Standing Waves and Normal Modes

In short

A standing wave is the superposition of two identical sinusoidal waves travelling in opposite directions. Its displacement is

y(x, t) = 2A\sin(kx)\cos(\omega t),

a pattern that does not translate — it oscillates in place. Points where \sin(kx) = 0 never move and are called nodes; points where |\sin(kx)| = 1 oscillate with the full amplitude 2A and are called antinodes. On a string of length L fixed at both ends, only a discrete set of wavelengths \lambda_n = 2L/n and frequencies f_n = nv/(2L) can form a standing wave. These are the normal modes of the string: f_1 is the fundamental, f_2, f_3, \ldots are the harmonics. Every note you have ever heard from a plucked or bowed string is a weighted sum of its normal modes.

Pluck the middle of the third string of a veena. In the first fraction of a second two pulses race off in opposite directions along the string, the one heading left eventually reaches the fixed end and bounces back inverted, the other does the same at the fixed end to the right, and within a few milliseconds the string has settled into a pattern that looks, to the eye, like nothing is moving horizontally at all. The whole string appears to just swell up and down — a single bulge in the middle that rises, falls, rises again, all in place. Touch the middle of the string with your fingertip while it is vibrating and the pitch of the note jumps up by an octave: the single bulge has split into two bulges, with a new stationary point right where your finger is. Touch the string a third of the way along and you get yet another pitch.

This is standing waves. It is what every vibrating string in every Indian classical instrument — the sitar, the veena, the sarangi, the tanpura, the violin — is doing, all the time, all at once. The strings carry waves that travel at tens of metres per second along their length, but because each wave is reflecting off the fixed ends and superposing with itself on the way back, the pattern of displacement on the string is stationary. The waves are moving; the pattern is not.

This article takes you from two counter-travelling sinusoids to the full standing-wave formula, derives the condition that only certain frequencies — the normal modes — can live on a bounded string, and shows why Indian musical instruments sound the way they do. By the end you will see that every note of every melody on every string is a superposition of a discrete set of stationary wave patterns, and the art of playing the instrument is, in physics, the art of choosing which of those patterns to excite.

Two sinusoids, opposite directions — the setup

Start with two identical sinusoidal waves on a long string, one travelling in the +x direction and one in the -x direction. Both have amplitude A, wave number k = 2\pi/\lambda, and angular frequency \omega = 2\pi f.

y_1(x, t) = A\sin(kx - \omega t), \qquad y_2(x, t) = A\sin(kx + \omega t)

The first is a wave you sent down the string; the second is what comes back after reflecting from a fixed end. (The sign conventions — why the reflected wave has +\omega t rather than -\omega t, and whether its overall sign flips — are the topic of Reflection and Transmission of Waves. For now we will work with the two sinusoids as given and let the boundary conditions decide the signs later.)

By the principle of superposition, the net displacement at every point is the sum:

y(x, t) = y_1(x, t) + y_2(x, t) = A\sin(kx - \omega t) + A\sin(kx + \omega t)

That is two SHMs added at every point in space. The question is: what does the sum look like?

Deriving the standing-wave formula

Step 1. Expand each sine using the angle-sum identity \sin(P \pm Q) = \sin P \cos Q \pm \cos P \sin Q.

\sin(kx - \omega t) = \sin(kx)\cos(\omega t) - \cos(kx)\sin(\omega t)

\sin(kx + \omega t) = \sin(kx)\cos(\omega t) + \cos(kx)\sin(\omega t)

Why: the identity separates the x-dependence from the t-dependence. You will see in the next step that the \cos(kx)\sin(\omega t) terms — which mix the two — cancel when you add, leaving a product of a pure function of x and a pure function of t. That product structure is the whole secret of standing waves.

Step 2. Add the two expansions.

y_1 + y_2 = A\big[\sin(kx)\cos(\omega t) - \cos(kx)\sin(\omega t)\big] + A\big[\sin(kx)\cos(\omega t) + \cos(kx)\sin(\omega t)\big]

The \cos(kx)\sin(\omega t) terms cancel; the \sin(kx)\cos(\omega t) terms add.

