Tuned oscillator design & calculation & barkhausen criteria ? ?

Discussion in 'General Electronics Chat' started by hkntrt, Jul 5, 2012.

  1. hkntrt

    Thread Starter New Member

    Jun 20, 2012
    13
    0
    Hi

    Tuned collector oscillator is simple and have less companent than others , easy to build and a lot of simple transmitter circuits have tuned collector oscillator.I've got no problem with calculating frequency 1/2pi(sqrt L * C ).
    Problem is feedback and barkhausen criteria ( A * β >= 1 ) and this oscillator don't have a collector resistor (Rc). impedance of tank circuit have a effect on current of collector like resistor(DC). Should dc biasing design of transistor include that impedance ? there is no really telling everyting about tuned collector oscillator and no calculations on web ! ..... and many questions in my mind.

    Can anyone please give me a example with all calculations ? I don't want to learn more theory.

    and sorry for bad english :)
     
  2. t_n_k

    AAC Fanatic!

    Mar 6, 2009
    5,448
    782
    DC design would be done on the basis of the tank inductance being assumed an effective short circuit or very low resistance. Set Rb & Re accordingly with assumed HFE for required Ic. Single Rb bias is simple option, but probably OK for this 'theoretical' situation with the Re giving reasonable Q point stability with HFE variation for the BJT.

    You have a common base configuration Colpitts in this instance. Other Colpitts forms are even more difficult to predict in terms of outcomes - e.g the common emitter version may even 'run' in effective Class C mode when oscillations stabilize.

    Barkhausen is indicative only and may not guarantee oscillation will occur in practice. Oscillators of this type are effectively non linear. Barkhausen is based on linear systems theory. Along with the non linear issue it's probably why it is difficult to find simple design models guaranteed to work - it's probably a case of 'build and tweak'.
     
    hkntrt likes this.
  3. hkntrt

    Thread Starter New Member

    Jun 20, 2012
    13
    0
    I designed a tuned collector oscillator , i think it's won't work but can you tell me my mistakes ? I uploaded asc and png file.
     
  4. t_n_k

    AAC Fanatic!

    Mar 6, 2009
    5,448
    782
    I haven't looked at your overall design. However, if it's a Common Base mode you need a bypass capacitor from Base to ground.
     
  5. hkntrt

    Thread Starter New Member

    Jun 20, 2012
    13
    0
    I did tried everything but i couldn't succeed it :confused:
     
  6. t_n_k

    AAC Fanatic!

    Mar 6, 2009
    5,448
    782
    While you claim to have researched the topic without success it would surprise me if you couldn't find several well documented examples of a Common Base Colpitts oscillator - with values. The example you gave earlier purports to be a Colpitts but it does not have the classic Colpitts structure around the inductor / capacitive divider topology.
    I'd suggest doing some further research and evaluating the more reliable standard forms.
     
  7. t_n_k

    AAC Fanatic!

    Mar 6, 2009
    5,448
    782
  8. hkntrt

    Thread Starter New Member

    Jun 20, 2012
    13
    0
    I build and simulate colpitts oscillator schema in wikipedia and other examples i found using ltspice but it is not working.Presumably i am doing something wrong ?
     
  9. Jony130

    AAC Fanatic!

    Feb 17, 2009
    3,957
    1,097
    The first think that you need to do is to learn how to properly use LTspice.
    If you learn this then you can start using LTspice to analyse oscillator.
     
    hkntrt likes this.
  10. takao21203

    Distinguished Member

    Apr 28, 2012
    3,577
    463
    In reality such circuits don't always work 1:1 but it is useful to adjust components in software, to see the margins where values start or stop to work.
     
    hkntrt likes this.
  11. hkntrt

    Thread Starter New Member

    Jun 20, 2012
    13
    0
    @Jony130 your design is working i learned what must be stop time , maximum timestep from your design but when i adjust stop time , time step correctly to wikipedia's circuit it did not work , 100 pF capacitor are too high ?
     
  12. Jony130

    AAC Fanatic!

    Feb 17, 2009
    3,957
    1,097
    Well don't bother that this circuit don't work in simulation.
    Simulations always have a problem with all sorts of generators.
    Also this circuit from wiki look suspicious and, in particular coil value 150nH.
    And I think that LC generators are the most difficult of all the basic analog electronic circuits. the Colpittsa generator is very difficult circuit, you might even say a tricky circuit (even more tricky are other basic generators - Hartley and Meissner). Unfortunately, textbooks are reluctant to write about it, because in theory everything works.
     
  13. hkntrt

    Thread Starter New Member

    Jun 20, 2012
    13
    0
    You are right.Lc oscillators are diffucult to calculate and textbooks aren't including "exactly" practical informations.What can you suggest about generating easily high frequency sine wave without tank circuit ?

    Thanks
     
Loading...