Flow Lab

Syllabus / Module 14: Applications: Physics & Engineering

Lesson 14.1

Coupled Oscillators & Normal Modes

Two and N masses on springs, normal modes as eigenvectors, beats between modes, and the continuum limit to the wave equation.

Written byClaude Sonnet

🌊
Why this matters for simulation: Mass-spring systems are also how cloth and soft bodies are simulated in games.

By the end of this lesson you can…

  • Write $M\ddot{\mathbf x} = -K\mathbf x$ for masses on springs and read the mass and stiffness matrices off the picture.
  • Find normal modes as solutions of $K\mathbf u = \omega^2 M\mathbf u$, and expand any motion as a sum of them (modal decomposition).
  • Solve a two-mass system by hand, and explain beats and energy exchange between weakly coupled oscillators.
  • Derive the chain frequencies $\omega_j = 2\sqrt{k/m}\,\sin\frac{j\pi}{2(N+1)}$ and the sine-shaped modes, and connect them to the finite-difference Laplacian and the sine transform.
  • Take the continuum limit to the wave equation, derive the dispersion relation, and say why short waves are slower than $c$.
  • Use energy conservation per mode, uniform damping, and the highest frequency to bound a time step.
On this page
·
  1. Masses, springs and two matrices
  2. Normal modes
  3. Worked example: two masses by hand
  4. Beats and energy exchange
  5. A chain of N masses
  6. From chain to string: the wave equation
  7. Energy, damping and the highest frequency
  8. The lab
  9. The exercise: normal modes from KKK and MMM
  10. Concept checks
  11. Problems
  12. What comes next
  13. Key Takeaways

The last lesson ended with a matrix pair, a stiffness matrix KK and a mass matrix MM, and the eigenproblem Ku=λMuK\mathbf u = \lambda M\mathbf u. That was a method for approximating a PDE. Here the same objects appear without any approximation: a handful of real masses on real springs. 5.3 Mechanical Vibrations: Spring–Mass–Damper solved one mass; this lesson solves many, and the answer turns out to be one mass many times over, in the right coordinates. Those coordinates are the eigenvectors of 1.6 Change of Basis & Diagonalization, called normal modes. Let the number of masses grow and the chain becomes a string obeying the wave equation of 9.4 The Wave Equation & Vibrating Strings, with an effect a finite-difference wave solver cannot avoid.

