Oscillators

Electronics · 77 questions and templates for CSIR NET 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

    An amplifier of gain A with a feedback network β sustains a steady sinusoidal oscillation at the frequency where
    1. Aβ=−1 with |A|=1
    2. |Aβ|=0
    3. |Aβ|=1 and the phase of Aβ is 0 (or a multiple of 360∘)
    4. |Aβ|→∞
    Answer and solution

    Answer: C

    For a signal to keep itself going around the loop it must come back exactly as it left: same amplitude (|Aβ|=1) and same phase (total phase shift 0∘ or 360∘n). This is the Barkhausen criterion; the frequency is fixed by the phase condition.

    Why people get this wrong: Aβ=−1 is the condition for a feedback amplifier written with a minus sign at the mixer (1+Aβ=0); be clear which convention a question uses.

  2. MCQWarm-up

    A sinusoidal oscillator is best described as
    1. an amplifier with positive feedback at one frequency, with loop gain one there
    2. an amplifier with strong negative feedback at all frequencies
    3. a filter with no amplifier
    4. a rectifier followed by a capacitor
    Answer and solution

    Answer: A

    The feedback network selects one frequency at which the fed-back signal is in phase with the input (positive, regenerative feedback) and the amplifier makes up the network's loss. Everywhere else the loop gain is below one or the phase is wrong, so only that frequency survives.

    Why people get this wrong: Negative feedback stabilises amplifiers; oscillators need the fed-back signal to add to the input.

  3. MCQWarm-up

    A Wien bridge oscillator uses equal resistors R and equal capacitors C in its lead-lag network. It oscillates at
    1. 1/(2π6RC)
    2. 1/(2πLC)
    3. 1/(RC) hertz
    4. 1/(2πRC)
    Answer and solution

    Answer: D

    The series RC and parallel RC arms give a transfer function with zero phase at ω0=1/RC, where it equals 1/3. So f0=1/(2πRC), and the amplifier must have a gain of 3.

    Why people get this wrong: 1/(2π6RC) belongs to the RC phase-shift oscillator.

  4. NATWarm-up

    A Colpitts oscillator has L=10μH and two tank capacitors of 2 nF each. It oscillates at about ______ MHz (to two decimal places).
    Answer and solution

    Answer: 1.59 MHz (±0.01)

    The two capacitors are in series around the tank: Ceq=C1C2/(C1+C2)=1 nF. f=1/(2πLCeq)=1/(2π10−5×10−9)=1/(2π×10−7)=1.59 MHz.

    Why people get this wrong: Adding the capacitors (4 nF) as if in parallel gives 0.80 MHz.

  5. MCQWarm-up

    Quartz crystals are used to fix the frequency of clocks and radio transmitters mainly because
    1. they amplify the signal
    2. their resonance has an extremely high quality factor, so the frequency hardly drifts
    3. they have no capacitance
    4. they work only at audio frequencies
    Answer and solution

    Answer: B

    A crystal is a piezoelectric mechanical resonator with Q of 104 to 106. The phase of its impedance changes extremely steeply near resonance, so changes elsewhere in the circuit shift the frequency by only parts per million.

    Why people get this wrong: The crystal is passive: the transistor or op-amp supplies the gain.

  6. MSQWarm-up

    Which of these are RC oscillators (frequency set by resistors and capacitors, no inductor)?
    1. Wien bridge oscillator
    2. RC phase-shift oscillator
    3. Colpitts oscillator
    4. Twin-T oscillator
    Answer and solution

    Answer: A, B, D

    Wien bridge, phase-shift and twin-T oscillators use RC networks and are common at audio frequencies, where inductors would be bulky. The Colpitts oscillator uses an LC tank (a tapped capacitor across an inductor).

    Why people get this wrong: The Colpitts circuit has two capacitors, but its frequency needs the inductor.

  7. MCQWarm-up

    The tank capacitance of an LC oscillator is made four times larger, with the inductance unchanged. The oscillation frequency
    1. doubles
    2. halves
    3. falls to a quarter
    4. falls by a factor of 2
    Answer and solution

    Answer: B

    f=1/(2πLC), so f∝C−1/2. Four times the capacitance gives 1/4=1/2 of the frequency.

    Why people get this wrong: Treating f as inversely proportional to C (a quarter) forgets the square root.

  8. NATStandard

    A Wien bridge oscillator uses a non-inverting op-amp amplifier with R1=10 kΩ from the inverting input to ground. The feedback resistor Rf needed for a loop gain of exactly one is ______ kΩ.
    Answer and solution

    Answer: 20 kΩ

    At f0 the lead-lag network passes 1/3 with zero phase, so the amplifier gain must be 1+Rf/R1=3: Rf=2R1=20 kΩ. In practice it is set slightly higher to start, with a non-linear element to settle.

