Nozzle thermodynamics — isentropic expansion from chamber to exit: Visual walkthrough

This page rebuilds the parent result — the isentropic nozzle equations — from absolute zero. No formula appears before we can see it. By the end you will understand, in pictures, why hot gas in a chamber turns into a fast jet, and why the tube has to first squeeze, then flare.

We build toward one big idea: thermal chaos in → directed speed out, and area is the knob that controls it.

Prerequisites we will lean on, each introduced when needed: Ideal Gas Law, Compressible Flow Fundamentals, Mach Number and Sound Speed, Entropy and Reversibility.

Step 1 — What "gas in a chamber" actually is

WHAT. Picture a sealed hot box: the combustion chamber. Inside are countless molecules flying in random directions, banging on the walls. That banging is pressure; the average speed of the banging is temperature. Nothing yet points backward — the motion is pure chaos.

WHY. Before we accelerate anything, we must name the two "stored" quantities we will spend: pressure $P$ and temperature $T$ . We also name their chamber values with subscript $0$ — the state where the gas is (almost) at rest. These are the stagnation conditions: $P_{0}$ , $T_{0}$ .

PICTURE. Look at the red box: arrows point every which way. That randomness is the fuel tank of energy we are about to redirect.

Figure — Nozzle thermodynamics — isentropic expansion from chamber to exit

Step 2 — One conservation law: energy can't vanish

WHAT. As gas leaves the chamber and speeds up, we track its energy. Two buckets exist:

thermal energy per kilogram, written $h$ (called enthalpy) — the chaos,
kinetic energy per kilogram, $\frac{1}{2} V^{2}$ — the directed motion, where $V$ is speed.

The total is called the stagnation enthalpy $h_{0}$ — the enthalpy the gas would have if brought fully to rest, so all its energy sat in the thermal bucket. For an ideal gas this total is $h_{0} = c_{p} T_{0}$ (thermal energy of the at-rest, chamber state). The total never changes:

$thermal (chaos) h + directed motion \frac{V ^{2}}{2} = total, fixed = c_{p} T_{0} h_{0}$

WHY this tool and not another? We are asking "where does the speed come from?" Energy conservation is the only law that says one bucket fills exactly as the other empties. It answers the question directly: speed is bought with thermal energy.

WHY $h$ and not just $T$ ? For an ideal gas the two are proportional: $h = c_{p} T$ , where $c_{p}$ is the specific heat at constant pressure — how much energy raises one kilogram of gas by one degree while it is free to expand. Substituting $h = c_{p} T$ and $h_{0} = c_{p} T_{0}$ :

$thermal now c_{p} T + \frac{V ^{2}}{2} = thermal at rest c_{p} T_{0}$

PICTURE. The see-saw: as the kinetic bucket (red) fills, the thermal bucket empties by exactly the same amount. Total bar height is constant.

Step 3 — Measuring speed the smart way: Mach number

WHAT. Raw speed in m/s is awkward — "fast" depends on the gas. Instead compare speed to the speed of sound $a$ in that gas. Their ratio is the Mach number:

$M = \frac{V}{a}, a = γ R T$

$V$ = flow speed, $a$ = local sound speed,
$R$ = gas constant (how "springy" the gas is), $T$ = local temperature,
$γ = c_{p} / c_{v}$ = heat capacity ratio, a fixed number for the gas (≈1.2 for rocket exhaust). Here $c_{v}$ is the specific heat at constant volume — the energy to raise one kilogram by one degree when the gas is not allowed to expand. Because a freely-expanding gas also does pushing work, always $c_{p} > c_{v}$ , so $γ > 1$ .

WHY this tool? Sound is the fastest a pressure signal can travel. Whether flow is $M < 1$ (subsonic, signals outrun it) or $M > 1$ (supersonic, they can't) flips the entire behaviour of the nozzle — see Step 6. Mach number is the switch. See Mach Number and Sound Speed.

PICTURE. A runner (flow) racing a sound wave. Below Mach 1 the wave keeps ahead; above Mach 1 the runner leaves the wave behind, forming a cone.

