0
\$\begingroup\$

I am trying to test a simple op-amp integrator (from my textbook) in LTSpice with a square-wave input.

As can be seen, the output is an (almost) triangular waveform for a resistance of 1 kΩ and a capacitance of 10 μF:

enter image description here

enter image description here

However, when I change R and C to 10 kΩ and 100 μF respectively, the output gets saturated:

enter image description here

enter image description here

I don't understand why the output would saturate given that the voltage on being integrated should not exceed the supply at all:

$$V_o = -\frac{1}{RC}\int_0^{t} V_i(t)dt$$

For R = 10 kΩ and C = 100 μF, RC should only be 1 and Vi(t) is only 0.5 V for the ON interval, so the integral should not exceed supply. My textbook does not impose any restrictions or constraints on the values R and C and simply provides the above equation.

Additionally, is there a reason why the output waveform in the first case is not perfectly triangular, and appears to have some distortion around the rising and falling edges of the pulse? Does it have to do with high frequency rejection at those places due to the frequency response of the particular model being used?

\$\endgroup\$
2
  • 1
    \$\begingroup\$ What's the offset voltage rating of your op amp? \$\endgroup\$
    – Hearth
    Jan 20, 2022 at 20:59
  • 1
    \$\begingroup\$ Run the simulation for a lot longer to see what happens. \$\endgroup\$
    – Andy aka
    Jan 20, 2022 at 21:08

2 Answers 2

4
\$\begingroup\$

Spice computes the DC operating point before starting out. The DC operating point of this circuit includes a perfect, lossless capacitor, which will cause the OPAMP to hit its rails.

This also occurs in your original simulation, but the RC network is much faster than the PULSE frequency. The 1k is able to fully discharge the 10u before the next charging pulse occurs. If you were to rerun the original simulation with the 1k/10u, and set the transient stop time to 3ns, you'll notice the C voltage start charged even though you "just turned it on."

Change the value of the capacitor to

"100u ic=0v"

to tell SPICE that the capacitor starts discharged. the IC directive is "initial condition"

\$\endgroup\$
2
\$\begingroup\$

I think your circuit exceeds the "maximum common-mode input voltage range" of 5V less than a supply voltage and the Jfet inputs produce "phase inversion" causing the output tro go as high as it can. TL07x audio opamps also do it.

opamp

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.