Doppler Effect & Shock Waves · Guide · glidepathscience.org
GuideDoppler Effect & Shock Waves
Contents
This Guide explains the ideas and equations behind every Walkthrough step, Concept Challenge question and Predict round, in the order a textbook takes them. Part 1 prepares the Intro track and Part 2 the Advanced track; each takes about a quarter of an hour to read. Each section ends with a worked example that is not one of the sim’s own problems.
Part 1 · The essentials
Math you’ll need: algebra, solving one equation for one unknown, and the sine of an angle; one optional box uses a derivative.
Why a moving siren changes its pitch
Stand beside a road as an ambulance goes by. While it comes toward you its siren sounds higher than it really is; the moment it passes, the pitch drops, and as it drives away it sounds lower. The siren itself never changes. What changes is how often its crests of pressure reach your ear. This is the Doppler effect, named for Christian Doppler, who described it in 1842, and it works for every kind of wave: sound, water waves and light.
The effect is more than a curiosity. Police radar guns use it to time cars, weather radar to see the wind inside a storm, doctors to watch blood flow, and astronomers to learn how fast stars and galaxies move. Pushed to its limit, when a source moves faster than its own waves, it becomes a shock wave: the cone of a supersonic jet, heard on the ground as a sonic boom. This Guide builds all of that from one picture: crests spreading out as circles from the points where they left.
Crests from a still and a moving source
A source of sound at rest sends out a crest of pressure every period $T = 1/f$. Each crest spreads as a circle (a sphere, in three dimensions) at the speed of sound $v$, about 343 m/s in air at 20 °C. Neighboring crests are one wavelength apart, the distance a crest travels in one period:
$λ = v/f$
Now let the source move at speed $v_s$. Each crest still spreads at $v$ from the point where it left, because once a crest is launched the air carries it and the source has no further say. But between one crest and the next, the source moves $v_sT$ forward. Ahead of it, the next crest starts that much closer to the last; behind it, that much farther away:
$λ = (v ∓ v_s)/f$ (ahead, behind)
A common slip is to think the sound ahead travels faster, at $v + v_s$. It does not: the speed of sound is set by the air. The source changes the spacing of its crests, never their speed. The next section turns that spacing into the pitch a listener hears.
The moving-source formula
A listener at rest ahead of the source meets crests that are $(v − v_s)/f$ apart, arriving at speed $v$. The number arriving each second is the heard frequency $f′$:
$f′ = fv/(v ∓ v_s)$
with the minus sign while the source approaches and the plus sign once it moves away. A source moving at a tenth of the speed of sound raises the pitch by about a tenth while it approaches and lowers it by a little less as it leaves.
Notice what the formula does not say. While the source approaches at a steady speed, $f′$ is steady: the pitch does not rise as the source gets closer. What rises is the loudness. Many people remember a passing siren as “rising, then falling”, but the rise they remember is the loudness; the pitch is high and steady, slides down quickly as the source passes, then stays low.
A moving listener, and the sign rule
Now hold the source still and move the listener toward it at speed $v_o$. The crests are evenly spaced, $λ = v/f$, but the listener runs into them, meeting them at the closing speed $v + v_o$. Moving away, the crests have to catch up, at $v − v_o$:
$f′ = f(v ± v_o)/v$
The two cases combine into one formula for any source and listener moving along the line between them:
$f′ = f(v ± v_o)/(v ∓ v_s)$
The signs follow one rule: use the upper sign for a motion toward the other, the lower sign for a motion away. Check each motion on its own. A source moving toward the listener raises the pitch, so its sign must make the denominator smaller: $v − v_s$. A listener moving toward the source raises the pitch, so its sign must make the numerator bigger: $v + v_o$. When the source and listener move the same way at the same speed, the two factors cancel exactly and there is no shift at all.
A source moving at 30.0 m/s toward a still listener would give 500 × 343/313 = 547.9 Hz, a little more than the moving listener’s 543.7 Hz. At low speeds the two are almost the same, but not exactly; the reason is that the air itself picks out who is moving.
The passing siren, and a speed from a pitch
A siren passing beside you, rather than straight at you, gives the pattern of the chart below. Far up the road, it moves almost straight toward you and the pitch is $fv/(v − v_s)$; far down the road, $fv/(v + v_s)$. In between, the pitch slides from one plateau to the other, and it slides fastest just as the siren goes by. The closer you stand to the road, the more sudden the slide; the plateaus themselves depend only on the speed.