    Why people get this wrong: Setting Rf/R1=3 gives a gain of 4: the output grows until it clips.

  9. MCQStandard

    An RC phase-shift oscillator uses three identical high-pass RC sections in a ladder driven by an inverting amplifier. The minimum magnitude of amplifier gain is
    1. 3
    2. 29
    3. 8
    4. 1/29
    Answer and solution

    Answer: B

    Solving the ladder gives β=1/(1−5α2−j(6α−α3)) with α=1/ωRC. The phase is 180∘ when α2=6, and then β=1/(1−30)=−1/29. The inverting amplifier supplies the other 180∘ and a gain of at least 29.

    Why people get this wrong: 8 is the gain for three buffered (isolated) sections of 60∘ each; the loaded ladder needs 29.

  10. NATStandard

    An RC phase-shift oscillator with three identical high-pass sections uses R=10 kΩ and C=10 nF. Its frequency is ______ Hz (to the nearest whole number).
    Answer and solution

    Answer: 650 Hz (±2)

    f=12π6RC=12π×2.449×10−4=650 Hz.

    Why people get this wrong: Using 1/(2πRC)=1592 Hz (the Wien frequency).

  11. NATStandard

    A Hartley oscillator has a tapped coil with sections of 2 mH and 0.5 mH (no mutual coupling between them) and a 1 nF tank capacitor. Its frequency is ______ kHz (to one decimal place).
    Answer and solution

    Answer: 100.7 kHz (±0.3)

    With no coupling, Leq=L1+L2=2.5 mH. f=1/(2π2.5×10−3×10−9)=1/(2π×1.581×10−6)=100.7 kHz.

    Why people get this wrong: Combining the coil sections in parallel.

  12. MCQStandard

    In a Colpitts oscillator, the fraction of the tank voltage fed back to the amplifier input is set mainly by
    1. the value of the inductor
    2. the supply voltage
    3. the ratio of the two tank capacitors
    4. the load resistance alone
    Answer and solution

    Answer: C

    The two capacitors form a capacitive voltage divider across the tank; the tap between them feeds the amplifier. Changing their ratio changes the feedback fraction (and the gain needed), while their series value with L sets the frequency.

    Why people get this wrong: The inductor sets the frequency, not the feedback fraction.

  13. MCQStandard

    Practical oscillators are designed with
    1. loop gain exactly 1 from the moment they are switched on
    2. loop gain well below 1, so the oscillation is stable
    3. loop gain slightly above 1 at start-up, with a non-linearity that brings it down to exactly 1 at the final amplitude
    4. no feedback at all
    Answer and solution

    Answer: C

    Oscillation starts from noise, so the small-signal loop gain must exceed 1 for the amplitude to grow. As it grows, a non-linear element (amplifier saturation, a lamp, thermistor, diodes or an AGC loop) reduces the gain until the average loop gain is exactly 1. A lamp or thermistor gives a cleaner sine than clipping.

    Why people get this wrong: Exactly 1 at switch-on: any small loss and the oscillation would never start.

  14. MCQStandard

    A quartz crystal's motional arm has L=0.5 H and C=0.02 pF. Its series resonant frequency is about
    1. 159 kHz
    2. 15.9 MHz
    3. 1.59 MHz
    4. 5.0 MHz
    Answer and solution

    Answer: C

    fs=1/(2πLC)=1/(2π0.5×2×10−14)=1/(2π×10−7)=1.59 MHz. The huge L and tiny C model the mechanical mass and stiffness of the quartz plate.

    Why people get this wrong: Slipping a power of ten under the square root.

  15. MSQStandard

    About a Wien bridge oscillator with equal R and equal C, which statements are correct?
    1. At f0 the lead-lag network has zero phase shift
    2. At f0 the lead-lag network passes one third of its input
    3. The amplifier must be inverting
    4. An NTC thermistor in place of Rf can stabilise the amplitude
    Answer and solution

    Answer: A, B, D

    The network transfer is 1/[3+j(ωRC−1/ωRC)]: real and equal to 1/3 at ωRC=1. Zero phase means the amplifier must be non-inverting with gain 3. If the amplitude grows, the thermistor warms, its resistance falls, the gain falls: automatic level control.

    Why people get this wrong: Inverting amplifiers go with the phase-shift oscillator, whose network gives 180∘.

