Quincke interference tube (3D bench)

One sound wave splits into two tube arms and recombines at a microphone; pull the trombone slide out and the level rises and falls every half wavelength.

Switch to 2D

Presets

Playback

Experiment

Tube and sound

View

Camera

Layers

Quality

Model check

The science

What Quincke's tube does

A single sound wave of frequency f enters a tube and is split at a junction into two arms. The arms meet again at a second junction in front of a microphone. One arm has a trombone-style slide: pulling the slide out by x lengthens that arm by 2x, because the sound has to travel down and back along the slide. The detector adds the two waves, so its level depends on the path difference dL = L2 - L1. Rudolph Quincke described this interference tube in 1866.

The model

  • Plane waves in each arm, p(s, t) = a cos(wt - ks) e^(-alpha s) with k = w / c, w = 2 pi f.
  • Detector pressure: p = a1 e^(-alpha L1) e^(-ikL1) + a2 e^(-alpha L2) e^(-ikL2), so the intensity is I = A1^2 + A2^2 + 2 A1 A2 cos(k dL) with A_i = a_i e^(-alpha L_i).
  • Maxima when dL = n lambda, minima when dL = (n + 1/2) lambda. With dL = 2(xB - xA) the minima are spaced lambda / 2 apart in slide position, so c = 2 f x spacing. That is the classic way to measure the speed of sound.
  • With equal power split the minima are perfect zeros; with an unequal split they only reach (a1 - a2)^2. Fringe visibility is 2 a1 a2 / (a1^2 + a2^2).
  • Speed of sound from kinetic theory of an ideal gas: c = sqrt(gamma R T / M) (air 343.2 m/s at 20 C, helium about 1007 m/s, carbon dioxide about 268 m/s).
  • Wall loss (Kirchhoff boundary layer): alpha = (1 + (gamma - 1) / sqrt(Pr)) sqrt(w nu / 2) / (r c); the slider multiplies it so you can switch it off.
  • Plane-wave validity: below the first cross-mode cut-off f_c = 1.8412 c / (pi d). Above it the model is no longer valid and the readout says so.

Honest simplifications

  • No reflections at the junctions or bends: a real T-junction reflects some sound and adds a small fixed phase (end correction), which shifts all minima by the same distance but does not change their spacing.
  • The slide is drawn at a fixed tube size; the diameter slider only changes the loss and the cut-off. Rounded elbows in the drawing are not subtracted from the path length.
  • The air particles are shown with a hugely exaggerated displacement (real amplitudes are micrometres) so that compressions and rarefactions are visible. Red = compression, blue = rarefaction.
  • Viscosity of the gases is taken at 20 C and not rescaled with temperature; humidity and CO2 content of air are ignored.
  • The "measured" speed of sound uses the same wave model as the display, so it checks the method (minima spacing, parabolic refinement) and the bookkeeping, not nature. The independent checks are the time-domain simulation, the frequency comb and the energy balance.

Four independent checks

(1) A sample-based time-domain delay-line simulation reproduces the closed-form level. (2) The speed of sound found from the spacing of minima along the slide equals the kinetic-theory value. (3) The comb of minima along the frequency axis at a fixed path difference gives the same speed (df = c / dL). (4) Energy: averaged over one fringe the intensity equals a1^2 + a2^2, the maximum is (a1 + a2)^2 and the minimum (a1 - a2)^2.

Loading model...
Loading Blender model...
Level vs measurement-slide extension (minima marked)
Level vs frequency at the present path difference
Arm pressures and their sum at the detector
Made with 🟠 Blender · Three.js