Wave drag — transonic and supersonic: Visual walkthrough

Step 1 — A tiny sound pulse is just a spreading ring

WHAT. Imagine dropping a pebble in still water. A ring spreads outward at a fixed speed. Air does the same thing with pressure: any small poke sends a pressure ring outward. That speed is called the speed of sound, written $a$ .

WHY we start here. Air is warned that something is coming only by these pressure rings. If the rings reach the air ahead before the body does, the air gets time to gently step aside. If they don't, the air is taken by surprise — and surprise is where shocks (and drag) are born. So the whole story is a race between body and its own rings.

PICTURE. A single point makes one pulse. After a time $t$ the pulse has grown into a circle of radius $a t$ — speed times time, the plain "distance = speed × time" rule.

Step 2 — Many pulses, and the number that decides everything

WHAT. The body doesn't poke the air once — it pokes it continuously as it moves. So at every instant it drops a new expanding ring, while itself sliding forward. We compare, in the same time $t$ :

how far a ring has spread: $a t$ ,
how far the body has travelled: $u t$ .

WHY. Their ratio is the only thing that matters, because the picture looks identical whether $t$ is a second or a year — only the ratio $u / a$ changes the geometry. We give that ratio its own name, the Mach number:

$M = body speed u / ring speed a .$

$M < 1$ : body slower than its rings → rings escape ahead → air is warned.
$M = 1$ : body keeps pace with the front of its own rings.
$M > 1$ : body outruns its rings → the air ahead is never warned.

PICTURE. Three panels: rings escaping ahead (subsonic), rings just touching the body (sonic), rings left behind piling up (supersonic).

Recall Why a ratio, not the raw speeds?

Why does $M = u / a$ , and not just " $u$ ", decide the flow? ::: Because the geometry of body-vs-rings only depends on how the two distances $u t$ and $a t$ compare — their ratio. Same $M$ ⇒ same picture, at any scale.

Step 3 — When the body outruns its rings: the Mach cone

WHAT. Take the supersonic case ( $M > 1$ ). Draw all the rings the body has emitted along its path. Because the body is ahead of every ring's leading edge, all the rings tuck neatly under one straight line touching them all — a tangent to the whole family. In 3-D that tangent line sweeps into a cone: the Mach cone.

WHY the cone matters. Outside the cone the air has felt nothing — total silence, no warning. Inside, the pulses have arrived. The cone is the exact boundary between "warned" and "un-warned" air. Its opening tells us how far supersonic flow reaches; see Sonic boom and the Mach cone.

PICTURE — where the angle comes from. Build a right triangle from one old ring:

the ring's radius $a t$ is the side opposite the cone's half-angle $μ$ ,
the body's travel $u t$ is the hypotenuse (straight-line distance from where that ring was born to where the body now sits).

The sine of an angle is opposite over hypotenuse — and that is exactly the ratio we need:

$sin μ = \frac{opposite}{hypotenuse} = \frac{a t}{u t} = \frac{a}{u} = \frac{1}{M} .$

All the cases (never skip one):

$M$	$1/ M$	$μ = arcsin (1/ M)$	Meaning
$\to 1^{+}$	$\to 1$	$\to 90°$	Cone opens flat — barely supersonic
$2$	$0.5$	$30°$	A clean example
$\to \infty$	$\to 0$	$\to 0°$	Cone wraps tight to the body
$< 1$	$> 1$	undefined	$arcsin$ of a number $> 1$ has no answer — no cone exists, matching "rings escape ahead"

Step 4 — Why silence-then-arrival forces a shock

WHAT. Zoom into a slender body's nose in supersonic flow. The air just ahead is inside the "silent" zone — it has heard nothing, so it is still flowing straight at full speed. A hair's-width later it must suddenly bend around the body. Nature has no smooth ramp available (the warning never came), so the change happens across a razor-thin front: a shock wave.

WHY it is abrupt. In subsonic flow, the pressure rings running ahead let the air ease into its new direction over a long distance. Remove that early warning (supersonic) and the entire turn is crammed into an almost-zero thickness. Big change over tiny distance = a shock. See Normal and oblique shock waves.

