Propulsion (Turbojets, Turbofans, Rockets) — Engineering Reference

See also (Tier 3 family index): Jet Engine Types

1. At a glance

Propulsion is the engineering of devices that produce thrust — the force that accelerates a vehicle against drag, gravity, and its own inertia. The discipline divides cleanly on the question of where the working fluid comes from:

• Air-breathing propulsion ingests atmospheric air, adds energy to it (combustion, electric heating, or shaft work), and exhausts it at higher velocity. Family: turbojet, turbofan (low- and high-bypass, geared), turboprop, turboshaft, piston-prop, ramjet, scramjet, pulsejet, electric propeller, and combined cycles (TBCC, RBCC).
• Non-air-breathing propulsion carries both fuel and oxidiser (or a reaction-mass alone). Family: chemical rockets (liquid bipropellant, solid, hybrid, monopropellant), electric propulsion (Hall-effect, gridded ion, MPD, arcjet, resistojet, pulsed plasma), nuclear thermal, nuclear electric, solar sails, and exotic concepts (laser-pumped, fusion, antimatter).

Propulsion is half of every airplane — by mass, by acquisition cost, by maintenance burden, and by a large fraction of the airframer’s engineering effort — and it is essentially all of every rocket: an Apollo Saturn V was 92 % propellant by mass at lift-off, and a Falcon 9 first stage is ≈ 96 %. Every kilogram of payload to LEO that costs ≈ $2 000 in 2026 (Falcon 9 commercial price, partially reusable) is a direct consequence of how good — or how bad — the engine is. The metric that controls everything for a rocket is specific impulse I_sp; for an air-breather it is TSFC (thrust specific fuel consumption) or PSFC (power specific, for shaft engines). All other figures of merit (T/W, NOₓ, noise, cost-per-flight) follow.

Where propulsion sits in the design stack: it consumes thermodynamics (Brayton cycle, rocket combustion, real-gas effects), fluid-mechanics (compressible inlet and nozzle flow, mixing), heat-transfer (turbine blade cooling, regen-cooled rocket chambers, ablative liners), pumps-turbomachinery (multi-stage axial compressor, centrifugal turbopump), materials-ceramics (single-crystal Ni superalloys, thermal-barrier coatings, CMC), and aerodynamics for installed performance (inlet–airframe interaction, nacelle drag, intake distortion). It feeds vehicle structural-dynamics (vibration, gimbal loads, POGO) and ultimately every guidance and control loop.

2. Why it matters

Three economic and physical facts make propulsion the dominant variable:

1. The rocket equation has no escape. Δv = I_sp · g₀ · ln(m₀/m_f) (Tsiolkovsky 1903). Going to LEO requires ≈ 9.4 km/s of Δv after gravity and drag losses; reaching escape velocity from LEO another ≈ 3.2 km/s; landing on Mars after that another 4–6 km/s. Each ln term multiplies propellant mass; small improvements in I_sp produce exponential savings in vehicle gross mass. A 30 s gain in I_sp on a single-stage cryogenic upper stage cuts vehicle dry mass by ≈ 15 %.

2. TSFC sets airline economics. Fuel is 25–35 % of a major airline’s operating cost. A 1 % cut in cruise TSFC on a Boeing 787 saves ≈ 250 000 in fuel per aircraft per year (2026 prices). That is why GE, Pratt & Whitney, and Rolls-Royce each spend ≈ 1 B/yr on engine R&D, and why the CFM RISE open-fan programme (target 20 % cut vs. LEAP) and the Pratt PW1100G geared turbofan (delivered ≈ 16 % cut vs. CFM56) are existential bets.

3. Without affordable propulsion there is no aviation, no space access, no exploration. Every constraint on commercial air travel — range, payload, ticket price, contrail-driven climate impact, NOₓ near airports — and every constraint on space activity — launch cadence, payload mass, cost per kilogram, deep-space mission duration — traces back to the engine. The propulsion budget is the budget.

