Connect with us

What is modulator?

Discussion in 'Electronic Basics' started by boki, Sep 15, 2003.

Scroll to continue with content
  1. boki

    boki Guest

    What is modulator?
     
  2. grahamk

    grahamk Guest

  3. Modulation is the addition of information (or the signal) to an electronic
    or optical signal carrier. Modulation can be applied to direct current
    (mainly by turning it on and off), to alternating current, and to optical
    signals. One can think of blanket waving as a form of modulation used in
    smoke signal transmission (the carrier being a steady stream of smoke).
    Morse code, invented for telegraphy and still used in amateur radio, uses a
    binary (two-state) digital code similar to the code used by modern
    computers. For most of radio and telecommunication today, the carrier is
    alternating current (AC) in a given range of frequencies. Common modulation
    methods include:
    a.. Amplitude modulation (AM), in which the voltage applied to the carrier
    is varied over time
    b.. Frequency modulation (FM), in which the frequency of the carrier
    waveform is varied in small but meaningful amounts
    c.. Phase modulation (PM), in which the natural flow of the alternating
    current waveform is delayed temporarily
    These are sometimes known as continuous wave modulation methods to
    distinguish them from pulse code modulation (PCM), which is used to encode
    both digital and analog information in a binary way. Radio and television
    broadcast stations typically use AM or FM. Most two-way radios use FM,
    although some employ a mode known as single sideband (SSB).

    More complex forms of modulation are Phase Shift Keying (PSK) and Quadrature
    Amplitude Modulation (QAM). Optical signals are modulated by applying an
    electromagnetic current to vary the intensity of a laser beam.

    A computer with an online or Internet connection that connects over a
    regular analog phone line includes a modem. This term is derived by
    combining beginning letters from the words modulator and demodulator. In a
    modem, the modulation process involves the conversion of the digital
    computer signals (high and low, or logic 1 and 0 states) to analog
    audio-frequency (AF) tones. Digital highs are converted to a tone having a
    certain constant pitch; digital lows are converted to a tone having a
    different constant pitch. These states alternate so rapidly that, if you
    listen to the output of a computer modem, it sounds like a hiss or roar. The
    demodulation process converts the audio tones back into digital signals that
    a computer can understand. directly.



    More information can be conveyed in a given amount of time by dividing the
    bandwidth of a signal carrier so that more than one modulated signal is sent
    on the same carrier. Known as multiplexing, the carrier is sometimes
    referred to as a channel and each separate signal carried on it is called a
    subchannel. (In some usages, each subchannel is known as a channel.) The
    device that puts the separate signals on the carrier and takes them off of
    received transmissions is a multiplexer. Common types of multiplexing
    include frequency-division multiplexing (FDM) and time-division multiplexing
    (TDM). FDM is usually used for analog communication and divides the main
    frequency of the carrier into separate subchannels, each with its own
    frequency band within the overall bandwidth. TDM is used for digital
    communication and divides the main signal into time-slots, with each
    time-slot carrying a separate signal.
     
  4. Steve

    Steve Guest

    The simple explanation;

    A modulator "multiplys" two signals together into one. It is
    sometimes called a "multiplier".

    For example; When you tune into a radio station, you are tuning your
    radio to receive the "carrier frequency." Mixed with this carrier is
    the audio signal that you hear (music or voice). The radio extracts
    the audio signal from the "modulated" carrier.

    The purpose of the modulator is to combine the carrier and the audio
    (message) into one signal.

    In AM (amplitude modulation) the carriers AMPLITUDE varies in
    proportion to the audio signal.

    In FM (frequency modulation) the carriers FREQUENCY varies in
    proportaion to the audio signal.

    The basic reason for modulation is so that the low frequency audio
    signal can be transmitted using a high frequency carrier.

    There are many forms of modulation, and hence many types of modulator.
    Further info might be found by searching via www.google.com.
     
