Back to the on-screen lesson ·

Oscillation model limits

The simple harmonic model assumes small swings and no energy loss; large angles lengthen the period, and damping shrinks the amplitude each cycle.

Paper packet. Every task here also exists on screen, where it is checked automatically; answers written on paper are not assessed by Nydus. When you are back at a device, enter your answers there.

1. What you will learn

By the end of this lesson you will be able to judge when the simple harmonic model applies, correct a pendulum's period for large swings, and describe damping with a decay factor.

2. What you already have

You can model springs and small-angle pendulums as simple harmonic oscillators, find their periods, and track their energy. This lesson tests the two assumptions behind that model, a force exactly proportional to displacement and no energy lost, and measures how wrong the model becomes when they fail.

3. Words for this lesson

TermWhat it means
Small-angle approximationUsing $\sin\theta \approx \theta$, accurate only for small angles in radians.
DampingThe loss of an oscillator's energy to friction and drag.
Decay factorThe fraction $r$ of amplitude kept each cycle.
Elastic limitThe stretch beyond which a spring no longer obeys Hooke's law.
ResonanceThe growth of amplitude when pushes match the natural frequency.
Natural frequencyThe frequency at which a system oscillates on its own.

4. Two assumptions, two ways to fail

The simple harmonic model assumes:

  1. A linear restoring force, $F = -kx$. For a pendulum the true force is $mg\sin\theta$, which matches $mg\theta$ only for small angles. At large angles the pull is weaker, so the period grows: $T \approx T_0(1 + \theta_0^2/16)$.
  2. No energy loss. Real oscillators lose energy to friction and drag. The amplitude shrinks by a fraction each cycle, $A_n = A_0r^n$, while the period stays nearly the same.

Knowing when these assumptions hold tells you when the model's predictions can be trusted.

Another way: picture

Picture a child on a playground swing pulled back only a little: the swing keeps steady time, but each arc is a bit lower than the last until it stops. Pulled back very high, the swing takes noticeably longer to come back. Both effects are outside the simple model.

Another way: steps

  1. Check the size of the swing: is the angle small, under about fifteen degrees?
  2. If not, use $mg\sin\theta$ for the force and correct the period.
  3. Check for energy loss: does the amplitude shrink?
  4. If so, find the decay factor per cycle.
  5. Use $A_n = A_0r^n$ and $E_n = E_0r^{2n}$.

5. An oscillation dying away

A graph of position against time for a pendulum swinging in air. The blue curve rises and falls like a wave, but each peak is lower than the one before: every cycle keeps four fifths of the previous amplitude, so the peaks shrink from 80 to 64 to about 51 to about 41 units. Two dashed gray curves above and below trace the shrinking peaks. Green tick marks along the top mark the time of each peak: they are evenly spaced, because the period stays the same while the swing dies away.
A graph of position against time for a pendulum swinging in air. The blue curve rises and falls like a wave, but each peak is lower than the one before: every cycle keeps four fifths of the previous amplitude, so the peaks shrink from 80 to 64 to about 51 to about 41 units. Two dashed gray curves above and below trace the shrinking peaks. Green tick marks along the top mark the time of each peak: they are evenly spaced, because the period stays the same while the swing dies away.

The graph shows a pendulum's position against time as air drag slowly steals its energy. Each peak is lower than the one before: every cycle keeps four fifths of the previous amplitude. The dashed curves trace the shrinking peaks.

The green tick marks at each peak are evenly spaced. Even as the swing dies away, the period stays the same. Light damping changes the amplitude but hardly touches the period, which is why a pendulum clock keeps good time even as its swing varies.

6. The small-angle approximation

For a pendulum, the restoring force is $mg\sin\theta$. The simple model replaces $\sin\theta$ with $\theta$ in radians. At ten degrees the two differ by about half a percent; at thirty degrees, by about five percent; at sixty degrees, by about seventeen percent.

Because $\sin\theta$ is always less than $\theta$, the real restoring force is weaker than the model predicts. A weaker pull means slower motion and a longer period at large angles.

7. Large-angle periods

For moderate angles, the period grows roughly as $T \approx T_0(1 + \theta_0^2/16)$, with $\theta_0$ the release angle in radians. At fifteen degrees the correction is under half a percent; at sixty degrees, about seven percent.