3. First principles

3.1 Thrust equation

For a generalised reaction engine, the uninstalled net thrust is:

F  =  ṁ_e · V_e  −  ṁ_0 · V_∞  +  (P_e − P_∞) · A_e

with subscripts 0 = freestream (intake), e = nozzle exit. For a rocket, ṁ_0 = 0 and V_∞ does not appear — only mass flow out matters. The first two terms are momentum thrust; the third is pressure thrust, which vanishes when the nozzle is perfectly expanded (P_e = P_∞) and goes negative when over-expanded (a real concern at sea-level for an upper-stage bell nozzle).

For an air-breather where ṁ_e ≈ ṁ_0 + ṁ_f and ṁ_f ≪ ṁ_0, the thrust simplifies to:

F  ≈  ṁ_0 · (V_e − V_∞)  +  (P_e − P_∞) · A_e

3.2 Performance metrics

Specific impulse (the universal rocket metric):

I_sp  =  F / (ṁ_p · g₀)    [s]       where g₀ = 9.80665 m/s² by definition

Numerically, I_sp in seconds equals the effective exhaust velocity V_eq in m/s divided by 9.80665. So an LH2/LOX engine with I_sp = 452 s has V_eq = 4 432 m/s. The choice of g₀ (not local g) makes I_sp a property of the propellant + engine, not the launch site.

Thrust specific fuel consumption (TSFC, for air-breathers):

TSFC  =  ṁ_f / F     [kg/(N·h)  or  lb/(lbf·h)]

Modern high-bypass turbofan at cruise (M = 0.78, 11 km alt): TSFC ≈ 0.55 lb/(lbf·h) = 1.56 × 10⁻⁵ kg/(N·s). Older low-bypass like JT8D: 0.8 lb/(lbf·h). A turbojet (Olympus 593, Concorde): 1.2 lb/(lbf·h) at supersonic cruise.

Power specific fuel consumption (PSFC, for turboprops/shaft engines): kg/(kW·h) or lb/(hp·h). PT6A turboprop: ≈ 0.50 lb/(hp·h).

Propulsive efficiency (Froude):

η_p  =  2 · V_∞ / (V_e + V_∞)

Peaks at unity when V_e = V_∞ — but then thrust = 0. Practical engines trade thrust against η_p. A high-bypass turbofan moves a large air mass at modest ΔV (high η_p ≈ 0.75 at cruise); a turbojet moves a small mass at huge ΔV (η_p ≈ 0.50 at cruise).

Thermal efficiency:

η_t  =  (V_e² − V_∞²) / (2 · q_in)        q_in ≡ heat added per unit mass of working fluid

Overall efficiency η = η_p · η_t. Modern airliner engine cruise: η_p ≈ 0.78, η_t ≈ 0.50, η_overall ≈ 0.39. Best automobile diesel: ≈ 0.45 overall — but in a stationary application. The fact that aircraft engines reach ≈ 40 % overall while pushing through atmosphere at 250 m/s is remarkable.

3.3 Tsiolkovsky rocket equation

For a vehicle expending propellant at constant exhaust velocity V_eq with no external forces (gravity, drag) acting:

Δv  =  V_eq · ln(m₀ / m_f)  =  I_sp · g₀ · ln(m₀ / m_f)

Re-arranged: propellant mass fraction = 1 − exp(−Δv / V_eq). For Δv = 9.4 km/s to LEO and V_eq = 3 050 m/s (LOX/RP-1):

m_p/m₀ = 1 − exp(−9.4/3.05) = 1 − 0.0457 = 0.954 → 95.4 % propellant by mass.

This is why staging exists: no single tank-engine combination can carry 95 % of its own mass as fuel and still hold structural margin.

