Skip to content

Unit 7: Oscillations

Physics C Mech cheatsheet

Open ↗

Loading…


Oscillations will mostly deal with one specific type of motion: Simple Harmonic Motion (SHM).

Definition (Simple Harmonic Motion). A system is in simple harmonic motion when its acceleration is proportional to displacement and points toward equilibrium, a=−ω2xa = -\omega^2 x, equivalently d2xdt2+ω2x=0\dfrac{d^2x}{dt^2} + \omega^2x = 0.

The general solution is given as:

x(t)=Acos⁡(ωt+ϕ)x(t)=A\cos(\omega t+\phi)

or equivalently

x(t)=Asin⁡(ωt+ϕ).x(t)=A\sin(\omega t+\phi).

A rigorous proof of the formula (one that does not involve knowing the solution beforehand) requires multivariable calculus and will not be shown here. The constants AA and ϕ\phi are set by initial position and velocity. Concretely, from x(t)=Acos⁡(ωt+ϕ)x(t)=A\cos(\omega t+\phi) and v(t)=−Aωsin⁡(ωt+ϕ)v(t)=-A\omega\sin(\omega t+\phi), evaluating at t=0t=0 gives

x0=Acos⁡ϕ,v0=−Aωsin⁡ϕ.x_0=A\cos\phi,\qquad v_0=-A\omega\sin\phi.

Solving these two equations for the two unknowns yields

A=x02+(v0ω)2,tan⁡ϕ=−v0ωx0.A=\sqrt{x_0^2+\left(\frac{v_0}{\omega}\right)^2},\qquad \tan\phi=-\frac{v_0}{\omega x_0}.

There are many types of simple harmonic motion (the most common of which are the spring and the simple pendulum), and most problems will state that it involves SHM.

If

x(t)=Acos⁡(ωt+ϕ),x(t)=A\cos(\omega t+\phi),

then taking the derivative gets

v(t)=dxdt=−Aωsin⁡(ωt+ϕ),v(t)=\frac{dx}{dt}=-A\omega\sin(\omega t+\phi),

and taking it again gets

a(t)=d2xdt2=−Aω2cos⁡(ωt+ϕ)=−ω2x.a(t)=\frac{d^2x}{dt^2}=-A\omega^2\cos(\omega t+\phi)=-\omega^2x.

The maximum speed is

vmax⁡=Aω,v_{\max}=A\omega,

at a point where sin⁡x=1\sin x = 1 and the maximum acceleration magnitude is

amax⁡=Aω2.a_{\max}=A\omega^2.

Speed is greatest at equilibrium and zero at the turning points, both of which are pretty intuitive.

Example. A mass in SHM has amplitude A=0.20 mA=0.20\ \text{m}. At x=0.12 mx=0.12\ \text{m}, its speed is 0.80 m/s0.80\ \text{m/s}. Find its maximum speed, without using the period or frequency. For SHM, it may be useful to know that v2=ω2(A2−x2).v^2=\omega^2(A^2-x^2).

For SHM,

v2=ω2(A2−x2)v^2=\omega^2(A^2-x^2)

and

vmax⁡=Aω.v_{\max}=A\omega.

Eliminate ω\omega:

vvmax⁡=A2−x2A.\frac{v}{v_{\max}}=\frac{\sqrt{A^2-x^2}}{A}.

So

vmax⁡=vAA2−x2=0.800.200.202−0.122=1.0 m/s.v_{\max}=v\frac{A}{\sqrt{A^2-x^2}} =0.80\frac{0.20}{\sqrt{0.20^2-0.12^2}}=1.0\ \text{m/s}.

The angular frequency ω\omega is related to frequency and period by

ω=2πf=2πT.\omega = 2\pi f = \frac{2\pi}{T}.

Thus

T=2πω.T = \frac{2\pi}{\omega}.

Period is the time for one full cycle. Frequency is cycles per second, measured in hertz.

Angular frequency is not “cycles per second”; it is radians of phase per second. One cycle corresponds to 2π2\pi radians of phase, which is why ω=2πf\omega=2\pi f. In SHM, ω\omega is usually the most natural quantity because it appears directly in a=−ω2xa=-\omega^2x. Large ω\omega means a strong restoring acceleration for a given displacement, so the oscillator turns around quickly and has a short period.

Example. An oscillator has position x(t)=0.12cos⁡(8.0t+π/3)x(t)=0.12\cos(8.0t+\pi/3) in meters. Find its amplitude, angular frequency, period, and frequency.