  5. A modulator is an object that converts the information encoded in a
    signal into another code. It can either be digital or analogue. The main
    reason for doing this is that transmission of the signal is usually more
    optimum in the coded form than the non-encoded form.

    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.
     
  6. Dbowey

    Dbowey Guest

    niftydog posted, in part:
    << In AM (amplitude modulation) the carriers AMPLITUDE varies in
    proportion to the audio signal. >>

    In AM modulation, the carrier amplitude does NOT vary - it is constant. The
    process of modulation produces sidebands.

    To answer "what is a modulator" depends on your perspective.

    A "modulator" for a class C RF amplifier is nothing more nor less a voiceband
    amplifier having a suitable output transformer to match the transmitter's Class
    C output stage.

    An end-user Video "modulator" is a device that will accept a video signal and
    output, usually, a chan 3 or 4 TV band signal.

    And the list goes on......

    You should try to refine your question if you want to receive the correct
    answer.

    Don
     
  7. Steve

    Steve Guest


    I understand, but given the broken english in the OPs question I was
    trying to give as simple an answer as possible so they might glean the
    basic concept. I can appreciate how my statement would have misled
    people.

    Looking at a simple AM envelope on a CRO gives a beginner the
    impression that the amplitude is changing in proportion to the
    message. However, when you superimpose the message on the envelope,
    it's easier to 'see' what's actually happening. Having a spec. an.
    would be handy too! Explaining the concepts and maths behind it all
    seems to be beyond the understanding of the OP at this time.

    Maybe folks would benefit from reading this;

    http://www.whatis.com/definition/0,,sid9_gci214073,00.html
     
  8. This is more a matter of semantics. If a signal is Vpk(t).Sin(wt), if we
    don't call the Vpk(t) the amplitude of Sin(wt), just what do we call it?
    What do we call the Sin(wt) term if not the "carrier"?
    That's because it is. The peak to peak amplitude of the complete signal
    is indeed changing. Thats what AM means. Sure, you can make an argument
    that the signal is composed of different components or "signals" in the
    frequency domain, only some of which has changing amplitudes, but in the
    time domain, there is only one "signal", i.e the one voltage/current at
    any instant of time.

    The main point here is a mathematical identity relating sin(x).sin(y) ->
    sin(x-y) + sin(x+y)

    I don't see how it is technically any more valid, language wise, to say
    that the sin(x-y) and sin(x+y) terms have amplitude terms that are
    varying, rather than saying the sin(y) term has an amplitude term that
    varies. Mathematically, they are identical descriptions. For lack of any
    other word, one is pretty much compelled to refer to the sin(y) term as
    the "carrier". "The frequency in absence of any modulation frequency",
    seems a bit much.

    From a practical point of view, one can certainly measure the average
    frequency term that forms sin(y), and separately its Vp(t) amplitude
    term, just as easily, in principle, as one can measure the amplitudes
    and frequencies in the conventional descriptive. Both measurements sets
    form a correct description of the signal. What is the more "real" one?

    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.
     
  9. Dbowey

    Dbowey Guest

    Kevin posted much irrelevant STUFF, but I choose to comment on the following:
    << The peak to peak amplitude of the complete signal
    is indeed changing. Thats what AM means. Sure, you can make an argument
    that the signal is composed of different components or "signals" in the
    frequency domain, only some of which has changing amplitudes, but in the
    time domain, there is only one "signal", i.e the one voltage/current at
    any instant of time.
    Sure, if we ignore that a scope is essentially a not-very-smart AC voltmeter,
    we might be satisfied that we have the entire answer.

    That's what AM means? I believe the first guys that modulated an oscillator
    and heard the result on their diode detector did think the ampltude of the
    carrier varied and "Amplitude Modulation" was a descriptive term at the time.
    It doesn't mean we should wear blinders and ignore the realities of the
    frequency domain view.

    Today there are many ways to observe that the sidebands are separate from the
    carrier, and one does not require a spectrum analyzer to do this.


