3.3.13Rocket Propulsion

Optimum expansion — P_e = P_a for maximum thrust

3,079 words14 min readdifficulty · medium

Overview

The optimum expansion condition for a rocket nozzle occurs when the exit pressure equals the ambient pressure (Pe=PaP_e = P_a). This maximizes thrust by ensuring all pressure energy converts to kinetic energy with no pressure imbalance at the exit plane.

For rockets: Thrust = (momentum flux) + (pressure force). If PePaP_e \neq P_a, the pressure term adds or subtracts from ideal momentum transfer. When Pe=PaP_e = P_a, that pressure term vanishes, and you get maximum efficient conversion of combustion energy → directed kinetic energy.

The Physics: Deriving Why P_e = P_a is Optimal

WHY this form? Thrust is the rate of momentum change of expelled gas PLUS the net pressure force acting on the exit plane. The gas pushes on the nozzle with force PeAeP_e A_e (outward), and the atmosphere pushes back with force PaAeP_a A_e (inward).

Derivation from First Principles

Step 1: Control Volume Analysis

Consider a control volume around the rocket nozzle. By Newton's second law for a control volume:

F=ddtCVvρdV+CSvρ(vn^)dA\sum \vec{F} = \frac{d}{dt}\int_{CV} \vec{v} \rho \, dV + \int_{CS} \vec{v} \rho (\vec{v} \cdot \hat{n}) \, dA

For steady flow, the first term vanishes. At the exit plane, exhaust leaves with velocity vev_e:

Fthrust=m˙ve+(pressure forces)F_{thrust} = \dot{m} v_e + \text{(pressure forces)}

WHY the pressure term? The gas at the exit plane has pressure PeP_e. If this doesn't match PaP_a, there's a NET pressure force on the imaginary exit plane:

  • Gas pushes on the plane with force PeAeP_e A_e (from inside)
  • Atmosphere pushes with force PaAeP_a A_e (from outside)
  • Net: (PePa)Ae(P_e - P_a) A_e

Step 2: The CORRECT optimization — geometry is the design variable

Here is the crucial subtlety. We CANNOT treat PeP_e as a free independent variable and set dFdPe=0\frac{dF}{dP_e} = 0 with everything else fixed. Why? Because PeP_e, vev_e, and AeA_e are not independent — they are all locked together by the nozzle geometry (the area ratio Ae/AA_e/A^*) through the isentropic flow relations.

What is actually free to choose? The nozzle geometry, i.e. the expansion (area) ratio ε=Ae/A\varepsilon = A_e/A^*. Once you fix ε\varepsilon and the chamber conditions, then PeP_e, vev_e, and AeA_e are ALL determined. So the real optimization is: choose ε\varepsilon to maximize FF.

For fixed chamber conditions (P0P_0, T0T_0), m˙\dot{m} is fixed by the throat (which is choked). The exit velocity is:

ve=2γγ1RT0[1(PeP0)γ1γ]v_e = \sqrt{\frac{2\gamma}{\gamma - 1} R T_0 \left[1 - \left(\frac{P_e}{P_0}\right)^{\frac{\gamma-1}{\gamma}}\right]}

WHY this equation? It comes from isentropic expansion (conservation of energy + entropy). The term in brackets is the fraction of thermal energy converted to kinetic energy.

Step 3: Differentiate with respect to the design variable PeP_e (along the isentrope)

As we open up the nozzle (increase ε\varepsilon), the exit pressure PeP_e drops. So we can parametrize the design by PeP_e itself, but crucially AeA_e must vary along with it according to the mass-conservation relation m˙=ρeveAe\dot{m} = \rho_e v_e A_e. Differentiating the thrust,

dFdPe=m˙dvedPe+ddPe[(PePa)Ae]=0.\frac{dF}{dP_e} = \dot{m}\frac{dv_e}{dP_e} + \frac{d}{dP_e}\big[(P_e - P_a)A_e\big] = 0.

To evaluate this we use two exact facts along the isentrope:

  1. Euler's momentum relation for steady 1-D flow: ρevedve=dPe\rho_e v_e\, dv_e = -\,dP_e, i.e. m˙dve=AedPe\dot{m}\,dv_e = -A_e\, dP_e (using m˙=ρeveAe\dot{m} = \rho_e v_e A_e).
  2. The pressure-force term differentiates by the product rule: d[(PePa)Ae]=AedPe+(PePa)dAed[(P_e-P_a)A_e] = A_e\,dP_e + (P_e-P_a)\,dA_e.