Compare the equation to x(t)=Acos⁡(ωt+ϕ)x(t)=A\cos(\omega t+\phi):

A=0.12 m,ω=8.0 rad/s.A=0.12\ \text{m},\qquad \omega=8.0\ \text{rad/s}.

The period is

T=2πω=2π8.0=0.785 s.T=\frac{2\pi}{\omega} =\frac{2\pi}{8.0} =0.785\ \text{s}.

The frequency is

f=1T=ω2π=8.02π=1.27 Hz.f=\frac{1}{T} =\frac{\omega}{2\pi} =\frac{8.0}{2\pi} =1.27\ \text{Hz}.

The phase π/3\pi/3 shifts where the oscillator starts, but it does not change amplitude, period, or frequency.


For a mass on an ideal, massless spring, the force follows Hooke’s Law:

F=−kx.F=-kx.

The angular frequency is defined as ω=km\omega=\sqrt{\frac{k}{m}} and the period is T=2πω=2πmk.T=\frac{2\pi}{\omega}=2\pi\sqrt{\frac{m}{k}}. As a reminder, kk is the spring constant, and has units of Nm\frac{N}{m}. A larger spring constant means the spring is more stiff and thus takes more force to push or pull on it.

Proof (Mass-Spring Period). Hooke’s law gives

F=ma=−kx.F=ma=-kx.

Using the differential form of Newton’s second law gives

md2xdt2=−kx.m\frac{d^2x}{dt^2}=-kx.

Rearrange:

d2xdt2+kmx=0.\frac{d^2x}{dt^2}+\frac{k}{m}x=0.

Compare this to the SHM form (since all springs undergo SHM)

d2xdt2+ω2x=0.\frac{d^2x}{dt^2}+\omega^2x=0.

So

ω2=km,\omega^2=\frac{k}{m},

meaning

ω=km.\omega=\sqrt{\frac{k}{m}}.

Since T=2π/ωT=2\pi/\omega,

T=2πmk.T=2\pi\sqrt{\frac{m}{k}}.

Many AP setups attach a mass to more than one spring. Each combination behaves like a single ideal spring with an effective spring constant keffk_{\text{eff}} that models the whole system, and the period is then T=2πm/keffT=2\pi\sqrt{m/k_{\text{eff}}}. Similarly, U=12keffx2U=\frac{1}{2}k_{\text{eff}}x^2.

Springs are in parallel when they attach to the same moving object and stretch or compress by the same amount. If the mass moves by xx, each parallel spring changes length by xx, so their restoring forces add.

Springs are in series when they are connected end-to-end and the same force passes through each spring. The total stretch is split between them: one spring may stretch more than the other, but both carry the same tension. A good test is: same displacement means parallel; same force through each spring means series.

seriesparallel

Proof (series and parallel spring constants). Lets say you have two ideal springs with spring constant k1k_1 and k2k_2 supporting one mass.

Parallel. Each pulls back, and the forces add:

F=−k1x−k2x=−(k1+k2)x.F=-k_1 x-k_2 x=-(k_1+k_2)x.

Comparing to F=−keffxF=-k_{\text{eff}}x,

keff=k1+k2.k_{\text{eff}}=k_1+k_2.

Parallel springs are stiffer than either alone. If you have more than two springs in parallel, you simply add all of the spring constants together to get the effective spring constant.

Series. Each spring stretches by xi=F/kix_i=F/k_i in magnitude, and the total stretch is

x=x1+x2=Fk1+Fk2=F(1k1+1k2).x=x_1+x_2=\frac{F}{k_1}+\frac{F}{k_2}=F\left(\frac{1}{k_1}+\frac{1}{k_2}\right).

Since keff=F/xk_{\text{eff}}=F/x,

1keff=1k1+1k2.\frac{1}{k_{\text{eff}}}=\frac{1}{k_1}+\frac{1}{k_2}.

Alternatively,

keff=k1k2k1+k2k_{\text{eff}}=\frac{k_1 k_2}{k_1 + k_2}

Series springs are softer than either alone. If you have more than two springs in parallel, you simply add all of the reciprocals of the spring constants together and take the reciprocal of the sum to get the effective spring constant.

Example. A 2.0 kg2.0\ \text{kg} block is connected to two springs with k1=300 N/mk_1=300\ \text{N/m} and k2=600 N/mk_2=600\ \text{N/m}. Find the period (a) if the springs act in parallel and (b) if they act in series.

