{
  "id": "harmonics",
  "title": "Harmonics",
  "lang": "en",
  "source_lang": "en",
  "summary": "Harmonics are oscillatory components at integer multiples of a fundamental frequency, central to wave analysis, musical sound, electrical power, and nonlinear optics.",
  "category": "science",
  "subcategory": "physics",
  "type": "concept",
  "aliases": [
    "Harmonic components",
    "harmonic frequency components"
  ],
  "generated_at": "2026-11-06",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/harmonics/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/harmonics/",
      "json": "https://aiwiki.page/zh/harmonics.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Harmonics are sinusoidal components whose frequencies are positive integer multiples of a common fundamental frequency. If the fundamental frequency is \\(f_0\\), the harmonic frequencies are \\(f_n=nf_0\\), where \\(n=1,2,3,\\ldots\\). The first harmonic is the fundamental itself; the second harmonic has twice its frequency, the third has three times its frequency, and so on. Harmonics provide a common language for describing periodic vibrations, musical tones, electrical waveforms, and the generation of new frequencies by nonlinear interactions.\n\nThe importance of harmonics rests on two complementary facts. A periodic waveform can, under broad mathematical conditions, be represented by sinusoidal components at harmonic frequencies. Physical systems can also select, suppress, or generate these components through their geometry, boundary conditions, and response to excitation. Consequently, the harmonic content of a signal can reveal how it was produced and how it will affect another system.\n\nThe word has related but distinct uses. A vibrating system’s natural frequencies are not necessarily harmonic, even when its motion is oscillatory. Spherical harmonics and other mathematical harmonic functions belong to a broader theory of spatial structure rather than simply to integer multiples of a temporal frequency. These distinctions are essential when moving between acoustics, mechanics, electronics, and mathematical physics."
    },
    {
      "heading": "Frequency relationships, terminology, and historical development",
      "text": "For an exactly periodic signal, the fundamental period \\(T_0\\) is the smallest positive interval after which the entire waveform repeats. Its fundamental frequency is \\(f_0=1/T_0\\), and its harmonic frequencies are \\(nf_0\\). Frequency is commonly expressed in hertz, equivalent to reciprocal seconds in the International System of Units. Angular frequency is \\(\\omega_0=2\\pi f_0\\), with harmonic angular frequencies \\(n\\omega_0\\). Thus, a waveform with a 10-millisecond fundamental period has a fundamental frequency of 100 Hz and possible harmonics at 200, 300, 400 Hz, and higher integer multiples.\n\nNot every possible harmonic must actually occur. A signal containing components at 200 and 300 Hz repeats every 0.01 seconds and therefore has a 100 Hz fundamental frequency, even though its spectrum contains no component at 100 Hz. By contrast, a signal containing only 200 and 400 Hz has a fundamental frequency of 200 Hz, unless a separate external reference is being used. In engineering, “harmonic order” is often defined relative to a specified supply or drive frequency rather than inferred from the shortest repetition period of an observed signal.\n\nA partial is any sinusoidal component of a complex tone. Harmonic partials have integer-multiple frequencies; inharmonic partials do not. An overtone is a partial above the lowest component, so the first overtone is the second harmonic only when the spectrum follows a harmonic series and its fundamental is present. A subharmonic has a frequency such as \\(f_0/2\\) relative to a reference oscillation. An interharmonic lies between integer multiples of that reference. These terms distinguish different frequency relationships rather than different kinds of sinusoidal motion.\n\nThe subject has roots in ancient investigations of vibrating strings and musical intervals. For an ideal string under unchanged tension, halving its vibrating length doubles its fundamental frequency, producing an octave. Simple length ratios also yield the frequency ratios associated with the fifth and fourth. Such relationships were important in ancient Greek mathematical accounts of music, although traditional stories assigning particular experiments to individual ancient thinkers are not all equally secure.\n\nDuring the seventeenth and eighteenth centuries, studies of vibrating strings connected musical ratios with mechanics. Joseph Sauveur helped develop acoustic terminology around 1700, while Daniel Bernoulli argued in the eighteenth century that string motion could be represented as a superposition of normal vibrations. Joseph Fourier’s work on heat conduction, presented in 1807 and developed in his 1822 treatise, established the broader importance of trigonometric expansions. In the nineteenth century, Hermann von Helmholtz related complex tones, resonance, and auditory perception, helping connect harmonic analysis with experimental acoustics."