Now rewrite Step 2's energy law using $M$ . Substitute $c_{p} = \frac{γ R}{γ - 1}$ (which follows from $c_{p} - c_{v} = R$ and $γ = c_{p} / c_{v}$ ) and $V = M a$ with $a^{2} = γ R T$ :

$\frac{γ R T}{γ - 1} + \frac{M ^{2} γ R T}{2} = \frac{γ R T _{0}}{γ - 1}$

Divide every term by $\frac{γ R T _{0}}{γ - 1}$ and factor $T$ :

$\frac{T}{T _{0}} (1 + \frac{γ - 1}{2} M^{2}) = 1$

which rearranges to:

Step 4 — The "no waste" rule: isentropic flow

WHAT. We assume two things about the journey:

Adiabatic — no heat leaks through the walls (the gas is gone in milliseconds; there's no time).
Reversible — no friction, no shockwaves scrambling the flow.

Together these mean entropy $s$ stays constant: $Δ s = 0$ . This is isentropic flow. See Entropy and Reversibility.

WHY. Constant entropy is the bridge that links pressure to temperature. Without it, knowing $T$ tells you nothing about $P$ . With it — and provided the gas is calorically perfect, meaning $c_{p}$ and $c_{v}$ (and therefore $γ$ ) are constant over the whole expansion — they are locked together:

$\frac{T}{T _{0}} = (\frac{P}{P _{0}})^{\frac{γ - 1}{γ}}$

left: temperature ratio (Step 3), right: pressure ratio raised to a special power,
the exponent $\frac{γ - 1}{γ}$ is what "constant entropy" costs.

PICTURE. Two dials, $T$ and $P$ , wired together by a rigid rod labelled " $s =$ const". Turn one, the other must follow along a fixed curve.

Invert (raise to power $γ / (γ - 1)$ ) and plug in Step 3:

Step 5 — Why the tube must change shape: the area rule

WHAT. Mass can't pile up: whatever flows in per second flows out per second. With density $ρ$ , speed $V$ , area $A$ :

$ρ V A = \overset{m}{˙} = constant (the co n t in u i t y law)$

Take a tiny step down the nozzle and demand this stay constant. We now build the master shape equation ourselves, from three facts we already have.

WHY (the derivation, step by step).

(a) Continuity in "percent-change" form. If $ρ V A$ never changes, then the fractional changes of its three factors must cancel. Taking the differential of $ρ V A =$ const and dividing by $ρ V A$ :

$\frac{d ρ}{ρ} + \frac{d V}{V} + \frac{d A}{A} = 0$

Each term is a "percent change": if area grows by 1% and density drops by 1%, speed is untouched. This is just bookkeeping — the product is fixed, so the percent changes sum to zero.

(b) Momentum (Euler) for a smooth, frictionless flow. Pushing gas forward costs a pressure drop. For steady flow along the nozzle, $d P = - ρ V d V$ . Rearranged into percent form:

$\frac{d ρ}{ρ} = \frac{d P}{ρ} \cdot \frac{1}{a ^{2}} \cdot \frac{a ^{2} ρ}{ρ} = - \frac{V ^{2}}{a ^{2}} \frac{d V}{V} = - M^{2} \frac{d V}{V}$

Here we used two things we already own: for isentropic flow the sound speed obeys $a^{2} = d P / d ρ$ (a pressure bump travels at the sound speed), and $M = V / a$ from Step 3. In words: faster flow means thinner gas, and the thinning is amplified by $M^{2}$ — this is exactly why supersonic gas thins so violently.

(c) Substitute (b) into (a). Replace $d ρ / ρ$ with $- M^{2} d V / V$ :