(a) Parallel:

keff=k1+k2=300+600=900 N/m,k_{\text{eff}}=k_1+k_2=300+600=900\ \text{N/m}, T=2πmkeff=2π2.0900≈0.30 s.T=2\pi\sqrt{\frac{m}{k_{\text{eff}}}}=2\pi\sqrt{\frac{2.0}{900}}\approx 0.30\ \text{s}.

(b) Series:

1keff=1300+1600=2+1600=1200⟹keff=200 N/m,\frac{1}{k_{\text{eff}}}=\frac{1}{300}+\frac{1}{600}=\frac{2+1}{600}=\frac{1}{200}\quad\Longrightarrow\quad k_{\text{eff}}=200\ \text{N/m}, T=2π2.0200≈0.63 s.T=2\pi\sqrt{\frac{2.0}{200}}\approx 0.63\ \text{s}.

The series arrangement is softer, so it oscillates more slowly. A quick sanity check: 200 N/m200\ \text{N/m} is smaller than either individual spring, and 900 N/m900\ \text{N/m} is larger than either — exactly as the proof predicts.

Not every spring configuration is purely series or parallel. A general method is to displace the mass by a small coordinate xx (which simulates SHM), write the total spring potential energy, and match it to

U=12keffx2.U=\frac{1}{2}k_{\text{eff}}x^2.

Example. A block is attached symmetrically to two identical springs of constant kk. Each spring makes angle θ\theta with the horizontal at equilibrium. If the block is displaced a small distance xx horizontally, find the effective spring constant for horizontal oscillations.

mxkk

For a small horizontal displacement, each spring’s length changes by the component of the block’s displacement along that spring:

Δℓ=xcos⁡θ.\Delta \ell=x\cos\theta.

Each spring stores energy 12k(Δℓ)2\tfrac12k(\Delta\ell)^2, so the total spring energy is

U=2(12k(xcos⁡θ)2)=kx2cos⁡2θ.U=2\left(\frac12k(x\cos\theta)^2\right)=kx^2\cos^2\theta.

Match this to 12keffx2\tfrac12k_{\text{eff}}x^2:

12keffx2=kx2cos⁡2θ.\frac12k_{\text{eff}}x^2=kx^2\cos^2\theta.

Therefore

keff=2kcos⁡2θ.k_{\text{eff}}=2k\cos^2\theta.

This configuration is not simply series or parallel; the geometry determines how much each spring actually stretches.

A simple pendulum is defined as a point mass (or equivalent, explained in the next section) hanging on a massless rod that swings at small angles.

Definition (Simple Pendulum). For a simple pendulum of length LL and small angular displacement θ\theta,

ω=gL,\omega=\sqrt{\frac{g}{L}},

and

T=2πLg.T=2\pi\sqrt{\frac{L}{g}}.

This formula assumes the small-angle approximation sin⁡θ≈θ\sin\theta\approx\theta when θ\theta is measured in radians. Like the ideal mass-spring period, the simple-pendulum period is independent of mass and approximately independent of amplitude.

Proof (Simple Pendulum Period). Gravity creates a restoring torque about the pivot:

τ=−mgLsin⁡θ.\tau=-mgL\sin\theta.

For small angles, sin⁡θ≈θ\sin\theta\approx\theta (using a Taylor series expansion), so

τ≈−mgLθ.\tau\approx-mgL\theta.

Using I=mL2I=mL^2 and ∑τ=I d2θ/dt2\sum\tau=I\,d^2\theta/dt^2,

mL2d2θdt2=−mgLθ.mL^2\frac{d^2\theta}{dt^2}=-mgL\theta.

Rearrange:

d2θdt2+gLθ=0.\frac{d^2\theta}{dt^2}+\frac{g}{L}\theta=0.

Compare with d2θ/dt2+ω2θ=0d^2\theta/dt^2+\omega^2\theta=0:

ω=gL,\omega=\sqrt{\frac{g}{L}},

so

T=2πLg.T=2\pi\sqrt{\frac{L}{g}}.

Note that this only works for small angles of θ\theta! Without the small angle approximation, the solution has no closed form and requires elliptical integrals to solve!

Example. Suppose a pendulum has a length of 11 meter and a period of 22 seconds. Find gg in this scenario (do NOT assume that g=9.8g=9.8!).

Use

T=2πLg.T=2\pi\sqrt{\frac{L}{g}}.

