Connect with us

crystal oscillator

Discussion in 'Electronic Design' started by max, Jan 18, 2007.

Scroll to continue with content
  1. max

    max Guest

    Hi,

    Did anybody know how to design a colpitts or clapp crystal oscillator fo
    3rd overtone?? The frequency desired is about 100MHz.
     
  2. You're going at it backwards.

    You pick a type of oscillator to match what you want to do.

    So if it's an overtone crystal, you make sure you use an oscillator
    that will have it oscillating in the overtone mode.

    It might help, besides asking in sci.electronics.basics where this
    question likely really belongs, is to ponder what an overtone is,
    and what you are trying to do when getting a crystal to oscillate
    in overtone mode.

    Understand that, and you'll have a better idea of what you are
    seeking.

    Michael
     
  3. Don Bowey

    Don Bowey Guest


    Google do.
     
  4. I would not call crystal oscillator design basic...
     
  5. James Arthur

    James Arthur Guest

    Especially not at 100MHz, which is likely to be a 5th overtone
    design.

    James Arthur
     
  6. Ken Smith

    Ken Smith Guest

    You don't even need to go 3rd harmonic to get 100MHz. You can get
    fundamental 100MHz crystals.

    http://xtal.conwin.com/pages/crystals.html


    You can get 100MHz with a transistor like the 2N2222, making a fairly
    simple oscillator. The transistor doesn't have enough gain to take off at
    the 3rd.
     
  7. LVMarc

    LVMarc Guest

    yes
     
  8. Guest


    If you have installed LTspice SwitcherCadIII look in this directory,
    "C:\Program Files\LTC\SwCADIII\examples\Educational"

    There's a few example designs of colpitts and clapp oscillators.

    If you never heard of this free software get it here;
    http://www.linear.com/company/software.jsp
     
  9. James Arthur

    James Arthur Guest

    Those inverted mesa (I-M) crystals are cool, but for that kind of
    money why not buy Prince Albert in a can?

    http://catalog.digikey.com/scripts/partsearch.dll?Detail?name=CW507TR-ND

    Of course it all depends on our guerilla (hit-and-run) poster's
    needs: stability, quantity, power, accuracy, cost, designer's time,
    etc. For now, 100MHz AT overtone crystals have ~5-10x higher Qs, age
    better and cost less.

    OTOH, a 3rd overtone I-M looks pretty slick at 300MHz, and they look
    really interesting as filters.

    Thanks for the reminder -- last time I considered I-M they were
    boutique, exotic parts. I just read that they're down to only 3x the
    price, which isn't really all that bad.

    Best,
    James Arthur
     
  10. max

    max Guest

    Due to certain reason, I had to used the 3rd crystal. And I need thi
    oscillator circuit to drive my IC which had a build in transistor inside.

    right now, I can't make the circuit work...so need some help on th
    calculation for the capacitance and inductance.
     
  11. max

    max Guest

    Due to certain reason, I had to used the 3rd crystal. And I need thi
    oscillator circuit to drive my IC which had a build in transistor inside.

    right now, I can't make the circuit work...so need some help on th
    calculation for the capacitance and inductance.
     
  12. Ken Smith

    Ken Smith Guest

    Only because the OP said he wanted to make a crystal oscillator.

    The GreenRay OCXO is a lot better, if cost is no object.

    [...]
    I think that will be true basically forever. The Q of an overtone will
    always come out higher. You do have to be very careful in your frequency
    selector section to make sure that it doesn't end up adding a bunch of
    close in noise to take advantage of this though.
     
  13. Ken Smith

    Ken Smith Guest

    Due to certain reason, I had to used the 3rd crystal. And I need this
    oscillator circuit to drive my IC which had a build in transistor inside.

    right now, I can't make the circuit work...so need some help on the
    calculation for the capacitance and inductance.[/QUOTE]

    I assume that the L and C in question are in the frequency selector
    network.

    The first step is to ignore the issue and replace the LC circuit
    with a magic resistor. This magic resistor has no resistance at DC and
    some fixed value at high frequencies.

    You may need to place a real capacitor or inductor in series or parallel
    with this magic resistor to make the oscillator design workable.

    Now, you work out what resistance the magic resistor must have for proper
    operation.

    Now the problem is one of making an LC circuit with the needed resistance
    at resonance and the right resonant frequency. Generally, you want to go
    for a low Q tuned circuit. If this is AT cut, the Q can be quite low.
    Lets say 3. With SC cut crystals the Q needs to be higher. Lets say 20.

    The exact Q you want is enough to ensure that every crystal the meets spec
    must run on the desired frequency but not much more. A higher Q tuned
    circuit will have a rapid phase vs frequency slope. This will allow it to
    start causing drift.
     
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

-