Substituting fact 1 into the first term:

m˙dvedPe=Ae.\dot{m}\frac{dv_e}{dP_e} = -A_e.

So

dFdPe=Ae+Ae+(PePa)dAedPe=(PePa)dAedPe.\frac{dF}{dP_e} = -A_e + A_e + (P_e - P_a)\frac{dA_e}{dP_e} = (P_e - P_a)\frac{dA_e}{dP_e}.

WHY this matters: The momentum-thrust change (AedPe-A_e\,dP_e) exactly cancels the "AedPeA_e\,dP_e" piece of the pressure term. What SURVIVES is (PePa)dAe(P_e - P_a)\,dA_e. Since dAedPe0\dfrac{dA_e}{dP_e}\neq 0 (opening the nozzle changes area), the only way to get dFdPe=0\frac{dF}{dP_e}=0 is:

At this condition:

  • Pressure thrust term: (PePa)Ae=0(P_e - P_a)A_e = 0
  • ALL thrust comes from momentum: F=m˙veF = \dot{m} v_e (maximum efficiency)
  • Nozzle is "perfectly expanded" or "optimally expanded"

The deeper insight: If Pe>PaP_e > P_a (under-expanded), the exhaust can still expand outside the nozzle, gaining more velocity, but this happens OUTSIDE where it doesn't push against the nozzle walls → wasted potential. If Pe<PaP_e < P_a (over-expanded), atmospheric pressure compresses the exhaust, creating flow separation and back-pressure that REDUCES net thrust. Only Pe=PaP_e = P_a sits at the peak.

Three Expansion Regimes

  1. Optimally expanded (Pe=PaP_e = P_a): Perfect match. Clean exit, no external expansion/compression. Maximum thrust.

  2. Over-expanded (Pe<PaP_e < P_a): Nozzle too long. Ambient pressure exceeds exit pressure, causing flow separation inside nozzle. NEGATIVE pressure thrust, significant losses — you've gone past the peak.

WHY can't we always operate at optimum? Rockets fly through varying altitudes. PaP_a decreases with altitude but nozzle geometry is fixed. A nozzle optimized for sea level is under-expanded in vacuum; one optimized for vacuum is over-expanded at sea level.

Worked Examples

Calculate thrust at: (a) Vacuum: Pa=0P_a = 0 (b) Sea level: Pa=1.01P_a = 1.01 bar

Solution:

(a) Vacuum (Pa=0P_a = 0): Fvac=m˙ve+(Pe0)AeF_{vac} = \dot{m} v_e + (P_e - 0) A_e Fvac=(300)(2800)+(0.1×1050)(2.0)F_{vac} = (300)(2800) + (0.1 \times 10^5 - 0)(2.0) Fvac=840,000+20,000=860,000 N=860 kNF_{vac} = 840{,}000 + 20{,}000 = 860{,}000 \text{ N} = 860 \text{ kN}

WHY the positive pressure term? In vacuum, Pe>PaP_e > P_a means the exhaust pressure actively pushes against the nozzle exit rim, adding thrust.

(b) Sea level (Pa=1.01P_a = 1.01 bar): FSL=m˙ve+(PePa)AeF_{SL} = \dot{m} v_e + (P_e - P_a) A_e FSL=(300)(2800)+(0.1×1051.01×105)(2.0)F_{SL} = (300)(2800) + (0.1 \times 10^5 - 1.01 \times 10^5)(2.0) FSL=840,000+(91,000×2.0)F_{SL} = 840{,}000 + (-91{,}000 \times 2.0) FSL=840,000+(182,000)=658,000 N=658 kNF_{SL} = 840{,}000 + (-182{,}000) = 658{,}000 \text{ N} = 658 \text{ kN}

WHY negative pressure term? The atmosphere pushes back harder than the exhaust pushes out. This nozzle is OVER-EXPANDED at sea level.

Percentage loss: 860658860×100%=202860×100%23.5%\frac{860 - 658}{860} \times 100\% = \frac{202}{860}\times 100\% \approx 23.5\%

The engine loses about 23.5% thrust at sea level compared to vacuum!

Key insight: Vacuum-optimized nozzles (high expansion ratios, low PeP_e) perform poorly at sea level. This is why first-stage engines use moderate expansion ratios.