    },
    {
      "heading": "Fourier representation of periodic waveforms",
      "text": "The mathematical foundation of harmonic decomposition is the Fourier series. A real periodic function \\(x(t)\\) with period \\(T\\) can be written, subject to appropriate convergence conditions, as\n\n\\[x(t)=a_0+\\sum_{n=1}^{\\infty} \\left[a_n\\cos(n\\omega_0t)+b_n\\sin(n\\omega_0t)\\right], \\qquad \\omega_0=\\frac{2\\pi}{T}.\\]\n\nHere \\(a_0\\) is the mean, or zero-frequency component. It is commonly called the DC component in electrical contexts and is not ordinarily included among the positive-order harmonics. Each pair \\(a_n,b_n\\) specifies the amplitude and phase of the \\(n\\)th harmonic.\n\nWith this convention, the coefficients follow from integration over any complete period:\n\n\\[a_0=\\frac{1}{T}\\int_0^T x(t)\\,dt,\\]\n\n\\[a_n=\\frac{2}{T}\\int_0^T x(t)\\cos(n\\omega_0t)\\,dt, \\qquad b_n=\\frac{2}{T}\\int_0^T x(t)\\sin(n\\omega_0t)\\,dt.\\]\n\nThese formulas work because different sinusoidal harmonics are orthogonal over a whole period: their cross-products integrate to zero. This orthogonality allows one component to be extracted without contamination from the others in an ideal, exactly periodic measurement.\n\nAn equivalent amplitude–phase form is\n\n\\[x(t)=a_0+\\sum_{n=1}^{\\infty}A_n\\cos(n\\omega_0t+\\phi_n),\\]\n\nwhere \\(A_n=\\sqrt{a_n^2+b_n^2}\\). Amplitudes alone do not determine a waveform; relative phases matter as well. Two signals can have identical harmonic magnitudes but different peak values and shapes. Shifting the time origin changes the phase of the \\(n\\)th harmonic by an amount proportional to \\(n\\), without changing its amplitude.\n\nSymmetry strongly constrains harmonic content. If a zero-mean signal satisfies \\(x(t+T/2)=-x(t)\\), its even harmonics vanish. An ideal symmetric square wave consequently contains only odd harmonics, with amplitudes proportional to \\(1/n\\). A symmetric triangular wave also contains only odd harmonics, but their amplitudes decrease as \\(1/n^2\\). A sawtooth wave generally contains both even and odd harmonics, with amplitudes proportional to \\(1/n\\). The faster decay of the triangular wave’s coefficients reflects its greater smoothness: it is continuous, whereas the ideal square and sawtooth waves have jump discontinuities.\n\nFor a square-integrable periodic signal, Parseval’s identity gives\n\n\\[\\frac{1}{T}\\int_0^T |x(t)|^2\\,dt = a_0^2+\\frac{1}{2}\\sum_{n=1}^{\\infty}(a_n^2+b_n^2).\\]\n\nThus the mean square is the sum of independent harmonic contributions. At a jump, a Fourier series satisfying the usual piecewise-smoothness conditions converges to the midpoint of the two limiting values. A finite truncation displays the Gibbs phenomenon: oscillations near the jump whose region narrows as more terms are added, although their relative overshoot does not disappear. These are properties of the representation, not additional physical frequencies generated by the original system."