\boxed{y(x, t) = 2A\sin(kx)\cos(\omega t)}

Why: the combination \sin(kx)\cos(\omega t) factors into a spatial part \sin(kx) times a temporal part \cos(\omega t). This is completely different from a travelling wave, where position and time appear only in the single combination kx - \omega t. In a standing wave the two roles are split: the shape in space is a fixed sinusoid \sin(kx), and the amplitude of that shape oscillates uniformly in time as \cos(\omega t).

Step 3. Read the equation as physics.

The factor \sin(kx) tells you the shape of the string — which points are high and which are low — at the moment when \cos(\omega t) = 1 (i.e. at t = 0, T, 2T, \ldots). The factor \cos(\omega t) tells you how strongly that shape is expressed as time passes. Every point on the string oscillates in SHM with its own amplitude |2A\sin(kx)|, but every point oscillates at the same angular frequency \omega and, critically, in phase — they all reach their maximum at the same instant, and all pass through zero at the same instant.

This is not a travelling wave. Nothing is moving along the x-direction. The string just pulses, with the pattern \sin(kx) modulated in amplitude by \cos(\omega t). Call this a standing wave or a stationary wave — both names describe exactly this behaviour.

Nodes and antinodes

The amplitude of oscillation at position x is |2A\sin(kx)|. Two special kinds of point exist.

Nodes — points that never move

A node is a point where \sin(kx) = 0, so the amplitude of oscillation is zero. The string at that point does not move at any time, ever, regardless of how the wave evolves.

\sin(kx) = 0 \quad\iff\quad kx = n\pi \quad\iff\quad x = \frac{n\pi}{k} = \frac{n\lambda}{2}, \qquad n = 0, 1, 2, \ldots

Why: sine is zero at integer multiples of \pi. Converting to x using k = 2\pi/\lambda gives x = n\pi/k = n\lambda/2. Nodes are therefore spaced half a wavelength apart.

Antinodes — points that oscillate maximally

An antinode is a point where |\sin(kx)| = 1, so the amplitude of oscillation is 2A — twice the amplitude of either of the original travelling waves. At these points the string swings fully up and fully down.

|\sin(kx)| = 1 \quad\iff\quad kx = \frac{(2n+1)\pi}{2} \quad\iff\quad x = \frac{(2n+1)\lambda}{4}, \qquad n = 0, 1, 2, \ldots

Antinodes sit exactly halfway between consecutive nodes. The spacing between two consecutive nodes is \lambda/2, and the spacing between a node and its neighbouring antinode is \lambda/4.

A snapshot of a standing wave. Solid and dashed curves show the string's extreme positions — it oscillates between them. Nodes (N) are on the axis; between every pair of nodes sits an antinode (A) which swings through the full amplitude $2A$. Node spacing is $\lambda/2$; node-to-antinode spacing is $\lambda/4$.

Watch the pattern oscillate

Eight probe points on a standing wave with wavelength $\lambda = 2$ m and frequency $1$ Hz. The four dark points sit at nodes ($x = 0, 1, 2, 3$ m) and do not move. The three red points sit at antinodes ($x = 0.5, 1.5, 2.5$ m) and swing through $\pm 2$ m. The two pink intermediate points show that the amplitude varies smoothly between these extremes. Click replay to watch again.

Notice three features of what you just watched:

Nodes are absolutely stationary. Their trails collapse to single dots.
All points move together. Every red and pink trail reaches its peak at the same instant, then passes through zero together, then reaches the opposite peak together. This is the defining feature of a normal mode: every particle oscillates in phase with every other.
The spatial pattern never shifts. Unlike a travelling wave, there is no sideways motion of the bulges. The pattern of nodes and antinodes is fixed in space.

The bounded string — why frequencies become discrete

So far the two travelling waves could have had any wavelength. But what happens when the string is a finite length, say L, and is clamped at both ends? A real sitar string is fixed where it passes over the bridge and where it loops around the tuning peg. A real veena string is held at the nut at one end and the bridge at the other. The string must have zero displacement at both of its endpoints, at all times, for all modes of vibration. That is the boundary condition.

Assumptions: The string has uniform tension and mass per unit length along its length. Its ends are fixed (zero displacement, not clamped at arbitrary height). The small-amplitude approximation for the wave equation holds (displacements are small compared to the string length).

