3.1.6 · D5Compressible Flow & Aerodynamics
Question bank — Area-Mach number relation A - A - = f(M) — isentropic flow


True or false — justify
Every value of corresponds to exactly one Mach number.
False. Because falls then rises (its derivative changes sign at ), a horizontal line at any cuts the U-shaped curve twice — one subsonic, one supersonic root; only (the valley floor at ) is unique.
can change from station to station along an isentropic streamtube.
False. For isentropic flow the mass flow and stagnation state are conserved, so is a constant — that is exactly why it works as a yardstick.
At the throat of a choked converging–diverging nozzle, and .
True. When the nozzle is choked the actual minimum (throat) area is the sonic area, so the geometric minimum coincides with the reference and there.
If a purely subsonic flow never reaches anywhere, then does not exist.
False. is a reference (the area the flow would need to hit ); it exists mathematically as a constant even when no physical point in the duct is sonic.
can be less than somewhere in a real duct.
False. The curve's minimum value is ; every other Mach gives , so the ratio never dips below one.
As , the required area ratio approaches .
False. As the factor in blows up, so : to carry a fixed mass flow at nearly-zero speed you'd need an enormous area.
For a given , the subsonic root and supersonic root share the same static pressure.
False. They share the same area ratio only. Through , higher means much lower , so the supersonic root sits at far lower static pressure (and temperature).
The area–Mach relation was derived using momentum conservation.
False. It comes from continuity ( const), the definition of at , and the isentropic stagnation relations — momentum is what gives the separate area–velocity relation.
Spot the error
"Since , the exit Mach number is ."
The area ratio is not the Mach number. Plugging into gives two roots (≈0.31 subsonic, ≈2.2 supersonic); you must solve the equation, not read off directly.
"To speed up any flow, shrink the area — that's why nozzles narrow."
True only for subsonic flow. From , once the factor turns positive, so area and speed rise together — you must diverge (widen) the area to accelerate. See Area-Velocity Relation dA-A = (M^2-1) dV-V.
"Lowering the back-pressure below choking pumps more mass through the throat."
Once choked ( at the throat), is fixed by reservoir and . The throat can't "hear" downstream changes because pressure signals travel only at the speed of sound; see Choked Flow & Maximum Mass Flow.
"The formula gives at , so is whichever area happens to be the smallest."
The smallest actual area equals only when the flow is sonic there. In a subsonic-only duct the minimum area still has , so there and its area exceeds .
"Air accelerating in the diverging part of a nozzle must be losing energy since it spreads out."
In supersonic flow the density and temperature drop fast enough that speed increases as area grows — total (stagnation) enthalpy is conserved; the flow trades thermal energy for kinetic energy, not loses it.
"Because looks like a proper function, I can just invert it algebraically for ."
has no closed-form inverse and is non-monotonic; you must iterate numerically or use tables, and choose the branch (sub- or supersonic) from the physics.
Why questions
Why does go to infinity at both and ?
At the flow is nearly still (), so =const demands huge (the factor blows up); at the density collapses toward zero (the bracket's power of blows up), so again must grow enormously to pass the same mass. This is the two rising arms of the U in figure s01.
Why is the minimum of the curve at and not somewhere else?
Because (the -terms cancelled to leave exactly ), which is zero only at and changes from negative to positive there — the sole turning point (the red dot in figure s01). Physically it is where the area–velocity sign flips, so the area is stationary (a minimum) exactly at the sonic point.
Why do we compare every station to the sonic state rather than the reservoir?
The sonic point is the unique state fixed for the whole streamtube by two conserved quantities together — mass flow and stagnation state — so its area is a single constant. The reservoir has , giving no throughput per unit area to normalise by; the sonic state is the natural "one flow unit" that makes collapse continuity into a clean function of alone.
Why does the boundary/back-pressure decide the branch, when geometry already fixes ?
Geometry alone gives a symmetric double answer (the two intersections of the green line with the U-curve in figure s01); the back-pressure — the downstream push sketched in figure s02 — sets the pressure gap that determines whether the flow stayed subsonic or passed through a sonic throat and went supersonic. Only the physics of the ends breaks the tie.
Why does the area–Mach relation assume isentropic flow?
Constant relies on unchanging stagnation state; friction (viscosity) and shocks raise entropy and drop stagnation pressure, which increases downstream, so the single-yardstick derivation no longer holds. In a real nozzle, boundary-layer growth and heat transfer make the effective drift, which is why designers apply discharge-coefficient corrections. See Normal Shock Waves.
Why does knowing from unlock every other property?
Through the Isentropic Stagnation Relations, alone fixes , , and relative to the fixed reservoir values; the area ratio is the one hard-to-get piece, and everything thermodynamic follows once is in hand.
Edge cases
At exactly , how many Mach numbers does give?
Exactly one — this is the single point where the subsonic and supersonic branches meet at the valley floor (the lone turning point of ), so the double-root ambiguity disappears.
What happens to if a normal shock forms inside the diverging section?
The flow is no longer isentropic across the shock: entropy rises, stagnation pressure drops, and the downstream becomes larger than the upstream one — the yardstick jumps. See Normal Shock Waves.
Can a converging-only nozzle produce supersonic exit flow?
No. A converging duct can accelerate subsonic flow at most to at its narrowest exit; to go supersonic you need a diverging section afterward, i.e. a Converging-Diverging (de Laval) Nozzle.
If two nozzles have the same but different absolute sizes, do they exit at the same Mach?
Yes for the same — depends only on the ratio, not absolute area, so exit Mach is identical (mass flow scales with size, but Mach does not).
What is the exit Mach number as the exit area ratio from above?
Both roots converge toward ; the subsonic root rises toward 1 and the supersonic root falls toward 1, meeting exactly at the choked throat condition — the valley floor.
For a duct where the flow stays subsonic throughout, is ever physically occupied?
No — no point in the duct actually reaches , so is a purely mathematical reference smaller than every real area present.
Do the isentropic edge cases above ever hold exactly in a real nozzle?
Never perfectly — viscosity (friction along the walls), boundary-layer growth, and any heat exchange all add entropy, so real flows only approximate the isentropic ; the relation is an idealised backbone corrected by empirical factors in practice.
Recall One-line self-test
Why can't geometry alone tell you the Mach number at a nozzle station? ::: Because is U-shaped and maps each area ratio above 1 to a subsonic and a supersonic Mach; only the back-pressure/boundary conditions pick the real branch.