Home / Courses / GATE Physics / Practice / Operational amplifiers

Operational amplifiers

Electronics · 40 questions and templates for GATE Physics, in order from warm-up to top rank. Check each answer, then read the full solution.

Loading questions… (they need JavaScript; nothing you type is sent anywhere)

Written questions with solutions

The same questions as a list, so you can print them or read without the practice box. Templates (fresh numbers) appear only in the practice box above.

  1. MCQWarm-up

    The two "golden rules" for an ideal op-amp circuit (the inputs draw no current, and the two input voltages are equal) can be used
    1. whenever the op-amp is powered
    2. only when there is negative feedback and the output is not saturated
    3. only when there is positive feedback
    4. only for ac signals
    Answer and solution

    Answer: B

    The rule "no input current" is a property of the device. The rule V+=V− is not: it is what negative feedback achieves, by adjusting the output until the tiny difference times the huge gain gives the needed output. Without feedback (a comparator), with positive feedback (a Schmitt trigger) or once the output hits a supply rail, the inputs can differ by volts.

    Why people get this wrong: Applying V+=V− to a comparator or a saturated amplifier.

  2. NATStandard

    An op-amp has a slew rate of 0.5 V/μs. The highest frequency at which it can deliver an undistorted sine wave of 10 V peak is ______ kHz (to two decimal places).
    Answer and solution

    Answer: 7.96 kHz (±0.05)

    For Vpsin⁡2πft the steepest slope is 2πfVp (at the zero crossings). It must not exceed the slew rate: fmax=SR/(2πVp)=5×105/(2π×10)=7958 Hz =7.96 kHz.

    Why people get this wrong: Using peak-to-peak (20 V) halves the answer; the slope uses the peak amplitude.

  3. NATStandard

    An op-amp has a gain-bandwidth product of 1 MHz. Used as a non-inverting amplifier with gain 20, its −3 dB bandwidth is ______ kHz.
    Answer and solution

    Answer: 50 kHz (±0.1)

    For a single-pole op-amp, closed-loop gain times bandwidth stays equal to the gain-bandwidth product: f3dB=1 MHz/20=50 kHz.

    Why people get this wrong: For an inverting amplifier the divisor is the noise gain 1+Rf/Rin, not Rf/Rin; here the stage is non-inverting so it is just 20.

  4. NATStandard

    Two inputs, 2.0 V and 4.0 V, are each connected through a 10 kΩ resistor to the non-inverting input of an ideal op-amp (nothing else connects there). The op-amp has Rf=20 kΩ from output to inverting input and R1=10 kΩ from inverting input to ground. The output is ______ V (to one decimal place).
    Answer and solution

    Answer: 9 V (±0.05)

    No current enters the non-inverting input, so the two equal resistors form a divider that averages the inputs: V+=(2.0+4.0)/2=3.0 V. The stage is a non-inverting amplifier of gain 1+Rf/R1=3, so Vout=9.0 V. This non-inverting summer gives Vout=V1+V2 when the gain equals the number of inputs.

    Why people get this wrong: Treating it like the inverting summer (adding currents into a virtual ground) gives −12 V.

  5. MCQStandard

    An ideal inverting amplifier has Rin=10 kΩ and Rf=100 kΩ. The input resistance seen by the signal source is
    1. infinite, because the op-amp draws no current
    2. zero, because of the virtual ground
    3. 10 kΩ
    4. 100 kΩ
    Answer and solution

    Answer: C

    The inverting input is held at (virtual) ground, so the source sees Rin going to ground: the input current is Vin/Rin and the input resistance is 10 kΩ. This is a weakness of the inverting stage: raising its input resistance means raising both resistors. The non-inverting stage, by contrast, has a very high input resistance.

    Why people get this wrong: "Op-amps have infinite input resistance" describes the op-amp's own terminals, not this circuit's input.