Applying the boundary conditions

The standing wave on the string must look like

y(x, t) = 2A\sin(kx)\cos(\omega t)

and it must vanish at x = 0 and x = L for every t.

At x = 0: \sin(k \cdot 0) = 0 automatically. The choice of \sin(kx) (rather than \cos(kx)) has already built in the left-end condition.

At x = L: \sin(kL) = 0 must hold. This forces

kL = n\pi, \qquad n = 1, 2, 3, \ldots

Why: \sin is zero only at integer multiples of \pi. The case n = 0 gives k = 0, which corresponds to a string that isn't moving at all — the trivial solution. Integer n \geq 1 gives the physically interesting modes.

Allowed wavelengths and frequencies

Solve kL = n\pi for the wavelength, using k = 2\pi/\lambda.

\frac{2\pi L}{\lambda_n} = n\pi \quad\implies\quad \boxed{\lambda_n = \frac{2L}{n}}, \qquad n = 1, 2, 3, \ldots

And since the wave speed on the string is v (set by the tension and the mass per unit length — see Speed of Waves on a String for the derivation v = \sqrt{T/\mu}), the corresponding frequencies are

f_n = \frac{v}{\lambda_n} = \frac{v}{2L/n} = \boxed{\frac{n v}{2L}}, \qquad n = 1, 2, 3, \ldots

These are the normal modes of the string. Only these discrete frequencies — and no others — can sustain a standing wave on a string of length L fixed at both ends.

n = 1: f_1 = v/(2L), the fundamental or first harmonic. Wavelength \lambda_1 = 2L. Only one antinode, right at the centre; nodes at both ends.
n = 2: f_2 = 2f_1, the second harmonic. Wavelength \lambda_2 = L. Two antinodes, one node in the middle (in addition to the two at the ends).
n = 3: f_3 = 3f_1, the third harmonic. Three antinodes, two internal nodes.
n = 4: f_4 = 4f_1. Four antinodes, three internal nodes.

And so on indefinitely. The frequencies form an arithmetic sequence: each harmonic is an integer multiple of the fundamental. This is the reason stringed instruments sound harmonic rather than arbitrary — the frequencies that the string naturally supports stand in simple integer ratios, and the human ear perceives integer-ratio combinations as musically consonant.

The first four normal modes of a string of length $L$ clamped at both ends. Each mode fits an integer number of half-wavelengths between the fixed ends. Internal nodes appear wherever the shape crosses the axis; the number of internal nodes is $n - 1$.

Counting half-wavelengths

A clean way to remember the mode number is: n counts the number of half-wavelengths that fit in the string. The fundamental (n = 1) is one half-wavelength. The second harmonic is two half-wavelengths (one full wavelength). The third harmonic is three half-wavelengths (1.5 wavelengths). You can read this directly off the diagrams.

Equivalently, n equals the number of antinodes, and also equals the number of internal + boundary nodes minus one.

Explore the modes

Drag the red dot along the lower axis to sweep the mode number $n$ from $1$ to $6$. The red curve shows the envelope $\sin(n\pi x/L)$ — the extreme shape of the $n$-th normal mode on a string of length $L = 6$ (in arbitrary units). Integer values of $n$ give exact standing-wave patterns; non-integer values show how waves that *fail* to satisfy the boundary condition cannot form stable modes.

Why only these frequencies survive

A real string, when plucked or struck, briefly contains a noisy mixture of all possible wave shapes. Within milliseconds the shapes that do not fit the boundary conditions — frequencies f for which \sin(kL) \neq 0 — get into destructive interference with their own reflections and cancel themselves out. Only the shapes satisfying \sin(kL) = 0 — the normal modes — survive, because these are the only ones for which the wave's reflection adds in phase to the original on every round trip.

You can see this by a round-trip argument. A wave on the string bounces between the two fixed ends, and a fixed end flips the sign of the wave on reflection (a result proved in Reflection and Transmission of Waves). After one complete round trip — down and back — the wave has travelled distance 2L, acquired a phase shift 2kL from the propagation, and a factor of (-1)(-1) = +1 from the two reflections.