▶ Video Two masses and a coupling spring, the two normal modes, beats between them, a chain of masses with its sine-shaped modes, and the chain turning into a stringcaptions only
Transcript
  1. Two masses sit between two walls. Outer springs of stiffness k, and a weak coupling spring, k c, joins the masses. Each mass feels a force from each neighbor equal to the spring's stiffness times its stretch. Collect the two positions in one vector x, and the equations become M times x double dot equals minus K times x. The mass matrix M is diagonal. The stiffness matrix K has the total stiffness at each mass on its diagonal and minus the joining stiffness off it, so K is symmetric, and the walls make it positive definite. It is the same pair of matrices that finite elements gave us, now with real springs.
  2. Ask for motions where every mass oscillates at the same frequency, with a fixed shape u. Substituting gives K u equals omega squared M u, a generalized eigenproblem. For two masses, take m and k equal to 1 and k c equal to 0.05. The first shape is 1, 1: the masses move together, the coupling spring is never stretched, and the frequency is 1.0000. The second is 1, minus 1: the masses move oppositely, the coupling spring is stretched twice as much, and the frequency is higher, omega squared equals 1 plus 2 k c, which is 1.1, so omega is 1.0488. Any motion is a sum of these two modes, and each mode is a single independent oscillator.
  3. Pull mass 1 aside and let go. That start is half of each mode, so mass 1 moves as cosine of the mean frequency times a slow envelope, and mass 2 as sine of the mean frequency times the complementary envelope. These are beats. The two frequencies differ by 0.0488, so the slow envelope takes pi over delta, 64.37 time units, to hand the whole motion from mass 1 to mass 2, against a single oscillation period of 2 pi. Then it flows back. The weaker the coupling, the slower the exchange. Notice that the energy of each mode never changes. What moves is energy between the masses.
  4. Now let N equal masses sit on equal springs between fixed walls. K is k times the tridiagonal matrix minus 1, 2, minus 1: the very matrix of the finite difference second derivative. A sine is an eigenvector, because sine of i plus 1 times theta plus sine of i minus 1 times theta equals 2 cosine theta times sine of i theta, and the wall at the far end fixes theta equal to j pi over N plus 1. So the frequencies are 2 root k over m times sine of j pi over 2 times N plus 1. Mode 1 swings all masses one way, and the top mode alternates. For N equal to 5 the frequencies are 0.5176, 1.0000, 1.4142, 1.7321 and 1.9319. Splitting a motion into modes is the discrete sine transform.
  5. Let the number of masses grow, with spacing a, and the correction to the wave equation vanishes. A plane wave solves the chain exactly when omega equals 2 c over a times sine of kappa a over 2, below the straight line of the string, omega equals c kappa. For N equal to 24 the slowest mode is 0.12558 against 0.12566, a ratio of 0.9993. Mode 12 is 1.36909 against 1.5080, a ratio of 0.9079, and the top mode 1.99605 against 3.0159, a ratio of 0.6618. The harmonics go out of tune: the chain is not the string. The slope of the curve is the group speed, c cosine of kappa a over 2, which is 0.5 c at kappa a equal to 2 pi over 3. A finite-difference wave solver is exactly this chain.
  6. Add the same damper to every mass and each modal equation stays independent: q double dot plus gamma q dot plus omega squared q equals zero. Each mode is the damped oscillator we already know. For N equal to 24, gamma equal to 0.1, mode 12, the exact energy ratio at t equal to 20 is 0.131, against e to the minus 2, which is 0.135. At t equal to 40 it is 0.0187 against 0.0183. With gamma equal to 0.3, gamma over 2 is 0.15, larger than omega 1, which is 0.126, so the slowest mode no longer oscillates at all.
  7. The highest mode also sets the time step. A leapfrog step on one mode stays bounded only when h is at most 2 over omega. For the chain of 24 masses that is 2 over 1.99605, which is 1.0020, while the slowest motion has period 2 pi over 0.12558, which is 50. The ratio of fastest to slowest is 15.9, and it grows with N: add masses and the step must shrink. Next, the lesson on electrical circuits takes this equation into electronics. An RLC circuit is the same second-order equation, with inductance in the place of mass, resistance for the damper, and one over C for the spring. I expect coupled loops to give a matrix equation like this one, and so to have normal modes.

Masses, springs and two matrices

Put NN masses on a line, joined to each other and to two fixed walls by N+1N+1 springs, and let each move a small distance xix_i sideways or along the line (the equations are identical). Mass ii feels a force from each neighbor equal to the spring’s stiffness times its stretch. For two masses between two walls, with outer springs kk and a coupling spring kck_c between the masses:

mx¨1=−kx1−kc(x1−x2),mx¨2=−kx2−kc(x2−x1).m\ddot x_1 = -k x_1 - k_c(x_1 - x_2), \qquad m\ddot x_2 = -k x_2 - k_c(x_2 - x_1).

Collect the unknowns in x=(x1,x2)\mathbf x = (x_1, x_2). Then Mx¨=−KxM\ddot{\mathbf x} = -K\mathbf x with

M=(m00m),K=(k+kc−kc−kck+kc).M = \begin{pmatrix} m & 0 \\ 0 & m\end{pmatrix}, \qquad K = \begin{pmatrix} k + k_c & -k_c \\ -k_c & k + k_c\end{pmatrix}.

The pattern holds for any chain: MM is diagonal with the masses; KiiK_{ii} is the total stiffness of the springs touching mass ii; Kij=−(stiffness joining i and j)K_{ij} = -(\text{stiffness joining } i \text{ and } j). So KK is symmetric. The walls make it positive definite: xTKx\mathbf x^{\mathsf T}K\mathbf x is twice the stored spring energy. Everything below uses just these two properties. The same KK and MM are the finite element matrices of 10.13 Introduction to Finite Elements: there the springs were elements and MM came from integrating products of hat functions.

Normal modes

A single oscillator has one natural motion, cos⁡(ωt−φ)\cos(\omega t - \varphi). Ask whether the coupled system has motions where every mass oscillates at the same frequency: x(t)=ucos⁡(ωt−φ)\mathbf x(t) = \mathbf u\cos(\omega t - \varphi) with a fixed shape u\mathbf u. Substituting gives

Ku=ω2Mu.K\mathbf u = \omega^2 M\mathbf u .