\;\Longrightarrow\; (1-M^2)\frac{dV}{V}=-\frac{dA}{A}$$ Divide through by $(1-M^2)$ and flip the sign: > [!formula] The area–velocity rule > $$\frac{dV}{V} = \frac{1}{\,M^2 - 1\,}\,\frac{dA}{A}$$ > $dV$ = tiny speed change, $dA$ = tiny area change, and $M^2-1$ = the sign-flipping gatekeeper we just built. The whole story hides in that denominator. **PICTURE (of the logic).** Two competing effects fight inside the tube: making the tube wider *slows* gas (more room), but the gas thinning out ($-M^2\,dV/V$) *speeds* it. Below Mach 1 the room-effect wins; above Mach 1 the thinning-effect wins. Their tie is $M=1$. **WHY it matters.** The denominator $M^2-1$ *flips sign* when $M$ crosses 1, so the *same* change in area produces opposite effects on speed: | Regime | $M^2-1$ | To speed up ($dV>0$) we need... | Tube shape | |---|---|---|---| | Subsonic $M<1$ | negative | $dA<0$ | **converging** (squeeze) | | Supersonic $M>1$ | positive | $dA>0$ | **diverging** (flare) | | Sonic $M=1$ | zero | $dA=0$ | **throat** (narrowest) | **Read the table off the formula.** When $M<1$, $M^2-1<0$; to get $dV>0$ the right side must be positive, so $dA$ must be *negative* (squeeze). When $M>1$, $M^2-1>0$; now $dA>0$ (flare) gives $dV>0$. The formula and the table now agree exactly. **PICTURE.** The classic hourglass profile: squeeze to a waist, then flare. The red waist is the ==throat==, the one spot where $M=1$. ![[deepdives/dd-physics-3.3.11-d2-s05.png]] > [!intuition] The counter-intuitive bit > Below Mach 1, squeezing speeds gas up (like a garden hose). Above Mach 1 the gas *thins out* so fast that widening the tube is what accelerates it. The throat is the unique handover point. --- ## Step 6 — The throat is forced to Mach 1 **WHAT.** At the throat $dA=0$. The area rule then demands either $dV=0$ or $M^2-1=0$. For a flow that *keeps accelerating* through, the only consistent choice is $M=1$ exactly. We mark throat quantities with a star: $T^*, P^*, \rho^*$. **WHY.** This is the *choke point*: it fixes the maximum mass flow and sets the reference area $A^*$ everything else is measured against. Set $M=1$ in Steps 3–4: > [!formula] Critical (throat) ratios > $$\frac{T^*}{T_0}=\frac{2}{\gamma+1},\quad > \frac{P^*}{P_0}=\left(\frac{2}{\gamma+1}\right)^{\frac{\gamma}{\gamma-1}},\quad > \frac{\rho^*}{\rho_0}=\left(\frac{2}{\gamma+1}\right)^{\frac{1}{\gamma-1}}$$ For $\gamma=1.2$: $T^*/T_0 = 0.909$, $P^*/P_0 = 0.564$. The gas at the waist is still 91% as hot but only 56% of chamber pressure. **PICTURE.** A gauge sweeping from chamber (M=0) to throat (M=1): temperature bar barely dips, pressure bar drops hard, and the red needle lands exactly on Mach 1 at the waist. ![[deepdives/dd-physics-3.3.11-d2-s06.png]] --- ## Step 7 — Turning geometry into performance: the area–Mach map **WHAT.** Combine continuity at any point with continuity at the throat ($\rho V A = \rho^* V^* A^*$). Substituting every isentropic ratio yields the equation a designer actually uses: > [!formula] Area–Mach relation > $$\frac{A}{A^*}=\frac{1}{M}\left[\frac{2}{\gamma+1}\left(1+\frac{\gamma-1}{2}M^2\right)\right]^{\frac{\gamma+1}{2(\gamma-1)}}$$ - $A/A^*$ = ==area ratio==, the geometry you carve into metal, - $M$ = the exit Mach number, the performance you crave, - the exponent $\frac{\gamma+1}{2(\gamma-1)}$: for $\gamma=1.2$ this equals $\frac{2.2}{0.4}=5.5$ — a large power, so the area grows steeply away from the throat. **WHY.** It runs both directions: *pick* a desired exit Mach → *get* the required flare ratio. Notice $A/A^*$ has a **minimum of 1 at $M=1$** and grows for *both* $M<1$ and $M>1$ — the same area appears twice, once subsonic, once supersonic. Which one you get depends on whether you're before or after the throat. (The minimum value of exactly 1 is machine-checked in VERIFY.) **PICTURE.