For the wave to constructively reinforce itself after each round trip, the total phase acquired per round trip must be a multiple of 2\pi:

2kL = 2n\pi \quad\implies\quad kL = n\pi

That is exactly the same condition you got from the boundary condition on the standing wave. Two derivations, same answer — the normal-mode condition is the condition for self-consistent resonance on the bounded string.

Standing waves in Indian musical instruments

The veena string — length, tension, and pitch

The Saraswati veena has four playing strings; each one is a length of metal wire clamped at the nut (near the top of the instrument) and again at the bridge (on the resonator gourd at the bottom), with a typical string length of about L \approx 0.7 m. The wave speed on a metal string under tension T with mass per unit length \mu is v = \sqrt{T/\mu}. Taking a characteristic tension of 80 N and a string of mass per length 3 \times 10^{-3} kg/m gives v = \sqrt{80/0.003} \approx 163 m/s. The fundamental frequency is

f_1 = \frac{v}{2L} = \frac{163}{1.4} \approx 116\ \text{Hz}

— which is close to the low A (A2) used as a drone reference in many ragas. The frets along the veena's neck shorten the effective string length L, each fret raising f_1 by a known ratio. The physics of fretting is just the normal-mode formula with a smaller L.

The sitar and the "touch point" trick

A sitar player sometimes lightly rests a fingertip on the string exactly halfway between the nut and the bridge without pressing it down. This forces a node at that point. The only modes that can now vibrate are those for which the string has a node at the midpoint: that is, modes n = 2, 4, 6, \ldots — the even harmonics. The odd harmonics are damped out because they demand displacement at the midpoint, and the finger prevents it. The note you hear jumps up by exactly one octave, because the new fundamental is f_2 = 2 f_1. Touch the string a third of the way along and only modes n = 3, 6, 9, \ldots survive — the note jumps up by a musical fifth above the octave (an interval Indian musicians call the pancham shifted up a saptak). This technique, called playing harmonics, is a direct audible demonstration of normal-mode selection by boundary conditions.

The tanpura and the sympathetic drone

A tanpura's four strings are tuned to the tonic and its fifth (typically s, P, s, S in the sargam — the two middle strings at the tonic and the fifth, the outer strings doubling them at different octaves). Each string vibrates as a superposition of many normal modes — its fundamental plus a rich stack of harmonics. The bridge of the tanpura has a specially curved shape (the jawari) that couples the harmonics together and extends their decay, producing the shimmering, continuously-sounding drone that underlies every Indian classical performance. The rich timbre is the sum of many normal modes excited together; the sustain is the near-perfect isolation of the string from internal damping.

Worked examples

Example 1: The fundamental of a sitar string

The main kharaj string of a sitar has length L = 0.85 m, is made of steel, has mass per unit length \mu = 4.5 \times 10^{-3} kg/m, and is under tension T = 92 N. Find the fundamental frequency and the frequencies of the second and third harmonics. Sketch the shapes of these three modes.

The first three normal modes on the kharaj string. The fundamental is a single bulge; the second harmonic has a central node; the third harmonic has two internal nodes. Solid and dashed curves show the extreme positions.

Step 1. Compute the wave speed.

v = \sqrt{\frac{T}{\mu}} = \sqrt{\frac{92}{4.5 \times 10^{-3}}} = \sqrt{20444} \approx 143\ \text{m/s}

Why: the wave speed on a tensioned string is v = \sqrt{T/\mu}, derived in Speed of Waves on a String from applying Newton's second law to a small curved segment. The speed depends only on tension and linear density, not on the amplitude or wavelength of the wave.

Step 2. Apply the normal-mode formula.

f_n = \frac{n v}{2L}, \qquad n = 1, 2, 3, \ldots

For the fundamental (n = 1):

f_1 = \frac{1 \times 143}{2 \times 0.85} = \frac{143}{1.70} \approx 84\ \text{Hz}

Why: setting n = 1 gives the lowest allowed frequency — the mode that fits exactly one half-wavelength in the string. Everything else is an integer multiple of this.

Step 3. Second and third harmonics.

f_2 = 2 f_1 = 2 \times 84 = 168\ \text{Hz}

f_3 = 3 f_1 = 3 \times 84 = 252\ \text{Hz}

Step 4. Translate into musical language.