This is why the period of a large swing depends on the amplitude, breaking the isochronism of the simple model. A clock whose pendulum is kicked into a larger swing runs slow.

8. Damping

Friction at the pivot and air drag on the bob remove a little energy each cycle. For light damping, each cycle keeps roughly the same fraction $r$ of the amplitude, so after $n$ cycles $A = A_0r^n$. The energy, proportional to $A^2$, keeps $r^2$ each cycle.

A swing that keeps ninety percent of its amplitude each cycle keeps eighty-one percent of its energy, and halves its amplitude in fewer than seven cycles.

9. Exponential decay

Multiplying by the same factor every cycle gives exponential decay: the amplitude falls quickly at first and then more and more slowly, never quite reaching zero in the model. In practice, static friction eventually stops the motion.

The number of cycles to halve the amplitude is $\ln 0.5/\ln r$, independent of the starting amplitude. This half-life of the swing is a convenient way to describe how strongly an oscillator is damped.

10. The method, step by step, and how to check it

  1. Angle check: small-angle model only below about fifteen degrees.
  2. Large angles: use $mg\sin\theta$ and $T_0(1 + \theta_0^2/16)$.
  3. Damping: find $r$ and use $A_0r^n$.
  4. Energy: multiply by $r^2$ per cycle.

Checking an answer. A large-angle period must exceed the small-angle one. A damped amplitude must shrink every cycle. The energy fraction must be smaller than the amplitude fraction.

11. Why each step is allowed

The sine of a small angle in radians is very close to the angle itself, because the arc of a circle and its chord nearly coincide for small angles. That is the only step the simple pendulum model needs, and it fails gradually as the angle grows.

The constant fraction per cycle follows when the damping force is proportional to the velocity, as for slow motion through air or oil. Each cycle then removes the same fraction of the energy.

12. Springs past their limits

Hooke's law holds only for small stretches. Stretch a spring too far and its coils deform permanently, and the force no longer grows in proportion to the stretch. The spring no longer returns to its original length.

Real materials also stiffen or soften at large deformations. Rubber bands, for example, stretch easily at first and then resist strongly. Their oscillations are not simple harmonic.

13. Resonance

A swing pushed at just the right moment each cycle builds up a large amplitude. Pushing at the oscillator's natural frequency adds energy every cycle faster than damping removes it, an effect called resonance.

Resonance can be useful, as in radio tuning and musical instruments, or dangerous. Engineers design bridges and buildings so their natural frequencies avoid the rhythms of wind, footsteps and traffic.

14. The Tacoma Narrows Bridge

In 1940, the Tacoma Narrows Bridge in Washington State twisted itself apart in a moderate wind only four months after opening. The wind drove large twisting oscillations that grew until the deck collapsed.

Engineers now test bridge designs in wind tunnels and add stiffening and damping to prevent such runaway oscillations. The collapse, caught on film, is still shown in engineering and physics classes.

15. Shock absorbers

A car's springs alone would let it bounce for many cycles after every bump. Shock absorbers add strong damping, pushing oil through small holes so the energy becomes heat. A good suspension settles in about one cycle.

Too little damping gives a bouncy ride; too much gives a harsh one. Engineers aim for damping just below the point where the car would return without overshooting at all.

16. Keeping oscillators going

Clocks, swings and museum pendulums lose energy every cycle and need small, well-timed pushes to keep going. A pendulum clock's escapement gives the pendulum a tiny kick each swing from a falling weight or wound spring.

Museum Foucault pendulums use an electromagnet that gives the bob a nudge as it passes. The pushes replace exactly the energy lost to drag, keeping the amplitude steady.

17. When the model is good enough

Every model has limits, and knowing them is part of using it well. For a grandfather clock swinging a few degrees, the simple model is accurate to a fraction of a percent. For a playground swing pumped high, it can be off by ten percent.

Before trusting a prediction, check the angle and the damping. If both are small, use the simple formulas with confidence; if not, apply the corrections from this lesson.