This is the generalized eigenproblem, with eigenvalue λ=ω2\lambda = \omega^2. For equal masses it is the ordinary eigenproblem of M−1KM^{-1}K. In general, MM is diagonal and positive, so M1/2M^{1/2} exists, and the substitution y=M1/2x\mathbf y = M^{1/2}\mathbf x turns the problem into

y¨=−Sy,S=M−1/2KM−1/2.\ddot{\mathbf y} = -S\mathbf y, \qquad S = M^{-1/2} K M^{-1/2}.

SS is symmetric and positive definite. The spectral theorem (from memory; 1.5 Eigenvectors & Eigenvalues gives the geometry) says a real symmetric matrix has real eigenvalues λj>0\lambda_j \gt 0 and orthonormal eigenvectors vj\mathbf v_j. Positive eigenvalues mean ωj=λj\omega_j = \sqrt{\lambda_j} is real: the system oscillates and never grows. Going back, the mode shapes are uj=M−1/2vj\mathbf u_j = M^{-1/2}\mathbf v_j, and they satisfy uiTMuj=δij\mathbf u_i^{\mathsf T}M\mathbf u_j = \delta_{ij}: orthogonal in the mass-weighted sense.

Modal decomposition. Write x(t)=∑jqj(t) uj\mathbf x(t) = \sum_j q_j(t)\,\mathbf u_j. Because the uj\mathbf u_j diagonalize both matrices, the equations decouple completely:

q¨j+ωj2qj=0(j=1,…,N),\ddot q_j + \omega_j^2 q_j = 0 \qquad (j = 1, \dots, N),

NN independent one-mass oscillators. The solution is qj=Ajcos⁡ωjt+Bjsin⁡ωjtq_j = A_j\cos\omega_j t + B_j\sin\omega_j t, with Aj=ujTMx(0)A_j = \mathbf u_j^{\mathsf T}M\mathbf x(0) and Bjωj=ujTMx˙(0)B_j\omega_j = \mathbf u_j^{\mathsf T}M\dot{\mathbf x}(0). So every motion is a sum of normal modes, and the coupling has not disappeared: it has been moved into the choice of coordinates. In the first-order form z˙=Az\dot{\mathbf z} = A\mathbf z of 7.2 Linear Systems & the Eigenvalue Method, these are the eigenvalues ±iωj\pm i\omega_j on the imaginary axis of 7.4 Complex & Repeated Eigenvalues, one pair per mode.

Energy. The total energy E=12x˙TMx˙+12xTKxE = \tfrac12\dot{\mathbf x}^{\mathsf T}M\dot{\mathbf x} + \tfrac12\mathbf x^{\mathsf T}K\mathbf x becomes E=∑j12(q˙j2+ωj2qj2)E = \sum_j\tfrac12(\dot q_j^2 + \omega_j^2q_j^2) in modal coordinates. Each term is conserved separately. Energy never moves from one mode to another. It moves between masses, which is what beats are.

Worked example: two masses by hand

Take m=k=1m = k = 1 and a weak coupling spring kc=0.05k_c = 0.05. Then K−λMK - \lambda M has determinant (1+kc−λ)2−kc2=0(1 + k_c - \lambda)^2 - k_c^2 = 0, so λ=1+kc∓kc\lambda = 1 + k_c \mp k_c, that is

ω12=1,ω22=1+2kc=1.1.\omega_1^2 = 1, \qquad \omega_2^2 = 1 + 2k_c = 1.1 .

The eigenvectors: for λ=1\lambda = 1, Ku=uK\mathbf u = \mathbf u reads kc(u1−u2)=0k_c(u_1 - u_2) = 0, so u1=(1,1)\mathbf u_1 = (1, 1). For λ=1.1\lambda = 1.1 it gives u2=−u1u_2 = -u_1, so u2=(1,−1)\mathbf u_2 = (1, -1). In the first mode the masses move together and the coupling spring is never stretched, so the frequency is that of one mass on one spring. In the second they move oppositely and the coupling spring is stretched twice as much as either end, so the frequency is higher. So ω1=1\omega_1 = 1 and ω2=1.1=1.0488\omega_2 = \sqrt{1.1} = 1.0488.