A frequency of 84 Hz is close to a low E (E2 ≈ 82.4 Hz in 12-tone equal temperament; in Indian classical music, the kharaj saptak of a male singer typically sits around this region). The ratios 1 : 2 : 3 correspond to the musical intervals unison, octave, and octave-plus-fifth — the so-called harmonic series that gives the string its characteristic rich, consonant tone.

Result: f_1 \approx 84 Hz, f_2 \approx 168 Hz, f_3 \approx 252 Hz. The string supports infinitely many such modes, each at an integer multiple of f_1.

What this shows: Given the three physical parameters of a string — length, tension, and linear density — the allowed frequencies are completely determined by the normal-mode formula. Tuning a sitar is physically the act of adjusting T (by turning the pegs) to make f_1 match the desired pitch.

Example 2: The harmonic trick — touching the midpoint

Extending Example 1: the sitar player now lightly touches the kharaj string exactly at its midpoint, x = L/2 = 0.425 m. Which normal modes survive? What is the new fundamental frequency heard, and how does this relate to the octave harmonic technique?

Top: the fundamental has its antinode at the midpoint — the finger forces zero displacement there, destroying the mode. Bottom: the second harmonic has a natural node at the midpoint — the finger changes nothing, and the mode is free to vibrate.

Step 1. State the new boundary condition.

The finger forces y(L/2, t) = 0 for all t.

Why: a lightly-touching finger cannot pull the string upward or press it downward, but it damps any vibration at its contact point into stillness. Modes whose natural shape has a node at that point are unaffected; modes whose natural shape has anything else are killed.

Step 2. Check each mode.

Mode n has shape \sin(n\pi x/L). At x = L/2:

\sin\!\left(\frac{n\pi}{2}\right) = \begin{cases} 0 & \text{if } n \text{ is even} \\ \pm 1 & \text{if } n \text{ is odd} \end{cases}

Why: n\pi/2 is a multiple of \pi when n is even (and then \sin = 0) and a half-integer multiple of \pi when n is odd (and then \sin = \pm 1). So the even modes have a natural node at the midpoint; the odd modes have an antinode there.

Step 3. Identify the surviving modes.

The odd modes (n = 1, 3, 5, \ldots) require nonzero displacement at x = L/2 — the finger forbids this, and these modes damp to zero within a few tens of milliseconds.

The even modes (n = 2, 4, 6, \ldots) have a natural node at x = L/2 — the finger touches a point that was already still, and these modes vibrate freely.

Step 4. New fundamental and pitch shift.

The lowest surviving mode is n = 2, with frequency 2 f_1 = 168 Hz. The ear interprets this as a note an octave above the original fundamental — the frequency has exactly doubled. Subsequent surviving harmonics are at 4 f_1, 6 f_1, 8 f_1, \ldots — all at even multiples of the original fundamental, which is the same as integer multiples of the new fundamental 2 f_1.

Result: Only even harmonics survive. The perceived fundamental is f_2 = 168 Hz, exactly one octave above f_1 = 84 Hz.

What this shows: The normal-mode structure is not just mathematics — it can be manipulated physically by imposing extra boundary conditions anywhere on the string. This is how sitar players produce octave harmonics, violin players find natural harmonics on their strings, and physicists build modal filters to select desired frequency components in engineered systems.

Example 3: The melting dome — a standing-wave diagnostic

A physics student visits the Taj Mahal in Agra and notices a curious acoustic effect in the central chamber: certain whistled pitches produce a striking resonance, as if the dome were singing back. The main chamber of the Taj's inner dome has a diameter of about D = 17.5 m. Treating the chamber as a simple one-dimensional tube between two walls (a rough but useful model), estimate the lowest standing-wave frequency and the first few harmonics of the chamber. Take the speed of sound in air at 25°C as v = 346 m/s.

The Taj Mahal central chamber, modelled as a pair of parallel walls separated by $D$. Three lowest standing-wave pressure modes are sketched, each an integer number of half-wavelengths between the walls.

Step 1. Apply the standing-wave condition.