    },
    {
      "heading": "Natural modes, boundary conditions, and mechanical vibration",
      "text": "Harmonics are closely associated with normal modes, but the concepts are not interchangeable. A normal mode is a characteristic pattern of motion in which a linear system oscillates at one natural frequency with a fixed spatial shape. Whether those natural frequencies form a harmonic series depends on the governing equations and boundary conditions. Harmonic frequency ratios therefore reveal something about a system’s structure, not merely the fact that it vibrates.\n\nAn ideal flexible string of length \\(L\\), fixed at both ends, illustrates an exactly harmonic system. If its tension is \\(F\\) and its mass per unit length is \\(\\mu\\), transverse waves travel at\n\n\\[v=\\sqrt{\\frac{F}{\\mu}}.\\]\n\nThe end conditions allow standing waves with wavelengths \\(\\lambda_n=2L/n\\), giving\n\n\\[f_n=\\frac{nv}{2L}=nf_1.\\]\n\nThe first mode has no interior node, the second has one, and the third has two. The harmonic series follows from the combination of fixed-end conditions and a wave speed independent of frequency.\n\nThe excitation determines which modes are appreciably present. Plucking a string at its midpoint does not excite even-numbered modes in the ideal symmetric model because those modes have a node there. Plucking nearer an end generally gives a different balance of higher harmonics. The subsequent motion follows the superposition principle: each excited mode evolves according to its amplitude, phase, and damping. A system can therefore have a harmonic set of available modes without producing them all at comparable strength.\n\nAir columns offer another important example. An ideal tube open at both ends supports resonances approximately at \\(nc/(2L)\\), where \\(c\\) is the speed of sound. A tube closed at one end and open at the other supports approximately\n\n\\[f_m=\\frac{(2m-1)c}{4L}, \\qquad m=1,2,3,\\ldots,\\]\n\nso only odd multiples of its lowest resonance occur. Real instruments require corrections for open-end radiation, nonuniform bore shape, tone holes, and losses. Conical bores behave differently from simple closed cylindrical tubes, showing why boundary conditions must be considered together with geometry.\n\nMany mechanical systems are not harmonically tuned. A bending beam has natural frequencies governed by flexural stiffness, not the ideal string equation. A circular membrane has mode frequencies related to zeros of Bessel functions, which do not generally form integer ratios. Stiff piano strings likewise have upper partials slightly above exact harmonic frequencies; this contributes to the stretched tuning used across a piano’s range.\n\nA resonance occurs when forcing couples strongly to a mode near its natural frequency. A harmonic of a rotating machine’s operating frequency can therefore excite a structural resonance even when the rotation frequency itself does not. Conversely, a linear structure driven by a pure sinusoid responds at that forcing frequency in steady state; additional natural-frequency motion appears in transients but is not automatically a persistent harmonic series."
    },
    {
      "heading": "Musical tones, timbre, and auditory perception",
      "text": "In music, harmonic content helps distinguish instruments playing nominally the same note. A flute, violin, and trumpet can share a fundamental frequency while producing very different distributions of energy among their harmonics. These distributions contribute to timbre, together with the attack, decay, noise components, vibrato, and changes in spectrum over time. A sustained spectrum alone is therefore an incomplete description of an instrument’s sound.\n\nFor a note with a fundamental of 110 Hz, the first several harmonics occur at 110, 220, 330, 440, and 550 Hz. Their frequency ratios explain important relationships between the harmonic series and musical intervals. The second harmonic lies one octave above the first; the third lies a perfect fifth above the second, with ratio \\(3:2\\); and the fifth lies a major third above the fourth, with ratio \\(5:4\\). These simple ratios do not coincide exactly with every interval in twelve-tone equal temperament. For example, the \\(5:4\\) major third is about 386.3 cents, whereas an equal-tempered major third is 400 cents.\n\nThe perceived pitch of a complex tone is not always the frequency of its strongest spectral component. When several harmonics support a common repetition rate, listeners may hear the corresponding fundamental even if that frequency is physically absent. This “missing fundamental” effect is especially clear for suitable groups of lower harmonics. It reflects auditory processing of spectral relationships and temporal periodicity, rather than requiring a strong sound wave at the perceived fundamental.\n\nIndividual harmonics are not equally distinguishable by the ear. At lower harmonic orders, neighboring components may stimulate sufficiently separate auditory filters to be partly resolved. At higher orders, several harmonics often fall within one filter, creating temporal fluctuations in its output. Auditory models consequently combine place-based information along the cochlea with timing information. Harmonicity can also help the auditory system group components as belonging to one source, although grouping depends on other cues, including simultaneous onset and common modulation.\n\nInstrumental techniques exploit particular harmonics. Lightly touching a string at its midpoint suppresses modes that require motion there while permitting a strong second harmonic. Touching at other appropriate fractional positions selects different subsets of modes. On wind instruments, changes in blowing conditions and fingering can favor higher resonances. The produced tone nevertheless depends on the entire coupled system: excitation mechanism, resonator, radiation, and losses.\n\nHarmonic richness is not simply a scale from pleasant to unpleasant sound. A nearly sinusoidal tone can be musically useful, and a strongly harmonic sound can be harsh or smooth depending on its spectral envelope and level. Bells and gongs frequently contain prominent inharmonic partials yet have recognizable musical identities. The relationship between harmonic ratios and consonance involves auditory interaction, learned musical practice, register, and context rather than a single universal rule."
