Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

How do you multiply the frequency of an oscillator?

  1. Jan 1, 2013 #1
    I am building a HAM radio transmitter. I have noticed most crystal oscillators above 100mhz are very hard to find. Is there any way to multiply an oscillator's output, say, four times?
     
  2. jcsd
  3. Jan 1, 2013 #2

    berkeman

    User Avatar

    Staff: Mentor

    A typical approach is to use a PLL-based freqency synthesizer IC.
     
  4. Jan 1, 2013 #3
    I have been searching, and searching, and searching the internet on PLL devices. I can't find anything that simply explains how to make and use one. It's just too complicated for me to understand. Is there a simpler way to do it? I can recall something about using using a non-linear device to produce harmonics then using a bandpass filter to filter the correct frequency?
     
  5. Jan 1, 2013 #4

    Averagesupernova

    User Avatar
    Gold Member

    Xtal oscillators around 100 Mhz will use a 3rd or 5th overtone scheme. After that the signal can be multiplied. Common is X2, X3, X5.
     
  6. Jan 1, 2013 #5

    berkeman

    User Avatar

    Staff: Mentor

    Please try a bit harder. I gave you the search terms to use. If you need spoon feeding, I can give you direct links. Sorry to be harsh, but it's not that hard.
     
  7. Jan 1, 2013 #6
    Yeah but how do you multiply it?
     
  8. Jan 1, 2013 #7

    berkeman

    User Avatar

    Staff: Mentor

    With a PLL frequency synthesizer IC. Why are you ignoring my posts?
     
  9. Jan 1, 2013 #8
    Well, as I said in reply to your post, it was too much for me to comprehend how a PLL works. Is there an alternative?
     
  10. Jan 1, 2013 #9

    berkeman

    User Avatar

    Staff: Mentor

    No. I will try to post some helpful links tomorrow. All you have to do is Google my search terms though...
     
  11. Jan 1, 2013 #10
  12. Jan 1, 2013 #11

    dlgoff

    User Avatar
    Science Advisor
    Gold Member

    [offtopic]
    Aren't there some ham class license that require crystal controlled oscillators? Been a long time since I was a novice; CW keying, crystal controlled.
    [/endofftopic]
     
  13. Jan 1, 2013 #12

    NascentOxygen

    User Avatar

    Staff: Mentor

    That is a multiplier based on a PLL, certainly. But it's a clock multiplier, and "clock" implies digital electronics and square waves. You are needing an oscillator for ham radio. Oscillators for radio involve sine waves, not square waves. Sine waves are smooth, undulating waveforms containing a single radio frequency. Square waves have sharp sides and flat tops, and contain dozens of radio frequencies, many more than you would believe. (Yes, it is possible to convert a square wave to a sinewave, but for this you need another PLL arrangement, and to get a pure sinewave is not so simple.)

    I think you should look for a ready-made transmitter that has been designed exactly for what you want, or at least a published circuit specific to what you need. There must be plenty of people in the same boat as yourself. I'm sure you'll find an overtone crystal oscillator designed for the bands you will be licensed to use. A crystal oscillator means that the frequency is tightly fixed so you cannot* accidently transmit outside the ham band, perhaps causing havoc. "Oscillator" means it produces a nice sine wave (providing it is working properly, of course).

    Your first transmitter should be simple to understand and low-powered; it should not be your own design. RF circuits often need a lot of troubleshooting and tweaking to get them operating properly, and for this you need test equipment, patience, and an experienced mentor to guide you.

    Have fun! :smile:
     
    Last edited: Jan 1, 2013
  14. Jan 1, 2013 #13

    vk6kro

    User Avatar
    Science Advisor

    Any waveform that is not a sine wave contains harmonics of the fundamental frequency. A 1000 Hz signal that is not a sine wave contains exact multiples of 1000 Hz.

    So, to produce multiples of a sinewave's frequency, you just have to distort it.

    A square wave contains the fundamental frequency plus odd harmonics. Other waveforms contain different proportions of odd and even harmonics.

    If you took a transistor with a tuned circuit resonant at 100 MHz in the collector and applied a large 50 MHz sine wave to the base, the result would be 100 MHz output from the collector.
    Simple parallel tuned circuits are not perfect, though, and there would also be detectable signals at 50, 150, 200, 250 (and so on) MHz.

    There are other types of frequency multiplier.
    A full wave rectifier produces large amounts of second harmonic.
    Diodes whose capacitance varies with voltage, but in a non linear manner, can also be used to generate harmonics.
     
    Last edited: Jan 1, 2013
  15. Jan 2, 2013 #14
    A ready-made transmitter would be the right choice, but one good enough for my purposes will be too expensive. I am going to attach the transmitter to a high altitude balloon so I was hoping to make it myself. I won't be getting the transmitter back so a disposable module is what I am trying to achieve. This is what I am trying to copy: http://www.swharden.com/blog/2010-07-14-high-altitude-balloon-transmitter/
    But the can oscillator he uses is below the frequency range of my base station. I thought it would be simple enough to find a VHF can oscillator, but it's not. And after learning about these transmitters I found out that, as you mentioned, square waves are a big no-no.

    Here is what I was going to do: https://www.sparkfun.com/products/9261 + https://www.sparkfun.com/products/9089 + http://www.radioshack.com/product/index.jsp?productId=2062616
     
  16. Jan 2, 2013 #15

    berkeman

    User Avatar

    Staff: Mentor

    Hi Don,

    Frequency synthesizers are crystal controlled. From the datasheet linked to in tacky's post:

    You choose the input crystal frequency and the multiplier-divider ratios in the synthesizer to give you the desired output frequency.

    Although, as correctly pointed out by others, the output of the synthesizer is digital, and needs filtering to be used as the TX carrier waveform.
     
  17. Jan 2, 2013 #16
    Using a digital "clock" signal followed by 4th order filter is not uncommon in QRP designs.

    You will have to design and test the filter, do you have an oscilloscope/spectrum analyzer?

    Here are two examples.
    http://www.amqrp.org/kits/38spcl/schematic.html [Broken]
    http://www.qsl.net/qrp/tx/logi-tx.htm


    Those geniuses at that swharden.com link you provided paralleled outputs of a 74HC244 and sent this directly to their antenna, unterminated and unfiltered. Yikes!
     

    Attached Files:

    Last edited by a moderator: May 6, 2017
  18. Jan 2, 2013 #17
    Interesting. I do have an oscilloscope but I don't think it would be capable of reading 200 mhz signals.
     
    Last edited by a moderator: May 6, 2017
  19. Jan 2, 2013 #18
    Exactly which HAM band are you targeting?

    Why not use 10meters (28MHz)?
     
  20. Jan 2, 2013 #19
    1.25 meters. I would use 10 meters, but my receiver can only receive between 52 - 2200 MHz.
     
  21. Jan 2, 2013 #20
    Your balloon is TX only right? So on the ground you just need a receiver.

    You can get a 10meter receiver kit for $50.

    Plus, it is much easier to work at 28MHz vs. 222MHz (less worries about layout, parasitics etc.).

    Since it is a balloon the antenna can be pretty long.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: How do you multiply the frequency of an oscillator?
  1. Frequency Multiplier (Replies: 13)

  2. Frequency multiplier (Replies: 12)

  3. Frequency Multiplier? (Replies: 4)

  4. Frequency multiplier (Replies: 7)

Loading...