18. Measuring damping in the lab

Damping is easy to measure. Release a pendulum, and record the amplitude of each swing against a ruler or with a video camera. Dividing each amplitude by the one before gives the decay factor, which should stay roughly constant.

Adding a piece of cardboard to the bob increases air drag and lowers the decay factor. Plotting the amplitudes on a graph shows the smooth exponential decay that the model predicts.

19. Models as useful approximations

Every physics model leaves something out. The simple harmonic model ignores friction and treats every restoring force as a perfect spring. That makes it easy to use and surprisingly accurate for small, gentle oscillations, which is why engineers reach for it first.

The skill is knowing when the leftovers matter. A good habit is to estimate the size of each neglected effect, such as the angle error or the fraction of energy lost per cycle, and keep the simple model only when those effects are smaller than the accuracy the problem needs.

20. In the world: museum pendulums

The Foucault pendulums at Griffith Observatory, the Franklin Institute and the United Nations swing heavy bobs on long cables. Even so, air drag steals a little energy every swing. Left alone, such a pendulum loses about a quarter of its amplitude in an hour and would stop within a day.

To keep them going, museums use a Charron ring or an electromagnet that gives the bob a tiny push each swing, replacing exactly the energy drag removes. The swing angle is kept small, a few degrees, so the small-angle model holds and the period stays steady. The long period and heavy bob keep the damping gentle, so the pushes can be very small.

21. In the world: the Tacoma Narrows Bridge

On November 7, 1940, the first Tacoma Narrows Bridge in Washington State began twisting in a steady wind. The motion grew until the deck tore apart and fell into Puget Sound. Film of the collapse is one of the most famous engineering failures in history.

The wind fed energy into a twisting oscillation faster than the structure's damping could remove it, so the amplitude grew far beyond the small-motion regime the designers expected. The replacement bridge, opened in 1950, used a deeper, stiffer deck with openings to let wind pass, and modern bridges are tested in wind tunnels to make sure their oscillations stay small and well damped.

22. The small-angle period is not good for every swing

It is tempting to use $T = 2\pi\sqrt{L/g}$ for any pendulum, however far it swings. But that formula assumes $\sin\theta \approx \theta$, which fails at large angles. At sixty degrees the true period is about seven percent longer, and the error grows quickly beyond that.

A related error is to think damping changes the period a lot. Light damping shrinks the amplitude steadily but leaves the period nearly unchanged, which is why the tick marks in the damped graph stay evenly spaced.

23. How wrong is the small-angle model?

  1. Convert $30°$ to radians.

    $\theta = 30 \times \dfrac{\pi}{180} = 0.524\ \text{rad}$

    Radians for comparison.

  2. Find the sine.

    $\sin 30° = 0.500$

    Exact.

  3. Find the percent error of the approximation.

    $\dfrac{0.524 - 0.500}{0.500} = 4.7\%$

    Noticeable.

  4. Repeat at $10°$.

    $\dfrac{0.1745 - 0.1736}{0.1736} = 0.5\%$

    Negligible.

  5. Repeat at $60°$.

    $\dfrac{1.047 - 0.866}{0.866} = 21\%$

    Large.

24. A large-angle pendulum

  1. A $2.0$ m pendulum is released from $60°$. Find the small-angle period.

    $T_0 = 2\pi\sqrt{\dfrac{2.0}{9.8}} = 2.84\ \text{s}$

    Simple model.

  2. Convert the angle.

    $\theta_0 = 1.047\ \text{rad}$

    Radians.

  3. Find the correction factor.

    $1 + \dfrac{1.047^2}{16} = 1.069$

    Seven percent.

  4. Find the corrected period.

    $T = 2.84 \times 1.069 = 3.03\ \text{s}$

    Longer.

  5. Find the difference over an hour.

    $\dfrac{3600}{2.84} - \dfrac{3600}{3.03} = 80\ \text{swings}$

    Fewer swings.

  6. Repeat for a $10°$ release.

    $T = 2.84 \times 1.002 = 2.85\ \text{s}$

    Hardly changed.

25. A swing dying away

  1. A swing starts with a $1.2$ m amplitude and keeps $0.85$ of it each cycle. Find the amplitude after one cycle.

    $A_1 = 1.2 \times 0.85 = 1.02\ \text{m}$

    Multiply once.

