A drum is a wave equation with a circular boundary
A stretched circular membrane, like a drumhead, obeys the two-dimensional wave equation: the acceleration of the membrane's displacement at any point is proportional to how curved the surface is around that point, scaled by the wave speed c, which itself depends on the membrane's tension and its mass per unit area.
∂²u/∂t² = c² · ∇²u in polar coordinates: ∂²u/∂t² = c² [ ∂²u/∂r² + (1/r)·∂u/∂r + (1/r²)·∂²u/∂θ² ]
Because the drum is clamped at the rim, displacement must be exactly zero all the way around the boundary circle at every moment — that single constraint is what selects a discrete family of allowed shapes out of infinitely many possible vibration patterns.
Bessel functions: the drum's natural shapes
Separating variables in polar coordinates splits the solution into a radial part and an angular part, and the radial part turns out to satisfy Bessel’s differential equation, whose well-behaved solutions are the Bessel functions of the first kind, Jₙ(r). Each natural vibration mode of the drum is indexed by two integers: m, the number of angular nodal diameters (straight lines through the centre where the membrane stays still), and n, the number of nodal circles (concentric rings of stillness). A mode's shape is Jₙ(kₖₙr) · cos(mθ), and its frequency is fixed by kₖₙ, the n-th zero of Jₙ, divided by the drum radius and scaled by the wave speed c.
u_{m,n}(r,θ,t) = J_m(k_{m,n} · r) · cos(mθ) · cos(ω_{m,n} · t)
ω_{m,n} = c · k_{m,n} / R (R = drum radius, k_{m,n} = n-th zero of J_m)
This is the reason a real drum sounds nothing like a guitar string. A string's overtones fall at neat integer multiples of the fundamental (2f, 3f, 4f...), but a circular drum's frequencies are set by the irregularly spaced zeros of Bessel functions, which are not integer multiples of each other at all — the second mode is roughly 1.59 times the fundamental, the third roughly 2.14 times, and so on. That inharmonic spacing is exactly why an unmodified drumhead has a pitch that is much less clearly defined than a plucked string, and why timpani need carefully tuned, non-uniform membranes to bring their strongest modes closer to a harmonic relationship.
Chladni patterns made visible
Scatter fine sand or salt on a real vibrating membrane and the grains bounce away from the antinodes (where the surface moves the most) and collect exactly along the nodal lines — the diameters and circles where displacement stays zero — tracing out the mode's shape directly. These are Chladni patterns, named after Ernst Chladni, who demonstrated them on vibrating plates in 1787 using a violin bow, decades before anyone had the mathematics of Bessel functions to explain why the sand collected exactly where it did.
Beats: what happens when two modes are close in pitch
Excite two modes with frequencies close to each other and the membrane doesn't simply show a blend of the two shapes at every instant; because the two oscillations drift in and out of phase, the combined amplitude at any point rises and falls at the difference of the two frequencies, an audible or visible slow pulsing called beating. It's the same phenomenon two guitar strings produce when very slightly out of tune with each other, and on a drum it shows up as parts of the Chladni-style pattern brightening and dimming rhythmically as the two nearby modes trade energy back and forth in relative phase.
Superposition: real drum hits are many modes at once
Strike a real drum anywhere off-centre and you excite a whole weighted mixture of m,n modes simultaneously, each ringing at its own natural frequency and each decaying at its own rate (higher modes typically damp out faster). What you hear as a single percussive thud is the linear superposition of dozens of these eigenmodes, and which modes get excited most strongly, and therefore what the drum actually sounds like, depends heavily on exactly where and how hard you strike it — striking near the centre favours the low axisymmetric modes, striking off-centre brings in the asymmetric ones with angular nodal lines.
Frequently asked questions
Why doesn't a drum have a clear musical pitch the way a guitar string does?
A string's natural frequencies are exact integer multiples of its fundamental, which the ear hears as a single clear pitch. A circular drumhead's frequencies are set by the zeros of Bessel functions, which are not evenly spaced multiples of each other, so its overtones sound inharmonic rather than blending into one clean note.
What causes the patterns that sand traces out on a vibrating drum?
Sand grains bounce away from regions of large vibration (antinodes) and settle along the nodal lines, the diameters and circles where the membrane's displacement stays at zero for that mode. The resulting shape, called a Chladni pattern, directly traces out the geometry of whichever eigenmode is being excited.
Why does hitting a drum in different spots change its sound?
Every strike excites a mixture of many eigenmodes at once, and how strongly each mode is excited depends on where the membrane is struck relative to that mode's nodal lines. A centre strike favours simple axisymmetric modes, while an off-centre strike brings in modes with angular nodal diameters too, changing the overall timbre.
Try it live
Everything above runs in your browser — open Drum Membrane and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.
▶ Open Drum Membrane simulation