  6. NATExam level

    An op-amp with open-loop gain A=1000 is used in a non-inverting amplifier whose resistor divider feeds back β=0.1 of the output. The closed-loop gain is ______ (to three decimal places).
    Answer and solution

    Answer: 9.901 (±0.002)

    From Vo=A(Vin−βVo): Vo/Vin=A/(1+Aβ)=1000/(1+100)=9.901. The ideal value 1/β=10 is off by about 1% because the loop gain Aβ=100 is not infinite.

    Why people get this wrong: Using the ideal value 10 is wrong when the question gives a finite A.

  7. MCQExam level

    A Wien bridge oscillator uses a non-inverting amplifier with resistors Rf (feedback) and R1 (to ground). For oscillations to just start and be sustained, the resistors must satisfy (at least)
    1. Rf=R1
    2. Rf=2R1
    3. Rf=3R1
    4. Rf=29R1
    Answer and solution

    Answer: B

    At its centre frequency f=1/(2πRC) the Wien network has zero phase shift and passes 1/3 of the voltage. Barkhausen: the loop gain must be 1, so the amplifier needs gain 3=1+Rf/R1, i.e. Rf=2R1.

    Why people get this wrong: Rf=3R1 gives gain 4 (it confuses Rf/R1 with the gain); 29 belongs to the RC phase-shift oscillator.

  8. MSQExam level

    Which statements about the ideal op-amp integrator (input resistor R, feedback capacitor C) are correct?
    1. Vo=−1RC∫Vindt.
    2. The magnitude of its gain rises with frequency.
    3. At DC its gain is unlimited, so even a tiny input offset slowly drives it into saturation.
    4. A large resistor is often placed across C to limit its gain at low frequencies.
    Answer and solution

    Answer: A, C, D

    The input current Vin/R flows on into the capacitor, so CdVo/dt=−Vin/R. In terms of frequency the gain is −1/(jωRC), whose size FALLS as 1/ω (that is the differentiator that rises). At ω→0 the gain is unbounded, so offsets integrate up to a rail; a resistor across C caps the DC gain at −Rf/R.

    Why people get this wrong: Integrator and differentiator are easy to swap: the integrator is a low-pass circuit.

  9. NATExam level

    A three-op-amp instrumentation amplifier has input-stage resistors R=25 kΩ on each side, a gain-setting resistor RG=1 kΩ between them, and a unity-gain difference stage. Its differential gain is ______.
    Answer and solution

    Answer: 51 (±0.01)

    The two input followers force the input difference across RG, so the current (V2−V1)/RG flows through R+RG+R. The difference between the two buffer outputs is (V2−V1)(1+2R/RG), and the unity difference stage passes it on: gain 1+2×25/1=51.

    Why people get this wrong: 1+R/RG=26 forgets that there is a resistor R on each side.

  10. MCQExam level

    An op-amp with input offset voltage 2 mV is used in an inverting amplifier with Rf/Rin=100. With the input grounded, the output offset is about
    1. 200 mV
    2. 202 mV
    3. 2 mV
    4. 20 mV
    Answer and solution

    Answer: B

    The offset acts like a small source in series with the non-inverting input. For that source the circuit is a non-inverting amplifier with gain 1+Rf/Rin=101 (the "noise gain"). Output offset =101×2 mV =202 mV.

    Why people get this wrong: The signal gain is 100, but the offset sees the noise gain, 101.

  11. MCQExam level

    An op-amp with gain-bandwidth product 1 MHz is wired as a non-inverting amplifier of gain 100. At 100 kHz the magnitude of its closed-loop gain is about
    1. 10
    2. 100
    3. 1
    4. 50
    Answer and solution

    Answer: A

    The closed-loop corner is at 1 MHz/100=10 kHz. Ten times above it, a single-pole response has fallen by a factor of about 10 (exactly 100/1+102=9.95). Equivalently, above the corner the curve follows the open-loop gain ≈GBW/f=10.

    Why people get this wrong: The gain is flat only up to the corner frequency, not up to the GBW.