Solve for gg:

T2π=Lg,\frac{T}{2\pi}=\sqrt{\frac{L}{g}}, T24π2=Lg,\frac{T^2}{4\pi^2}=\frac{L}{g}, g=4π2LT2.g=\frac{4\pi^2L}{T^2}.

With L=1 mL=1\ \text{m} and T=2 sT=2\ \text{s},

g=4π2(1)22=π2 m/s2.g=\frac{4\pi^2(1)}{2^2}=\pi^2\ \text{m/s}^2.

Approximately,

g≈9.87 m/s2.g\approx9.87\ \text{m/s}^2.

You may find it curious that the solution (π2\pi^2) is very close to gg, the gravitational acceleration on Earth! This is no coincidence. In the late 18th century, scientists defined the meter as the length of a “seconds pendulum,” where the time it takes for it to go from end to end is one second (and thus the period is two seconds).

However, as time went on, scientists changed the definition of the meter since gravity was different depending on where you stand from, resulting in a geodetic definition: one ten-millionth of the distance from the equator to the North Pole along the meridian passing through Paris. The old definition ended up being around 1% off of the current accepted value for gg!

While simple pendulums are very easy to solve, not all pendulums act like point masses. A rigid body swinging about a pivot is known as a physical pendulum and is subject to rotation about the pivot point. If the center of mass is distance dd from the pivot,

ω=mgdI,\omega=\sqrt{\frac{mgd}{I}},

and

T=2πImgd.T=2\pi\sqrt{\frac{I}{mgd}}.

Proof (Physical Pendulum Period). Gravity acts at the center of mass, distance dd from the pivot. The torque is

τ=−mgdsin⁡θ.\tau=-mgd\sin\theta.

For small angles, sin⁡θ≈θ\sin\theta\approx\theta, so

τ≈−mgdθ.\tau\approx-mgd\theta.

Using ∑τ=I d2θ/dt2\sum\tau=I\,d^2\theta/dt^2 about the pivot,

Id2θdt2=−mgdθ.I\frac{d^2\theta}{dt^2}=-mgd\theta.

Rearrange:

d2θdt2+mgdIθ=0.\frac{d^2\theta}{dt^2}+\frac{mgd}{I}\theta=0.

Thus

ω=mgdI,\omega=\sqrt{\frac{mgd}{I}},

and

T=2πImgd.T=2\pi\sqrt{\frac{I}{mgd}}.

Here II is the rotational inertia about the pivot, not about the center of mass; use the parallel-axis theorem when needed. The simple pendulum is the special case where all mass sits at distance LL from the pivot (for all purposes, this is a point mass), giving I=mL2I=mL^2 and d=Ld=L, which reduces TT back to 2πL/g2\pi\sqrt{L/g}.

Example. A uniform rod of mass MM and length L=1.2 mL=1.2\ \text{m} is pivoted at one end and swings as a physical pendulum through small angles. Find its period. Use g=9.8 m/s2g=9.8\ \text{m/s}^2.

The rotational inertia of a uniform rod about one end is

I=13ML2.I=\frac{1}{3}ML^2.

The center of mass is at the rod’s midpoint, so d=L/2d=L/2. Substitute into the physical-pendulum period:

T=2πIMgd=2π13ML2Mg (L/2)=2π2L3g.T=2\pi\sqrt{\frac{I}{Mgd}}=2\pi\sqrt{\frac{\tfrac{1}{3}ML^2}{Mg\,(L/2)}}=2\pi\sqrt{\frac{2L}{3g}}.

The mass MM cancels, as it must for a gravity-driven pendulum. Plugging in numbers:

T=2π2(1.2)3(9.8)=2π2.429.4=2π0.0816≈1.8 s.T=2\pi\sqrt{\frac{2(1.2)}{3(9.8)}}=2\pi\sqrt{\frac{2.4}{29.4}}=2\pi\sqrt{0.0816}\approx 1.8\ \text{s}.

It is instructive to compare this to a simple pendulum of the same length L=1.2 mL=1.2\ \text{m}, which would have T=2πL/g=2π1.2/9.8≈2.2 sT=2\pi\sqrt{L/g}=2\pi\sqrt{1.2/9.8}\approx2.2\ \text{s}. The rod swings faster because its mass is distributed closer to the pivot than a point mass at the far end — equivalently, the rod behaves like a simple pendulum of effective length 23L=0.80 m\tfrac{2}{3}L=0.80\ \text{m}.

Not every oscillator is a spring or a pendulum, but the same methods still apply.