4. Reference data — engine family taxonomy

Family	Working fluid	V_e [m/s]	I_sp [s, vac]	Typical TSFC [lb/(lbf·h)]	Mach regime	Example
Turbojet	Air + combustion	600 – 1 000	—	0.8 – 1.2	0 – 3	J85, Olympus 593
Low-bypass turbofan	Air + combustion (BPR 0.3 – 2)	400 – 700	—	0.7 – 0.9	0 – 2.5	F100, F119, F135
High-bypass turbofan	Air + combustion (BPR 5 – 12)	250 – 400	—	0.55 – 0.65	0 – 0.95	CFM56, GEnx, Trent XWB
Geared turbofan	Air (BPR 12 – 17)	230 – 350	—	0.50 – 0.58	0 – 0.85	PW1100G, RR UltraFan
Turboprop	Air via propeller	100 – 200	—	0.45 – 0.55 (TSFC equiv.)	0 – 0.7	PT6A, TP400, T56
Turboshaft	Air, output to shaft	—	—	0.45 – 0.55	—	T700, CT7, GE38
Ramjet	Air + combustion	1 000 – 1 800	—	1.5 – 2.5	2 – 5	SA-6, ASMP-A
Scramjet	Air + supersonic combustion	2 000 – 3 000	—	2 – 4	5 – 15	X-43A, X-51
Solid rocket	AP + Al + HTPB	2 200 – 2 700	230 – 280	—	0 – ∞	SLS SRB, Antares Castor 30XL
Storable bipropellant	N₂O₄ / MMH or UDMH	2 800 – 3 200	280 – 330	—	0 – ∞	Soyuz upper, Aestus, Apollo SPS
LOX/RP-1	Kerosene	2 900 – 3 350	300 – 340	—	0 – ∞	Merlin 1D, RD-180, F-1
LOX/CH4	Methane	3 250 – 3 700	330 – 380	—	0 – ∞	Raptor, BE-4, Vulcain (LH2 variant)
LOX/LH2	Hydrogen	4 200 – 4 460	440 – 465	—	0 – ∞	RS-25, Vulcain 2, RL10, BE-3U
Hall-effect (EP)	Xenon / krypton	15 000 – 25 000	1 500 – 2 500	—	(space only)	SPT-100, BHT-1500, X3
Gridded ion (EP)	Xenon	30 000 – 50 000	3 000 – 5 000	—	(space only)	NSTAR, NEXT-C, BepiColombo T6
Nuclear thermal	H₂ heated by reactor	7 800 – 9 000	800 – 920	—	(in-space)	NERVA (1968 ground test), DRACO

I_sp_vac column omitted for air-breathers (it does not capture the physics — they breathe atmosphere).

5p. The Brayton cycle for jet engines

A jet engine is a Brayton cycle (see thermodynamics) with the work output dumped into accelerating the working fluid through a nozzle rather than extracted at a shaft. Station numbering (SAE AS755):

• 0 — freestream
• 1 — engine face
• 2 — fan / LP compressor inlet
• 2.5 — LP compressor exit / HP compressor inlet
• 3 — HP compressor exit (combustor inlet)
• 4 — combustor exit / HP turbine inlet (“TIT”)
• 4.5 — HP turbine exit / LP turbine inlet
• 5 — LP turbine exit
• 6 — jet pipe / mixer
• 8 — nozzle throat
• 9 — nozzle exit
• 18 — bypass duct exit (turbofan)

Each station carries a stagnation temperature T_t and pressure P_t. The cycle is laid out on a T-s diagram exactly like a stationary gas turbine, with two added subtleties: the inlet does free-stream-to-engine-face compression (M-dependent recovery η_d ≈ 0.97 subsonic, 0.85 – 0.95 supersonic with shocks); the nozzle converts thermal-to-kinetic energy at exit.

Modern parameter ranges:

• Overall pressure ratio (OPR = P_t3 / P_t2): 30 – 45 in current airliners; GE9X 60+; CFM RISE target 70+.
• Turbine inlet temperature (T_t4): 1 700 – 1 950 K for current cores; cooled single-crystal blades with TBC.
• Bypass ratio (BPR = ṁ_bypass / ṁ_core): 4 – 6 (legacy CFM56), 8 – 11 (LEAP-1A/B, Trent XWB), 12 – 14 (PW1100G), 16 – 19 (RR UltraFan demonstrator, CFM RISE).
• Fan pressure ratio (FPR): 1.6 (high-BPR conventional), 1.3 – 1.4 (geared / open-fan).