The plateaus measure the speed. If the approaching pitch is $f_a$ and the receding pitch $f_r$, dividing one formula by the other removes the unknown $f$, and solving gives the speed and the siren’s own frequency:
$v_s = v(f_a − f_r)/(f_a + f_r)$
If you know the siren’s true pitch, one reading is enough: solve $f′ = fv/(v − v_s)$ for the speed,
$v_s = v(1 − f/f′)$
Mach 1, and the shock cone
As the source speeds up toward the speed of sound, the spacing ahead, $(v − v_s)/f$, shrinks toward zero. Every crest the source sends forward lands almost on top of the one before. In the linear picture the pitch ahead, $fv/(v − v_s)$, grows without limit, but real air stops following that formula here: crests this crowded are no longer a small wave, a strong compression travels a little faster than a weak one, and the front steepens into a shock, a nearly sudden jump in pressure. Early jet designers called this front the “sound barrier”. The ratio of the source’s speed to the speed of sound is its Mach number, $M = v_s/v$.
Past Mach 1 the source outruns its own sound. In a time $t$, a crest it sent spreads to radius $vt$ while the source flies a distance $v_st$. Every crest just touches two lines from the source, the edges of a cone, and the half-angle $θ$ of that cone has the crest’s radius opposite it and the source’s path as the hypotenuse:
$\sin θ = 1/M$, $M = v_s/v$
The faster the source, the narrower the cone. A common slip is to use cosine or tangent; draw the triangle and the sine follows.
The sonic boom
Along the cone’s edge real air carries a shock, a sudden jump in pressure, and when it sweeps over you, you hear it as a sonic boom: a bang, not a pitch. The boom is not made at one moment when the jet “breaks the sound barrier”: the cone travels with the jet for the whole of its supersonic flight, and everyone under its path hears the boom in turn, like the wake of a boat reaching the shore.
It also arrives late. A jet flying level at altitude $h$ passes over you in silence; the cone’s edge trails behind it and reaches the ground a distance $h/\tan θ$ behind the jet. By the time it reaches you, the jet has flown that far past, which takes
$Δt = h\sqrt{M^2 − 1}/v_s$
(using $\tan θ = 1/\sqrt{M^2 − 1}$, which follows from $\sin θ = 1/M$). The formula assumes the air has the same temperature everywhere.
Radar guns: a shift taken twice
A radar gun sends radio waves at a car. Light and radio waves travel at $c$ = 3.00 × 10⁸ m/s, and for speeds far below $c$ they shift just as sound does. The car first acts as a moving observer, receiving waves at a shifted frequency; then it reflects them as a moving source, shifting them again. For a car closing at $u$, the echo returns shifted by twice the single shift:
$Δf ≈ 2fu/c$
That shift is tiny, about one part in ten million of the gun’s frequency. The gun does not measure the echo’s frequency directly: it mixes the echo with its own outgoing wave, and the sum rises and falls at the difference, a beat at $Δf$, which an electronic counter times easily. A car pulling away gives the same beat with the echo below the outgoing frequency.
Redshift
Light from a source that moves away from us arrives with its waves stretched: every wavelength in its spectrum is longer by the same fraction. Astronomers call that fraction the redshift,
$z = Δλ/λ ≈ v/c$
where the approximation holds for speeds far below $c$. An approaching source is blueshifted, with $z$ negative. The word “red” describes the direction of the shift, toward longer wavelengths, not a color: a blue line moves toward green, and an infrared line moves farther into the infrared. Because every line moves by the same fraction, astronomers can recognize a familiar pattern of lines, such as hydrogen’s, and read off $z$.
At speeds that are a large fraction of $c$, $z ≈ v/c$ is no longer accurate, and the formula of special relativity takes over. The redshift of very distant galaxies also has another cause, the stretching of space itself, which this sim leaves out.
Part 2 · Going further
Math you’ll need: algebra with vectors and the dot product, the quadratic formula, first-order approximations; one section uses a derivative and one special relativity’s $γ$.
Deriving the shift, and why source and observer differ
Let a source moving toward a still listener at $v_s$ emit one crest at $t = 0$ and the next at $t = T$. When the second leaves, the first is $vT$ ahead of where it left, and the source is $v_sT$ along, so the two crests are $(v − v_s)T$ apart and both travel at $v$: they reach the listener $(v − v_s)T/v$ apart in time, and $f′ = fv/(v − v_s)$. For a moving listener the crests are unchanged, $λ = vT$, and they are met at the rate $(v + v_o)/λ$.
The two results agree to first order in the speeds, $f′ ≈ f(1 + u/v)$, but not beyond. At a fifth of the speed of sound, a moving source gives 1/(1 − 0.20) = 1.25 times the pitch and a moving listener 1.20 times. The reason is that the air is a preferred frame for sound: what matters is who moves through it. With the source and listener both moving, everything is measured in the air’s frame. Light has no medium, so for light only the relative velocity can matter, and the formula must be the same whichever one you call the mover.