  16. NATStandard

    A UJT relaxation oscillator has R=10 kΩ, C=0.1μF and intrinsic stand-off ratio η=0.6 (diode drop and discharge time neglected). Its frequency is ______ Hz (to the nearest whole number).
    Answer and solution

    Answer: 1091 Hz (±3)

    The capacitor charges from about 0 towards VBB and the UJT fires at ηVBB: 1−e−T/RC=η, so T=RCln⁡11−η=10−3ln⁡2.5=0.916 ms and f=1091 Hz.

    Why people get this wrong: Using ln⁡(1/η) instead of ln⁡[1/(1−η)].

  17. MCQStandard

    At the frequency where its loop phase is zero, an oscillator circuit has loop gain 0.98. After it is disturbed, the oscillation
    1. grows until the amplifier clips
    2. dies away gradually
    3. continues at constant amplitude
    4. jumps to twice the frequency
    Answer and solution

    Answer: B

    Each trip around the loop multiplies the amplitude by 0.98, so the oscillation decays (slowly, like a lightly damped resonator). Sustained oscillation needs at least 1.

    Why people get this wrong: Close to 1 is not 1: the amplitude still decays exponentially.

  18. MCQStandard

    Which pairing of oscillator and its feedback tap is correct?
    1. Colpitts: tapped inductor; Hartley: tapped capacitor
    2. Hartley: tapped inductor; Colpitts: tapped capacitor
    3. Both use a tapped capacitor
    4. Both use an RC ladder
    Answer and solution

    Answer: B

    Hartley taps the coil (two inductors in series across one capacitor); Colpitts taps the capacitance (two capacitors in series across one inductor). The Clapp is a Colpitts with an extra capacitor in series with the inductor.

    Why people get this wrong: A memory aid: Colpitts and capacitor both start with C.

  19. MCQStandard

    A frequency synthesiser locks a voltage-controlled oscillator to a 10 MHz crystal reference. The VCO output passes through a divide-by-125 counter before the phase detector compares it with the reference. In lock, the VCO runs at
    1. 80 kHz
    2. 10 MHz
    3. 125 MHz
    4. 1.25 GHz
    Answer and solution

    Answer: D

    In lock the two inputs of the phase detector have the same frequency: fVCO/125=10 MHz, so fVCO=1.25 GHz. Changing the division ratio steps the output in 10 MHz steps, each as stable as the crystal.

    Why people get this wrong: 80 kHz is the reference divided by 125; the divider is in the feedback path, so it multiplies.

  20. NATStandard

    A two-transistor astable multivibrator has timing components R1=10 kΩ, C1=10 nF for one half-cycle and R2=30 kΩ, C2=10 nF for the other. Each half-cycle lasts RCln⁡2. Its frequency is ______ kHz (to two decimal places).
    Answer and solution

    Answer: 3.61 (±0.02)

    Half-periods: 0.693×10−4=69.3 μs and 0.693×3×10−4=207.9 μs. Period 277.3 μs, so f=3.61 kHz. The output is high for 25% or 75% of the cycle depending on which collector is used.

    Why people get this wrong: Using only one time constant, or 1/RC without the ln⁡2.

  21. MCQStandard

    A two-transistor monostable multivibrator has timing resistor 47 kΩ and timing capacitor 10 nF; its pulse lasts RCln⁡2. When triggered it gives an output pulse of about
    1. 0.47 ms
    2. 3.3 ms
    3. 0.33 ms
    4. 0.65 ms
    Answer and solution

    Answer: C

    RC=47×103×10×10−9=0.47 ms, and the pulse is 0.693RC=0.33 ms. The circuit then returns to its one stable state and waits for the next trigger.

    Why people get this wrong: Quoting RC itself (0.47 ms) leaves out the ln⁡2 from the capacitor recharging through half its swing.

  22. NATExam level

    A Wien network has a series arm R1=10 kΩ, C1=10 nF and a parallel arm R2=20 kΩ, C2=10 nF. The non-inverting amplifier gain needed for oscillation is ______ (to two decimal places).
    Answer and solution

    Answer: 2.5 (±0.01)

    1/β=1+Z1/Z2=1+R1R2+C2C1+j(ωR1C2−1ωR2C1). The imaginary part vanishes at ω0=1/R1R2C1C2 (1125 Hz here), where 1/β=1+0.5+1=2.5. So the gain must be 2.5.

    Why people get this wrong: Assuming the gain is always 3; that needs equal R and equal C.

