Common Base Colpitts Oscillator

Discussion in 'Homework Help' started by aaval002, Jul 28, 2012.

  1. aaval002

    Thread Starter New Member

    Jul 22, 2012
    16
    3
    I'm trying to build a Common Base Colpitts Oscillator for a project in my university. I've been failing at this for somewhat over a month now. I've tried various times to build the actual circuit, but the oscilloscope isn't reading anything from my circuit. :(


    The only specification is that the peak should be at around 450-550KHz.


    I've also tried the circuit in various PSPICE simulations.


    I was previously trying a different configuration, but I kept getting an overdamped sinusoid displayed below:


    [​IMG]


    The current schematic I am using is this one:


    [​IMG]

    This is my output:

    [​IMG]

    For some reason that I cannot make sense of, at 21ms the oscillations disappear. I can't seem to understand why, and this is part of the problem.


    http://www.electronics-tutorials.ws/oscillator/colpitts.html


    I've been using the above link as a guide / reference.


    I played around with some numbers and I'll try them on PSPICE sometime this week, but the circuit didn't work in the lab and the oscilloscope still didn't get any readings, so I doubt they'll work on PSPICE.


    Please let me know if you see something that I am doing wrong, if you have any suggestions for a different configuration, anything would be of great help. I'm really perplexed on this one. :confused:
     
  2. mlog

    Member

    Feb 11, 2012
    276
    36
    What if you added a 10 mH inductor in series with the emitter resistor? What does that do for you?
     
  3. aaval002

    Thread Starter New Member

    Jul 22, 2012
    16
    3

    I did as you said, and what this did was increase the voltage, as shown below. Thanks for the suggestion though. I've never seen a degenerative inductor though? :)

    [​IMG]
     
  4. mlog

    Member

    Feb 11, 2012
    276
    36
    Are you saying it fixed your problem?

    I thought maybe the emitter resistor was dissipating the high frequency energy. I suggested a decoupling inductor that was about 10x the impedance of the resistor at the frequency of intended oscillation. If you get too wild with the inductor, it might play tricks with the oscillator frequency.
     
  5. aaval002

    Thread Starter New Member

    Jul 22, 2012
    16
    3
    Do you think the emitter resistor is too high then and should be decreased, or use a wire instead? The inductor did increase my oscillation amplitude, but not the duration. It's good to have a larger amplitude either ways, but the duration is still a problem.:(
     
  6. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    Damped ringing is a gain problem. You are under unity gain, which is allowing your oscillation to die.

    C1 is the feedback path, it is a bit low. You might try increasing it. Or not.

    This is not a classic Colpitts design. You might want to start from scratch with a different arrangement. A classic Colpitts uses a split capacitor to invert the signal, which creates the positive feedback when used in conjunction with a inverting amplifier.

    [​IMG]
     
    Last edited: Jul 29, 2012
  7. aaval002

    Thread Starter New Member

    Jul 22, 2012
    16
    3
    Thank you for the response. By unity gain, do you mean that the voltage of collector over voltage of emitter should be greater than one? If so, should I add a resistor to the collector to allow this increment?

    I was using 10pF and 100pF since it's what I calculated from 1/[2*pi*sqrt(L* C1||C2)] with an inductor of 10mH. (If you suggest I use a different value inductor to be able to use higher capacitors to satisfy my frequency, I'll do so. I just use this inductor since I have the most of it.)


    When both capacitors are equal though, my oscillation does last a bit longer.


    Yes, what you refer to by a classic Colpitts would be something such as this:

    [​IMG]


    The problem is that I failed to mention that I cannot use op-amps in my project, otherwise it would be much easier.


    I'm also hoping to have the oscillator work without the need of a function generator as in my previous LC circuit, it should work with strictly DC input.


    The schematic I was following was this one:

    [​IMG]
    Which I obtained from this website:
    https://docs.google.com/viewer?a=v&q=cache:iwb646iuuz4J:venus.ece.ndsu.nodak.edu/~yuvaraja/EE421/E421-N-Colpitts%2520Design1.DOC+common+base+colpitts+oscillator&hl=en&gl=us&pid=bl&srcid=ADGEESiEGhy925ZF-OwUwrpYQV1maRezzB27gOGbrfMVZoHoYw7Hn-aBygLIHTK-EjrkY425P1CM6Jn20tmfvNC66x-_l4wzQxVK1aUbrHkOJ-FT2h-wQwiG3tTptWIe_tsQCyMUnse-&sig=AHIEtbRdYSNN76MMLvwl-Bsx707vBjEYYQ

    I've recalculated the values trying to follow all the given equations. I'll organize my work and scan it in a day or two. :)

    From my new calculations I obtained

    C1 = 9.2725pF
    C2 = 328.96pF
    L = 10mH
    R1 = 2KΩ
    R2 = 2KΩ
    RL = 1KΩ
    RE = 65Ω
    CB = 150pF

    When simulated on PSPICE I failed to get any oscillations. :(

    I'm willing to use a different schematic than this one, maybe there's something inherently wrong with the equations, but I was following it since it was basically a step by step guide type of thing. :)
     
  8. Jony130

    AAC Fanatic!

