op amp circuit explanation

Discussion in 'The Projects Forum' started by Denesius, Feb 27, 2015.

  1. Denesius

    Thread Starter Member

    Feb 5, 2014
    I came across the attached schematic sometime back, and saved it for future reference. Well, the future is now, and I uploaded to a Multisim program which shows the output as sawtooth, but with unstable levels, and a freq of 38KHz (multisim jpeg attached). I built the circuit on a breadboard, and the output appears to be a nice sawtooth, with the frequency around 20KHz (screenshot jpeg attached). The 50pF capacitor appears to control the output frequency, up to a point. Above 35KHz there is significant clipping.
    Can someone explain what's going on? Is there a name for this circuit? I understand the feedback path between pins 7 & 1 and pin 3, but what's with the two diodes? And is there a way to get the frequency up around 50KHz (which is what I need) without the distortion and clipping? (ie- faster response diodes, or different op amps, or something else).
    As always- thank you to all the knowledgeable experts out there!
  2. MikeML

    AAC Fanatic!

    Oct 2, 2009
    You have the right voltages on pins 4 and 11, but usually, the more positive voltage is called Vdd and the more negative is Vss.

    U1a is a Schmitt trigger. R3, D1 and D2 is a symmetric clipper. U1b is an integrator. U1c is redundant, and can be removed.

    The Multisim waveform distortion is an artifact of how Mulisim implemented their stupid oscilloscope "instrument". It is not real. Another nail in the coffin of Multisim!
    planeguy67 likes this.
  3. AnalogKid

    AAC Fanatic!

    Aug 1, 2013
    The overall circuit is one of the classic low cost "function generator" circuits, called so because it makes both square and triangle waves. Add a sine shaping circuit to the triangle output to complete the trio. It also is the basis for many pulse width modulators, formed by adding a comparator stage to the triangle output. I agree with Mike - if you tie the two diodes directly to GND you can eliminate U1C. The circuit frequency is directly related to the R4-C1 time constant. Clipping and other high-frequency problems are caused by the opamp's limited gain-bandwidth product. At higher freqs there is less negative feedback around the integrator (U1b), causing its output impedance to increase, causing a non-constant impedance driving the capacitor, effectively changing its value.

    A TL084 has a gain bandwidth product (GBW) of 3 MHz. A NE5532 has a GBW of 10 MHz. Faster amps = higher freqs.

    Denesius likes this.
  4. Denesius

    Thread Starter Member

    Feb 5, 2014
    I saw the GBW in the rating, but would appreciate an explanation of why with the 3MHz limit I'm getting clipping at 35KHz? Also, is there a quad version of the NE5532? Thanks