Cycle-analysis essentials (real-fluid, non-isentropic):

Compressor:   T_t3 / T_t2  =  1 + (1/η_c)·[(P_t3/P_t2)^((γ−1)/γ) − 1]
Combustor:    ṁ_f · LHV · η_b  =  (ṁ_a + ṁ_f) · c_p · (T_t4 − T_t3)
Turbine:      T_t5 / T_t4  =  1 − η_t · [1 − (P_t5/P_t4)^((γ−1)/γ)]
Nozzle:       V_e² / 2  =  c_p · T_t5 · [1 − (P_∞ / P_t5)^((γ−1)/γ)]

The closure constraint is shaft work balance: turbine power = compressor power + fan power + accessories. For a two-spool turbofan, two separate balances (LP shaft and HP shaft).

6p. Inlets, combustors, nozzles

Inlets (intake / diffuser):

• Subsonic pitot intake (modern airliner nacelle): pressure recovery η_d = P_t2/P_t0 = 0.97 – 0.99; diffusion in expanding duct, lip rounded to handle 30° α at static.
• Supersonic, fixed compression: cone (F-104 with movable spike, SR-71 Pratt J58 with translating spike). External-compression inlet for M ≤ 2; mixed-compression for M 2 – 3.
• Variable geometry: Concorde (movable ramps, auxiliary spill doors, η_d ≈ 0.92 at M 2), F-15 (rectangular two-ramp), F-22 (caret, fixed geometry but matched to M < 2.5).
• Hypersonic: scramjet inlets compress through oblique shocks; isolator section absorbs combustor back-pressure transients.

Combustors:

• Can (legacy, e.g. J47, Avon): individual cans around the shaft.
• Annular (modern, e.g. CFM56, V2500, LEAP, Trent): single ring chamber — best volume, weight, and pattern factor.
• Can-annular (transition, e.g. JT9D, F100): cans with shared casing.
• Zones: primary (rich, hot, stabilises flame), secondary (combustion completion), dilution (cool the gas to ≤ T_t4 set by turbine).
• NOₓ control: lean-premixed (LP / DLE for stationary GTs); RQL (rich-quench-lean, used in CFM56-7B Tech Insertion, V2500-A5); TAPS (twin annular premixing swirler, GE GEnx, GE9X — staged pilot + main); lean-direct injection (Pratt’s GTF combustor). Modern airliner cruise NOₓ ≈ 7 – 12 g/kg fuel.
• Materials: nickel sheet (Hastelloy X) on legacy combustors, Haynes 188 on hot rings, CMC (SiC/SiC) on GE9X combustor inner liner and on the LEAP HP turbine shroud — first commercial CMC hot-section parts.

Nozzles:

• Convergent (subsonic-discharge, all civil turbofans): exit M ≤ 1 for choked or unchoked; simple, light, fixed.
• Convergent-divergent (C-D): required for supersonic exit. F-22 (round, slightly squashed), B-2 (slot, low-observable), SR-71 (axisymmetric, ejector-augmented), all rockets.
• Variable area: military reheat (afterburner) engines need A_8 to open when AB lights; F119, F135, EJ200, M88.
• Thrust vectoring: paddle (X-31), axisymmetric (F-22 with 20° pitch), 3D (F-35B STOVL with three-bearing swivel duct and lift fan).
• Rocket nozzles: bell (TIC / Rao optimum truncated ideal contour), conical (simple but heavy and low η_n), aerospike / plug (altitude-compensating, never flown operationally — XRS-2200 cancelled with X-33), dual-bell (compromise sea-level + vacuum, demonstrated by DLR), expansion-deflection (compact, used on RL10A-4-2 derivative studies). Bell-nozzle expansion ratio ε = A_e/A_t typically 25 – 80 for booster engines, 80 – 280 for upper stages (RL10C-1 ε = 285, Vinci ε = 240). Nozzle efficiency η_n = 0.97 – 0.99 in vacuum, drops sharply if over-expanded at sea level.

Cycle / open-cycle / closed-cycle comparison

For air-breathers, an open Brayton cycle dumps exhaust to atmosphere; closed Brayton (nuclear electric, sCO₂ Brayton) recirculates the working fluid through a cooler. For rockets, the analogous distinction is whether turbine-drive gas is dumped (gas-generator, open) or re-injected into the main chamber (staged, closed) — see §9p below. Closed cycles always have higher I_sp but require higher Pc → higher turbopump discharge → heavier pumps and more complex cooling.