    },
    {
      "heading": "Nonlinear generation, symmetry, and frequency mixing",
      "text": "A linear, time-invariant system can change the amplitudes and phases of frequencies already present in its input, but it cannot create new frequencies in steady state. Harmonic generation therefore commonly indicates nonlinear behavior. If the output depends nonlinearly on the input, a sinusoidal excitation can produce oscillations at multiples of its frequency. Examples include an amplifier driven toward saturation, a magnetic core with nonlinear magnetization, and a fluid wave whose propagation speed varies with amplitude.\n\nA simple model expresses output \\(y\\) as a power series in input \\(x\\):\n\n\\[y=a_1x+a_2x^2+a_3x^3+\\cdots.\\]\n\nFor \\(x=A\\cos\\omega t\\),\n\n\\[x^2=\\frac{A^2}{2}\\left(1+\\cos2\\omega t\\right),\\]\n\nand\n\n\\[x^3=\\frac{A^3}{4}\\left(3\\cos\\omega t+\\cos3\\omega t\\right).\\]\n\nThe quadratic term produces a mean shift and a second harmonic. The cubic term changes the fundamental amplitude and generates a third harmonic. These relationships show why a nonlinear response can alter both a waveform’s shape and its average value.\n\nSymmetry determines which terms are permitted. An odd-symmetric transfer law, satisfying \\(y(-x)=-y(x)\\), has no even powers in its expansion around the symmetry point. For a zero-offset sinusoidal input, such a memoryless nonlinearity produces odd harmonics but no even harmonics. Unequal clipping of positive and negative excursions breaks that symmetry and can produce even harmonics. A DC bias can similarly allow an otherwise symmetric device to generate even-order components.\n\nWith two or more input frequencies, nonlinearities also produce intermodulation. A quadratic response to frequencies \\(f_1\\) and \\(f_2\\) generates components at \\(f_1+f_2\\) and \\(|f_1-f_2|\\), in addition to second harmonics. Cubic terms produce combinations such as \\(2f_1-f_2\\) and \\(2f_2-f_1\\). These products need not be harmonics of either input separately. In communications systems they can fall into neighboring channels, making intermodulation a different practical concern from single-tone harmonic distortion.\n\nNonlinear wave propagation provides a spatial mechanism for harmonic generation. In a sufficiently intense sound wave, compression and rarefaction can propagate differently, progressively steepening the waveform and transferring energy to higher frequencies. Dissipation becomes increasingly important as gradients sharpen; in strong cases a shock wave forms. Related processes occur in water waves and elastic solids.\n\nNot all new-frequency behavior is ordinary harmonic generation. Periodically driven nonlinear systems can develop subharmonic oscillations, quasiperiodicity, or chaos. Period doubling produces a response with twice the drive period and therefore half its frequency. Such behavior lies outside a simple expansion containing only integer multiples of the drive frequency, although it can still be analyzed using spectral methods."