    << The frequency in absence of any modulation frequency",
    seems a bit much. >>

    What's the problem? A frequency in absence of any modulation is simply an
    unmodulated signal; a "carrier" awaiting modulation or a signal waiting to be
    keyed to send code.

    Don
     
  10. Its deeper than this, and yes, I saw this answer coming.
    All due respect here, this is old hat, and not the point I am
    addressing.
    I didn't say that. What I indicated was that a Fourier expansion is no
    more *physically* valid than many other views.
    Not the point. An expansion in Fourier co-efficients is not a unique way
    to analyse a function.
    You really missed the point here. Its much more involved. It about what
    is actually real, verses what is an artefact of the particular measuring
    method. If I design a piece of equipment to pick out the amplitudes of
    single frequency sine waves, i.e the co-efficients of a Fourier
    expansion, it does not mean that they are any more real then if I design
    a piece of equipment to, say, pick out Bessel co-efficients from a
    signal. It also does not imply that they are not real. Its entirely
    arbitrary.

    There is nothing unique in deciding to resolve f(t) into an orthogonal
    expansion of sine(nw) and cosine(nw) waves, and unilaterally declare
    that such co-efficients of the expansion are the real truth. Its just
    one way of many, to look at the problem.

    For example, I could equally well decide to expand f(t) in Walsh
    functions, construct an analyser that outputs Walsh function
    co-efficients and declare that they represent the real truth. In fact,
    for a periodic waveform, I could use any suitable set of periodic
    functions, e.g a set of orthogonal elliptic functions.

    Indeed, if I decide to claim that the Walsh function co-efficient
    analyser are the real truth, no real experiment will contradict that
    view. f(t) can be described exactly by those co-efficients just as
    validly as the Fourier co-efficients can. The only real blemish is that
    it is a bit more complicated because circuits with L's and C's,
    preferentially pick out Fourier co-efficients, not Walsh co-efficients.
    However, this does not mean that Fourier co-co-efficients have any
    better validity or stronger physical meaning. As noted, a Fourier
    analysis is just one of an infinite number of ways of describing f(t) by
    a sum of orthogonal functions. The prejudice caused by familiarly of
    using Fourier co-efficients just doesn't make it any more valid than any
    other view.

    In the real world we construct observations of objects to to match how
    we suppose those objects to behave. Our description of such objects
    turns out to be not unique. For example, the shrodinger and hiesenburg
    pictures of Quantum Mechanics, are fundamentally different, yet
    mathematically, they are identical in that they predict exactly the same
    results.

    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.
     
  11. Its deeper than this, and yes, I saw this answer coming.
    All due respect here, this is old hat, and not the point I am
    addressing.
    I didn't say that. What I indicated was that a Fourier expansion is no
    more *physically* valid than many other views.
    Not the point. An expansion in Fourier co-efficients is not a unique way
    to analyse a function.
    You really missed the point here. Its much more involved. It about what
    is actually real, verses what is an artefact of the particular measuring
    method. If I design a piece of equipment to pick out the amplitudes of
    single frequency sine waves, i.e the co-efficients of a Fourier
    expansion, it does not mean that they are any more real then if I design
    a piece of equipment to, say, pick out Bessel co-efficients from a
    signal. It also does not imply that they are not real. Its entirely
    arbitrary.

    There is nothing unique in deciding to resolve f(t) into an orthogonal
    expansion of sine(nw) and cosine(nw) waves, and unilaterally declare
    that such co-efficients of the expansion are the real truth. Its just
    one way of many, to look at the problem.

    For example, I could equally well decide to expand f(t) in Walsh
    functions, construct an analyser that outputs Walsh function
    co-efficients and declare that they represent the real truth. In fact,
    for a periodic waveform, I could use any suitable set of periodic
    functions, e.g a set of orthogonal elliptic functions.