  12. MCQExam level

    An op-amp circuit has a resistor from the input to the inverting input and a forward-biased diode in the feedback path. Its output is proportional to
    1. Vin2
    2. eVin
    3. 1/Vin
    4. −ln⁡Vin (for positive Vin)
    Answer and solution

    Answer: D

    The current Vin/R must flow through the diode, whose current is ISeVD/VT. The diode voltage is the output with a minus sign, so Vo=−VTln⁡(Vin/(RIS)): a logarithmic amplifier. Swapping the diode and resistor gives the antilog (exponential) amplifier.

    Why people get this wrong: The exponential law of the diode, sitting in the feedback path, is inverted: exponential in the element means logarithm in the circuit.

  13. NATExam level

    A unity-gain Sallen-Key low-pass filter has R1=R2=10 kΩ, C1=20 nF (from the junction of the resistors to the output) and C2=10 nF (from the non-inverting input to ground). Its natural frequency f0=12πR1R2C1C2 is ______ Hz (to the nearest whole number).
    Answer and solution

    Answer: 1125 Hz (±5)

    R1R2C1C2=(104)2(2.0×10−8)(1.0×10−8)=2.0×10−8 s2, whose square root is 1.414×10−4 s. f0=12π×1.414×10−4=1125 Hz. Its quality factor is Q=R1R2C1C2C2(R1+R2)=1.414×10−42.0×10−4=0.707: a Butterworth response, chosen by making C1=2C2.

    Why people get this wrong: Using 1/2πRC with one of the capacitors gives 796 Hz or 1592 Hz.

  14. MSQExam level

    Which statements about active filters are correct?
    1. A second-order Butterworth low-pass filter has Q=1/2.
    2. A unity-gain Sallen-Key low-pass with equal resistors and equal capacitors has Q=0.5.
    3. An active filter cannot have a pass-band gain greater than 1.
    4. A first-order low-pass filter falls off at 20 dB per decade above its cutoff.
    Answer and solution

    Answer: A, B, D

    Each pole adds −20 dB/decade, so first order gives 20 and second order 40 dB/decade. The Butterworth (maximally flat) second-order response has Q=0.707. For the unity-gain Sallen-Key, Q=R1R2C1C2C2(R1+R2)=RC2RC=0.5 with equal parts (a bit droopy; choose C1=2C2 for Butterworth). Op-amp filters can easily have gain, e.g. by setting the amplifier gain above 1; that is one of their advantages over passive RC filters.

    Why people get this wrong: Thinking "filter" means "passive": the op-amp can provide gain and a low output impedance at the same time.

  15. MCQExam level

    A non-inverting op-amp amplifier uses which kind of negative feedback, and what does it do to the input and output resistances?
    1. Series-shunt (voltage-series): it raises the input resistance and lowers the output resistance.
    2. Shunt-shunt (voltage-shunt): it lowers both the input and output resistances.
    3. Series-series (current-series): it raises both resistances.
    4. Shunt-series (current-shunt): it lowers the input resistance and raises the output resistance.
    Answer and solution

    Answer: A

    The feedback network samples the output voltage (shunt connection at the output) and returns a voltage that sits in series with the input signal, between the two op-amp inputs. Series mixing multiplies the input resistance by (1+Aβ); shunt sampling divides the output resistance by (1+Aβ). That makes it a near-ideal voltage amplifier. The inverting amplifier is the shunt-shunt case.

    Why people get this wrong: Naming the topology by the output connection only; the input mixing (series here) decides what happens to the input resistance.

  16. NATHard

    A difference amplifier has R1=10 kΩ from V1 to the inverting input, R2=50 kΩ feedback, R3=10 kΩ from V2 to the non-inverting input and R4=40 kΩ from the non-inverting input to ground. With V1=V2=1 V the output is ______ V.
    Answer and solution

    Answer: -0.2 V (±0.002)

    V+=V2R4/(R3+R4)=0.8V2. Vo=−(R2/R1)V1+(1+R2/R1)V+=−5V1+6×0.8V2=−5V1+4.8V2. With both at 1 V: −0.2 V. A true difference amplifier needs R4/R3=R2/R1; here the mismatch lets a common-mode input through.