7p. Engine cooling and hot-section materials

Turbine inlet temperatures (1 700 – 1 950 K) routinely exceed the incipient-melting temperature of even single-crystal Ni superalloys (≈ 1 600 K). Continuous operation depends on three stacked technologies:

1. Single-crystal Ni-base superalloys: CMSX-4 (Cannon-Muskegon, 2nd-gen, 3 wt% Re), René N5 (GE, 2nd-gen), CMSX-10 (3rd-gen, 6 % Re), René N6, PWA 1497. The single-crystal solidification (helical selector or seeded process) eliminates grain boundaries, which dominate creep at > 0.7 T_melt. Re slows γ′-phase coarsening.
2. Internal serpentine cooling: cold compressor-bleed air (≈ 600 – 700 K, ≈ 20 % of core flow) enters the blade root, snakes through pin-fins and ribbed turbulators, exits via:
3. Film cooling: small holes (≈ 0.25 – 0.5 mm) on the leading edge and pressure/suction surfaces eject a thin cool layer that insulates the metal from the 1 800 K freestream. Hole shapes: cylindrical (legacy), shaped/fan-diffuser (modern, better blowing ratio coverage).
4. Thermal-barrier coating (TBC): ≈ 200 µm of yttria-stabilised zirconia (8YSZ) plasma-sprayed (APS) or electron-beam-deposited (EB-PVD) over a MCrAlY bond-coat. Adds ≈ 100 – 150 K of effective T capability. Failure mode: spallation triggered by thermally grown oxide (TGO) at the bond/TBC interface.

Combustor liners: Haynes 188, Nimonic 263, and increasingly SiC/SiC ceramic matrix composite (CFM RISE programme, GE9X HPT shroud — Norton-style melt-infiltration SiC matrix with Hi-Nicalon Type S fibre). CMC tolerates ≈ 200 K higher than the best Ni alloy and is half the density.

Hot-section life targets: ≈ 20 000 – 25 000 h on-wing for narrowbody (CFM56, LEAP), 15 000 – 20 000 for widebody (GE90, Trent 900). Life-limited parts (LLPs) per FAR 33.70 are tracked cycle-by-cycle and retired before fatigue-failure probability exceeds 1 in 10⁹.

8p. Rocket propulsion

8.1 Liquid bipropellant

Propellant pair	I_sp_vac [s]	T_c [K]	ρ_avg [kg/m³]	Notes / example engine
LOX / LH2	440 – 465	3 550	360	Best chemical I_sp; bulky tanks; RS-25 (SLS), Vulcain 2 (Ariane 5/6), RL10 (Centaur), BE-3U
LOX / CH4	330 – 380	3 550	830	Best volume × I_sp tradeoff; easy reignition; non-coking; Raptor (Starship), BE-4 (Vulcan/New Glenn), Prometheus (Ariane Next)
LOX / RP-1 (kerosene)	300 – 340	3 670	1 030	Dense, room-T fuel, coking risk; Merlin (Falcon 9), RD-180 (Atlas V), F-1 (Saturn V), RD-191 (Angara)
N₂O₄ / MMH or UDMH	280 – 333	3 400	1 200	Hypergolic (no ignition needed); storable for years; toxic; Aestus, Apollo SPS, Soyuz upper, Long March 2/3/4
H₂O₂ / RP-1	270 – 320	2 970	1 280	Black Arrow, AR1 demo, Skyrora; less toxic than N₂O₄
LOX / LH2 + LF₂ trace	480	3 800	380	Tripropellant studies (cancelled, materials issue)

8.2 Solid motors

Composite solids: HTPB binder (hydroxyl-terminated polybutadiene, 12 – 14 wt%) cast around a star-shaped or cylindrical port, loaded with AP oxidiser (ammonium perchlorate, 65 – 70 wt%) and Al powder (15 – 20 wt% — adds energy density, freezes some performance into Al₂O₃ smoke). I_sp_vac 250 – 280 s. The burn rate scales as r = a·P^n (Saint-Robert / Vieille), with n ≈ 0.3 – 0.5 — chosen low to avoid chamber-pressure runaway. Once lit, normally runs to completion; thrust-termination ports exist but are usually not flown.