Example. A solid cylinder of cross-sectional area AA, height hh, and density ρobj\rho_{\text{obj}} floats upright in a liquid of density ρliq\rho_{\text{liq}}. It is pushed down slightly and released. Show the motion is SHM and find ω\omega.

These formulas may be helpful:

FB=ρfluidgVsub,m=ρobjectV.F_B=\rho_{\text{fluid}}gV_{\text{sub}}, \qquad m=\rho_{\text{object}}V.

The first is Archimedes’ principle: buoyant force equals the weight of displaced fluid. The second is the density relation, mass equals density times volume.

At equilibrium the object floats with some submerged depth d0d_0, where buoyancy balances weight. Now push it down an extra distance yy. The submerged volume increases by A yA\,y, so the buoyant force grows by

ΔFB=ρliq g (A y),\Delta F_B=\rho_{\text{liq}}\,g\,(A\,y),

directed upward — opposite the displacement. Since the weight is unchanged, the net restoring force is

F=−ρliq g A y.F=-\rho_{\text{liq}}\,g\,A\,y.

This is Hooke’s law with effective constant keff=ρliqgAk_{\text{eff}}=\rho_{\text{liq}}gA. The object’s mass is m=ρobjAhm=\rho_{\text{obj}}Ah, so

ω=keffm=ρliqgAρobjAh=ρliq gρobj h.\omega=\sqrt{\frac{k_{\text{eff}}}{m}}=\sqrt{\frac{\rho_{\text{liq}}gA}{\rho_{\text{obj}}Ah}}=\sqrt{\frac{\rho_{\text{liq}}\,g}{\rho_{\text{obj}}\,h}}.

The area AA cancels, and the frequency depends only on the density ratio, gg, and the height — a clean result that comes entirely from the linear restoring force.

Example. A U-shaped tube of uniform cross-sectional area AA contains a liquid of density ρ\rho, with total liquid column length LL. The liquid is disturbed so one side rises by yy while the other falls by yy. Find the period of the resulting oscillation.

For a uniform tube, the liquid volume is cross-sectional area times length, V=ALV=AL (since the object is a cylinder). Weight is then the gravitational force on that mass.

When the left surface drops by yy and the right rises by yy, the height difference between the two columns is 2y2y. That excess column of height 2y2y and cross-section AA has weight

ΔW=ρg (A⋅2y),\Delta W=\rho g\,(A\cdot 2y),

and this unbalanced weight is the restoring force on the whole liquid:

F=−2ρgA y.F=-2\rho g A\,y.

The total moving mass is the entire liquid column, m=ρALm=\rho A L. Newton’s second law gives

ρALd2ydt2=−2ρgA y⟹d2ydt2+2gL y=0.\rho A L\frac{d^2y}{dt^2}=-2\rho g A\,y \quad\Longrightarrow\quad \frac{d^2y}{dt^2}+\frac{2g}{L}\,y=0.

So

ω=2gL,T=2πL2g.\omega=\sqrt{\frac{2g}{L}},\qquad T=2\pi\sqrt{\frac{L}{2g}}.

The density and area both cancel: only the column length matters. As with the floating object, the entire problem reduced to writing the restoring force as a constant times the displacement.


For a mass-spring oscillator,

E=K+U=12mv2+12kx2.E = K+U = \frac{1}{2}mv^2+\frac{1}{2}kx^2.

At maximum displacement, v=0v=0 and x=±Ax=\pm A, so

E=12kA2.E=\frac{1}{2}kA^2.

At equilibrium, x=0x=0 and speed is maximum:

E=12mvmax⁡2.E=\frac{1}{2}mv_{\max}^2.

Energy continuously transfers between kinetic energy and spring potential energy while total mechanical energy remains constant if the spring system is ideal. These equations lead to the following two equations:

12mv2+12kx2=12kA2⟹v(x)=±ωA2−x2.\frac{1}{2}mv^2+\frac{1}{2}kx^2=\frac{1}{2}kA^2 \quad\Longrightarrow\quad v(x)=\pm\omega\sqrt{A^2-x^2}.

This recovers vmax⁡=Aωv_{\max}=A\omega at x=0x=0 and v=0v=0 at x=±Ax=\pm A without ever solving the differential equation.

Proof (SHM from energy conservation). For an ideal mass-spring system the total mechanical energy is constant:

E=12mv2+12kx2=const.E=\frac{1}{2}mv^2+\frac{1}{2}kx^2=\text{const.}

Differentiate both sides with respect to time. Using v=dx/dtv=dx/dt and dv/dt=d2x/dt2dv/dt=d^2x/dt^2,

dEdt=mvdvdt+kxdxdt=mvd2xdt2+kxv=0.\frac{dE}{dt}=mv\frac{dv}{dt}+kx\frac{dx}{dt}=mv\frac{d^2x}{dt^2}+kxv=0.