    Feb 17, 2009
    3,963
    1,098
    Add simulation file. Then I try to see what is wrong with the simulation
     
  9. aaval002

    Thread Starter New Member

    Jul 22, 2012
    16
    3
    Okay so the attachments are the equations used, and I have also attached the simulation.

    I think the inductor suggestion actually works, I tested for 1second and it seemed to work, I didn't test it for higher times, but I think it'll work as well.


    The problem is that the oscillation seems a bit weird, the sinusoid is not a perfect sinusoid and it has a ripple at the start of every wave. Please let me know any suggestions you may have. Thank you.
     
  10. cork_ie

    Member

    Oct 8, 2011
    348
    58
    Sorry to intrude into another users thread but now seems a good time to ask a few questions . I hope these are also relevant to the OP's question

    As far as I am aware the resonant frequency for a Colpitts Oscillator is 1/2π√(L.C1.C2)/(C1+C2)

    1) Are there any rules of thumb for the values of L , C1 , C2 for a given frequency? (how many times greater or smaller) can C1 be with respect to C2.

    2) What is the ideal capacitance to inductance ratio. ie μF:μH ?

    3) What are the advantages & disadvantages of using either a Colpitts or Hartley oscillator for generating an AM carrier wave of approx 1MHz.
     
    Last edited: Jul 30, 2012
  11. aaval002

    Thread Starter New Member

    Jul 22, 2012
    16
    3
    Hi, by all means go ahead. Your questions have also created curiosity for me.

    I can only answer the 1st one. My professor actually mentioned that somewhere between 10 to 100 times larger is a good difference between the capacitors. Anything above that isn't as ideal. I'm not entirely sure why though. :)
     
  12. JoeJester

    AAC Fanatic!

    Apr 26, 2005
    3,373
    1,164
    Your topology matches this one at wiki. It is reputed to be a working model.

    [​IMG]
     
  13. WTP Pepper

    New Member

    Aug 1, 2012
    21
    6
    Try zooming in on the mush being shown. You may find a stable sinusoidal signal.
     
  14. aaval002

    Thread Starter New Member

    Jul 22, 2012
    16
    3

    Yes, you're right! Just needed a ground (they assumed negative was ground...) and it works. It says 50MHz on the schematic there, but the output is more like 2MHz when I ran it.

    The problem is that when I try to adjust the frequency by adjusting the capacitor and inductor, it becomes overdamped again. I'd have to know how they calculated the other values in order to be able to adjust it properly. :confused:
     
  15. aaval002

    Thread Starter New Member

    Jul 22, 2012
    16
    3
    Yes, I'm getting a somewhat stable somewhat sinusoidal looking signal.

    I mentioned it to the last person who requested my schematics. I circled in red where I have a bit of a ripple.

    The oscillations look a bit triangular, but I'm not sure how much of a problem that'll pose. If the ripple can be fixed, it should be good.

    This is what I have now.

    (Also: Thanks to everyone who has helped so far, I wouldn't have been able to get this far without your help.)

    [​IMG]
     
  16. aaval002

    Thread Starter New Member

    Jul 22, 2012
    16
    3
    As Michael Jackson once said: "This is It."


    It seems to be working fine in the simulation, perfectly fine. The experimental isn't as nice. I'm getting a frequency of 56.18MHz when it should be 500KHz. Not entirely sure why. Oh well.


    THANK YOU ALL VERY MUCH FOR THE ASSISTANCE! I'll probably be back if I try a Hartley oscillator and it doesn't work out. Thanks again guys! :D:D
     
  17. JoeJester

    AAC Fanatic!

    Apr 26, 2005
    3,373
    1,164
    Are you sure you didn't accidentally substitute a 1 uH coil for a 10 mH coil? 58 MHz is just a bit outside your design specification.

    As far as the real circuit not oscillating, you might want to review the application note attached and the datasheet attached.

    Solving the real life circuit is to your benefit.
     
  18. aaval002

    Thread Starter New Member

    Jul 22, 2012
    16
    3
    The schematic I have actually uses the 10mH coil to work properly. 10uH gives about 15MHz and 1uH doesn't even give a sinusoid, it's overdamped at 1uH.

    The real life circuit is oscillating, just not at the correct frequency. I've attached a picture. It should be at about 527KHz which is about 0.527MHz but it's way above this, it's at 56.42MHz which is way off. It's about 100 times larger than it should be, and I'm not entirely sure why.

    I looked at the PDFs you attached. Are these internal capacitances of the BJT? I vaguely remember taking these into account while doing small signal analysis. They should be bothersome only at really low frequencies or very high frequencies if I'm not mistaken?
     
Loading...