Why the wave form of my wien bridge oscillator is not a pure sine wave?

Discussion in 'The Projects Forum' started by SKMR168, May 23, 2016.

  1. SKMR168

    Thread Starter New Member

    Aug 8, 2015
    9
    1
    my circuit is aimed to generate a 1.5kHz(approximately) sine wave. The circuit in MULTISIM is-

    upload_2016-5-23_13-32-38.png

    Using oscilloscope i got this waveform-

    upload_2016-5-23_13-32-20.png

    I had checked the gain properly. It should be a pure sine wave at the gain i had set to because increaing the gain made it a square wave and decreasing the gain made it more and more distorted. what causes the upper portion of the sine wave to be distorted
     
  2. MrAl

    Well-Known Member

    Jun 17, 2014
    2,423
    490
    Hi,

    What the heck is that?

    It looks like a green line sine wave, but there is yellow in there too at the top. Is that fast oscillation?
    If it is fast oscillation then there is something wrong with the layout or you need to compensate somewhere for the speed of the op amp perhaps (or maybe diodes should be 1N4148 instead of 1N4007).

    In a circuit like this you have to have the gain set right (as you seem to know) so that the limiting part of the circuit does not cause too much clipping. With minimal clipping you can get pretty good distortion numbers.

    You may have to add more diodes however to increase the voltage clip point. 0.7v sounds too low. 1N4007 is not the right diode for the job either, the capacitance may be causing the extra oscillations. Use 1N4148 or similar.
     
    SKMR168 likes this.
  3. atferrari

    AAC Fanatic!

    Jan 6, 2004
    2,647
    759
    Curious about your power supplies: neg and neg_1, are they a common point in your circuit?

    Is that the way to identify them in Multisim?
     
    Last edited: May 23, 2016
  4. SKMR168

    Thread Starter New Member

    Aug 8, 2015
    9
    1
    Sorrry for the mistake.Actually both should be denoted with the same name - "neg"
     
    Last edited: May 23, 2016
    atferrari likes this.
  5. grahamed

    Member

    Jul 23, 2012
    99
    11
    Hi

    Yes that's hf oscillation right enough.

    Oscillation is (almost) always caused by feedback.

    Either the desired feedback of the Wein network or by undesired feedback through the power supply - try adding some 100nF near to the OPA supply pins

    Having said I suspect that this OPA is only conditionally stable, i.e. too much negative feedback, too low a gain in fact, will make it oscillate I can't find a datasheet so I can't be sure but less than 5 or 10 might be a problem and your gain could be quite low when the diodes conduct and the diodes have 10pF so exhibit quite low impedence at h.f. Try 1N4148 as suggested.

    Also have a search for grid- or gate-stopper resistors.
     
  6. SLK001

    Well-Known Member

    Nov 29, 2011
    812
    225
    I believe that it is low frequency oscillation. What you are seeing is a result of many traces being captured by your o'scope. Set your scope for a single shot and capture ONE trace.

    Also, where is a picture of your physical circuit? I would bet that your problems are layout specific.
     
  7. atferrari

    AAC Fanatic!

    Jan 6, 2004
    2,647
    759
    Maybe in a protoboard?
     
  8. MrAl

    Well-Known Member

    Jun 17, 2014
    2,423
    490
    Hi again,

    You quoted my post but i dont see it here, did you delete it later?

    No problem either way, but hope you made some progress with this now.
     
  9. grahamed

    Member

    Jul 23, 2012
    99
    11
    turn timebase right up, right down - see what that hairyness is....
     
  10. grahamed

    Member

    Jul 23, 2012
    99
    11
  11. SKMR168

    Thread Starter New Member

    Aug 8, 2015
    9
    1
    But It is OPA552 not OPA522
     
  12. crutschow

    Expert

    Mar 14, 2008
    12,993
    3,227
    Then that's your problem.
    The 552 is only stable for gains of >5 and can oscillate with a lower closed-loop gain (as is likely the case here).
    You need the 551 or other unity-gain stable op amp.
     
  13. grahamed

    Member

    Jul 23, 2012
    99
    11
  14. SKMR168

    Thread Starter New Member

    Aug 8, 2015
    9
    1
    Using fast switching diodes instead of 1n4007 have no effect on the waveform. Again the idea of using more diodes also has no effect.The problem was with the breadboard. Because after proper soldering the waveform is quite better than that found while testing in breadboard.
     
Loading...