Find the expansion ratio ε=Ae/At\varepsilon = A_e/A_t for optimum thrust.

Solution:

Step 1: For optimum, Pe=Pa=0.2P_e = P_a = 0.2 bar.

Step 2: Pressure ratio: PeP0=0.230=0.00667\frac{P_e}{P_0} = \frac{0.2}{30} = 0.00667

WHY use this ratio? The expansion from chamber to exit is isentropic, so pressure and area are related through isentropic flow relations.

Step 3: Find exit Mach number from the isentropic pressure relation: PeP0=(1+γ12Me2)γγ1\frac{P_e}{P_0} = \left(1 + \frac{\gamma-1}{2}M_e^2\right)^{-\frac{\gamma}{\gamma-1}}

0.00667=(1+0.252Me2)1.250.25=(1+0.125Me2)50.00667 = \left(1 + \frac{0.25}{2}M_e^2\right)^{-\frac{1.25}{0.25}} = (1 + 0.125\, M_e^2)^{-5}

WHY this equation? Isentropic relation: as gas accelerates and expands, pressure drops according to this power law from energy and entropy conservation.

Raise both sides to the power 1/5-1/5: (0.00667)0.2=1+0.125Me2(0.00667)^{-0.2} = 1 + 0.125\, M_e^2 2.72=1+0.125Me22.72 = 1 + 0.125\, M_e^2 Me2=2.7210.125=1.720.125=13.8M_e^2 = \frac{2.72 - 1}{0.125} = \frac{1.72}{0.125} = 13.8 Me=3.71M_e = 3.71

Step 4: Now find ε\varepsilon from the area-Mach relation: ε=AeA=1Me[2γ+1(1+γ12Me2)]γ+12(γ1)\varepsilon = \frac{A_e}{A^*} = \frac{1}{M_e} \left[\frac{2}{\gamma+1}\left(1 + \frac{\gamma-1}{2}M_e^2\right)\right]^{\frac{\gamma+1}{2(\gamma-1)}}

ε=13.71[22.25(1+0.125×13.8)]2.252×0.25\varepsilon = \frac{1}{3.71} \left[\frac{2}{2.25}\left(1 + 0.125 \times 13.8\right)\right]^{\frac{2.25}{2 \times 0.25}} ε=13.71[0.889×2.72]4.5\varepsilon = \frac{1}{3.71} \left[0.889 \times 2.72\right]^{4.5} ε=13.71×(2.42)4.5=13.71×15842.7\varepsilon = \frac{1}{3.71} \times (2.42)^{4.5} = \frac{1}{3.71} \times 158 \approx 42.7

Answer: ε42.7\varepsilon \approx 42.7 for optimum expansion at this altitude.

WHY such large area ratio? High expansion ratios (large exit area) are needed to drop pressure from 30 bar to 0.2 bar. Vacuum engines need even larger ratios (80-400).

Common Misconceptions

Why this feels right: More expansion → more vev_e → more thrust. Simple!

The steel-man: This IS true in vacuum, where Pa=0P_a = 0. There's no penalty for over-expanding. Space Shuttle main engines used ε77\varepsilon \approx 77 for this reason.

Why it's wrong (on Earth): At sea level, over-expansion causes flow separation. The exhaust can't push against ambient pressure, creates turbulent recirculation zones, and the nozzle walls experience back-pressure. The negative (PePa)Ae(P_e - P_a)A_e term overwhelms gains in vev_e.

The fix: Optimize expansion ratio for your operating altitude. Dual-bell nozzles or altitude-compensating nozzles solve this by changing effective geometry with altitude.

Why this feels right: It looks like standard calculus optimization — treat PeP_e as the variable, differentiate, set to zero.

The steel-man: The instinct to differentiate is correct; optimization DOES require calculus here.

Why it's wrong: PeP_e, vev_e, and AeA_e are NOT independent — they are all tied together by the nozzle geometry ε=Ae/A\varepsilon = A_e/A^* through isentropic flow. If you hold AeA_e fixed while varying PeP_e, you're describing a physically impossible nozzle. The correct free variable is the geometry ε\varepsilon; AeA_e must vary with PeP_e via m˙=ρeveAe\dot m = \rho_e v_e A_e.