    Why people get this wrong: Assuming "equal inputs give zero output" is true only for matched ratios.

  17. NATHard

    An op-amp has gain-bandwidth product 2 MHz and slew rate 1 V/μs. It is used as a non-inverting amplifier of gain 10 with a sine input of 1 V peak. The highest input frequency for an undistorted output is ______ kHz (to one decimal place).
    Answer and solution

    Answer: 15.9 kHz (±0.1)

    Two limits. Small-signal bandwidth: 2 MHz/10=200 kHz. Slew-rate limit for a 10 V peak output: f=SR/(2πVp)=106/(2π×10)=15.9 kHz. The lower one wins: 15.9 kHz.

    Why people get this wrong: Quoting the 200 kHz bandwidth ignores that the large output swing is slew-limited much earlier.

  18. MCQHard

    An RC phase-shift oscillator uses three identical sections, each a series capacitor C=10 nF followed by a resistor R=10 kΩ to ground, driving an inverting amplifier. Its frequency of oscillation is about
    1. 650 Hz
    2. 1.59 kHz
    3. 3.9 kHz
    4. 265 Hz
    Answer and solution

    Answer: A

    For this ladder the total phase shift reaches 180∘ at ω=1/(6RC), where the network passes 1/29 of the voltage (so the amplifier needs gain 29). f=1/(2π6RC)=1/(2π×2.449×10−4)=650 Hz.

    Why people get this wrong: 1/(2πRC)=1.59 kHz is the Wien bridge frequency, not the three-section ladder.

  19. NATHard

    An inverting amplifier with Rin=10 kΩ and Rf=100 kΩ uses an op-amp of open-loop gain 1000 (infinite input impedance, zero output impedance). Its closed-loop gain is ______ (to two decimal places).
    Answer and solution

    Answer: -9.89 (±0.01)

    Let V− be the inverting input voltage; Vo=−AV−. Current balance: (Vin−V−)/Rin=(V−−Vo)/Rf. Solving, Vo/Vin=−Rf/Rin1+(1+Rf/Rin)/A=−101+11/1000=−9.89.

    Why people get this wrong: The error term uses the noise gain 1+Rf/Rin=11, not 10.

  20. MCQHard

    On a Bode plot, the gain of an ideal integrator with R=10 kΩ and C=1 μF falls at 20 dB per decade. It crosses 0 dB (gain magnitude 1) at about
    1. 100 Hz
    2. 15.9 Hz
    3. 1.59 Hz
    4. 159 Hz
    Answer and solution

    Answer: B

    |A|=1/(ωRC)=1 when ω=1/RC=100 rad/s, i.e. f=100/(2π)=15.9 Hz.

    Why people get this wrong: 100 is the angular frequency in rad/s, not the frequency in Hz.

  21. MSQHard

    For a real op-amp amplifier delivering a large-amplitude sine wave, which of these limit the highest frequency at which the output is undistorted?
    1. The slew rate
    2. The gain-bandwidth product
    3. The input offset voltage
    4. The common-mode rejection ratio
    Answer and solution

    Answer: A, B

    The slew rate caps how fast the output can change, which matters for large amplitudes. The gain-bandwidth product sets where the closed-loop gain starts to fall. The offset voltage adds a DC shift and the CMRR controls how well common-mode signals are rejected; neither sets a frequency limit for the wanted signal.

    Why people get this wrong: Offset and CMRR are accuracy specs, not speed specs.

  22. MCQHard

    A diode is placed inside the feedback loop of a voltage follower: the op-amp output drives the diode's anode, the diode's cathode is the circuit output, and the output is fed back to the inverting input and has a load resistor to ground. For a 50 mV-peak sine input, the output is
    1. zero, since the input is far below the diode's 0.6 V turn-on
    2. the positive half-cycles reproduced accurately, with the negative half-cycles removed
    3. the negative half-cycles reproduced accurately, with the positive half-cycles removed
    4. the full sine wave, unchanged
    Answer and solution

    Answer: B

    For positive inputs the op-amp raises its own output to whatever is needed (about 0.6 V above the circuit output) to make the fed-back output equal the input: the diode drop is divided by the huge open-loop gain. For negative inputs the diode is reverse biased, the loop opens, and the load pulls the output to zero. This "super-diode" is a precision half-wave rectifier that works for millivolt signals.

