Home / Learn / Exam prep / IIT JAM Physics / Lesson 7.6
All lessons

Module 7: Solid state physics and electronics · Lesson 7.6 · about 45 min

Operational amplifiers

The two golden rules that solve almost every op-amp circuit, the inverting, non-inverting, summing, difference, integrator and differentiator circuits, what real op-amps get wrong (gain-bandwidth, slew rate, offsets, CMRR), and comparators, Schmitt triggers and oscillators.

Key points, meanings and formulas on one page: choose "Save as PDF" in the print box.

Module 7 · about 45 minutes. By the end you will be able to solve any linear op-amp circuit with two rules and Kirchhoff’s current law, write the output of summing, difference, integrating and differentiating circuits, predict when a real op-amp will saturate, slew or lose bandwidth, and work out the thresholds of a Schmitt trigger and the conditions for an oscillator.

An op-amp has so much gain that, with negative feedback, it does whatever it takes to make its two inputs equal; the resistors around it, not the chip, decide what the circuit does.

1. What an op-amp is

An operational amplifier is a high-gain differential amplifier on a chip (inside are a few dozen transistors of the kind met in lesson 7.5). It has two inputs, inverting (−) and non-inverting (+), one output, and power supply pins, often ±15 V. Its output is

Vo=A(V+−V−),A∼105 to 106 (open-loop gain)

limited to just inside the supply rails. Used like this, “open loop”, a difference of 0.1 mV is enough to slam the output to a rail. So op-amps are almost always used with negative feedback: part of the output is fed back to the inverting input.

The ideal op-amp has infinite gain, infinite input resistance, zero output resistance and infinite bandwidth. From these come the two golden rules (for a circuit with negative feedback that is not saturated):

  1. No current flows into either input.
  2. The output does whatever it must to make V+=V− (the virtual short).

Rule 2 is a consequence of feedback: if V− were a little below V+, the huge gain would push the output up, which (through the feedback) raises V− until the difference is a few microvolts.

2. The inverting amplifier

The signal goes through Rin to the inverting input; Rf connects the output back to that input; the non-inverting input is grounded. Apply the rules step by step:

Virtual ground. V₊ = 0, so rule 2 says V₋ = 0 too. The inverting input is at ground potential without being connected to ground.

Input current. The current through R_in is (V_in − 0)/R_in = V_in/R_in.

Nowhere else to go. Rule 1: no current enters the op-amp, so the same current must flow on through R_f.

Output. Across R_f the voltage drops from 0 to V_o, so V_in/R_in = (0 − V_o)/R_f.

Result. V_o = −(R_f/R_in)V_in. The gain is set by two resistors only, and the minus sign means the output is inverted.

VoVin=−RfRin,Rin, seen by source=Rin

The source sees just Rin, because its far end sits at 0 V. Note that the gain can be smaller than 1 in size.

Switch between the inverting and non-inverting circuits and change the resistors. The transfer curve is a straight line through the origin until the output reaches the ±11 V it can swing to, then it flattens: the op-amp saturates and the golden rules stop working.

Worked example. An inverting amplifier has Rin=4.7 kΩ and Rf=47 kΩ, on ±12 V supplies (output swing ±11 V).

  1. Gain =−47/4.7=−10.
  2. For Vin=0.5 V: Vo=−5 V. For Vin=1.5 V: the formula says −15 V, but the output stops at −11 V (saturated).
  3. The source delivers 0.5 V/4.7 kΩ=0.106 mA, whatever the op-amp is like.

3. Non-inverting amplifier and voltage follower

Now the signal goes straight to the + input, and the − input sits on a divider: Rf from the output, R1 to ground. Rule 2 says V−=Vin, and the divider says V−=VoR1/(R1+Rf). So

VoVin=1+RfR1

The gain is at least 1 and positive (no inversion), and the source sees the op-amp’s own huge input resistance. Remove R1 (or make Rf=0) and the gain is exactly 1: the voltage follower, a buffer that copies a voltage from a weak source to a heavy load.

A
open-loop gain: how strongly the output reacts to a difference between the inputs
β = R₁/(R₁ + R_f)
the fraction of the output fed back to the inverting input
Closed-loop gain ≈ 1/β
the gain with feedback: set by the resistors, not by A
Virtual short
V₊ ≈ V₋, true only with negative feedback and an unsaturated output
Physical meaning
The op-amp trades its enormous, unreliable gain for a smaller, precise one. The extra gain it does not use (the loop gain Aβ) is what keeps the inputs equal and the result accurate.

Worked example. A thermocouple gives 2.4 mV and must be amplified to 1.2 V without loading it.