Beats and energy exchange

Pull mass 1 aside by 11 and let go: x(0)=(1,0)=12(1,1)+12(1,−1)\mathbf x(0) = (1, 0) = \tfrac12(1,1) + \tfrac12(1,-1). Then

x1=12(cos⁡ω1t+cos⁡ω2t)=cos⁡ωˉt cos⁡Δt2,x2=12(cos⁡ω1t−cos⁡ω2t)=sin⁡ωˉt sin⁡Δt2,x_1 = \tfrac12(\cos\omega_1 t + \cos\omega_2 t) = \cos\bar\omega t\,\cos\tfrac{\Delta t}{2}, \qquad x_2 = \tfrac12(\cos\omega_1 t - \cos\omega_2 t) = \sin\bar\omega t\,\sin\tfrac{\Delta t}{2},

with ωˉ=(ω1+ω2)/2\bar\omega = (\omega_1 + \omega_2)/2 and Δ=ω2−ω1\Delta = \omega_2 - \omega_1. This is an oscillation at the mean frequency with a slow envelope cos⁡(Δt/2)\cos(\Delta t/2) on mass 1 and sin⁡(Δt/2)\sin(\Delta t/2) on mass 2: beats. Mass 1 comes to rest and mass 2 takes over the whole motion at t=π/Δt = \pi/\Delta; the energy then flows back. For kc=0.05k_c = 0.05, Δ=1.0488−1=0.04881\Delta = 1.0488 - 1 = 0.04881, so the transfer takes π/Δ=64.4\pi/\Delta = 64.4 time units against a single oscillation period of 2π2\pi. The weaker the coupling the slower the exchange. For small kc/kk_c/k, Δ≈ω1kc/k\Delta \approx \omega_1 k_c/k, because 1+2ε≈1+ε\sqrt{1 + 2\varepsilon} \approx 1 + \varepsilon. At kc=0.02k_c = 0.02 the exact value is π/(1.04−1)=158.6\pi/(\sqrt{1.04} - 1) = 158.6 against π/0.02=157.1\pi/0.02 = 157.1 from the approximation.

A chain of N masses

Now NN equal masses mm with equal springs kk and fixed ends. Then K=k tridiag(−1,2,−1)K = k\,\mathrm{tridiag}(-1, 2, -1) and M=mIM = mI: the same matrix as the finite difference second derivative of 10.1 Finite Differences: Discretizing Derivatives, and the same one whose eigenvectors the multigrid and spectral lessons found. Try a sine, ui=sin⁡(iθ)u_i = \sin(i\theta). The identity sin⁡((i+1)θ)+sin⁡((i−1)θ)=2cos⁡θ sin⁡(iθ)\sin((i+1)\theta) + \sin((i-1)\theta) = 2\cos\theta\,\sin(i\theta) gives

(Ku)i=k(2−2cos⁡θ)ui=4ksin⁡2θ2  ui(K\mathbf u)_i = k\bigl(2 - 2\cos\theta\bigr)u_i = 4k\sin^2\tfrac\theta2\;u_i

for every interior ii, so the sine is an eigenvector as soon as the end conditions hold. The wall at i=0i = 0 is satisfied automatically; the wall at i=N+1i = N+1 needs sin⁡((N+1)θ)=0\sin((N+1)\theta) = 0, so θj=jπ/(N+1)\theta_j = j\pi/(N+1). Therefore

uj=2m(N+1)(sin⁡ijπN+1)i=1N,ωj=2km sin⁡jπ2(N+1),j=1,…,N.\mathbf u_j = \sqrt{\tfrac{2}{m(N+1)}}\Bigl(\sin\tfrac{ij\pi}{N+1}\Bigr)_{i=1}^N, \qquad \omega_j = 2\sqrt{\tfrac km}\,\sin\frac{j\pi}{2(N+1)}, \quad j = 1, \dots, N.

Mode 1 swings all masses one way; mode NN alternates sign between neighbors. Modal decomposition of the initial displacement is a discrete sine transform: the coefficients qj(0)q_j(0) are the sine transform of x(0)\mathbf x(0), the transform the spectral lesson used to solve Poisson’s equation on a box, and the frequencies are the square roots of the finite-difference Laplacian’s eigenvalues, up to the factor k/mk/m. For N=5N = 5 and k=m=1k = m = 1, ωj=2sin⁡(jπ/12)\omega_j = 2\sin(j\pi/12):

jj12345
ωj\omega_j0.51760.51761.00001.00001.41421.41421.73211.73211.93191.9319

With the consistent mass matrix m6 tridiag(1,4,1)\tfrac m6\,\mathrm{tridiag}(1,4,1) of the finite element lesson instead, the same eigenproblem is the finite element one. Lumped beads are finite differences; smeared beads are finite elements.