Applications: ICBM/SLBM (Minuteman III, Trident D5), launch SRBs (Space Shuttle, SLS, Ariane 5/6 EAP/P120C, Antares Castor 30XL upper, H3 SRB-3), tactical missiles (AMRAAM, Sidewinder, Stinger), spacecraft AKM/PAM.

8.3 Hybrid

Solid fuel grain + liquid (or gaseous) oxidiser. Inherent throttle / stop / restart. I_sp 250 – 320 s. SpaceShipOne (HTPB + N₂O, 2004), SpaceShipTwo (HTPB + N₂O, 2014 fatal break-up partly due to feathering, not engine), Stratolaunch concepts, Gilmour Space Eris. Combustion regression rate is mass-flux-limited; design has wrestled with low regression rate (small Δm per pass) for decades — hence multi-port grains, paraffin-wax fuels.

8.4 Electric propulsion

Thrust is low (mN – N), I_sp very high. Mission delta-v over months-to-years, not minutes.

Type	I_sp [s]	Thrust [mN]	Thrust/Power [mN/kW]	Power class	Example
Resistojet	250 – 400	100 – 800	700 – 900	0.1 – 1 kW	INTELSAT V, Iridium NEXT N2H4
Arcjet	400 – 800	100 – 500	100 – 200	0.5 – 5 kW	Lockheed A2100 (legacy)
Hall thruster	1 500 – 2 500	50 – 500	50 – 75	0.5 – 25 kW	SPT-100 (Russian), BHT-200/1500 (Busek), X3 (NASA), Starlink Hall (krypton then argon)
Gridded ion (Kaufman)	3 000 – 5 000	25 – 250	30 – 50	0.5 – 10 kW	NSTAR (Deep Space 1), NEXT-C (DART), T6 (BepiColombo, Artemis), RIT (Astrium)
MPD (magnetoplasmadynamic)	2 000 – 6 000	1 000 – 100 000	20 – 60	100 kW – 10 MW	Princeton applied-field MPD; not yet flown at scale
PPT (pulsed plasma)	500 – 1 500	0.1 – 1	5 – 20	1 – 100 W	EO-1, NovaSAR-S
FEEP (field-emission)	4 000 – 12 000	0.001 – 0.1	5 – 15	1 – 100 W	LISA Pathfinder, microNewton drag-free

Trade: very high I_sp ⇒ very small propellant mass ⇒ ideal for long-life GEO station-keeping and deep-space probes. SpaceX Starlink uses krypton Hall thrusters for orbit-raise and end-of-life de-orbit (since 2019); next-gen Starlink V2 uses argon Hall — cheaper feedstock at the cost of ≈ 20 % I_sp.

8.5 Nuclear thermal

H₂ propellant heated by passing through a fission reactor core (no combustion, no oxidiser). I_sp 800 – 920 s — roughly 2 × the best chemical, with thrust at chemical-rocket levels (10 – 100+ kN). NERVA / KIWI / Phoebus programmes (1955 – 1972) ground-fired engines up to 1 GW reactor thermal power and demonstrated 25 minutes of burn at 870 s I_sp. The 2023 DRACO programme (DARPA + NASA + Lockheed + BWX) targets a 2027 in-space demo. Trade vs. nuclear-electric: NTP is high-thrust short-burn (Mars trajectory in 6 months); NEP is low-thrust long-burn (more efficient mass-wise, slower).

8.6 Ramjet, scramjet, and combined cycles

A ramjet uses freestream compression through the inlet — no compressor, no turbine. Self-starts only above M ≈ 2 (otherwise inlet recovery and combustor pressure are too low to sustain a flame). Operating range M 2 – 5. Solid-fuel ramjet (SFRJ) and ducted-rocket variants exist for missiles (ASMP-A, BrahMos, Meteor air-to-air). The Lockheed SR-71 Pratt J58 is a hybrid turbojet/ramjet: at M > 2.2 the inlet bleed-bypass shifts 80 % of the air around the core into the afterburner, effectively becoming a ramjet around a windmilling turbojet.