  1. Needed gain: 1.2/0.0024=500; it must be positive and draw no current, so use a non-inverting stage.
  2. 1+Rf/R1=500: with R1=1 kΩ, Rf=499 kΩ.
  3. An inverting stage would need Rin very large to avoid loading the sensor, and would invert the signal.

4. Summing and difference amplifiers

Add more input resistors to the virtual ground of an inverting amplifier and their currents add in Rf:

Vo=−Rf(V1R1+V2R2+V3R3)

Equal resistors give −(V1+V2+V3); Rf=R/n gives minus the average. Audio mixers and simple digital-to-analogue converters (binary-weighted resistors) are summing amplifiers.

For circuits with signals on both inputs, use superposition: switch on one input at a time (grounding the others), find the output from the inverting or non-inverting formula, and add. With V1 through R1 to the − input, feedback R2, and V2 through R3 to the + input with R4 to ground:

Vo=−R2R1V1+(1+R2R1)R4R3+R4V2=R2R1(V2−V1)  when R4R3=R2R1

That is the difference amplifier. It ignores anything common to both inputs, such as mains hum picked up equally on two wires, but only if the resistor ratios match. A mismatch lets the common-mode signal through, and the circuit’s common-mode rejection ratio (CMRR =Ad/Acm, often quoted in dB as 20log10⁡ of the ratio) is set by resistor matching as much as by the op-amp. The instrumentation amplifier puts a follower-type stage in front of each input (very high input resistance) with gain 1+2R/RG set by one resistor.

Worked example. V1=1 V goes through 10 kΩ to the inverting input, Rf=20 kΩ, and V2=0.5 V goes straight to the non-inverting input.

  1. From V1 alone: inverting, −(20/10)×1=−2 V.
  2. From V2 alone: non-inverting, (1+20/10)×0.5=1.5 V.
  3. Together: Vo=−2+1.5=−0.5 V. The common slip is to give V2 a gain of 2 instead of 3.

5. Integrator and differentiator

Replace Rf by a capacitor C. The input current Vin/R still flows on, now charging the capacitor, and CdVo/dt=−Vin/R:

Vo=−1RC∫Vindt(integrator),Vo=−RCdVindt(differentiator: C and R swapped)
Input Integrator output Differentiator output
constant ramp (opposite sign) zero
square wave triangle wave spikes at the edges
triangle wave parabolic arcs square wave
sin⁡ωt cos⁡ωt/(ωRC), gain falls with frequency −ωRCcos⁡ωt, gain rises with frequency

The integrator is a low-pass circuit: its gain is unlimited at DC, so a tiny offset slowly drives it into saturation. A large resistor across C caps the DC gain. The differentiator boosts high-frequency noise; a small resistor in series with its capacitor tames it. With both Rf and Cf in parallel in the feedback path you get a first-order active low-pass filter with corner fc=1/(2πRfCf).

Worked example. A ±1 V, 1 kHz square wave drives an integrator with R=10 kΩ, C=0.1 μF (RC=1 ms).

  1. During each half-period (0.5 ms) the input is constant, so the output ramps at 1 V/1 ms=1000 V/s.
  2. In 0.5 ms it moves 0.5 V, then ramps back.
  3. The output is a triangle wave of 0.5 V peak-to-peak (centred wherever the starting condition left it).

6. What real op-amps get wrong

“The two inputs of an op-amp are always at the same voltage.” Only with negative feedback, and only while the output is inside its range. In a comparator (no feedback) or a Schmitt trigger (positive feedback), the inputs differ by volts and the output sits at a rail. In an amplifier whose formula asks for more than the supply allows, the output saturates, the feedback can no longer keep up, and V+≠V−. Always check the output is within the rails before using the virtual short.

  • Finite gain. With feedback fraction β, the gain is Acl=A/(1+Aβ), not quite 1/β. The fractional error is about 1/(Aβ), so high-gain stages are the least accurate.
  • Gain-bandwidth product. The open-loop gain falls as 1/f above a few hertz. A stage with “noise gain” 1/β (=1+Rf/Rin, for inverting and non-inverting stages alike) has bandwidth f3dB=GBW×β. A 1 MHz op-amp at gain 100 gives 10 kHz.
  • Slew rate. The output cannot change faster than the slew rate SR (e.g. 0.5 V/μs). A sine of peak Vp has a largest slope 2πfVp, so large signals are limited to the full-power bandwidth fmax=SR/(2πVp), often far below the small-signal bandwidth.
  • Offset voltage and bias currents. A few millivolts of input offset is amplified by the noise gain 1+Rf/Rin. Bias currents flowing through unequal resistances add more offset; a resistor equal to Rin∥Rf on the + input balances them.
  • Output limits. The output swings only to within a volt or two of the supplies (rail-to-rail types do better) and can supply only tens of milliamps.

