Distortion in an Oscillator

Discussion in 'Homework Help' started by Malsch, Jun 7, 2011.

  1. Malsch

    Thread Starter New Member

    Mar 19, 2011
    23
    0
    I built a Wien bridge oscillator (see schematic below) and was wondering what factors and which of the components used mostly contribute to the distortion of the oscillator. My first guess is capacitors maybe due to phase shifts...

    [​IMG]

    Thank you.
     
  2. t_n_k

    AAC Fanatic!

    Mar 6, 2009
    5,448
    782
    The capacitors aren't the problem.

    I guess you are referring to the amplitude distortion which causes the output to deviate from an ideal sinusoidal waveform.
     
  3. Malsch

    Thread Starter New Member

    Mar 19, 2011
    23
    0
    yes thats what i was refering to. i was thinking that this happens due to he noise produced by the resistors and the op-amp
     
  4. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,347
    Hello,

    The limit in frequency of the opamp.
    The adjustment of the feedback resistor (RV1).

    This PDF will give you a lot of information:
    sloa087_sinewave_oscillators.pdf

    Bertus
     
    eromlig1991 likes this.
  5. t_n_k

    AAC Fanatic!

    Mar 6, 2009
    5,448
    782
    Nice source Bertus.

    To Malsch:

    One might argue that the network comprising RV1, D1 & D2 has been included in the circuit to notionally reduce the degree of distortion, rather than contributing to the distortion. It's a moot point.

    These elements form a non-linear network. The original Wien Bridge design included an incandescent lamp as the non-linear element for reduction of distortion.

    Did you try Google with a search phrase something like - "Causes of Wien bridge oscillator distortion" ....? Wikipedia has a section on the oscillator which mentions the distortion matter.
     
  6. Adjuster

    Well-Known Member

    Dec 26, 2010
    2,147
    300
    You may find what I'm about to say more fully and rigorously described in Bertus' link, but perhaps some simpler words may also help.

    The difficulty lies in achieving a working oscillator with a stable output amplitude, without this resulting in excessive waveform distortion. You might think that it was possible to obtain a loop gain of exactly one, then apply a starting signal and expect to see its level maintained, but this cannot be done. There will inevitably be some error in the gain, and the oscillation will therefore either grow or decay.

    In order to make a practical oscillator which can be relied on to start up and work, the loop gain must begin appreciably more than one at switch-on, and this must be predictably so despite temperature and supply voltage variations, component tolerances, and ageing. As a result the starting loop gain typically has to be quite a lot more than one.

    This situation results in the amplitude building up rapidly when the oscillator is turned on, but if nothing else is done the amplitude will increase until non-linear effects in some part of the system prevent it rising further. In some types of oscillator using sharply tuned resonators it is possible for the amplifying device to limit quite hard without the waveform being too obviously distorted, but this will not work with the Wien Bridge.

    The Wien bridge therefore employs some other means of amplitude regulation. One possibility is a diode limiter in the feedback loop, arranged to be a bit less brutal than simply letting the amplifier hit the end-stops. The result may nevertheless be mediocre.

    A possibly better but dated method utilised the positive temperature coefficient of resistance of a filament lamp (more developed versions used a tiny thermistor) in the feedback, arranged so that as the rising signal level heats the device, the loop gain falls. Provided that the period of the oscillation is much shorter than the thermal time constant of the controlling device, the distortion can be quite low.

    Finally more elaborate oscillators might use an automatic level control system, where the output amplitude is measured and used to produce a control signal, which governs some king of adjustable amplifier or attenuator. Simple versions of this use a rectifier feeding the gate of a FET used as a voltage controlled resistor - here the distortion depends mainly on the control element: the simple FET versions may not be very good, integrated VCAs may do better.
     
    Last edited: Jun 7, 2011
  7. eromlig1991

    New Member

    Jun 10, 2011
    4
    0
    So guys, the components that contribute mostly to the distortion are?
     
  8. t_n_k

    AAC Fanatic!

    Mar 6, 2009
    5,448
    782
    As Adjuster explained, it is an inherent property of the oscillator design / topology, rather than something attributable to any particular component or components. As Adjuster also explained, the designer takes steps of varying complexity (according to the required output signal purity specification) to reduce the effects of the distortion.
     
Loading...