Op-amp circuit gamut
Quickly walk through comparator, integrator, differentiator and Schmitt trigger. What goes wrong with each in practice?
| Level | ●● onsite technical |
| Track | General board |
| Really testing | Breadth plus practical failure modes. |
Going deeper
Extra explanation that is not in the book: worked numbers, where people go wrong, and how to say it in the room.
Schmitt trigger — worked. Rail-to-rail output on 5 V (swing 5 V), R1 = 10 kΩ to ground, R2 = 100 kΩ feedback: hysteresis = 5 × 10/110 ≈ 0.45 V. Make it larger than the peak-to-peak noise on the input, smaller than the signal swing.
Integrator — how big the "leak" resistor is. R = 10 kΩ, C = 10 nF, leak 1 MΩ across C: below 1/(2π·1 MΩ·10 nF) ≈ 16 Hz it is an amplifier with gain 100 (1 MΩ/10 kΩ), above it an integrator. Offset now gives a fixed error instead of a ramp into the rail.
Differentiator made practical. A series R in the input and a small C across Rf turn it into a band-pass: it differentiates in the band you care about and stops amplifying noise above it.
Comparator traps beyond speed
- Saturation recovery of an op-amp can take microseconds.
- Some op-amps show phase reversal when an input goes past the common-mode range — the output flips the wrong way.
- Many op-amps have back-to-back diodes across the inputs, which conduct with a large differential voltage — fine in a feedback circuit, wrong in a comparator.
Errata
Correction: Schmitt-trigger hysteresis formula now names its circuit (inverting type, R1 from the + input to ground). Fixed before the first printing. All errata
Ask about this question
Stuck on a step, or think something is wrong? Send it with AN-05 in the subject. Good questions get answered on this page, and corrections go into the next printing.
Related
← AN-04 Current source and current sinkAN-06 AC-coupled BJT amplifier →