/proaudio/tubes/tubes.htm




    martin

    Serious error.
    All shortcuts have disappeared.
    Screen. Mind. Both are blank.
     
  11. The mics used in that test are (I believe) all condenser types, and
    would be expected to output robust levels.


    An advantage of preamp designs such as the SSM2019 is that their gain
    can be reduced to 0dB, which avoids clipping on such high level
    inputs.
    By switching only one resistor, the same preamp can be used for both
    mic and line level signals.

    Regards,
    Allan
     
  12. martin griffith wrote...
    "... peak readings of 1 volt or more were common, especially
    from close up microphones on voice and drums." Nice ref.

    Speaking of tube vs transistor sound under overload, what's a
    good comprehensive article that's newer than that 1973 tome?
     
  13. santec

    santec Guest

    Have a look at Rod Elliott's Webpage. I've build his Low Balanced
    Microphone Preamp (http://sound.westhost.com/project66.htm) myself. I'm
    running it from 2 9V Blocks (one for +9V and one for -9V) and it works
    just perfect. Added two 2k resistors from +9V to balanced inputs
    followed by two capacitors as phantom power supply.

    Cheers, Nico
     
  14. All the mics apart from the RX77 are condenser mics, the Rx77 is a
    ribbon.
    One problem with the 2019 is the high level of noise at unity gain

    from the data sheet
    f = 1 kHz; G = 1 50 nV/ vHz
    f = 1 kHz; G = 100 1.7 nV/ vHz

    So the less gain you use the more noisy the micamp is.
    Somewhat back to front!

    BTW I found a good write up on protection of the front end of mic amps
    from 48V glitches in TI's PGA2500 data sheet.



    martin

    Serious error.
    All shortcuts have disappeared.
    Screen. Mind. Both are blank.
     
  15. Ban

    Ban Guest

    I think this is really a very old article with a lot of unsubstantial myths.
    It is not true that you find tubes in mixing consoles today. Where you find
    them is in guitar amps and certain microphone preamps. The mics mentioned in
    the article are super expensive big sized condensor mikes, which have a lot
    more sensitivity than small 1/2 or 1/4" electrets.
    Generally spoken the discussion has moved to analog vs. digital recording
    gear not tubes vs. transistors any more.


    ciao Ban
     
  16. This is normal for an amplifier with the noise referred back to the
    input. It just means that (at low gains) the noise is dominated by
    something further down the chain, e.g. the output amplifier. Output
    referred noise doesn't change much with the gain.

    I'm not sure if this "high" noise is a problem. G = 1 would be used
    with +10dBu inputs, resulting in an SNR of about 110dB (assumes 20kHz
    noise bw).
    Good find.

    Regards,
    Allan
     
  17. I agree, I was only pointing out the that mics can pump out a
    reaonable level, and this has nothing to do with transistors or valve
    technology.
    if you ebay for akg microphones, you can get big sized condensrs for
    less than 200$, so not so "super expensive"

    If I had a studio ( those were the days!), I would like to run the
    same tests with slightly more elaborate equipment.


    martin

    Serious error.
    All shortcuts have disappeared.
    Screen. Mind. Both are blank.
     
  18. I always associated the "excess" noise at unity gain with the highish
    5K feedback resistor causing the noise. Change this to zero ohms, as
    in a voltage follower, and the noise would (might?) decrease
    significantly ,(I havent done the sums,)


    martin

    Serious error.
    All shortcuts have disappeared.
    Screen. Mind. Both are blank.
     
  19. Sorry Win
    I gave up my AES membership years ago, Icouldnt keep up with the maths
    Maybe Woodgate would have something more up to date.

    Urban myth...About triggering an old type photgraphic flash bulb from
    a mic in a bass drum, I wonder if they are true?



    martin

    Serious error.
    All shortcuts have disappeared.
    Screen. Mind. Both are blank.
     
  20. Try this link- http://sound.westhost.com/projects-8a.htm
    Elliot has many other links to designs on the site including limiters etc,
    all of which are low cost
    He also has good answers to the questions you asked.
    Regards
    Anthony
     
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

-