    Why people get this wrong: Treating the diode as if it were outside the loop, where its 0.6 V threshold would block small signals.

  23. MCQHard

    An op-amp with gain-bandwidth product fT is used (a) as a voltage follower and (b) as an inverting amplifier of gain −1. Their closed-loop −3 dB bandwidths are about
    1. fT for both
    2. fT/2 for (a) and fT for (b)
    3. fT for (a) and fT/2 for (b)
    4. fT/2 for both
    Answer and solution

    Answer: C

    The bandwidth is fTβ, where β is the feedback fraction (equivalently fT divided by the noise gain 1+Rf/R1). The follower has β=1: bandwidth fT. The inverting amplifier with Rf=R1 has β=12 (noise gain 2), even though its signal gain is only −1: bandwidth fT/2.

    Why people get this wrong: Dividing fT by the signal gain magnitude (1 in both cases) instead of by the noise gain.

  24. NATHard

    An inverting amplifier has Rin=10 kΩ. Its feedback path is a T-network: 100 kΩ from the inverting input to a node X, 100 kΩ from X to the output, and 1.0 kΩ from X to ground. The magnitude of its gain is ______.
    Answer and solution

    Answer: 1020 (±1)

    Current I=Vin/10 k flows through the first 100 k, so node X is at −100I⋅(kΩ) =−10Vin. The 1 k to ground then draws 10Vin/1 k =100I, so the second 100 k carries I+100I=101I. Output: VX−100 k×101I=−10Vin−1010Vin=−1020Vin. In general |A|=R2+R3+R2R3/R4Rin, high gain without huge resistors.

    Why people get this wrong: Adding only the series resistors (gain 20), ignoring the current drawn by the resistor to ground.

  25. NATHard

    A practical integrator has input resistor 10 kΩ, feedback capacitor 0.10μF, and a 1.0 MΩ resistor across the capacitor to stop the output drifting. Below a corner frequency it acts as an amplifier of gain −100 rather than an integrator. That corner frequency is ______ Hz (to two decimal places).
    Answer and solution

    Answer: 1.59 Hz (±0.02)

    The feedback impedance is Rf‖(1/jωC). The capacitor dominates when 1/ωC<Rf, i.e. above fc=12πRfC=12π(106)(10−7)=1.59 Hz. Below it the gain levels off at −Rf/R=−100; above it the circuit integrates. (The input resistor does not set this corner.)

    Why people get this wrong: Using the input resistor in the corner formula (1/2πRC=159 Hz is where the integrator's gain magnitude is 1).

  26. NATHard

    A non-inverting Schmitt trigger feeds the input through R1=10 kΩ to the non-inverting input, with R2=40 kΩ from the output back to the same input. The inverting input is held at +1.0 V, and the output saturates at ±12 V. Starting with the output low, the input must rise to ______ V before the output switches high (to two decimal places).
    Answer and solution

    Answer: 4.25 V (±0.02)

    The output flips when V+ crosses 1.0 V. By superposition V+=VinR2+VoutR1R1+R2. With Vout=−12 V: 1.0=40Vin−12050, so Vin=50+12040=4.25 V. The lower threshold (output high) is 50−12040=−1.75 V, so the hysteresis is 6.0 V wide and centred on 1.25 V, not on the 1.0 V reference.

    Why people get this wrong: Assuming the thresholds sit symmetrically about the reference (1.0±3.0 V); the reference is scaled by (R1+R2)/R2.