Treating the chamber as a one-dimensional tube of length D closed at both ends (walls are rigid to sound pressure, so pressure antinodes sit at each wall — the mathematical analogue of a string clamped at both ends for displacement), the allowed wavelengths are still

\lambda_n = \frac{2D}{n}, \qquad n = 1, 2, 3, \ldots

Why: "rigid wall" is the pressure-antinode boundary for a sound wave. A tube of length D with pressure antinodes at both ends is mathematically the same problem as a string of length D with displacement nodes at both ends — the same integer condition n\pi emerges. The details are in Vibrations of Strings and Pipes.

Step 2. Compute frequencies.

f_n = \frac{n v}{2D} = \frac{n \times 346}{2 \times 17.5} = \frac{346 n}{35}

f_1 \approx 9.9\ \text{Hz}, \quad f_2 \approx 19.8\ \text{Hz}, \quad f_3 \approx 29.7\ \text{Hz}

Step 3. Interpret.

The fundamental at 9.9 Hz is infrasonic — you cannot hear it directly, but you can feel it in your chest as a low rumble, and large musical instruments like the pipe organ or an orchestral bass drum can drive such frequencies audibly. The higher harmonics f_2, f_3 also lie in the deep-bass range (below the 20 Hz threshold of human hearing). Whistled pitches in the human range (500 Hz and up) correspond to mode numbers n \approx 50 and higher — and at those n the crude "rectangular tube" model breaks down: the actual dome geometry produces a dense forest of modes that create the famously long reverberation times reported by visitors.

Step 4. Reality check.

The Taj's central chamber is approximately spherical, not cylindrical. A full three-dimensional analysis (spherical Bessel functions, not \sin(n\pi x/L)) gives a richer mode structure, and accounts for the observation that whistling at particular pitches causes sustained acoustic rings. But the order of magnitude — fundamental modes in the tens of hertz, dense mode spacing in the audible range — is set correctly by the one-dimensional estimate. That is what this kind of back-of-envelope calculation is for.

Result: f_1 \approx 9.9 Hz, f_2 \approx 19.8 Hz, f_3 \approx 29.7 Hz.

What this shows: Any bounded region of air — a pipe, a chamber, a whole dome — has its own discrete set of normal-mode frequencies set by its size and the speed of sound. The characteristic "sound" of a space is the signature of its normal modes. Architectural acoustics is the practice of tuning those mode frequencies (by choice of shape, absorbing materials, and dimensions) to enhance or suppress different parts of the spectrum.

Common confusions

"A standing wave is not a real wave." It is. At every point on the string, the displacement is a perfectly good oscillating function of time, and the wave equation \partial^2 y/\partial t^2 = v^2 \partial^2 y/\partial x^2 is satisfied everywhere. What is different from a travelling wave is that the pattern of displacement does not move — but the displacement itself is still a function of position and time, and it is still a solution of the wave equation. "Standing" refers to the pattern, not to the physics.
"Nodes are where the wave does not exist." No — the wave very much exists at the nodes; it just happens to sum to zero there because the two counter-travelling components cancel each point exactly. If you could remove the reflected wave instantly, the original travelling wave would still pass through the former node at full amplitude. Nodes are destructive interference points, not absences.
"All modes of a string vibrate at the same amplitude." Very much not. When you pluck or bow a string, the initial shape determines how strongly each mode is excited — this is a Fourier decomposition of the pluck shape onto the mode shapes. A plucked string has strong n = 1, decent n = 2, weaker n = 3 and so on. A bowed string produces a different weighting (the famous Helmholtz motion). The timbre of the instrument — what makes a sitar sound different from a violin sound different from a guitar — is precisely this weighting.
"The second harmonic has a higher wavelength than the fundamental." Backwards. The second harmonic has a shorter wavelength (\lambda_2 = L < \lambda_1 = 2L) and a higher frequency (f_2 = 2 f_1). The rule is the reverse of common first-guesses: higher mode number means shorter wavelength, higher frequency.
"Energy flows along a standing wave." No — the time-averaged energy flux (Poynting-type for any wave) of a pure standing wave is exactly zero. Energy at each point oscillates between kinetic (at \cos(\omega t) = 0, maximum transverse velocity) and potential (at \cos(\omega t) = \pm 1, maximum stretch), but there is no net transport. Two equal and opposite travelling waves carry equal and opposite energy flux, and the two exactly cancel. This is derived in the Going Deeper section.
"Touching a string at any point kills all vibrations." Only modes that require displacement at that point are killed. Modes with a natural node at the touch point are unaffected. This is the entire basis for playing harmonics on stringed instruments — you select the surviving modes by choosing the touch point.