** A U-shaped curve of $A/A^*$ vs $M$, bottoming at $(1,1)$ in red. Left arm = converging (subsonic), right arm = diverging (supersonic). ![[deepdives/dd-physics-3.3.11-d2-s07.png]] --- ## Step 8 — Worked numbers, seen on the curve Using the parent's Example 2: $T_0=3500\text{ K}$, $P_0=10\text{ MPa}$, $\gamma=1.2$, exit $M_e=3$. First note the recurring group $1+\frac{\gamma-1}{2}M_e^2 = 1+0.1\times9 = 1.9$. > [!example] Exit conditions at $M_e=3$ > **Temperature:** $\displaystyle T_e=\frac{3500}{1.9}=1842\text{ K}$ > **Pressure:** $\displaystyle P_e=\frac{P_0}{(1.9)^{\gamma/(\gamma-1)}}=\frac{10}{(1.9)^{6}}\text{ MPa}=0.213\text{ MPa}=213\text{ kPa}$ > **Area ratio:** $\displaystyle \frac{A_e}{A^*}=\frac13\left[\tfrac{2}{2.2}(1.9)\right]^{5.5}=\frac13(1.727)^{5.5}=6.7$ **WHAT it means.** To reach Mach 3, the exit must be **≈6.7×** the throat area. Since $P_e=213\text{ kPa} > P_a=101\text{ kPa}$, the jet is still pushing outward at the lip — ==overexpanded== is the wrong word here; the exhaust pressure *exceeds* ambient, so the nozzle is **under-expanded**: the jet keeps expanding *after* leaving the lip. See [[Nozzle Flow Regimes]] and [[Specific Impulse Optimization]] for why matching $P_e$ to $P_a$ is the goal. --- ## The one-picture summary Everything above, on a single chamber-to-exit strip: the tube squeezes to a red throat then flares; underneath, three tracks show temperature falling, pressure crashing, and Mach number climbing straight through 1 at the waist. ![[deepdives/dd-physics-3.3.11-d2-s08.png]] > [!recall]- Feynman retelling — say it in plain words > Start with a hot box of gas flailing in every direction — that chaos is pressure and temperature. Energy can't disappear, so if I make the gas move *one direction*, its randomness (temperature) has to drop by exactly that much. To keep the same amount of gas flowing every second through a changing pipe, when the gas is slow I speed it up by *squeezing* the pipe; but once it hits the speed of sound at the narrowest waist, the rules flip — now I have to *widen* the pipe to keep it accelerating. That waist is always Mach 1. Because I assumed no friction and no heat leaks (constant entropy) and a gas whose $\gamma$ never drifts, temperature, pressure and speed are all locked to one number, the Mach number: pick the exit Mach I want, and one formula tells me exactly how much wider than the throat the exit must be. Cool the gas, drop its pressure, and out shoots a supersonic jet — that jet is the thrust. > [!recall]- Quick self-test > Where in the nozzle is Mach number exactly 1? ::: At the throat, the minimum-area point. > Why does temperature drop as the gas speeds up? ::: Energy conservation — thermal enthalpy converts into kinetic energy, so $T$ falls as $V^2$ rises. > Below Mach 1, to accelerate gas do you squeeze or flare the tube? ::: Squeeze (converging): $dA<0$ gives $dV>0$ when $M<1$. > Above Mach 1, squeeze or flare? ::: Flare (diverging): $dA>0$ gives $dV>0$ when $M>1$. > What does the assumption "isentropic" buy us? ::: Constant entropy (with constant $\gamma$) locks $P$, $T$, $\rho$ together, letting one Mach number fix all three. > [!mnemonic] Remember the flip > **"Slow squeeze, fast flare, sonic at the waist."** > Below Mach 1 you *squeeze* to speed up; above Mach 1 you *flare* to speed up; the handover — Mach 1 — sits at the narrowest waist, the throat. --- **Continue:** [[Thrust Equation Derivation]] uses the exit velocity found here · [[Combustion Chamber Thermodynamics]] supplies $P_0,T_0$ · [[Nozzle Flow Regimes]] explains over/underexpansion.