A scramjet burns fuel in supersonic flow inside the combustor (no normal shock to subsonic). Operating range M 5 – 15+ (theoretical). Flow-residence time ≈ 1 ms ⇒ requires reactive fuels (H₂, ethylene, JP-10) and elaborate flameholders (cavity, strut, ramp injectors). Demonstrated:

• NASA X-43A (Hyper-X, 2004): scramjet on a Pegasus booster; reached M 9.6 at 33 km — fastest air-breather ever, still the record.
• USAF X-51A Waverider (2010 – 2013): hydrocarbon (JP-7) scramjet, M 5.1 for 210 s — first sustained hydrocarbon scramjet flight.
• HIFiRE (US/Australia), HyShot (Queensland), DF-17/HGV (Chinese), Avangard (Russian) — hypersonic-weapon vehicles use scramjets or boost-glide vehicles.

Combined cycles stack a turbomachine with a ramjet or rocket to bridge takeoff (M 0) and hypersonic cruise (M 5+):

• TBCC (turbine-based combined cycle): turbojet/turbofan from M 0 to ≈ M 3, then transitions to ramjet/scramjet. Conceptual; SR-71 J58 is the closest realised. Lockheed SR-72 concept.
• RBCC (rocket-based combined cycle): rocket assists from M 0 to M 2 – 3, then air-breathing ramjet from M 3 to M 6+, then back to pure rocket for the trans-atmospheric ascent. NASA GTX, ISRO Avatar, Reaction Engines SABRE (precooled hybrid air-breather/rocket for Skylon).

9p. Rocket engine cycles

The cycle is how the turbopump is driven — i.e. where the energy comes from to pressurise propellants from tank (1 – 10 bar) to chamber (50 – 350 bar).

• Pressure-fed — no turbopump. Tank ullage pressure (helium or autogenous) pushes propellant. Simple, low-Pc (≤ 30 bar), heavy tanks because they must hold full chamber + ΔP. Examples: Apollo Lunar Module descent and ascent engines, Aestus (Ariane 5 EPS), AJ10 (Delta II second stage, Orion service module), Crew Dragon SuperDraco.

• Gas generator (open cycle) — a small precombustor burns a fuel-rich slip-stream (≈ 2 – 4 % of total flow), drives the turbopump turbine, and dumps the spent gas overboard (usually into the nozzle exit cone for a small thrust contribution). Simple, lower performance because that mass doesn’t see the main combustion chamber. Examples: F-1 (Saturn V S-IC, 6.9 MN sea-level, 263 s I_sp_SL), Merlin 1D (Falcon 9, 845 kN SL, 282 s SL), Vulcain 2 (Ariane 5/6 core, 1.36 MN vac, 432 s I_sp_vac, LOX/LH2), RS-68A (Delta IV, ablative chamber + GG).

• Staged combustion (closed cycle) — a fuel-rich or oxidiser-rich preburner produces hot gas, drives the turbine, and the spent gas is routed into the main chamber where it finishes burning. All the propellant reaches the chamber → highest performance. Higher Pc (200 – 300+ bar) ⇒ smaller, lighter engine at given thrust. Examples: RS-25 (Space Shuttle, SLS; 232 MPa Pc, 452 s I_sp_vac, fuel-rich SC), RD-180 (Atlas V; ox-rich SC, 268 bar Pc, 311 s SL — bought from NPO Energomash, supply cut after 2022), RD-191 (Angara), Vulcain 3 (proposed).

• Full-flow staged combustion (FFSC) — two preburners, one fuel-rich and one oxidiser-rich; each drives its own turbopump turbine; both gas streams enter the main chamber. Lowest turbine inlet temperatures (each preburner runs cool because of unburned excess of one reactant); highest Pc and Isp. Two engines have flown: the cancelled Soviet RD-270 (UDMH/N₂O₄, 1969) and SpaceX Raptor (LOX/CH4, 2019 first hop on SN5, 2025 Starship V3 with Raptor 3 at ≈ 350 bar Pc, ≈ 2.7 MN SL, 327 s SL / 380 s vac).