The fix: Let AeA_e vary along the isentrope. Then dFdPe=(PePa)dAedPe\frac{dF}{dP_e} = (P_e - P_a)\frac{dA_e}{dP_e}, which vanishes only at Pe=PaP_e = P_a.

Why this feels right: Intro courses often just show F=m˙veF = \dot{m} v_e, ignoring pressure.

The steel-man: For well-designed nozzles near their design altitude, (PePa)0(P_e - P_a) \approx 0, so the simplification is valid.

Why it's wrong: At off-design conditions, pressure thrust can be over 20% of total thrust (see Example 1). For upper-stage engines transitioning from atmosphere to vacuum, ignoring it means large errors in trajectory calculations.

The fix: Always use the complete thrust equation when analyzing performance across altitudes.

Connections

  • Thrust equation derivation - where (PePa)Ae(P_e - P_a)A_e comes from
  • Isentropic flow relations - how PeP_e, vev_e, and AeA_e relate
  • Specific impulse - Isp=F/(m˙g0)I_{sp} = F/(\dot{m} g_0) maximized when Pe=PaP_e = P_a
  • Nozzle expansion ratio - geometric parameter that determines PeP_e
  • Altitude compensation - aerospike and dual-bell nozzles that adapt to varying PaP_a
  • Rocket staging - why different stages use different expansion ratios
  • Supersonic flow separation - what happens when over-expanded
Recall

Explain this to a 12-year-old Imagine you're blowing up a balloon and then letting it go. The balloon flies around because air rushes out the hole, pushing the balloon forward. Now, here's the trick: how big should that hole be?

If the hole is too small (like pinching it), air doesn't come out very fast, but there's lots of pressure pushing. If the hole is too big, air comes out fast, but there's no pressure pushing anymore.

For a rocket, it's the same! The rocket engine's "hole" (the nozzle) needs to be just the right size so that when the hot gas comes out, its pressure matches the air pressure outside. If they match perfectly, all the energy goes into making the gas shoot out super fast (which pushes the rocket), instead of wasting energy on a pressure fight between the inside and outside.

When the rocket is in space (no air pressure outside), it wants a HUGE nozzle so the gas can spread out and go as fast as possible. But on the ground (with air pressure pushing back), a huge nozzle is bad—the air would push back and actually HURT the rocket's power. That's why the Space Shuttle had different engines for liftoff and for space!

Alternative: "Exit = Air, Thrust is Fair"


Flashcards

#flashcards/physics

What is the complete thrust equation for a rocket nozzle?
F=m˙ve+(PePa)AeF = \dot{m} v_e + (P_e - P_a) A_e, where the first term is momentum thrust and the second is pressure thrust.
What is the optimum expansion condition for maximum thrust?
Pe=PaP_e = P_a, where exit pressure equals ambient pressure, eliminating pressure thrust losses and maximizing momentum conversion efficiency.
Why is dFdPe=0\frac{dF}{dP_e}=0 with AeA_e held fixed the WRONG way to optimize?
Because PeP_e, vev_e, and AeA_e are not independent — they are coupled by nozzle geometry ε=Ae/A\varepsilon = A_e/A^*. The correct free variable is ε\varepsilon, with AeA_e varying along the isentrope.
What does dFdPe\frac{dF}{dP_e} correctly reduce to, and why does it give Pe=PaP_e=P_a?
Using Euler's relation m˙dve=AedPe\dot m\,dv_e=-A_e\,dP_e, the momentum change cancels part of the pressure term, leaving dFdPe=(PePa)dAedPe\frac{dF}{dP_e}=(P_e-P_a)\frac{dA_e}{dP_e}. Since dAedPe0\frac{dA_e}{dP_e}\neq 0, this vanishes only when Pe=PaP_e=P_a.
Why does the pressure term (PePa)Ae(P_e - P_a)A_e appear in the thrust equation?
The exhaust gas at exit has pressure PeP_e pushing outward with force PeAeP_e A_e, while the atmosphere pushes inward with PaAeP_a A_e. The NET pressure force is (PePa)Ae(P_e - P_a)A_e.
What happens in an under-expanded nozzle (Pe>PaP_e > P_a)?
The exhaust is still at higher pressure than ambient, so it continues expanding outside the nozzle (diamond shock patterns). Positive pressure thrust but missed opportunity for more vev_e inside.
What happens in an over-expanded nozzle (Pe<PaP_e < P_a)?
Ambient pressure exceeds exit pressure, causing flow separation inside the nozzle, turbulent recirculation, and NEGATIVE pressure thrust that reduces total thrust.
A vacuum-optimized nozzle has Pe=0.1P_e = 0.1 bar. What happens at sea level (Pa=1.01P_a = 1.01 bar)?
It becomes severely over-expanded. Pressure term (0.11.01)×105×Ae(0.1 - 1.01)\times 10^5 \times A_e is large and NEGATIVE — in Example 1 it cuts thrust from 860 kN to 658 kN, a ~23.5% loss.
How do multi-stage rockets handle the Pe=PaP_e = P_a optimization challenge?
Lower stages use moderate expansion ratios (ε = 10-40) for sea-level operation; upper stages use high ratios (ε = 50-400) optimized for vacuum where Pa0P_a \approx 0.
Why is maximum vev_e not the same as maximum thrust?
Maximum vev_e needs maximum expansion (lowest PeP_e), but if Pe<PaP_e < P_a the negative pressure term reduces total thrust. Thrust peaks when the COMBINATION of momentum and pressure terms is optimal, i.e. Pe=PaP_e = P_a.

