Visual walkthrough — Mass flow rate ṁ and its relation to throat area

---

Step 1 — What "mass flow rate" even counts

WHAT. Picture a straight pipe. Draw one vertical line across it — a cross-section. Ask a simple question: how many kilograms of gas cross that line each second? That number is the mass flow rate, written $\dot{m}$ (the dot means "per second"), measured in $\text{kg/s}$.

WHY. A rocket is a machine that throws mass backward. Before we can talk about how fast the throat lets gas out, we need a clean count of how much gas passes a line each second. That count is the whole game.

PICTURE. Look at the shaded slug of gas in the figure. In a tiny time $dt$ the gas drifts right by a length $vdt$, where $v$ is its speed. That slug is a cylinder: area $A$ across, length $vdt$ along. (One-dimensional assumption: the same $v$ everywhere across the face.)

The slug's volume is $A\times (vdt)$. Its mass is density times volume, $\rho Avdt$. Divide by the time $dt$:

$\dot{m}=\underset{\text{how packed}}{\underbrace{\rho }}\underset{\text{how wide}}{\underbrace{A}}\underset{\text{how fast}}{\underbrace{v}}$

Three honest factors — pack, width, speed. That is all $\dot{m}$ ever is.

---

Step 2 — The same $\dot{m}$ at every station (continuity)

WHAT. Draw the pipe now as a funnel that narrows and widens. Pick two lines: station 1 (wide) and station 2 (narrow). We claim the same $\dot{m}$ crosses both.

WHY. No gas is created or destroyed inside a sealed nozzle, and none piles up — this is exactly the steady assumption at work. Whatever kilograms-per-second enter a region must leave it. So $\dot{m}$ is a single constant along the whole channel.

PICTURE. In the figure, the wide station has fat slow arrows; the narrow station has thin fast arrows. Same number of kilograms squeeze through per second, so where the door is narrow the gas must move faster (or pack denser).

This single fact — narrow means fast — is the seed of everything. The narrowest point is called the throat, area $A^{\ast }$. It is the bottleneck we will focus all our attention on.

---

Step 3 — Two tools from gas physics (and why exactly these)

To turn "narrow means fast" into a number, we need two facts about how a gas behaves.

Tool 1 — the ideal gas law. It links the three slug-ingredients: $p=\rho RT⟹\rho =\frac{p}{RT}$ Here $p$ = pressure ($\text{Pa}$), $T$ = temperature ($\text{K}$), and $R$ = the gas constant for this gas ($\text{J/kg⋅K}$). WHY this tool: our slug formula needs $\rho$, but density is hard to measure directly. Pressure and temperature are easy. This law trades $\rho$ for $p$ and $T$.

Tool 2 — the speed of sound. How fast do small pressure nudges travel through the gas? $a=\sqrt{\gamma RT}$ $\gamma$ (Greek "gamma") = the ratio of specific heats, a fixed number for a gas (about $1.2$–$1.4$). WHY this tool: the whole drama of choking happens at the speed of sound. We must be able to compare the gas speed $v$ to $a$.

PICTURE. The figure shows a bell of gas; a small "ping" of pressure spreads out at speed $a$. If the gas itself already flows faster than $a$, the ping can never travel upstream — remember this, it is the punchline of Step 7.

---

Step 4 — Chamber vs local: the stagnation relations

WHAT. Inside the combustion chamber the gas is (almost) at rest — call these the stagnation conditions $p_0,T_0$ (subscript zero = "brought to rest"). Out in the flow the gas is moving, with local values $p,T$. We need the bridge between them.

WHY. We can only measure and control chamber conditions $p_0,T_0$. The throat's local $p,T$ are hidden inside a supersonic gas stream. So the final formula must speak only in $p_0,T_0$ — but the physics happens at local $p,T$. We need the translation.

PICTURE. The figure walks the temperature ratio out line-by-line, so you see exactly where $a^2=\gamma RT$ and $M=v/a$ enter. Follow it alongside the algebra below.

Temperature ratio — from energy (adiabatic), derived line by line. With no heat crossing the walls, the total energy per kilogram is the same in the chamber (all thermal, gas at rest) as out in the stream (thermal $+$ kinetic): $\underset{\text{all thermal, at rest}}{\underbrace{c_p{T}_0}}=\underset{\text{thermal}}{\underbrace{c_{p}T}}+\underset{\text{kinetic}}{\underbrace{\frac{1}{2}{v}^2}}$ Divide every term by $c_{p}T$ to make it a ratio: $\frac{T_0}{T}=1+\frac{v^2}{2c_{p}T}.$ Now feed in the two tools. Replace $c_p=\frac{\gamma R}{\gamma -1}$ (from the box above), and note $v^2=M^2{a}^2=M^2(\gamma RT)$ using $M=v/a$ and $a^2=\gamma RT$ from Step 3: $\frac{v^2}{2c_{p}T}=\frac{M^2\gamma RT}{2\cdot \frac{\gamma R}{\gamma -1}\cdot T}=\frac{M^2\gamma RT(\gamma -1)}{2\gamma RT}=\frac{\gamma -1}{2}{M}^2.$ The $\gamma RT$ cancels top and bottom — that is why the messy constants disappear and only $\frac{\gamma -1}{2}{M}^2$ survives. This step needs only the adiabatic assumption.