  2. Find it after four cycles.

    $A_4 = 1.2 \times 0.85^4 = 0.626\ \text{m}$

    Multiply four times.

  3. Find the energy fraction per cycle.

    $0.85^2 = 0.7225$

    Energy goes as $A^2$.

  4. Find the energy left after four cycles.

    $0.7225^4 = 0.27$

    About a quarter.

  5. Find the cycles to half amplitude.

    $n = \dfrac{\ln 0.5}{\ln 0.85} = 4.3$

    About four.

  6. Find the cycles to a tenth.

    $n = \dfrac{\ln 0.1}{\ln 0.85} = 14.2$

    About fourteen.

  7. Note the period.

    $\text{about the same throughout}$

    Light damping.

26. Your turn: a pendulum keeps $0.9$ of its amplitude each cycle, starting at $20$ cm. What is its amplitude after two cycles?

  1. Write the rule.

    $A_n = A_0r^n$

    One factor per cycle.

  2. Substitute the values.

    $A_2 = 20 \times 0.9^2$

    Two cycles.

  3. Your turn: work this step out. Its working is at the end of the packet.

    Evaluate the amplitude.

27. Guided practice

A $1.2$ kg pendulum bob is pulled aside until its string makes $50°$ with the vertical. With $g = 9.8$ m/s², what is the exact restoring force along its path, in N?

28. Guided practice

Complete the worked solution: a swing left alone keeps $0.95$ of its amplitude after each cycle. Find the fraction of its energy kept each cycle, the percent of energy lost each cycle, and the number of cycles until its amplitude has halved.

  1. Find the energy fraction kept.

    $\dfrac{E_1}{E_0} = r^2 =$ e

    Energy goes as $A^2$.

  2. Find the percent lost.

    $100(1 - r^2) =$ l

    What leaks away.

  3. Find the cycles to half amplitude.

    $n = \dfrac{\ln 0.5}{\ln r} =$ n

    Solve $r^n = 0.5$.

  4. Note the period.

    $\text{nearly unchanged}$

    Light damping.

29. Guided practice

Match each departure from the simple model to its effect.

the period grows with the amplitudethe amplitude shrinks each cyclethe force stops being proportional to the stretchthe amplitude builds up, resonance
a pendulum swinging through a large angle
friction and air drag
a spring stretched past its elastic limit
pushes timed to the natural frequency

30. Practice

A pendulum starts swinging with an amplitude of $5$ cm, and air drag leaves it $0.7$ of its amplitude after each cycle. Fill in the amplitude after one cycle in cm, the amplitude after three cycles in cm, and the fraction of its energy kept each cycle.

value
amplitude after one cycle (cm)
amplitude after three cycles (cm)
fraction of energy kept each cycle

31. Practice

A damped oscillator starts with an amplitude of $12$ cm and keeps $0.95$ of its amplitude each cycle. Write its amplitude, in cm, as a function of the number of completed cycles $n$.

Answer:

32. Practice

A pendulum $3$ m long is released from $50°$. Using the correction $T \approx T_0(1 + \theta_0^2/16)$, with $\theta_0$ in radians and $g = 9.8$ m/s², what is its period, in s?

Answer: s

33. Somewhere new

The pendulum at the Smithsonian's Museum of American History has a period of $7.978$ s and, left unpushed, keeps $0.9994$ of its amplitude each cycle. What percent of its starting amplitude is left after one hour?

Answer: %

34. Lesson test

Lesson test: one question per skill, one attempt each, no hints. Your answers are checked when you submit.

35. Test question

A damped oscillator starts with an amplitude of $20$ cm and keeps $0.9$ of its amplitude each cycle. Write its amplitude, in cm, as a function of the number of completed cycles $n$.

Answer:

36. What you can do now

You can recognize the limits of the oscillation model. Explain to someone why a pendulum swung very high takes longer to swing back.

Working for the steps left to you

26. Your turn: a pendulum keeps $0.9$ of its amplitude each cycle, starting at $20$ cm. What is its amplitude after two cycles?, step 3

$A_2 = 16.2\ \text{cm}$

Shrinking.