  23. MCQExam level

    In the three-section RC ladder of a phase-shift oscillator at its oscillation frequency, which statement is correct?
    1. Each section gives exactly 60∘
    2. The total shift is 180∘, but the sections load each other and their individual shifts are unequal
    3. The total shift is 90∘
    4. The ladder gives no attenuation
    Answer and solution

    Answer: B

    Only the total matters: 180∘, at ωRC=1/6 for high-pass sections. Because each section loads the one before, the shifts are not 60∘ apiece; that is why the attenuation is 29 rather than the 8 of buffered sections.

    Why people get this wrong: The equal-60∘ picture is true only with buffers between sections.

  24. NATExam level

    A Colpitts oscillator has L=100μH, C1=1 nF and C2=10 nF. Its frequency is ______ kHz (to the nearest whole number).
    Answer and solution

    Answer: 528 kHz (±2)

    Ceq=1×101+10=0.909 nF, dominated by the smaller capacitor. f=1/(2π10−4×9.09×10−10)=528 kHz.

    Why people get this wrong: Using only C2 gives 159 kHz.

  25. NATExam level

    A 10 MHz crystal (series resonance) has motional capacitance 0.02 pF and shunt capacitance C0=5 pF. Its parallel resonance lies above the series resonance by ______ kHz (to two decimal places).
    Answer and solution

    Answer: 19.98 kHz (±0.05)

    At parallel resonance the motional arm resonates with C in series with C0: fp=fs1+C/C0=fs1.004. fp−fs=107×0.001998=19.98 kHz, about fsC/2C0=20 kHz: the two resonances are very close.

    Why people get this wrong: Using C0/C instead of C/C0.

  26. MCQExam level

    A crystal's motional arm has L=0.1 H and series resistance 50Ω, resonating at 5 MHz. Its quality factor is about
    1. 6.3×104
    2. 1.0×104
    3. 6.3×102
    4. 3.1×106
    Answer and solution

    Answer: A

    Q=ωL/R=2π×5×106×0.1/50=6.28×104. An LC tank of discrete parts rarely exceeds a few hundred.

    Why people get this wrong: Forgetting the 2π gives 104.

  27. MSQExam level

    Which statements about the Barkhausen criterion are correct?
    1. At the oscillation frequency |Aβ|=1 and ∠Aβ=0∘ (mod 360∘)
    2. Meeting it guarantees a stable, constant amplitude without any non-linearity
    3. If |Aβ|>1 at the zero-phase frequency, the amplitude grows until a non-linearity limits it
    4. The oscillation frequency is fixed by the phase condition
    Answer and solution

    Answer: A, C, D

    Barkhausen describes the steady state. A linear loop with exactly |Aβ|=1 is marginal: the slightest change makes the amplitude grow or decay, so real oscillators rely on an amplitude-dependent gain. The phase condition picks the frequency; the magnitude condition fixes the amplitude through the non-linearity.

    Why people get this wrong: Barkhausen is necessary for steady oscillation, not a recipe for amplitude stability.

  28. NATExam level

    A phase-shift oscillator uses three identical low-pass (series R, shunt C) sections in a ladder, with R=10 kΩ and C=10 nF. It oscillates at ______ Hz (to the nearest whole number).
    Answer and solution

    Answer: 3898 Hz (±10)

    For the low-pass ladder 1/β=1−5x2+j(6x−x3) with x=ωRC. The phase is 180∘ at x2=6: f=6/(2πRC)=2.449/(6.283×10−4)=3898 Hz, and again β=−1/29.

    Why people get this wrong: Using the high-pass result 1/(2π6RC); the roles of R and C swap.

  29. MCQExam level

    A Hartley oscillator's tapped coil has sections L1=1 mH and L2=4 mH with coupling coefficient k=0.5, wound so that their fluxes aid. The effective tank inductance is
    1. 5 mH
    2. 7 mH
    3. 3 mH
    4. 6 mH
    Answer and solution

    Answer: B

    M=kL1L2=0.5×2=1 mH. Series-aiding coils: Leq=L1+L2+2M=7 mH.

    Why people get this wrong: Adding M once, or subtracting it (3 mH is series-opposing).

  30. MCQExam level

    Both capacitors of an equal-component Wien bridge oscillator drift up by 2%. Its frequency
    1. rises by about 2%
    2. falls by about 4%
    3. is unchanged
    4. falls by about 2%
    Answer and solution

    Answer: D

    f0=1/(2πRC), so f∝1/C: 1/1.02=0.980, a fall of 1.96%. The required gain stays 3 because the network stays balanced.

    Why people get this wrong: Thinking of 1/C as in an LC tank (that would give 1%).

← All topics · Timed mock test →