    Indeed, if I decide to claim that the Walsh function co-efficient
    analyser are the real truth, no real experiment will contradict that
    view. f(t) can be described exactly by those co-efficients just as
    validly as the Fourier co-efficients can. The only real blemish is that
    it is a bit more complicated because circuits with L's and C's,
    preferentially pick out Fourier co-efficients, not Walsh co-efficients.
    However, this does not mean that Fourier co-co-efficients have any
    better validity or stronger physical meaning. As noted, a Fourier
    analysis is just one of an infinite number of ways of describing f(t) by
    a sum of orthogonal functions. The prejudice caused by familiarly of
    using Fourier co-efficients just doesn't make it any more valid than any
    other view.

    In the real world we construct observations of objects to to match how
    we suppose those objects to behave. Our description of such objects
    turns out to be not unique. For example, the shrodinger and hiesenburg
    pictures of Quantum Mechanics, are fundamentally different, yet
    mathematically, they are identical in that they predict exactly the same
    results.

    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.
     
  12. Its deeper than this, and yes, I saw this answer coming.
    All due respect here, this is old hat, and not the point I am
    addressing.
    I didn't say that. What I indicated was that a Fourier expansion is no
    more *physically* valid than many other views.
    Not the point. An expansion in Fourier co-efficients is not a unique way
    to analyse a function.
    You really missed the point here. Its much more involved. It about what
    is actually real, verses what is an artefact of the particular measuring
    method. If I design a piece of equipment to pick out the amplitudes of
    single frequency sine waves, i.e the co-efficients of a Fourier
    expansion, it does not mean that they are any more real then if I design
    a piece of equipment to, say, pick out Bessel co-efficients from a
    signal. It also does not imply that they are not real. Its entirely
    arbitrary.

    There is nothing unique in deciding to resolve f(t) into an orthogonal
    expansion of sine(nw) and cosine(nw) waves, and unilaterally declare
    that such co-efficients of the expansion are the real truth. Its just
    one way of many, to look at the problem.

    For example, I could equally well decide to expand f(t) in Walsh
    functions, construct an analyser that outputs Walsh function
    co-efficients and declare that they represent the real truth. In fact,
    for a periodic waveform, I could use any suitable set of periodic
    functions, e.g a set of orthogonal elliptic functions.

    Indeed, if I decide to claim that the Walsh function co-efficient
    analyser are the real truth, no real experiment will contradict that
    view. f(t) can be described exactly by those co-efficients just as
    validly as the Fourier co-efficients can. The only real blemish is that
    it is a bit more complicated because circuits with L's and C's,
    preferentially pick out Fourier co-efficients, not Walsh co-efficients.
    However, this does not mean that Fourier co-co-efficients have any
    better validity or stronger physical meaning. As noted, a Fourier
    analysis is just one of an infinite number of ways of describing f(t) by
    a sum of orthogonal functions. The prejudice caused by familiarly of
    using Fourier co-efficients just doesn't make it any more valid than any
    other view.

    In the real world we construct observations of objects to to match how
    we suppose those objects to behave. Our description of such objects
    turns out to be not unique. For example, the shrodinger and hiesenburg
    pictures of Quantum Mechanics, are fundamentally different, yet
    mathematically, they are identical in that they predict exactly the same
    results.

    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.
     
  13. Jim Large

    Jim Large Guest

    Guess what guys, you're both right. The time-domain
    explanation and the frequency domain explanation are equally
    valid. Let's imagine, for example, that we've got a carrier
    wave of A Hz that is 100% modulated by a pure sine wave of B
    Hz where B << A. The time domain explanation says that the
    amplitude of the carrier wave slowly changes over time. The
    instantaneous value of the signal at time t is:

    COS(At)*(COS(Bt)+1)
    -------------------
    2

    The frequency domain explanation says something different.
    It says you're putting half of your power into an
    unmodulated carrier wave, and the rest into two symmetric
    sidebands at A+B and A-B Hz:

    2 COS(At) + COS((A-B)t) + COS((A+B)t)
    -------------------------------------
    4

    Well guess what. Both equations are equal. They both
    describe the exact same signal. I don't know the proof, but
    I do know that it's not exactly cutting edge science.
    Joseph Fourier could have proved it a hundred years before
    AM radio was invented.