From chain to string: the wave equation

Let ui(t)=u(xi,t)u_i(t) = u(x_i, t) with spacing aa. Newton’s law is mu¨i=k(ui+1−2ui+ui−1)m\ddot u_i = k(u_{i+1} - 2u_i + u_{i-1}). The Taylor series gives ui+1−2ui+ui−1=a2uxx+a412uxxxx+⋯u_{i+1} - 2u_i + u_{i-1} = a^2u_{xx} + \tfrac{a^4}{12}u_{xxxx} + \cdots, so

utt=c2(uxx+a212uxxxx+⋯ ),c=ak/m.u_{tt} = c^2\bigl(u_{xx} + \tfrac{a^2}{12}u_{xxxx} + \cdots\bigr), \qquad c = a\sqrt{k/m}.

Keep cc fixed while a→0a \to 0 (more and more lighter masses on stiffer springs: m=ρam = \rho a for line density ρ\rho, k=T/ak = T/a for tension TT, so c2=T/ρc^2 = T/\rho) and the correction vanishes: the wave equation utt=c2uxxu_{tt} = c^2u_{xx} of 9.4 The Wave Equation & Vibrating Strings. Its fixed-end modes are sin⁡(jπx/L)\sin(j\pi x/L) with ωj=jπc/L\omega_j = j\pi c/L, equally spaced, which is why a string has harmonics. See how the chain differs: a plane wave ei(κx−ωt)e^{i(\kappa x - \omega t)} solves the discrete equation exactly when

ω=2casin⁡κa2.\omega = \frac{2c}{a}\sin\frac{\kappa a}{2}.

This is the dispersion relation of the chain. For long waves (κa≪1\kappa a \ll 1), ω≈cκ(1−κ2a2/24)\omega \approx c\kappa(1 - \kappa^2a^2/24), the wave equation again. For short waves it bends below the line ω=cκ\omega = c\kappa. The phase speed ω/κ=c sin⁡(κa/2)/(κa/2)\omega/\kappa = c\,\sin(\kappa a/2)/(\kappa a/2) and the group speed dω/dκ=ccos⁡(κa/2)d\omega/d\kappa = c\cos(\kappa a/2) are both below cc, the group speed by more. At the shortest wave the chain can carry, κa=π\kappa a = \pi (neighbors alternating), ω\omega reaches its maximum 2k/m2\sqrt{k/m} and the group speed is zero. There is no wave above that frequency at all.

The fixed chain’s modes sit on this curve at κja=jπ/(N+1)\kappa_ja = j\pi/(N+1). For N=24N = 24 (sim values with k=m=a=1k = m = a = 1), ω1=0.12558\omega_1 = 0.12558 against the continuum value π/25=0.12566\pi/25 = 0.12566, a ratio of 0.99930.9993; ω12=1.36909\omega_{12} = 1.36909 against 1.50801.5080, a ratio of 0.90790.9079; and the top mode ω24=1.99605\omega_{24} = 1.99605 against 3.01593.0159, a ratio of 0.66180.6618. The nine percent loss at mode 12 is the harmonic series going out of tune: the chain is not the string, and it is a fact about beads, not an error.

Phonons glimpse. A crystal is a chain of atoms, with the same maximum frequency and zero group speed at the zone edge κ=π/a\kappa = \pi/a. Quantum mechanics (from memory, not derived here) makes each mode a harmonic oscillator with energies (n+12)ℏωj(n + \tfrac12)\hbar\omega_j, and one step n→n+1n \to n+1 is a phonon, a quantum of sound. Sound is the long-wavelength part, where the dispersion curve is straight.