    },
    {
      "heading": "Harmonics in electrical power and electronic systems",
      "text": "In an ideal alternating-current system, voltage and current are sinusoidal at the supply frequency. Common nominal frequencies are 50 and 60 Hz. Their fifth harmonics occur at 250 and 300 Hz respectively, while their seventh harmonics occur at 350 and 420 Hz. Actual systems contain harmonic currents and voltages because many loads do not draw current in direct proportion to instantaneous voltage.\n\nRectifiers, switched-mode power supplies, variable-speed drives, and saturating transformers are important sources. A capacitor-input rectifier, for example, can draw current mainly near the peaks of the supply voltage rather than throughout the cycle. That pulsed current contains harmonics. When it flows through the supply system’s frequency-dependent impedance, it creates harmonic voltage drops. Consequently, a nonlinear load can distort voltage at a shared connection and affect equipment elsewhere.\n\nAn ideal balanced six-pulse three-phase rectifier has characteristic line-current harmonic orders \\(h=6k\\pm1\\), including the fifth, seventh, eleventh, and thirteenth. Real installations can contain additional orders because of imbalance, control behavior, or departures from ideal operating conditions. In balanced three-phase systems, triplen harmonics—odd multiples of three such as the third and ninth—can be zero-sequence components. In four-wire systems they can add in the neutral rather than cancel, potentially producing substantial neutral current.\n\nHarmonics increase losses through several mechanisms. Their RMS currents add to conductor heating, while frequency-dependent skin and proximity effects can raise effective resistance. Transformers experience additional winding and stray losses, and motors can experience extra heating and pulsating torque. Capacitors have decreasing reactance as frequency rises, so harmonic voltages can cause appreciable harmonic currents. Interactions between capacitance and system inductance can also create resonances that amplify particular orders.\n\nA common index is total harmonic distortion, or THD. For a voltage waveform,\n\n\\[\\mathrm{THD}_V= \\frac{\\sqrt{\\sum_{n=2}^{N}V_n^2}}{V_1},\\]\n\nwhere \\(V_n\\) is the RMS voltage of harmonic \\(n\\) and \\(N\\) is the highest order included. THD is often expressed as a percentage. Its meaning depends on the denominator and included frequency range: a ratio referenced to total RMS differs from one referenced to the fundamental, and current total demand distortion uses a specified demand-current reference instead.\n\nDistortion also changes the interpretation of power factor. For periodic voltage and current, average real power is\n\n\\[P=\\sum_n V_nI_n\\cos\\phi_n,\\]\n\nwith a DC contribution added if present. Apparent power is \\(V_{\\mathrm{rms}}I_{\\mathrm{rms}}\\), and true power factor is their ratio. Under sinusoidal voltage, current harmonics can lower true power factor even when fundamental voltage and current are in phase. Engineering responses include passive tuned filters, active filters, multipulse converter arrangements, and power-factor-corrected converters; their behavior depends on the surrounding network as well as the load."