Off-axis motion, the retarded position and the pitch sweep
When the source does not move straight at the listener, only its velocity along the line to the listener shifts the pitch, and that line must be taken from where the source was when the sound left it, the retarded position. If the source’s velocity made an angle $θ$ with that line at the moment of emission,
$f′ = f/(1 − (v_s/v)\cos θ)$
The retarded time $t_e$ of the sound heard at time $t$ satisfies $|\vec r_o(t) − \vec r_s(t_e)| = v(t − t_e)$; for straight-line motion at constant speed this is a quadratic in $t_e$, which the sim solves exactly. A consequence: you hear a subsonic jet’s sound from where it was, and by then it has flown on. A jet at Mach $M$ heard from straight overhead at altitude $h$ is now $x = Mh$ ahead.
$x = Mh$ (retarded position)
For a car passing at closest distance $d$, the angle sweeps from 0 to 180°. Near closest approach, $\cos θ ≈ −v_st/d$, so the pitch falls almost linearly, and at low speed the steepest slope of the pitch is
$|df′/dt|_{max} ≈ fv_s^2/(vd)$
Reading a recorded pass therefore gives everything: the plateaus give $v_s$ and $f$, and the steepness gives $d$.
Wind
A steady wind $w$ blowing from the source toward the listener carries the crests faster, at $v + w$, but also spreads them farther apart, $(v + w)/f$. The arrival rate is unchanged: with the source and the listener both at rest on the ground, wind alone never shifts the pitch. In the air’s frame, both move at the same velocity, and the shifts cancel exactly. Wind matters only when something moves through the air; then the general formula is used with every velocity measured relative to the air, $f′ = f(v + w + v_o)/(v + w − v_s)$ along the line.
Shock cones, booms and the N-wave
A cone’s angle is a speedometer that needs no clock: $M = 1/\sin θ$. Ernst Mach measured it this way in the 1880s, photographing bullets lit by electric sparks.
$M = 1/\sin θ$
The boom’s delay after a jet passes overhead, $h\sqrt{M^2 − 1}/(Mv)$, and the distance the jet has gone, $h\sqrt{M^2 − 1}$, assume uniform air. In the real atmosphere the air is colder aloft, so sound is slower there, and the boom’s rays bend upward. The band of ground that hears the boom, the boom carpet, is narrower than the cone suggests, and below a cutoff Mach number (roughly 1.1 to 1.3, depending on altitude and the air’s temperature) the boom may not reach the ground at all.
Two kinds of physics meet at a shock, and it helps to keep them apart. The circles and the cone are kinematics: where each crest of a small (linear) wave would be. That geometry is exact at any speed, so the cone’s angle and the boom’s arrival time come out right. The shock itself is nonlinear: its strength, its shape, how sharply it rises and its speed (a strong shock runs a little faster than sound) depend on how air responds to a large compression, which no drawing of circles contains. The Doppler formula belongs to the first kind. It counts the rate at which linear crests reach you, and well below Mach 1 that rate is the pitch you hear. Near Mach 1 the formula’s $f′$ runs off to infinity and past Mach 1 it gives two crest rates, one negative; neither is a sound anyone hears. What reaches the ground is a pressure jump.
A microphone on the ground records a boom as an “N”: a sudden rise in pressure as the nose’s shock arrives, a steady fall to below normal along the jet’s length, and a sudden return as the tail’s shock arrives. The two jumps, a tenth of a second or so apart for a large aircraft, are heard as a double bang.
Double Doppler: ultrasound and echoes
A reflector closing at speed $u$ first receives the wave as a moving observer and then re-emits it as a moving source. Exactly, $f″ = f(v + u)/(v − u)$, and to first order the shift is $2fu/v$. When the reflector moves at an angle $θ$ to the beam, only its component along the beam counts:
$Δf ≈ 2fu\cos θ/v$
Doppler ultrasound measures blood flow this way, with sound at about 1,540 m/s in soft tissue. The shift falls in the audible range, which is why the machines play the flow as a whooshing sound. The angle is the largest source of error: near 60°, a few degrees of error in $θ$ changes the speed by 10 % or more. For radar, the exact relativistic mirror formula is the same $(1 + β)/(1 − β)$.
Relativistic and transverse Doppler shifts
For light, the classical stretching must be combined with time dilation: the moving source’s clock runs slow by $γ = 1/\sqrt{1 − β^2}$, with $β = u/c$. For a source moving straight away the result is
$1 + z = \sqrt{(1 + β)/(1 − β)}$
which depends only on the relative velocity, as it must for light, and reduces to $z ≈ β$ at low speed. In general, with $θ$ the angle in the observer’s frame between the source’s velocity and the line to the observer, $f′ = f/[γ(1 − β\cos θ)]$. At $θ$ = 90° the classical shift vanishes but time dilation remains: light reaching you at right angles to the source’s path is redshifted by $γ$,
$λ′ = γλ$ (transverse)
This transverse Doppler effect has no classical counterpart. Where the right angle is measured matters: light emitted at right angles in the source’s own frame is instead blueshifted by $γ$.