Energy, damping and the highest frequency

Each mode keeps its own energy. For N=5N = 5, mode 11 started at peak amplitude 11 has E=12ω12⋅N+12=0.4019E = \tfrac12\omega_1^2\cdot\tfrac{N+1}{2} = 0.4019 at every instant; the sim shows the ratio to the starting value as 1.00001.0000 throughout. Add a damper of strength γ\gamma on each mass, x¨+γx˙+Kx=0\ddot{\mathbf x} + \gamma\dot{\mathbf x} + K\mathbf x = 0 with unit masses. In modal coordinates q¨j+γq˙j+ωj2qj=0\ddot q_j + \gamma\dot q_j + \omega_j^2q_j = 0, still decoupled. Each mode is the damped oscillator of 5.3 Mechanical Vibrations: Spring–Mass–Damper with ζj=γ/(2ωj)\zeta_j = \gamma/(2\omega_j): underdamped when ωj>γ/2\omega_j \gt \gamma/2, so energy falls as e−γte^{-\gamma t} on average, and overdamped for slow modes with ωj<γ/2\omega_j \lt \gamma/2. For N=24N = 24, γ=0.1\gamma = 0.1, mode 1212 (ω12=1.369\omega_{12} = 1.369), the exact energy ratio at t=20t = 20 is 0.1310.131 against e−2=0.135e^{-2} = 0.135, and at t=40t = 40 it is 0.01870.0187 against 0.01830.0183. With γ=0.3\gamma = 0.3, γ/2=0.15\gamma/2 = 0.15 exceeds ω1=0.126\omega_1 = 0.126, so mode 11 no longer oscillates. A general damper couples the modes again; only C=αM+βKC = \alpha M + \beta K (Rayleigh damping, from memory) keeps them separate.

Driving the chain at frequency Ω\Omega forces each qjq_j, and resonance appears whenever Ω\Omega meets some ωj\omega_j (5.6 Forced Oscillations & Resonance): NN masses, NN resonances.

The lab

◆ Interactive Pick a mode, mix two modes for beats, weaken the middle spring, drag the masses, add damping

Loading simulation…

The top strip animates the masses. The panels show mode jj‘s shape, the frequencies ωj\omega_j (dots, against the dashed continuum line) with each mode’s energy underneath, and two masses against time. The frequencies come from a numerical eigenvalue solver (Jacobi iteration), and for uniform springs agree with 2sin⁡(jπ/2(N+1))2\sin(j\pi/2(N+1)) to 4.9×10−154.9\times10^{-15} at N=24N = 24. The motion is the exact modal sum.

  1. With N=5N = 5 the frequencies are 0.5176,1.0000,1.4142,1.7321,1.93190.5176, 1.0000, 1.4142, 1.7321, 1.9319. Step through the modes.
  2. Set N=2N = 2, the middle spring to 0.050.05, and start from “modes jj and j+1j+1”. The readout gives ω=1.0000\omega = 1.0000 and 1.04881.0488 and a transfer time of 64.3764.37: the worked example above.
  3. Set N=24N = 24, uniform: ω1=0.1256\omega_1 = 0.1256, ω12=1.3691\omega_{12} = 1.3691, ω24=1.9961\omega_{24} = 1.9961, ratio fastest to slowest 15.915.9, step limit 1.00201.0020.
  4. Keep N=24N = 24 and set the middle spring to 0.050.05 (a nearly cut chain). Mode 11 stays at 0.125580.12558, since it never stretches the middle spring; mode 22 drops from 0.25070.2507 to 0.16970.1697. This near-degenerate pair has a beat transfer time of π/0.04416=71.2\pi/0.04416 = 71.2.
  5. Drag one mass and release: the energy bars show the displacement spread over many modes, each staying constant with no damping.

The exercise: normal modes from KK and MM

The playground below shows the beat envelope of the worked example. The exercise asks for the general routine: from masses and springs to frequencies and mode shapes.

🐍 PythonBeats from two modes
⌘/Ctrl + Enter
🧪 Coding exerciseExercise: normal modes of a spring-mass chain
⌘/Ctrl + Enter

Concept checks

✅ Concept check

A fixed-end chain has $N = 7$ equal masses. In units of $\sqrt{k/m}$, what is the frequency of mode $4$?

✅ Concept check

Two identical oscillators are joined by a weak spring and mass 1 is pulled aside and released. Why does mass 2 slowly take over the motion?

✅ Concept checkselect all that apply

Select all statements that are true for the undamped chain.

✅ Concept check

For the chain, the group speed is $c\\cos(\\kappa a/2)$. At wavenumber $\\kappa a = 2\\pi/3$, what is the group speed as a fraction of $c$?