Factor out vv:

v(md2xdt2+kx)=0.v\left(m\frac{d^2x}{dt^2}+kx\right)=0.

Since vv is not zero throughout the motion (it is instantaneously zero only at the turning points), the bracket must vanish:

md2xdt2+kx=0,m\frac{d^2x}{dt^2}+kx=0,

which is exactly the SHM equation with ω=k/m\omega=\sqrt{k/m}. This equation can apply to any type of SHM, not just mass-spring oscillators.

Example. A block oscillates on a spring with amplitude A=0.12 mA=0.12\ \text{m}. Find the displacement where the kinetic energy is three times the potential energy, and find the fraction of the maximum speed at that point.

Total energy is E=12kA2E=\tfrac{1}{2}kA^2. When K=3UK=3U,

E=K+U=4U.E=K+U=4U.

Thus U=E/4U=E/4:

12kx2=14(12kA2).\frac12kx^2=\frac14\left(\frac12kA^2\right).

Cancel 12k\tfrac12k:

x2=A24⟹x=±A2=±0.060 m.x^2=\frac{A^2}{4}\quad\Longrightarrow\quad x=\pm\frac{A}{2}=\pm0.060\ \text{m}.

Since K=3E/4K=3E/4 and Kmax⁡=E=12mvmax⁡2K_{\max}=E=\tfrac12mv_{\max}^2,

12mv2=34(12mvmax⁡2).\frac12mv^2=\frac34\left(\frac12mv_{\max}^2\right).

Therefore

v=32vmax⁡≈0.866vmax⁡.v=\frac{\sqrt3}{2}v_{\max}\approx0.866v_{\max}.

Small Oscillations About Any Potential Minimum

Section titled “Small Oscillations About Any Potential Minimum”

The mass-spring system is just the simplest member of a much larger family. Recall that a one-dimensional conservative force is Fx=−dU/dxF_x=-dU/dx and that a stable equilibrium sits at a local minimum of U(x)U(x), where U′(x0)=0U'(x_0)=0 and U′′(x0)>0U''(x_0)>0. The key result is that any such minimum looks like a spring for small displacements, with an effective spring constant equal to the curvature of UU (think of a ball rolling back and forth in a divot):

keff=U′′(x0),ω=U′′(x0)m.k_{\text{eff}}=U''(x_0),\qquad \omega=\sqrt{\frac{U''(x_0)}{m}}.

Proof (small oscillations are SHM). Let a particle of mass mm move in a potential U(x)U(x) with a stable equilibrium at x0x_0, so U′(x0)=0U'(x_0)=0 and U′′(x0)>0U''(x_0)>0. Expand UU in a Taylor series (refer to Calc BC Unit 10 if you need more information about this) about x0x_0, writing the small displacement s=x−x0s=x-x_0:

U(x)=U(x0)+U′(x0) s+12U′′(x0) s2+⋯U(x)=U(x_0)+U'(x_0)\,s+\frac{1}{2}U''(x_0)\,s^2+\cdots

The constant U(x0)U(x_0) does not affect forces, and U′(x0)=0U'(x_0)=0 by the equilibrium condition. For small ss we keep only the quadratic term, since all the other terms are essentially 00:

U(x)≈U(x0)+12U′′(x0) s2.U(x)\approx U(x_0)+\frac{1}{2}U''(x_0)\,s^2.

The force is then

F=−dUdx=−dUds=−U′′(x0) s.F=-\frac{dU}{dx}=-\frac{dU}{ds}=-U''(x_0)\,s.

This is exactly Hooke’s law with keff=U′′(x0)k_{\text{eff}}=U''(x_0). Newton’s second law gives

md2sdt2=−U′′(x0) s⟹ω=U′′(x0)m.m\frac{d^2s}{dt^2}=-U''(x_0)\,s \quad\Longrightarrow\quad \omega=\sqrt{\frac{U''(x_0)}{m}}.

So near the bottom of any smooth potential well, the motion is simple harmonic — the parabola 12keffs2\tfrac12 k_{\text{eff}}s^2 is the local approximation to every potential minimum. Another way to think of it is

ω=restoring effectinertia.\omega=\sqrt{\frac{\text{restoring effect}}{\text{inertia}}}.

