Connect with us

Confusing waveform on oscilloscope

Discussion in 'Electronic Design' started by Paul Burridge, Oct 17, 2003.

Scroll to continue with content
  1. Hi chaps,

    I'm getting two different displays for the same waveform. I'm looking
    at the output of a slightly dirty 40Mhz oscillator. I've come across
    this knob called "holdoff" which when in one position, displays the
    wave as a single, sharp, slightly irregular sine wave; but in another
    position shows about 4 or 5 distinct, regular sine waves all
    superimposed on each other, in phase but with slightly differing
    amplitudes. Which is the *real* waveshape that accurately reflects
    what the oscillator's actually putting out?
    I don't understand what half the knobs and buttons on my scopes are
    actually supposed to do. :-/
  2. Fred Bloggs

    Fred Bloggs Guest

    The HOLDOFF feature is there to allow you to zoom in on waveform details
    with a separate timebase. There are two timebase settings, the
    highlighted/intensified portion of the waveform is there to show you the
    extent of the HOLDOFF expanded (faster) timebase sweep which should also
    be displayed. The "4 or 5" distinct sine waves you observe are due to
    the fact that your scope timebase resets and retriggers the sweep on
    amplitude level in a shorter time period then the periodic amplitude
    modulation of the waveform. You can pick out a better idea of this
    amplitude modulation by increasing the timebase to close to an integer
    multiple of the suspected period then "uncalibrate" the timebase to
    vernier the display into a stable waveform.
  3. Baphomet

    Baphomet Guest

    Probably, the single sine wave is the proper display. It sounds like the
    scope sweep is free running or recurrent (not correctly triggered) when you
    are seeing multiple displays of the same waveform.
  4. Holdoff is a very useful but widely un-understood feature.

    If the signal of interest is a nice clean repetetive waveform, with each and
    every cycle identical, then you don't need Holdoff.

    But if you have a waveform that has say, four cycles, all slightly
    different, without holdoff your scope will tend to trigger randomly on cycle
    1, cycle 2, etc... and all the slightly different size or shape cycles will
    be superimposed, leading to a fuzzy looking line.

    But if you turn on a bit of Holdoff, at just the right setting the scope
    will be coerced into triggering on the SAME cycle of the sequence every
    time, so you'l lget a clean and stable display of all X cycles. "Holdoff"
    means the sweep is held off from triggering on the next avilable cycle,
    instead it waits for the amount of time you've dialed in.

    You can do roughly the same thing with the VAR sweep time, to force the
    sweep to be within one cycle of the repeat-length. But then you're limited
    to seeing exactly one or two or N full cycles. With Holdoff, you can zoom
    in on one cycle or less, and the holdoff will add the right amount of delay
    so the scope doesnt trigger on any of the cycles you DONT want to see right

    If the irregular waveform you're watching has a little extra peak on one
    cycle, you can do these tricks too with the trigger level control. By
    carefully setting it, you may be able to get it to uniquely trigger on the
    one cycle that stands out. But again it's not as stable as using Holdoff to
    do the job.

    You can send me any unused Tek knobs, my scopes need a few of those
    hard-to-find knobs.


  5. Ian Stirling

    Ian Stirling Guest

    You should know that the proper etiquette in this group is to therefore
    declare that you've discovered FTL, and put up some coax on ebay for
    some large amount of money.
  6. Fred Abse

    Fred Abse Guest

    You're thinking of the delayed timebase function, Fred. The holdoff
    control varies the time at the end of one sweep before the timebase will
    accept another trigger. If you like, the "dead" time.

    It can be useful if there are several events in a cycle that will trigger
    a sweep, which can't be selected out by level alone, and you want to trigger
    on the same event in each cycle. Using the variable "vernier" time control
    can sometimes do much the same thing, but at the expense of uncalibrated
  7. [snip]
    Thanks for the comprehensive explanation, George. Most interesting
    indeed. I'm sure there's a lot of use I'm not getting out of these
    instruments owing to my limited knowledge of how to use them.
    I'll bear it in mind. I may have an older Tek coming up for spares
    shortly. The one I'm currently using is a bit too up-to-date to start
    nicking bits off. :)
  8. FTL coax, eh? Sounds like a major contribution to speeding up
    communications. :)
  9. Fred Bloggs

    Fred Bloggs Guest

    Ehh- he threw me off with the "highlighted" trace- sounded like dual
    timebase sweep B DELAY from A.
  10. Have we got more than one contributor calling himself "Fred Bloggs"
    then? The first reference to a "highlighted trace" (sic) comes from a
    Fred Bloggs. Unless there's only one Fred Bloggs that happens to
    suffer from multiple personality disorder in referring to himself in
    the third-person.
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