Problems

✏️ Pen & paperUnequal masses●○○

Mass 22 and mass 11 hang in a row from a wall: wall, spring 11, mass 22, spring 11, mass 11, with the right end free. Find the frequencies and mode shapes, and check that the shapes are orthogonal with respect to MM.

Show worked solution

M=diag(2,1)M = \mathrm{diag}(2, 1) and K=(2−1−11)K = \begin{pmatrix} 2 & -1 \\ -1 & 1\end{pmatrix}. Then det⁡(K−λM)=(2−2λ)(1−λ)−1=2(1−λ)2−1=0\det(K - \lambda M) = (2 - 2\lambda)(1 - \lambda) - 1 = 2(1-\lambda)^2 - 1 = 0, so λ=1∓1/2=0.2929\lambda = 1 \mp 1/\sqrt2 = 0.2929 and 1.70711.7071, and ω=0.5412\omega = 0.5412 and 1.30661.3066. The second row gives u1=(1−λ)u2u_1 = (1-\lambda)u_2: for λ=0.2929\lambda = 0.2929, u=(1,2)\mathbf u = (1, \sqrt2) (both move the same way, the lighter one more); for λ=1.7071\lambda = 1.7071, u=(1,−2)\mathbf u = (1, -\sqrt2). Check: u1TMu2=2⋅1⋅1+1⋅2⋅(−2)=0\mathbf u_1^{\mathsf T}M\mathbf u_2 = 2\cdot1\cdot1 + 1\cdot\sqrt2\cdot(-\sqrt2) = 0.

✏️ Pen & paperThree masses, one pluck●●○

For N=3N = 3, k=m=1k = m = 1, find the frequencies and the three mode shapes, then write x1(t)x_1(t) for the start x(0)=(1,0,0)\mathbf x(0) = (1, 0, 0) at rest. Check that the energy equals 11.

Show worked solution

ωj=2sin⁡(jπ/8)\omega_j = 2\sin(j\pi/8) gives ω2=2−2, 2, 2+2\omega^2 = 2 - \sqrt2,\ 2,\ 2 + \sqrt2, so ω=0.7654, 1.4142, 1.8478\omega = 0.7654,\ 1.4142,\ 1.8478. The normalized shapes are v1=(12,12,12)\mathbf v_1 = (\tfrac12, \tfrac1{\sqrt2}, \tfrac12), v2=(12,0,−12)\mathbf v_2 = (\tfrac1{\sqrt2}, 0, -\tfrac1{\sqrt2}), v3=(12,−12,12)\mathbf v_3 = (\tfrac12, -\tfrac1{\sqrt2}, \tfrac12). The amplitudes qj(0)=vj⋅x(0)q_j(0) = \mathbf v_j\cdot\mathbf x(0) are 12, 12, 12\tfrac12,\ \tfrac1{\sqrt2},\ \tfrac12, so x1(t)=14cos⁡ω1t+12cos⁡ω2t+14cos⁡ω3tx_1(t) = \tfrac14\cos\omega_1t + \tfrac12\cos\omega_2t + \tfrac14\cos\omega_3t, which is 11 at t=0t = 0. The energy is 12∑ωj2qj2=12(2−24+1+2+24)=1\tfrac12\sum\omega_j^2q_j^2 = \tfrac12\bigl(\tfrac{2-\sqrt2}{4} + 1 + \tfrac{2+\sqrt2}{4}\bigr) = 1, which equals 12K11x12=12⋅2\tfrac12 K_{11}x_1^2 = \tfrac12\cdot2.

✏️ Pen & paperWeak coupling, long beats●●○

Two identical oscillators with m=k=1m = k = 1 are coupled by kc=0.02k_c = 0.02. Mass 1 is released at 11 with mass 2 at rest. After how long is mass 1 first (nearly) at rest with the motion on mass 2? Compare with the estimate π/(ω1kc/k)\pi/(\omega_1k_c/k).

Show worked solution

ω1=1\omega_1 = 1 and ω2=1+2kc=1.04=1.0198\omega_2 = \sqrt{1 + 2k_c} = \sqrt{1.04} = 1.0198, so Δ=0.0198\Delta = 0.0198 and the transfer time is π/Δ=158.6\pi/\Delta = 158.6. The estimate gives π/0.02=157.1\pi/0.02 = 157.1, within 1%1\%.