  27. MCQHard

    An op-amp whose open-loop response has two poles is stable as an amplifier with gain 10 but oscillates when wired as a voltage follower. The reason is that the follower
    1. draws more current from the source
    2. has positive feedback
    3. has a higher slew rate
    4. has the largest feedback fraction, β=1, so its loop gain Aβ crosses unity at a higher frequency where the phase shift is larger
    Answer and solution

    Answer: D

    Stability depends on the loop gain Aβ. A follower feeds back all of the output (β=1), the most any resistive network can, so |Aβ|=1 happens at the highest frequency, where the second pole has added the most phase lag. If the total lag there reaches 180∘ the loop oscillates. Higher-gain stages (β=0.1) cross over lower down, with more phase margin. "Unity-gain stable" op-amps are compensated for exactly this worst case.

    Why people get this wrong: Thinking low gain means "easy" for the op-amp: for stability the unity-gain follower is the hardest case.

  28. MSQHard

    An amplifier with open-loop gain A (a single pole) has negative feedback with feedback fraction β. Which statements are correct?
    1. A fractional change in A changes the closed-loop gain by that fraction divided by (1+Aβ).
    2. The bandwidth rises by the factor (1+Aβ).
    3. If the output voltage is sampled, the output resistance falls by (1+Aβ).
    4. In an inverting amplifier, the resistance looking into the op-amp's inverting-input node rises by (1+Aβ).
    Answer and solution

    Answer: A, B, C

    G=A1+Aβ gives dGG=11+AβdAA. With a single pole the gain-bandwidth product is conserved, so the bandwidth grows by (1+Aβ) as the gain falls by it. Voltage sampling divides the output resistance by (1+Aβ). At the inverting-input node the feedback resistor looks like Rf/(1+A) (Miller effect): the node is a virtual ground with very low resistance, not high.

    Why people get this wrong: Assuming feedback always raises input resistance: shunt mixing (as at the inverting input) lowers it.

  29. NATTop rank

    A difference amplifier built with an ideal op-amp has R1=R3=10 kΩ (input resistors), feedback R2=100 kΩ, and R4=99 kΩ from the non-inverting input to ground (a 1% mismatch). Defining the differential input as V2−V1 and the common-mode input as (V1+V2)/2, the CMRR of the circuit is ______ dB (to one decimal place).
    Answer and solution

    Answer: 60.7 dB (±0.15)

    Vo=−R2R1V1+(1+R2R1)R4R3+R4V2=−10V1+11×99109V2=−10V1+9.9908V2. Write V1=Vcm−Vd/2, V2=Vcm+Vd/2: Acm=−10+9.9908=−0.00917 and Ad=(10+9.9908)/2=9.9954. CMRR =9.9954/0.00917=1090, i.e. 20log10⁡1090=60.7 dB, even though the op-amp itself is ideal.

    Why people get this wrong: A perfect op-amp does not guarantee a perfect difference amplifier: resistor matching sets the CMRR.

  30. MSQTop rank

    An inverting amplifier (Rin, Rf) uses an op-amp with large but finite open-loop gain A (and ideal input and output impedances). Which statements are correct?
    1. The inverting input sits at −Vo/A, not exactly at zero.
    2. The magnitude of the closed-loop gain is slightly less than Rf/Rin.
    3. The relative gain error gets larger as Rf/Rin is made smaller.
    4. The input resistance is Rin+Rf/(1+A).
    Answer and solution

    Answer: A, B, D

    With V+=0, Vo=A(0−V−), so V−=−Vo/A. The gain is −(Rf/Rin)/[1+(1+Rf/Rin)/A], a little smaller in size, and the fractional error (1+Rf/Rin)/A GROWS with larger Rf/Rin (less loop gain is left over), so the claim that the error grows as Rf/Rin is made smaller is backwards. Seen from the inverting node, Rf is divided by 1+A (the Miller effect for a resistor), so the input resistance is Rin+Rf/(1+A).

    Why people get this wrong: High-gain stages are the less accurate ones: the loop gain Aβ shrinks as the closed-loop gain grows.

← All topics · Timed mock test →