Here the restoring effect is the curvature U′′(x0)U''(x_0): a sharper well pushes back harder for the same displacement. The inertia is the mass mm: a heavier object responds more sluggishly to the same restoring force.

Example. A particle of mass m=0.20 kgm=0.20\ \text{kg} moves in the potential

U(x)=αx4−βx2,α=2.0 J/m4,β=4.0 J/m2.U(x)=\alpha x^4-\beta x^2,\qquad \alpha=2.0\ \text{J/m}^4,\quad \beta=4.0\ \text{J/m}^2.

Find the angular frequency of small oscillations about the positive-xx equilibrium.

First locate the equilibrium from U′(x)=0U'(x)=0:

U′(x)=4αx3−2βx=2x(2αx2−β)=0.U'(x)=4\alpha x^3-2\beta x=2x(2\alpha x^2-\beta)=0.

The nonzero solutions are x0=±β/2α=±4.0/4.0=±1.0 mx_0=\pm\sqrt{\beta/2\alpha}=\pm\sqrt{4.0/4.0}=\pm1.0\ \text{m}. Take x0=+1.0 mx_0=+1.0\ \text{m}. Now the curvature:

U′′(x)=12αx2−2β,U''(x)=12\alpha x^2-2\beta, U′′(x0)=12(2.0)(1.0)2−2(4.0)=24−8=16 N/m.U''(x_0)=12(2.0)(1.0)^2-2(4.0)=24-8=16\ \text{N/m}.

Since U′′(x0)>0U''(x_0)>0 this is indeed a stable minimum (the point x=0x=0, where U′′(0)=−8<0U''(0)=-8<0, is an unstable maximum). The angular frequency is

ω=U′′(x0)m=160.20=80≈8.9 rad/s.\omega=\sqrt{\frac{U''(x_0)}{m}}=\sqrt{\frac{16}{0.20}}=\sqrt{80}\approx 8.9\ \text{rad/s}.

Notice we never needed the full shape of UU — only its second derivative at the equilibrium.


  1. A mass mm hangs from a spring of constant kk in a uniform gravitational field. It oscillates vertically about equilibrium with amplitude AA. Which quantity depends on gg?

(A) The angular frequency

(B) The period

(C) The equilibrium extension

(D) The speed at the equilibrium point measured relative to the oscillation amplitude

  1. A pendulum clock is taken to a planet where the gravitational field strength is g/4g/4. To keep the same small-angle period, the pendulum length should be changed from LL to

(A) 4L4L

(B) 2L2L

(C) L/2L/2

(D) L/4L/4

  1. A mass on a spring is released from rest at x=Ax=A. When it first reaches x=A/3x=A/3, what fraction of the total mechanical energy is kinetic?

(A) 1/91/9

(B) 2/32/3

(C) 8/98/9

(D) 8/3\sqrt{8}/3

  1. A block of mass mm is attached to a spring of constant kk on a frictionless horizontal surface. A small constant horizontal force F0F_0 is then applied and left on. Compared with the original oscillator, the new motion has

(A) the same angular frequency and an equilibrium shifted by F0/kF_0/k

(B) angular frequency (k+F0)/m\sqrt{(k+F_0)/m} and the same equilibrium

(C) angular frequency k/(m+F0/g)\sqrt{k/(m+F_0/g)} and an equilibrium shifted by F0/kF_0/k

(D) no simple harmonic motion because the net force is not proportional to xx

  1. A bead slides without friction on a circular hoop of radius RR in a vertical plane. Near the bottom of the hoop, the coordinate along the arc is s=Rθs=R\theta. The bead’s small-oscillation angular frequency is

(A) g/R\sqrt{g/R}

(B) R/g\sqrt{R/g}

(C) g/Rg/R

(D) 2g/R\sqrt{2g/R}

  1. A mass mm is attached between two horizontal springs with constants k1k_1 and k2k_2 on a frictionless track, one spring on each side. Both springs are relaxed when the mass is at x=0x=0. If the mass is displaced slightly, its angular frequency is

(A) k1+k2m\sqrt{\dfrac{k_1+k_2}{m}}

(B) k1k2m(k1+k2)\sqrt{\dfrac{k_1k_2}{m(k_1+k_2)}}

(C) k1−k2m\sqrt{\dfrac{k_1-k_2}{m}}

(D) k1m+k2m\sqrt{\dfrac{k_1}{m}}+\sqrt{\dfrac{k_2}{m}}

  1. A block attached to a spring oscillates on a frictionless table. The block is replaced by two identical blocks glued together, and the amplitude is doubled. The maximum acceleration changes by a factor of