✏️ Pen & paperShort waves are slow●●●

Starting from ω=(2c/a)sin⁡(κa/2)\omega = (2c/a)\sin(\kappa a/2), find the wavelength (in units of aa) at which the phase speed is 90%90\% of cc, using sin⁡y≈y−y3/6\sin y \approx y - y^3/6, and give the group speed there.

Show worked solution

With y=κa/2y = \kappa a/2, the phase speed ratio is sin⁡y/y≈1−y2/6=0.9\sin y/y \approx 1 - y^2/6 = 0.9, so y2=0.6y^2 = 0.6 and y=0.775y = 0.775 (the exact solution of sin⁡y/y=0.9\sin y/y = 0.9 is 0.7870.787), κa=1.55\kappa a = 1.55 and λ=2π/κ=4.05a\lambda = 2\pi/\kappa = 4.05a (exactly 3.99a3.99a). The group speed is ccos⁡y≈0.71cc\cos y \approx 0.71c either way, so a packet at that wavelength travels at about 71%71\% of cc: the energy lags the crests.

What comes next

The next lesson, 14.2 Electrical Circuits, takes this equation into electronics. An RLC circuit obeys Lq¨+Rq˙+q/C=V(t)L\ddot q + R\dot q + q/C = V(t), the same second-order equation as a damped, driven mass on a spring, with inductance in the place of mass, resistance in the place of the damper and 1/C1/C in the place of the spring. The syllabus lists that analogy with impedance, filters and multi-loop circuits as systems. I expect, from memory, that coupled loops give a matrix equation like this lesson’s, with inductance and 1/C1/C matrices, and so have normal modes; that is for the next lesson to check.

Key Takeaways

  • Masses on springs give Mx¨=−KxM\ddot{\mathbf x} = -K\mathbf x, MM diagonal, KK symmetric positive definite. Trying ucos⁡ωt\mathbf u\cos\omega t gives Ku=ω2MuK\mathbf u = \omega^2M\mathbf u; symmetrizing shows the ω2\omega^2 are real and positive and the shapes MM-orthogonal.
  • Modal coordinates give NN independent oscillators q¨j+ωj2qj=0\ddot q_j + \omega_j^2q_j = 0; every motion is a sum of normal modes. Each mode’s energy is conserved; energy moves between masses, not between modes.
  • Two equal masses with coupling kck_c have modes (1,1)(1,1) at k/m\sqrt{k/m} and (1,−1)(1,-1) at (k+2kc)/m\sqrt{(k+2k_c)/m}. With kc=0.05k_c = 0.05 the frequencies are 11 and 1.04881.0488, and the beat transfer time π/Δ\pi/\Delta is 64.464.4.
  • A fixed-end chain of NN equal masses has ωj=2k/msin⁡jπ2(N+1)\omega_j = 2\sqrt{k/m}\sin\frac{j\pi}{2(N+1)} and sine shapes. It is the finite-difference Laplacian, modal decomposition is the discrete sine transform, and N=5N = 5 gives 0.5176, 1, 1.4142, 1.7321, 1.93190.5176,\ 1,\ 1.4142,\ 1.7321,\ 1.9319.
  • The continuum limit is utt=c2uxxu_{tt} = c^2u_{xx} with c=ak/mc = a\sqrt{k/m}. The chain’s dispersion relation ω=(2c/a)sin⁡(κa/2)\omega = (2c/a)\sin(\kappa a/2) gives phase speed c sinc(κa/2)c\,\mathrm{sinc}(\kappa a/2) and group speed ccos⁡(κa/2)c\cos(\kappa a/2), both below cc, and a maximum frequency 2k/m2\sqrt{k/m} with zero group speed.
  • A finite-difference wave solver is this chain: numerical dispersion is the dispersion of beads. At N=24N = 24 mode 1212 runs at 0.9080.908 of its continuum frequency, mode 2424 at 0.6620.662.
  • Uniform damping γ\gamma keeps modes decoupled; slow modes with ωj<γ/2\omega_j \lt \gamma/2 are overdamped. The highest frequency sets the explicit step limit h≤2/ωmax⁡h \le 2/\omega_{\max} (1.00201.0020 at N=24N = 24), and the stiffness ratio is 15.915.9.
  • Next: 14.2 Electrical Circuits, where an RLC circuit is the same second-order equation.
Finished this lesson?

Mark it complete to track your progress (saved in this browser). Reading every section does this for you.