    },
    {
      "heading": "Optical harmonic generation and strong-field physics",
      "text": "In nonlinear optics, harmonic generation converts incident light into radiation at integer multiples of its frequency. The material response is described through an induced polarization that can be expanded in powers of the electric field:\n\n\\[P=\\epsilon_0\\left(\\chi^{(1)}E+\\chi^{(2)}E^2+\\chi^{(3)}E^3+\\cdots\\right).\\]\n\nThis scalar form suppresses the tensor character of the susceptibilities. The linear term describes ordinary optical response; higher-order terms provide sources for new frequencies. A field oscillating at \\(\\omega\\) can therefore induce polarization oscillating at \\(2\\omega\\), \\(3\\omega\\), and higher multiples.\n\nSecond-harmonic generation was demonstrated by Peter Franken and colleagues in 1961 using ruby-laser light and quartz. Frequency doubling halves the vacuum wavelength: 1064 nm radiation can be converted to 532 nm green light. In a photon description, two incident photons contribute to one outgoing photon at twice the frequency, conserving energy through \\(2\\hbar\\omega=\\hbar(2\\omega)\\). Efficient conversion, however, depends on a coherent interaction across many atoms or molecules, not simply on isolated photon encounters.\n\nSymmetry imposes important restrictions. In the electric-dipole approximation, the bulk second-order susceptibility vanishes in a centrosymmetric medium. Many useful frequency-doubling crystals therefore lack inversion symmetry. Interfaces can generate second-harmonic light even when the adjoining bulk material is centrosymmetric, because the interface itself breaks inversion symmetry. Third-order optical effects are not excluded by bulk inversion symmetry, making third-harmonic generation possible in a wider range of materials.\n\nEfficient conversion also requires phase matching. For a simple second-harmonic process, the wave-vector mismatch is\n\n\\[\\Delta k=k(2\\omega)-2k(\\omega).\\]\n\nIf this mismatch is large, harmonic fields generated at different positions interfere destructively. Birefringent phase matching uses polarization-dependent refractive indices to reduce the mismatch. Quasi-phase matching instead reverses the sign of the nonlinear coefficient periodically, as in periodically poled ferroelectric crystals. Absorption, beam geometry, pulse duration, and optical damage thresholds impose additional constraints.\n\nHigh-harmonic generation in intense laser fields extends beyond the lowest orders and often beyond a simple perturbative power-series treatment. In a widely used semiclassical description for gases, a strong field releases an electron by tunneling, accelerates it, and can drive it back toward its parent ion. Recombination releases a high-energy photon. The spectrum commonly displays a plateau followed by a cutoff, approximated in the basic atomic model by\n\n\\[E_{\\max}\\approx I_p+3.17U_p,\\]\n\nwhere \\(I_p\\) is the ionization energy and \\(U_p\\) the electron’s ponderomotive energy. Coherent combinations of these harmonics enable attosecond-scale pulses and measurements of ultrafast electronic dynamics. Harmonic emission also occurs in solids, where band structure and carrier motion require different microscopic descriptions."
    },
    {
      "heading": "Measurement, digital spectra, and distortion metrics",
      "text": "Harmonics are measured by estimating the amplitudes and phases of frequency components. Instruments include spectrum analyzers, audio analyzers, power-quality meters, and digitizing oscilloscopes. Digital methods commonly use a discrete Fourier transform, often calculated by a fast Fourier transform algorithm. The transform does not decide by itself whether a spectral peak is a harmonic; that classification requires a fundamental or reference frequency.\n\nA finite record limits what can be distinguished. For a record of duration \\(T_{\\mathrm{obs}}\\), discrete Fourier bins are separated by\n\n\\[\\Delta f=\\frac{1}{T_{\\mathrm{obs}}}.\\]\n\nA one-second record therefore has 1 Hz bin spacing, while a ten-second record has 0.1 Hz spacing. Bin spacing is not identical to the ability to resolve two nearby components, which also depends on the window and relative component strengths. Increasing record duration improves frequency discrimination only if the signal remains sufficiently stable during that interval.\n\nSampling imposes another limit. Under the conditions of the sampling theorem, a band-limited signal can be reconstructed when its sampling rate exceeds twice its highest frequency. Components above half the sampling rate can alias into lower frequencies unless suppressed before sampling. Thus a high-order harmonic outside an instrument’s intended measurement band can appear falsely at a lower frequency. Anti-alias filtering, analog bandwidth, and converter performance all matter when interpreting a digital spectrum.\n\nSpectral leakage occurs when finite-record analysis spreads a component’s energy across frequency bins. A record containing an integer number of cycles of a stable sinusoid can align with the Fourier basis, whereas a noninteger number generally does not. Window functions reduce sidelobes at the cost of broadening the central response and changing amplitude scaling. A strong fundamental can otherwise obscure weak nearby components or bias harmonic estimates.\n\nDistortion measurements require further distinctions. THD sums selected harmonics while excluding the fundamental. THD plus noise, often written THD+N, includes noise and other residual components within a specified bandwidth after removal of the fundamental. These quantities depend on bandwidth, weighting, harmonic limits, and reference amplitude. Equal THD values can describe very different spectra: one device may be dominated by a second harmonic, another by many higher orders.\n\nThe measuring system can itself generate harmonics through input overload, amplifier nonlinearity, or converter distortion. Reliable characterization therefore includes calibration, suitable signal levels, and an assessment of measurement uncertainty. Time-varying signals introduce additional complications. A drifting fundamental, amplitude modulation, or a short transient can produce broadened peaks or sidebands that should not automatically be counted as exact harmonics. Synchronous analysis and time–frequency methods address different parts of this problem, but each imposes its own trade-off between temporal and frequency detail."