    -- Jim L.
     

  14. Err..I know...That's exactly what I said, ho..humm....:)

    That's right, and there's more. There is no unique way of describing the
    universe. I expanded on this in a follow-up post, which, for some
    reason, has appeared 3 times.

    A point here is that the Fourier view of the universe in EE is so
    abundant, that to see things in any other way needs a paradigm shift.
    The use of the word bandwidth, implicitly implies that the Fourier view
    is being used. If I chose to expand f(t) in say, an orthogonal set of
    elliptic functions, bandwidth would not even be a valid term, yet the
    resulting description of f(t) would still be just as correct, although
    rather a bit obtuse.


    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.
     
  15. John Fields

    John Fields Guest

    ---
    It still is, and it still describes the process. For example, let's say
    we have a 1MHz carrier with an amplitude of 100VPP. Then let's say we
    lower the plate voltage of the final amplifier reduce the amplitude of
    the carrier to 50VPP. Then let's say we further lower the plate voltage
    and reduce the amplitude of the carrier to 0VPP. Have we not modulated
    the amplitude of the carrier? Of course we have!

    Now let's say that our plate voltage controller (our modulator) can vary
    the plate voltage of the final sinusoidally at 1000Hz. so that at the
    positive peak of the sinusoid the carrier voltage is at 100VPP and at
    the negative peak of the sinusoid the carrier voltage is at 0VPP. Have
    we modulated the amplitude of the carrier? You bet your ass we have!
    have we generated sidebands at 1MHz. +/- 1kHz? You bet your ass we
    have!

    Were those guys with the diode detectors detecting the amplitude
    variations of the carrier? You bet your ass they were, they didn't have
    filters selective enough not to!
     
  16. Dbowey

    Dbowey Guest

    John Fields posted:
    Never mind.... it didn't add anything to the discussion.

    Don
     
  17. John Fields

    John Fields Guest

     
  18. dyson

    dyson Guest

    The biggest problem with the simple 'Fourier' view of a time
    varying amplitude modulation is that the Fourier view changes
    in character (and cannot directly describe the f(t) AM process.)
    If everything is constant, then the Fourier view can make some
    sense.

    I agree with you that either the f(t), f(w) or whatever views
    are equivalent, but f(w) becomes quite complex when the f(t)
    signals aren't constant in character. It is best to keep full
    understanding of the various processes, and decide when the f(w)
    view is the best way of viewing things, and when the f(t) way
    is good.

    In the more general case, f(t) will help to solve problems that
    f(w) will give LOTS of trouble -- but the f(w) is a good set of
    shortcuts.

    I agree with Kevin, that in the steady state, f(w) and f(t) can
    be equivalent.

    John
     
  19. Dbowey

    Dbowey Guest

    jfields posted:
    [/QUOTE][/QUOTE]


    I don't believe you set me straight....... Hmmmm....Nope. In my humble
    opinion, you are not correct.

    Yes, sidebands can be generated by turning the carrier on and off or changing
    its amplitude by changing the power supply voltage, but that is not AM as it is
    generally used; it's closer kin is CW, as in International Morse transmission.


    Don
     
  20. Technically, f(t) and its Fourier transform g(w) cannot explain all of
    the same details at the same time. This is due to an inherent
    uncertainty of the transform pair. That is

    df.dt >=1/2

    That is, the standard deviation of the signals time distribution dt,
    times the standard distribution of its frequency distributing df has a
    minimum value. That is, knowing the frequency distribution accurately,
    means you have larger uncertainty in the time distribution.

    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.
     
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

-