(A) 1/21/\sqrt{2}

(B) 1/21/2

(C) 2\sqrt{2}

(D) 11

  1. A mass mm on a vertical spring oscillates about its equilibrium position with period TT. At the instant the mass passes through equilibrium moving downward, a second identical mass is gently attached. Immediately after attachment, the new equilibrium position is

(A) unchanged

(B) lower by mg/kmg/k

(C) lower by 2mg/k2mg/k

(D) higher by mg/kmg/k

  1. A pendulum bob of mass mm and length LL is also attached to a horizontal spring of constant kk that is relaxed when the bob hangs vertically. For small angles, compared with the same pendulum without the spring, the period is

(A) larger

(B) smaller

(C) unchanged

(D) zero because the forces cancel

  1. A particle moves in the potential U(x)=12kx2+ϵx4U(x)=\dfrac{1}{2}kx^2+\epsilon x^4, where k,ϵ>0k,\epsilon>0. For sufficiently small oscillations about x=0x=0, the angular frequency is

(A) k/m\sqrt{k/m}

(B) (k+4ϵ)/m\sqrt{(k+4\epsilon)/m}

(C) ϵ/m\sqrt{\epsilon/m}

(D) dependent on amplitude even in the small-amplitude limit

  1. A particle moves near x=0x=0 in the potential U(x)=U0+ax2+bx3+cx4U(x)=U_0+ax^2+bx^3+cx^4, where a>0a>0. For sufficiently small oscillations, the angular frequency is

(A) a/m\sqrt{a/m}

(B) 2a/m\sqrt{2a/m}

(C) 6b/m\sqrt{6b/m}

(D) 12c/m\sqrt{12c/m}

  1. A solid cylinder of mass MM and radius RR is attached at its center to a horizontal spring of constant kk and rolls without slipping. Its angular frequency is

(A) k/M\sqrt{k/M}

(B) 2k/M\sqrt{2k/M}

(C) 2k/(3M)\sqrt{2k/(3M)}

(D) 3k/(2M)\sqrt{3k/(2M)}

  1. A solid cylinder of mass MM and radius RR rests on a rough horizontal surface and rolls without slipping. A light spring of constant kk is attached to the cylinder’s center, and the other end is fixed to a wall. The cylinder is displaced a small distance AA from equilibrium and released from rest.

    (A)(A) Using energy, derive an expression for the angular frequency of the oscillation in terms of MM and kk.

    (B)(B) Determine the maximum static friction force needed during the motion.

    (C)(C) Find the minimum coefficient of static friction required for rolling without slipping for the entire motion.

    (D)(D) Suppose the cylinder is replaced by a thin hoop with the same MM and RR. Without redoing the full calculation, determine whether the period increases, decreases, or stays the same, and justify your answer.

  1. A bead of mass mm slides without friction on a rigid circular wire of radius RR fixed in a vertical plane. The bead is also attached to a light spring of constant kk and negligible relaxed length whose other end is fixed at the top of the circle. Let θ\theta be the bead’s angular displacement from the bottom of the circle.

    (A)(A) Write the bead’s gravitational potential energy and spring potential energy as functions of θ\theta, taking the bottom of the circle as zero gravitational potential.

    (B)(B) Find the condition on kk and RR for the bottom of the circle to be a stable equilibrium.

    (C)(C) For small oscillations about the bottom, derive the angular frequency in terms of mm, gg, RR, and kk.

    (D)(D) Describe qualitatively how the equilibrium position changes if the spring constant is made very large.

  1. A student studies a cart-spring oscillator on a horizontal track. The cart of mass MM has a small block of mass mm resting on top of it. The coefficient of static friction between the block and cart is μs\mu_s. The cart is pulled to amplitude AA and released from rest; the block does not slip at first.

    (A)(A) Derive the period of the combined motion while the block does not slip.

    (B)(B) Determine the maximum amplitude Amax⁡A_{\max} for which the block can remain at rest relative to the cart throughout the motion.

    (C)(C) The student measures the period for several added top-block masses mm. Describe a graph that could be used to determine the spring constant kk from the data, including what should be plotted on each axis.

    (D)(D) If the block begins to slip near the endpoints of the motion, explain whether the measured period should be expected to match the expression from part (A)(A). Your explanation should refer to the forces on the two objects, not just energy loss.