    },
    {
      "heading": "Spatial harmonics and related mathematical meanings",
      "text": "Harmonic ideas also apply to patterns in space. A structure periodic over distance \\(L\\) can be expanded in spatial sinusoids with wave numbers\n\n\\[k_n=\\frac{2\\pi n}{L}.\\]\n\nThese are spatial harmonics rather than temporal frequencies. A periodic grating, for example, has spatial Fourier components that determine how it redistributes incident waves through diffraction. A pattern can be spatially harmonic while remaining constant in time, or it can have both spatial and temporal periodicity.\n\nRotating machinery provides a direct connection between these descriptions. A rotor’s magnetic field can contain spatial harmonics around its circumference. At a fixed observation point, rotation converts those angular variations into time-dependent signals. Slot geometry, winding distribution, and magnetic saturation can therefore produce forces and torque pulsations at characteristic frequencies. The temporal frequency depends on both the spatial order and rotational speed; an integer spatial order does not by itself specify a frequency in hertz.\n\nSpherical harmonics describe angular patterns on a sphere. They are indexed by degree \\(\\ell\\) and order \\(m\\), rather than by a single temporal harmonic number. They arise in separated solutions of Laplace’s equation and in many rotationally symmetric problems. The lowest degree describes an angularly uniform component; higher degrees describe progressively finer angular structure. Their orthogonality allows a function defined over a sphere to be decomposed into independent angular components.\n\nThis framework is widely used in models of Earth’s gravitational and magnetic fields, as well as in maps of the cosmic microwave background. In gravitational modeling, different spherical-harmonic coefficients represent departures from a spherically symmetric mass distribution. In cosmology, an angular power spectrum measures the statistical strength of temperature or polarization patterns at different angular scales. These “harmonics” do not imply that Earth or the sky is oscillating at a sequence of integer-related temporal frequencies.\n\nA further distinction concerns harmonic functions: functions satisfying Laplace’s equation,\n\n\\[\\nabla^2u=0.\\]\n\nThey occur in source-free regions of electrostatics, steady-state heat conduction under appropriate assumptions, and other potential problems. The term shares a historical and mathematical connection with trigonometric analysis, but a harmonic function need not describe a periodic time signal.\n\nLikewise, a harmonic oscillator is a system with linear restoring force, such as \\(F=-kx\\). Its ideal classical motion is sinusoidal at a single natural frequency; it does not require the simultaneous presence of higher harmonics. The quantum harmonic oscillator has equally spaced energy levels, but these are energy eigenvalues rather than a claim that every observable oscillates through a full harmonic series. The meaning of “harmonic” therefore depends on whether the subject is frequency composition, spatial decomposition, a differential equation, or an oscillator model."
    }
  ],
  "links": [
    "international-system-of-units",
    "fourier-series",
    "gibbs-phenomenon",
    "normal-mode",
    "standing-wave",
    "superposition-principle",
    "music",
    "timbre",
    "musical-pitch",
    "alternating-current",
    "electrical-impedance",
    "electric-field",
    "energy",
    "electron",
    "photon",
    "fourier-transform",
    "fast-fourier-transform",
    "sampling-theorem",
    "calibration",
    "measurement-uncertainty",
    "diffraction",
    "spherical-harmonics",
    "magnetic-field",
    "cosmic-microwave-background",
    "quantum-harmonic-oscillator"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}