Pressure ratio — from isentropic (adiabatic + frictionless + no shocks). For a constant-entropy ideal gas, $p\propto T^{\gamma /(\gamma -1)}$. Raise the whole temperature ratio to that power: $\frac{p_0}{p}={\left(\frac{T_0}{T}\right)}^{\frac{\gamma }{\gamma -1}}={\left(1+\frac{\gamma -1}{2}{M}^2\right)}^{\frac{\gamma }{\gamma -1}}.$ This is where we spend the full isentropic assumption — friction or a shock would raise entropy and break $p\propto T^{\gamma /(\gamma -1)}$.

At $M=0$ (gas already at rest) both ratios are $1$ — local equals chamber, as it must.

---

Step 5 — Assemble $\dot{m}$ as a function of Mach number

WHAT. Take the slug formula $\dot{m}=\rho Av$ and rewrite every piece using our tools, so nothing hidden remains.

WHY. We want $\dot{m}$ expressed purely in things we know: area, chamber pressure/temperature, and the Mach number $M$. Then we can hunt for the $M$ that maximises flow.

PICTURE. The figure is a "cancellation map": it lays the assembled expression over a picture of the gas thinning as it accelerates, so you see the two competing powers of $M$ — the lone $M$ on top that boosts flow, and the growing bracket underneath that thins it — that will fight to a draw in Step 6.

Swap the density (Tool 1) and the velocity ($v=Ma=M\sqrt{\gamma RT}$, from Step 3):