If you came here to understand what standing waves are, how they form, and what the normal modes of a bounded string are, you have it. What follows is for readers who want the energy analysis, the Fourier decomposition of a pluck, degenerate modes in two dimensions, and the connection to quantum mechanics.

Energy in a standing wave — zero net flow

The total energy in a standing wave is the sum of kinetic and potential energy densities, integrated along the string. For the wave y = 2A\sin(kx)\cos(\omega t), the transverse velocity is \partial y/\partial t = -2A\omega\sin(kx)\sin(\omega t), and the slope is \partial y/\partial x = 2Ak\cos(kx)\cos(\omega t).

Step 1. Kinetic energy density.

u_K = \tfrac{1}{2}\mu\!\left(\frac{\partial y}{\partial t}\right)^2 = \tfrac{1}{2}\mu\!\left(2A\omega\right)^2\sin^2(kx)\sin^2(\omega t) = 2\mu A^2\omega^2\sin^2(kx)\sin^2(\omega t)

Step 2. Potential energy density.

u_P = \tfrac{1}{2}T\!\left(\frac{\partial y}{\partial x}\right)^2 = \tfrac{1}{2}T\!\left(2Ak\right)^2\cos^2(kx)\cos^2(\omega t) = 2TA^2 k^2 \cos^2(kx)\cos^2(\omega t)

Using v^2 = T/\mu and \omega = vk, T k^2 = \mu v^2 k^2 = \mu \omega^2:

u_P = 2\mu A^2\omega^2\cos^2(kx)\cos^2(\omega t)

Step 3. Total energy density.

u = u_K + u_P = 2\mu A^2 \omega^2\left[\sin^2(kx)\sin^2(\omega t) + \cos^2(kx)\cos^2(\omega t)\right]

Why: the kinetic and potential terms have opposite x-dependence (\sin^2 vs \cos^2) and opposite t-dependence. Kinetic energy peaks at antinodes (maximum velocity there) when \sin(\omega t) = \pm 1; potential energy peaks at nodes (maximum stretch there) when \cos(\omega t) = \pm 1. Energy sloshes back and forth between antinode regions (all kinetic) and node regions (all potential), but the total is conserved.

Step 4. Energy flux and its time average.

The instantaneous energy flux along a string is

S(x, t) = -T\,\frac{\partial y}{\partial x}\,\frac{\partial y}{\partial t} = -T(2Ak\cos kx\cos\omega t)(-2A\omega\sin kx\sin\omega t) = 4TA^2 k\omega\sin(kx)\cos(kx)\sin(\omega t)\cos(\omega t)

Using \sin(2\theta) = 2\sin\theta\cos\theta:

S(x, t) = TA^2 k\omega\sin(2kx)\sin(2\omega t)

Averaging over one period, \langle\sin(2\omega t)\rangle = 0. So

\boxed{\langle S \rangle = 0}

Why: there is no time-averaged energy flow in a pure standing wave. Energy sloshes in place between kinetic and potential, but does not propagate. This is the physical reason standing waves are called "standing" — it is not just the pattern that stands still; the energy does too, on average.

The plucked string — a Fourier decomposition

When you pluck a sitar string by pulling the midpoint upward by a small height h and releasing it, the initial shape is a triangle: linear from (0, 0) up to (L/2, h) and back down to (L, 0).

y_0(x) = \begin{cases} 2hx/L & 0 \leq x \leq L/2 \\ 2h(L - x)/L & L/2 \leq x \leq L \end{cases}

The initial velocity is zero (you released the string at rest). The string must vibrate as a sum of normal modes:

y(x, t) = \sum_{n=1}^{\infty} B_n \sin\!\left(\frac{n\pi x}{L}\right)\cos(\omega_n t)

The coefficients B_n are found by Fourier decomposition:

B_n = \frac{2}{L}\int_0^L y_0(x)\sin\!\left(\frac{n\pi x}{L}\right)dx

For the triangular pluck at the midpoint:

B_n = \frac{8h}{n^2\pi^2}\sin\!\left(\frac{n\pi}{2}\right)

Why: the integral splits into two pieces over the two linear segments, each gives an integration-by-parts that produces a 1/n^2 factor plus a \sin(n\pi/2) factor capturing the midpoint value. The detailed computation is standard Fourier-series work.