PICTURE. Left: subsonic, streamlines curve gently and early. Right: supersonic, streamlines run dead straight then kink sharply at a shock line.

Step 5 — A real shock leans MORE than the Mach cone

WHAT. Put a wedge of half-angle $θ$ (the amount the flow must turn) into a $M_{1} > 1$ stream. The attached shock sits at angle $β$ , and $β$ is fixed by the mass/momentum/energy books across the shock — the == $θ$ – $β$ – $M$ relation==:

$tan θ = 2 shock lean cot β \frac{M _{1}^{2} sin ^{2} β - 1 normal Mach > 1 \Rightarrow real shock}{M _{1}^{2} ( γ + c o s 2 β ) + 2} .$

$θ$ = how hard the wall turns the flow (the cause).
$β$ = the shock's tilt (the effect we solve for). Note: in this step $β$ means the oblique-shock angle; the same letter is reused for a different quantity in Step 7 — we flag it there.
$M_{1}$ = upstream Mach number.
$γ$ = ratio of specific heats ( $\approx 1.4$ for air) — how "springy" the gas is.
$M_{1} sin β$ = the Mach number measured across the shock; it must exceed $1$ for a real shock, which forces $β > μ$ .

WHY $β > μ$ always. Set $θ \to 0$ (a vanishingly gentle turn). Then the shock strength fades to nothing and $β \to μ = arcsin (1/ M_{1})$ — the Mach cone is the zero-strength limit. Any real turn ( $θ > 0$ ) needs real compression, so $β$ climbs above $μ$ . See Theta-beta-M relation and detached bow shocks.

PICTURE. The Mach cone (dashed, angle $μ$ ) drawn under a steeper solid shock (angle $β$ ) off a wedge nose.

Degenerate case — the shock detaches. For each $M_{1}$ there is a biggest turn $θ_{ma x}$ . Ask for more turn than that (a blunt nose), and no attached shock exists: it pops off the body as a curved bow shock standing ahead, locally normal (its steepest, strongest form) right on the centreline. Blunt = strongest shock = most wave drag.

Step 6 — Why a shock IS drag: the entropy ledger

WHAT. A shock conserves total energy (it's fast and adiabatic — no heat leaks out) but it is irreversible: it stirs the gas up. Irreversibility is measured by entropy $s$ , and across a shock $s_{2} > s_{1}$ . The bookkeeping cost of that entropy jump is a drop in stagnation pressure $p_{0}$ (the pressure the flow would reach if brought smoothly to rest):

$\frac{p _{0, 2}}{p _{0, 1}} = exp (always \leq 0 - Δ s / R), Δ s = s_{2} - s_{1} \geq 0.$

$p_{0, 1}, p_{0, 2}$ = stagnation pressure before / after the shock.
$Δ s$ = entropy rise across the shock (never negative — that's the second law).
$R$ = gas constant, just fixing the units.
The exponential of a negative number is less than 1 ⇒ stagnation pressure always drops.

WHY that equals a rearward force. Stagnation pressure is the flow's "push potential." A wake with less $p_{0}$ than the oncoming stream carries a momentum deficit. Balance momentum on a box around the body and that deficit shows up as a net backward force on the body — drag. No stickiness (viscosity) needed; the shock alone did it. See Entropy and stagnation pressure loss.

PICTURE. A control box: full stagnation pressure in front, a shock inside, reduced $p_{0}$ in the wake, with the arrow showing the reaction drag on the body.

Step 7 — The transonic spike, and why supersonic drag falls again

WHAT. Even below $M_{\infty} = 1$ , air speeds up over a curved wing top and can locally hit $M = 1$ . The lowest $M_{\infty}$ where that first happens is the critical Mach number $M_{cr}$ . Push a little faster to $M_{dd}$ and a supersonic pocket forms, capped by a shock that (Step 6) makes drag and thickens the boundary layer behind it. Drag then spikes — the drag-divergence rise near $M_{\infty} \approx 0.8$ – $1.0$ .