Concept Map

determines via isentropic flow

feed into

momentum term

pressure term

compared to

makes zero

maximizes

adds to

derives

fixes via throat

sets

optimized to satisfy

Nozzle geometry: expansion ratio ε

P_e, v_e, A_e locked together

Thrust equation F

m_dot times v_e

P_e minus P_a times A_e

Ambient pressure P_a

Optimum condition P_e = P_a

Thrust F

Control volume + Newton's 2nd law

Fixed chamber P0, T0

m_dot mass flow rate

Hinglish (regional understanding)

Intuition Hinglish mein samjho

Dekho beta, is note ka core idea bahut simple hai — rocket ka nozzle tab sabse zyada thrust deta hai jab exit pe gas ka pressure (PeP_e) aur bahar ke atmosphere ka pressure (PaP_a) barabar ho jaaye. Water balloon wala example yaad rakho: agar andar ka pressure aur bahar ka pressure match nahi karte, toh energy ka kuch hissa "pressure ki ladai" mein waste ho jaata hai, aur poori energy paani ki speed banane mein nahi lagti. Rocket mein bhi thrust ka formula hai F=m˙ve+(PePa)AeF = \dot{m}v_e + (P_e - P_a)A_e — jab Pe=PaP_e = P_a hota hai, toh woh pressure wala term zero ho jaata hai aur saari combustion energy directly exhaust ki kinetic energy (yaani speed) mein convert ho jaati hai. Isko hi optimum expansion condition kehte hain.

Ab ek important baat jo yeh note highlight karta hai — hum PeP_e ko seedha ek free variable maan ke dFdPe=0\frac{dF}{dP_e}=0 nahi kar sakte, kyunki PeP_e, vev_e aur AeA_e teeno aapas mein jude hue hain nozzle ki geometry se (area ratio ε=Ae/A\varepsilon = A_e/A^* se). Matlab real mein jo cheez hamare haath mein hai woh nozzle ka shape hai — kitna phailao (expansion ratio). Jaise-jaise nozzle ko zyada khola jaata hai, exit pressure girta jaata hai. Toh optimization ka asli sawaal yeh hai: nozzle ki geometry kaise design karein taaki thrust maximum aaye. Jab hum sahi tareeke se differentiate karte hain (Euler's momentum relation aur mass conservation use karke), tab jawaab nikalta hai ki thrust tabhi peak karta hai jab Pe=PaP_e = P_a.

Yeh baat kyun matter karti hai? Kyunki real rockets alag-alag altitude pe udte hain — sea level pe PaP_a zyada hota hai, aur upar space ki taraf PaP_a kam hota hai. Isliye ek hi nozzle har jagah perfect nahi ho sakta; engineers nozzle design karte time yeh decide karte hain ki kis altitude pe optimum expansion chahiye. Agar Pe>PaP_e > P_a ho toh "under-expanded" aur Pe<PaP_e < P_a ho toh "over-expanded" nozzle kehte hain — dono cases mein thodi efficiency loss hoti hai. Toh yeh concept samajhna zaroori hai kyunki yahi rocket engine design ka backbone hai, aur exam mein bhi thrust equation aur uska pressure term bahut poocha jaata hai.

Go deeper — visual, from zero

Test yourself — Rocket Propulsion

Connections