Now \sin(n\pi/2) = 0 for even n and \pm 1 for odd n — so a pluck at the midpoint excites only the odd harmonics. The even harmonics have a node at the midpoint, and the triangular shape has maximum displacement there; the two are orthogonal. If the player plucks at x = L/3 instead, the amplitude factor becomes \sin(n\pi/3), which is zero for n = 3, 6, 9, \ldots — so plucking at the third-point kills every third harmonic, producing a different timbre. This is why where you pluck on a sitar's string matters so much — the player is choosing which Fourier components get weighted.

Degenerate modes — the 2-dimensional membrane

On a vibrating 2D membrane (tabla skin, drumhead) clamped at its rectangular boundary, the mode shapes are \sin(n_x\pi x/L_x)\sin(n_y\pi y/L_y) and the mode frequencies are

f_{n_x, n_y} = \frac{v}{2}\sqrt{\left(\frac{n_x}{L_x}\right)^2 + \left(\frac{n_y}{L_y}\right)^2}

Two integers label each mode, not one. If the membrane is square (L_x = L_y = L), then f_{m, n} = f_{n, m}: two different modes have the same frequency — a phenomenon called degeneracy. Any linear combination of these two degenerate modes is another standing wave of the same frequency, and the membrane has a whole family of equal-frequency patterns with qualitatively different shapes. Tabla players exploit this by striking different points on the skin to excite different combinations, producing the remarkable variety of tones (the ringing nā, the muted tā, the sliding bolti) all from the same instrument.

The degeneracy is a feature of the symmetry of the boundary: a square's symmetry group includes reflections that exchange x and y, and these symmetries force matching frequencies. In general, greater symmetry means more degeneracy. On a circular drumhead, the modes are labelled by Bessel functions and an angular integer, and the rotational symmetry of the disc gives automatic degeneracy of every non-axisymmetric mode. The Indian classical tabla and mridangam, being circular, have this rich spectrum of degenerate modes by geometry alone.

The bridge to quantum mechanics

A quantum particle in a one-dimensional box of length L — a standard class-12 example — satisfies the time-independent Schrödinger equation -\hbar^2\psi''/(2m) = E\psi with boundary condition \psi(0) = \psi(L) = 0. That equation is the same mathematical problem as finding the normal modes of a clamped string. The solutions are \psi_n = \sqrt{2/L}\sin(n\pi x/L), and the allowed energies are

E_n = \frac{n^2\pi^2\hbar^2}{2mL^2}, \qquad n = 1, 2, 3, \ldots

The pattern is different from the string only because the wave equation for \psi relates E to k^2 rather than \omega to k, so the quantised energies scale as n^2 rather than the linear n of string frequencies. But the underlying reason for quantisation is identical: a bounded wave supports only a discrete set of allowed wavelengths, because the boundary conditions impose an integer condition on kL. Every time you hear a student describe quantised energy levels in a box and wonder "why integer?", the answer begins on a string fixed at both ends.

De Broglie's insight that matter behaves as waves, combined with this classical wave-in-a-box analysis, is how quantum mechanics replaced classical mechanics for bound systems: electrons in atoms, nucleons in nuclei, vibrational modes in molecules, and phonons in crystal lattices. The geometrical fact that standing waves only exist at discrete frequencies became, in the 20th century, the physical fact that bound quantum systems have discrete energy levels. The music of the string became the arithmetic of the atom.

Where this leads next

Vibrations of Strings and Pipes — the same analysis applied to open and closed pipes, where the boundary conditions are pressure (rather than displacement) nodes, and closed pipes select only odd harmonics.
Beats — what happens when two standing-wave notes at slightly unequal frequencies are heard together: periodic waxing and waning of loudness, the tuning trick every sitar player uses.
Reflection and Transmission of Waves — the source of the inverted wave that superposes with the incoming wave to make a standing wave in the first place.
Principle of Superposition — the linearity of the wave equation that makes the whole analysis possible.
The Wave Equation — the differential equation whose solutions include both travelling and standing waves as special cases.