LTSpice simulation not matching Reference Design

Hello,

I'm looking at the article Low-Distortion Sine Wave Oscillator with Precise RMS Amplitude Stability https://www.analog.com/en/technical-articles/low-distortion-sine-wave-oscillator-with-precise-rms-amplitude-stability.html# and I tried to make it in LTspice.  I'm getting an oscillation at 100kHz as intended but the amplitude isn't 3xVset as expected, it's railing.  Does anyone have ideas on what I did wrong, I've attached the file and a picture of the circuit?  Thank you!

Sine.asc

Parents
  • 0
    •  Analog Employees 
    on Mar 20, 2021 11:25 AM

    Hi Hbert, I don't support the LT1966 but have used it in the past. First problem is your VOUT and OUTRTN are swapped. But even after swapping... it still didn't work. 

    This family of parts can be a bit tricky, and LTspice may not accurately all operating conditions (a question best posted in the LTspice forum.) On a hunch, I put the LT1968 on a split +/-2.5V supply, and set the input common-mode to ground and... success. Red trace is output, blue is RMS converter output, orange is reference voltage, and green is the integrator output. If you build up this circuit, you'll want to verify that the integrator output is around the middle of its control range at nominal output amplitude. Note that the circuit should work fine in real life with a single 5V supply. But it would be a good idea to exercise the LT1968 by itself, with inputs biased as in the circuit, using a signal generator to drive the input.

    There are simpler ways to control the peak output amplitude vs. RMS, see Application Note 43.

    Application Note 106 is also an excellent reference for the LTC1966/7/8 family.

    Sine_updated.asc

Reply
  • 0
    •  Analog Employees 
    on Mar 20, 2021 11:25 AM

    Hi Hbert, I don't support the LT1966 but have used it in the past. First problem is your VOUT and OUTRTN are swapped. But even after swapping... it still didn't work. 

    This family of parts can be a bit tricky, and LTspice may not accurately all operating conditions (a question best posted in the LTspice forum.) On a hunch, I put the LT1968 on a split +/-2.5V supply, and set the input common-mode to ground and... success. Red trace is output, blue is RMS converter output, orange is reference voltage, and green is the integrator output. If you build up this circuit, you'll want to verify that the integrator output is around the middle of its control range at nominal output amplitude. Note that the circuit should work fine in real life with a single 5V supply. But it would be a good idea to exercise the LT1968 by itself, with inputs biased as in the circuit, using a signal generator to drive the input.

    There are simpler ways to control the peak output amplitude vs. RMS, see Application Note 43.

    Application Note 106 is also an excellent reference for the LTC1966/7/8 family.

    Sine_updated.asc

Children