= p\,A\,M\sqrt{\frac{\gamma}{R T}}$$ Now trade local $p,T$ for chamber $p_0,T_0$ using Step 4. Write $D = 1+\tfrac{\gamma-1}{2}M^2$ (the shared bracket). Then $p = p_0\,D^{-\gamma/(\gamma-1)}$ and $T = T_0\,D^{-1}$, so $\tfrac{1}{\sqrt T} = D^{1/2}/\sqrt{T_0}$. Substitute: $$\dot m = \underbrace{p_0\,D^{-\frac{\gamma}{\gamma-1}}}_{p}\;A\;M\;\sqrt{\frac{\gamma}{R}}\;\underbrace{\frac{D^{1/2}}{\sqrt{T_0}}}_{1/\sqrt T} = A\,p_0\,M\sqrt{\frac{\gamma}{R T_0}}\;\; D^{\,-\frac{\gamma}{\gamma-1}+\frac12}$$ **WHY the final exponent.** Add the two powers of $D$ we just collected — one from pressure, one from the $1/\sqrt T$ factor: $$-\frac{\gamma}{\gamma-1}+\frac12 = \frac{-2\gamma+(\gamma-1)}{2(\gamma-1)} = \frac{-(\gamma+1)}{2(\gamma-1)}.$$ That negative power on top is the same as a positive power in the denominator, which is exactly the exponent in the box: $$\boxed{\;\dot m(M) = \frac{A\,p_0\,M\sqrt{\dfrac{\gamma}{R T_0}}}{\left(1+\dfrac{\gamma-1}{2}M^2\right)^{\frac{\gamma+1}{2(\gamma-1)}}}\;}$$ Every symbol on the right is either fixed ($\gamma, R$), controllable ($A, p_0, T_0$), or the dial we will now turn ($M$). --- ## Step 6 — Turn the dial: which $M$ pushes the most mass? **WHAT.** Hold the area $A$ fixed and ask: as we increase $M$ from $0$ upward, does $\dot m$ keep rising forever? **WHY.** The numerator $\propto M$ *grows* the flow (faster gas). But the denominator also grows (the gas thins out as it speeds up). Growth on top, growth on bottom — there must be a sweet spot where they balance and $\dot m$ per unit area is *largest*. **PICTURE.** The figure plots the *flux per unit area*, $\dot m/A$, against $M$. It rises, peaks, and falls. The peak sits exactly at $M=1$ (marked with a chalk-pink dot). ![[deepdives/dd-physics-3.3.07-d2-s06.png]] > [!intuition] Why the peak is exactly at $M=1$ > Below $M=1$: speeding the gas up wins over the thinning — flux rises. Above $M=1$: the thinning wins — flux falls. The exact crossover, where the two effects cancel, is the sonic point $M=1$. This is not a coincidence of the algebra; it is the defining property of a throat. So the maximum flow through a *fixed* area happens when the gas there is exactly sonic. A nozzle's narrowest point is built to sit at that peak. --- ## Step 7 — Choke it: set $M=1$ at the throat **WHAT.** At the throat we set $A = A^\ast$ and $M=1$. Substitute $M=1$ into the Step-5 boxed formula. **WHY.** Two reasons collide here. (1) Step 6: $M=1$ is the mass-flow peak, so this is the most a given throat can pass. (2) Step 3's picture: at $M=1$ the pressure pings are frozen — they *cannot travel upstream* past the throat. So the chamber literally cannot "hear" anything happening downstream. Lower the exit pressure all you like; the message never reaches the chamber, and $\dot m$ **freezes**. This freezing is ==choking==. **PICTURE.** The figure shows two pings: a subsonic one crawling upstream (chamber hears it) and a sonic one pinned at the throat (chamber deaf). ![[deepdives/dd-physics-3.3.07-d2-s07.png]] > [!formula] How low must the back-pressure be to *force* $M=1$? > Set $M=1$ in the Step-4 pressure ratio. The throat pressure needed for sonic flow is the ==critical pressure==: > $$\frac{p^\ast}{p_0} = \left(1+\frac{\gamma-1}{2}\cdot 1^2\right)^{-\frac{\gamma}{\gamma-1}} = \left(\frac{2}{\gamma+1}\right)^{\frac{\gamma}{\gamma-1}}$$ > The throat actually chokes ($M=1$) whenever the downstream (exit/back) pressure is low enough: > $$\boxed{\;\frac{p_{\text{exit}}}{p_0} \le \left(\frac{2}{\gamma+1}\right)^{\frac{\gamma}{\gamma-1}}\;}$$ > For $\gamma=1.2$ this threshold is about $0.56$; for $\gamma=1.4$, about $0.53$. Drop the back-pressure below roughly half the chamber pressure and the throat locks at sonic. Push it *above* that ratio and the throat stays subsonic — that is the "not-yet-choked" Case D of Step 8. This same critical ratio drives [[Choked Flow and Sonic Conditions]]. Putting $M=1$ makes the numerator's $M$ become $1$ and the bracket become $(1+\tfrac{\gamma-1}{2})^{\ldots} = (\tfrac{\gamma+1}{2})^{\ldots}$, which flips to give: > [!formula] Choked (maximum) mass flow rate > $$\dot m = \underbrace{A^\ast}_{\text{door width}}\;\underbrace{p_0}_{\text{how hard we push}}\;\underbrace{\sqrt{\frac{\gamma}{R\,T_0}}}_{\text{cooler = denser}}\;\underbrace{\left(\frac{2}{\gamma+1}\right)^{\frac{\gamma+1}{2(\gamma-1)}}}_{\text{fixed choking coefficient}}$$ > $\dot m \propto$ ==$A^\ast$== (linear), $\propto$ ==$p_0$== (linear), $\propto$ ==$1/\sqrt{T_0}$==. This connects onward to [[Choked Flow and Sonic Conditions]], and the same $A^\ast$ drives the downstream expansion in [[Nozzle Area Ratio and Expansion]]. --- ## Step 8 — Edge and degenerate cases (never leave a gap) **WHAT & WHY.