Cherenkov light: a shock cone for light
Nothing outruns light in a vacuum, but in water light travels at $c/n$, with $n$ = 1.33, and a fast charged particle can outrun it. Its light then forms a cone, just as a supersonic jet’s sound does: the crests spread at $c/n$ while the particle moves at $βc$. The light travels at right angles to the cone’s surface, at an angle $θ$ to the particle’s path with
$\cos θ = 1/(nβ)$
Below the threshold β = 1/n there is no cone and no light. This is the blue glow of a reactor pool, and the ring of light that detectors use to find a particle’s direction and speed.
Reference
Common mistakes
Part 1
- Thinking the pitch rises as an approaching siren nears: it is steady; the loudness rises.
- Thinking a moving source’s sound travels faster: it changes the spacing, never the speed.
- Getting a sign backward: motion toward raises the pitch, so it must shrink the denominator or grow the numerator.
- Using the observer formula for a moving source, or the reverse: they differ beyond first order.
- Using cosine or tangent for the cone: $\sin θ = 1/M$.
- Thinking the boom is made once, when the jet passes Mach 1: the cone sweeps the ground for the whole supersonic flight.
- Counting a radar shift once: the reflection doubles it.
- Reading “redshift” as a color: it is a stretch of every wavelength by the same fraction.
Part 2
- Taking the angle at the source’s present position: the shift depends on the angle where the sound left.
- Thinking wind alone shifts the pitch: with both at rest it never does.
- Using $z ≈ β$ at high speed, or a classical formula for light.
- Expecting no transverse shift for light: time dilation gives $λ′ = γλ$.
- Using the cone’s half-angle for Cherenkov light: the light leaves at $\cos θ = 1/(nβ)$ to the path.
The equation sheet
| Part 1 | |
|---|---|
| a still source | $λ = v/f$ |
| a moving source’s crests | $λ = (v ∓ v_s)/f$ (ahead, behind) |
| a moving source | $f′ = fv/(v ∓ v_s)$ |
| a moving listener | $f′ = f(v ± v_o)/v$ |
| both, upper signs toward | $f′ = f(v ± v_o)/(v ∓ v_s)$ |
| speed from a pitch | $v_s = v(1 − f/f′)$ |
| the cone | $\sin θ = 1/M$, $M = v_s/v$ |
| the boom’s delay | $Δt = h\sqrt{M^2 − 1}/v_s$ |
| radar | $Δf ≈ 2fu/c$ |
| redshift | $z = Δλ/λ ≈ v/c$ |
| Part 2 | |
|---|---|
| off-axis | $f′ = f/(1 − (v_s/v)\cos θ)$ |
| the sweep | $|df′/dt|_{max} ≈ fv_s^2/(vd)$ |
| where the jet is | $x = Mh$ (retarded position) |
| Mach from a cone | $M = 1/\sin θ$ |
| ultrasound | $Δf ≈ 2fu\cos θ/v$ |
| light, receding | $1 + z = \sqrt{(1 + β)/(1 − β)}$ |
| light at right angles | $λ′ = γλ$ (transverse) |
| Cherenkov light | $\cos θ = 1/(nβ)$ |
Symbols: $f$ the emitted frequency and $f′$ the heard one; $λ$ the wavelength; $v$ the speed of sound (343 m/s at 20 °C); $v_s$, $v_o$ the source’s and listener’s speeds; $w$ the wind; $M$ the Mach number; $θ$ an angle (the cone’s half-angle, or the angle between a velocity and the line of sight); $h$ an altitude and $d$ a closest distance; $u$ a reflector’s speed; $c$ = 2.998 × 10⁸ m/s; $z$ the redshift; $β = u/c$ and $γ = 1/\sqrt{1 − β^2}$; $n$ a refractive index.
Sources and further reading
- OpenStax, University Physics Volume 1, chapter 17, “Sound,” §17.7 The Doppler Effect and §17.8 Shock Waves (CC BY-NC-SA 4.0, openstax.org): the Intro track follows their order and notation.
- OpenStax, University Physics Volume 3, the relativity chapter’s section on the relativistic Doppler effect (CC BY-NC-SA 4.0): further reading for Part 2.
- This sim’s Field Manual (H) for the model and its limits; Sound & Resonance Tubes for the speed of sound; Special Relativity & Spacetime for time dilation.
Doppler Effect & Shock Waves v1.1 · © 2026 Dr. David Meier · CC BY-NC-SA 4.0 · glidepathscience.org