WHY it falls afterwards. Push past $M_{\infty} = 1$ everywhere and shocks turn oblique (Step 5) — weaker per unit length, leaning back along the body. The wave-drag recipe below carries a factor $1/ M_{\infty}^{2} - 1$ that shrinks as $M_{\infty}$ grows. So the curve humps up near 1 and slides back down.

PICTURE. A drag-coefficient-vs- $M_{\infty}$ curve: flat, then a sharp hill at $M_{cr} \to M_{dd} \to 1$ , then a gentle decline.

Formula The supersonic pay-off (Ackeret thin-airfoil)

$c_{d, wave} = \frac{4}{M _{\infty}^{2} - 1} (α^{2} + \overline{(d y_{c} / d x)^{2}} + \overline{(d y_{t} / d x)^{2}})$

$\frac{4}{M _{\infty}^{2} - 1}$ = the " $1/ M$ -shrinks-it" factor. The quantity $M_{\infty}^{2} - 1$ is the Prandtl–Glauert compressibility factor (in many books written $β$ — a different $β$ from the shock angle in Step 5; same letter, unrelated meaning). See Prandtl-Glauert compressibility correction.
$α$ = angle of attack = the tilt of the airfoil's chord line relative to the oncoming flow; $α^{2}$ = drag from tilting to make lift (every term a square ⇒ drag never negative).
$\overline{(d y_{c} / d x)^{2}}$ = the chord-averaged square of the camber-line slope ( $y_{c}$ = mean line, so $d y_{c} / d x$ = how the mean line rises); the overbar means "average along the chord."
$\overline{(d y_{t} / d x)^{2}}$ = the chord-averaged square of the thickness slope ( $y_{t}$ = half-thickness); present even at zero lift ⇒ make wings thin.

The one-picture summary

One ring → many rings → a Mach cone (angle $μ$ , $sin μ = 1/ M$ ) → the real shock leaning steeper ( $β > μ$ ) → entropy rises across it → stagnation pressure drops → that deficit is drag.

Recall Feynman retelling — say it in plain words

A flying thing keeps tapping the air, and each tap spreads out as a pressure ring at the speed of sound. If the thing is slow, its rings run out ahead and the air politely steps aside — no problem. If the thing is faster than sound, it outruns its own rings: they can't warn the air in front, so all the rings bunch up under one slanted line, the Mach cone, whose steepness $sin μ = 1/ M$ is just "ring radius over how far the thing moved." Now the air ahead learns about the body only at the last instant and has no room to ease aside, so it slams its direction change into a razor-thin shock. A real shock is a genuine squeeze, so it leans more steeply than the silent Mach cone. Squeezing the air roughly stirs it up — entropy rises — and that stirring quietly steals the flow's push, dropping its stagnation pressure. Less push in the wake than in front means the air is pushing the body backward: that backward push, born purely from shocks and needing no friction at all, is wave drag. It screams upward right around the speed of sound (a shock first sprouts on the wing) and then eases off once you're fully supersonic and the shocks lean back into weaker slanted lines.

Recall Quick checks

What does $sin μ = 1/ M$ come from geometrically? ::: A right triangle with the ring radius $a t$ opposite $μ$ and the body travel $u t$ as the hypotenuse; sine = opposite/hypotenuse = $a / u = 1/ M$ . Why is a real shock steeper than the Mach cone? ::: The Mach cone is the zero-strength limit ( $θ \to 0$ ); any finite turn needs finite compression, forcing $β > μ$ . Where does wave drag physically come from? ::: Entropy rises across the shock ⇒ stagnation pressure drops ⇒ wake momentum deficit ⇒ net rearward force, with no viscosity needed. Why does supersonic wave drag decrease with Mach? ::: Because every term in the Ackeret formula is multiplied by $1/ M_{\infty}^{2} - 1$ , and that factor gets smaller as $M_{\infty}$ grows; physically the shocks become oblique and weaker per unit length.