Worked example. An op-amp with GBW = 2 MHz and SR = 1 V/μs is used at a gain of 10 with a 1 V peak input.

  1. Small-signal bandwidth: 2 MHz/10=200 kHz.
  2. The output is 10 V peak, so the slew limit is 106/(2π×10)=15.9 kHz.
  3. Above 15.9 kHz the sine output turns into a triangle; the slew rate, not the bandwidth, sets the limit here.

7. Positive feedback: comparators, Schmitt triggers and oscillators

A comparator is an op-amp with no feedback: the output goes to +Vsat when V+>V− and to −Vsat otherwise. With a slowly varying noisy input it chatters back and forth near the switching point.

A Schmitt trigger cures this with positive feedback. In the inverting version the input goes to the − terminal, and the output is fed back to the + terminal through R2, with R1 from the + terminal to ground (or to a reference Vref). The + terminal then sits at

V+=VrefR2R1+R2+VoR1R1+R2

which depends on which rail the output is at. So there are two thresholds, VUT and VLT, and a hysteresis gap VUT−VLT=2VsatR1/(R1+R2). Noise smaller than the gap cannot make the output switch back.

Worked example. Vsat=±12 V, R1=10 kΩ, R2=20 kΩ, Vref=0.

  1. V+=Vo×10/30=Vo/3.
  2. Output high: the input must rise above +4 V to switch it low. Output low: the input must fall below −4 V to switch it high.
  3. Hysteresis: 8 V. A signal wobbling by ±1 V about 3 V can never cause a false switch.

Oscillators feed back in phase on purpose. The Barkhausen condition: around the loop the gain must be 1 and the phase shift a whole number of cycles. A Wien bridge network passes 1/3 of the voltage with zero phase at f=1/(2πRC), so it needs a non-inverting amplifier of gain 3 (Rf=2R1). Three RC sections give 180° at f=1/(2π6RC) with a loss of 1/29, so a phase-shift oscillator needs an inverting amplifier of gain 29.

To try real component values, the op-amp circuit calculator works out gain, bandwidth and clipping for the classic circuits, the analog filter designer builds active filters from op-amp stages, and the circuit simulator gives the AC response of any network. The phasor arithmetic behind filters and the integrator’s 1/(jωRC) is in the complex numbers lesson.

  • With negative feedback and an unsaturated output: no input current, and V+=V− (virtual short). Then use Kirchhoff’s current law at the inverting input.
  • Inverting: −Rf/Rin, input resistance Rin. Non-inverting: 1+Rf/R1, input resistance very high. Follower: gain 1, a buffer.
  • Summing: −Rf∑Vi/Ri. Two inputs on both terminals: superposition. Difference amplifier (R2/R1)(V2−V1) only with matched ratios.
  • Integrator −1RC∫Vdt (square in, triangle out); differentiator −RCdV/dt (triangle in, square out).
  • Real limits: A/(1+Aβ), bandwidth = GBW / noise gain, fmax=SR/(2πVp), offsets times the noise gain, output saturation.
  • Positive feedback: comparators and Schmitt triggers (hysteresis 2VsatR1/(R1+R2)); oscillators need loop gain 1 (Wien: 3, phase shift: 29).

Practise: exam questions on this lesson

They start easy and climb to top-rank level. Check each answer, then read the full solution. There are more on the op-amps practice page.

Next: lesson 7.7 leaves the analogue world for logic gates, where transistors are only ever fully on or fully off. The GATE Physics course treats feedback amplifiers, active filters and oscillators in more depth.

Check yourself

1An inverting amplifier has R_in = 10 kΩ and R_f = 100 kΩ. What input resistance does the signal source see?
2An op-amp with a gain-bandwidth product of 1 MHz is used as a non-inverting amplifier of gain 50. What is its bandwidth?
3When can you assume that the two input terminals of an op-amp are at the same voltage?

Sources

  • P. Horowitz and W. Hill, The Art of Electronics, 3rd ed. (2015), chapter 4: operational amplifiers
  • S. Franco, Design with Operational Amplifiers and Analog Integrated Circuits, 4th ed. (2015), chapters 1 to 6
  • A. S. Sedra and K. C. Smith, Microelectronic Circuits, 7th ed. (2015), chapter 2: operational amplifiers
  • IIT JAM Physics syllabus (Solid State Physics, Devices and Electronics section), published by the organising IIT each year

Your ticks stay in this browser. Save them as a file on the course home.

Updated 2026-11-22. Not affiliated with IIT or the JAM organising institute. Exam pattern as published for recent years: always check the official JAM website before the exam. All questions are original.