** A formula is only trustworthy if it behaves sensibly at its extremes. Let us push each dial to its limit and check the picture still makes sense. ![[deepdives/dd-physics-3.3.07-d2-s08.png]] > [!example] Case A — throat shrinks to nothing, $A^\ast \to 0$ > $\dot m \propto A^\ast \to 0$. A closed pinch passes no gas. ✔ (The figure's leftmost bar.) > [!example] Case B — chamber pressure drops to zero, $p_0 \to 0$ > $\dot m \propto p_0 \to 0$. No push, no flow, no matter how hot. ✔ > [!example] Case C — chamber temperature soars, $T_0 \to \infty$ > $\dot m \propto 1/\sqrt{T_0} \to 0$. At fixed pressure, infinitely hot gas is infinitely thin ($\rho = p/RT \to 0$), so almost no *mass* crosses even though it screams out fast. Heat helps [[Specific Impulse|exhaust speed]], not raw kilograms. ✔ > [!example] Case D — not yet choked ($M<1$, subsonic throat) > The back-pressure sits *above* the critical ratio of Step 7, so $M<1$: we are *left* of the peak in Step 6, and lowering the back-pressure still raises $\dot m$. Only once $p_{\text{exit}}/p_0$ falls to $\left(\tfrac{2}{\gamma+1}\right)^{\gamma/(\gamma-1)}$ does $M$ hit $1$ and $\dot m$ lock. The choked formula is the *ceiling*, valid only after choking. ✔ > [!mistake] "The choking coefficient $\left(\tfrac{2}{\gamma+1}\right)^{\frac{\gamma+1}{2(\gamma-1)}}$ changes with pressure." > **Fix:** it depends **only** on $\gamma$ — a pure property of the gas. For $\gamma=1.2$ it is $\approx 0.585$; for $\gamma=1.4$ it is $\approx 0.578$. Chamber pressure and temperature live in the *other* factors. --- ## The one-picture summary ![[deepdives/dd-physics-3.3.07-d2-s09.png]] The whole derivation on one board — each arrow is tagged with its step number so you can trace it back: **S1** the honest count $\dot m = \rho A v$, **S2** continuity fixes it constant, **S3** the gas law and sound speed, **S4** translate to chamber conditions (isentropic), **S5** assemble $\dot m(M)$, **S6** find the flux peak at $M=1$, **S7** freeze it there to get the boxed choked-flow law, **S8** the edge cases all check out. Every arrow is a step above. > [!recall]- Feynman retelling — the walkthrough in plain words > I wanted to know how many kilograms of hot gas leave the engine each second. I promised myself the flow would be tidy: steady (not changing in time), one-dimensional (same across each slice), frictionless, and with no heat leaking in or out — that tidiness is what makes all the neat formulas true. So I drew a line across the pipe and counted the little slug crossing it: that gave $\dot m = \rho\,A\,v$ — how packed, how wide, how fast. Since no gas vanishes, the *same* count crosses every line, so where the pipe pinches, the gas must rush. To turn "rush" into numbers I borrowed two facts: the ideal-gas law (pressure and temperature tell me the packing) and the speed of sound (so I can say how fast *compared to sound*, the Mach number). I could only measure the calm chamber, so I used energy conservation — heating one kilogram by one kelvin costs $c_p$ joules — to get the temperature ratio, and the constant-entropy rule $p\propto T^{\gamma/(\gamma-1)}$ to get the pressure ratio. Plugging both in, the powers of that shared bracket added up to exactly $-(\gamma+1)/[2(\gamma-1)]$ — that is where the funny exponent comes from. Then I asked: if I keep the pinch the same width and speed the gas up, does more mass always get through? No — speeding it up thins it out, and the two effects tie exactly at the speed of sound. So the pinch pushes the most mass when the gas there is exactly sonic. And at that speed, pressure whispers can't crawl back upstream, so the chamber goes deaf to the outside — but only if the outside pressure has dropped below about half the chamber pressure; that is the critical ratio that *forces* the throat to lock. Freeze the formula there and I get the boxed law: mass flow grows with pinch width and push pressure, and shrinks with the square root of temperature. That's the whole story, and it also underlies the [[Thrust Equation and Effective Exhaust Velocity|thrust equation]] and [[Tsiolkovsky Rocket Equation|Tsiolkovsky's rocket equation]]. --- > [!recall]- > Which four assumptions underlie every relation in this walkthrough? ::: Steady, one-dimensional, frictionless, and adiabatic/no-shocks — together making the flow isentropic. > Which assumption specifically justifies the pressure ratio $p_0/p=(1+\tfrac{\gamma-1}{2}M^2)^{\gamma/(\gamma-1)}$? ::: Isentropic (constant entropy), via $p\propto T^{\gamma/(\gamma-1)}$; friction or a shock would break it. > What is $c_p$ and its units? ::: Specific heat at constant pressure — joules to raise one kilogram by one kelvin at fixed pressure, in $\text{J/kg·K}$; for an ideal gas $c_p=\gamma R/(\gamma-1)$. > Where does the exponent $-(\gamma+1)/[2(\gamma-1)]$ come from? ::: Adding the bracket's powers from the pressure factor ($-\gamma/(\gamma-1)$) and the $1/\sqrt T$ factor ($+\tfrac12$). > Which factor in the choked formula depends only on the gas, not on the engine settings? ::: The choking coefficient $\left(\tfrac{2}{\gamma+1}\right)^{\frac{\gamma+1}{2(\gamma-1)}}$ — a pure function of $\gamma$. > How low must the back-pressure be to force the throat to $M=1$? ::: $p_{\text{exit}}/p_0 \le \left(\tfrac{2}{\gamma+1}\right)^{\gamma/(\gamma-1)}$ (about $0.5$–$0.56$). > Why does $\dot m$ freeze once the throat is sonic? ::: At $M=1$ pressure signals can't travel upstream, so the chamber never hears the lower back-pressure. > At which Mach number is the flux per unit area maximum? ::: $M=1$, the throat/sonic point. > As $T_0 \to \infty$ at fixed $p_0$, what happens to $\dot m$ and why? ::: It $\to 0$; the gas becomes infinitely thin ($\rho = p/RT \to 0$).