• Expander cycle — regen-cooled chamber heats fuel (LH2 or CH4) into a gas, which drives the turbopump turbine before entering the injector. No preburner. Self-limiting in size (chamber surface area scales as L², heat extraction as L²·heat-flux, but flow scales as L³, so above ≈ 300 kN the heat budget falls short). Examples: RL10 (Centaur, Saturn IV; 110 kN, 465 s — flown since 1963, the GOAT of upper-stage engines), Vinci (Ariane 6 upper, 180 kN, 465 s), BE-3U (New Glenn upper, 710 kN, 445 s; “tap-off” variant). Bleed-expander hybrids: LE-9 (H3 first stage, 1.5 MN, expander-bleed).

Cycle selection trade summary:

Cycle	Typical Pc [bar]	Complexity	I_sp delta vs. pressure-fed	Notable use
Pressure-fed	5 – 30	Lowest	baseline	LM descent, Aestus, SuperDraco
Electric-pump-fed	100	Low (battery + brushless pump)	+10 – 15 %	Rutherford (Electron)
Gas-generator	70 – 200	Medium	+25 – 30 %	Merlin, F-1, Vulcain 2
Tap-off	80 – 150	Medium	+25 – 30 %	BE-3 (variant), Firefly Lightning
Expander	30 – 90	Medium	+25 – 35 % (LH2 only)	RL10, Vinci
Staged combustion (single preburner)	200 – 270	High	+35 – 40 %	RS-25, RD-180, RD-191, Vulcain 3
Full-flow staged combustion	300 – 350+	Highest	+40 – 45 %	Raptor (only operational FFSC)

Injector design is a parallel discipline that lives inside every cycle. Liquid-bipropellant face designs:

• Coaxial swirl — RS-25, Vulcain. LOX through central post, fuel sleeve outside, both swirled.
• Pintle — Merlin, TR-201, Apollo LM descent. Single moving central post (pintle); doubles as throttle. Simple, robust, naturally damping for combustion instability (Tom Mueller / TRW lineage).
• Impinging doublet / triplet — F-1, J-2. Liquid jets cross and break each other into a fine spray. Sensitive to acoustic instability; F-1 stabilisation took 13 baffle iterations.
• Showerhead — early experimental, generally inferior atomisation.
• Gas-gas — both propellants enter the chamber as hot gas (full-flow staged combustion). Raptor only. Wider stability margin, but a complete plumbing redesign.

4a. Reference data — modern airliner engines (cruise, comparative)

Engine	Aircraft	Thrust [kN] (SL takeoff)	BPR	OPR	Fan dia [m]	Cruise TSFC [lb/(lbf·h)]	Notes
CFM56-7B27	737NG	121	5.1	32	1.55	0.65	Reference legacy narrowbody
V2500-A5	A320ceo	133	4.8	35	1.61	0.58	IAE consortium
CFM LEAP-1A	A320neo	143	11	40	1.98	0.51	CMC HPT shroud, woven CF fan blades
CFM LEAP-1B	737 MAX	130	8.6	40	1.76	0.53	Lower BPR — ground clearance
PW1100G	A320neo	147	12.5	50	2.06	0.51	Geared turbofan (1:3 planetary)
Trent XWB-84	A350-900	374	9.6	50	3.00	0.49	Three-spool RR architecture
GE9X	777-9	489	9.9	60	3.40	0.49	World’s largest fan; CMC combustor + LPT shrouds
GEnx-1B	787	320	9.6	45	2.82	0.51	Composite fan case
Trent 1000	787	360	10	50	2.85	0.51	Three-spool
PW1900G	E190-E2/E195-E2	102	12	40	1.85	0.50	Smallest GTF
RR UltraFan (demo)	(none — demo)	110 000 hp class	15+	70	3.56	(target −25 %)	First-run 2023, CTi fan, gearbox
CFM RISE (demo)	(none — demo)	(target 150 kN class)	70+	65	4.2 (open-fan)	(target −20 %)	Open rotor, CMC, hydrogen-capable

4b. Reference data — propellant comparison

Propellant pair	T_c [K]	I_sp_vac [s, ε≈40]	Bulk ρ [kg/m³]	Storability	Toxicity	Notes
LOX / LH2	3 550	460	360	Cryo (boil)	Low	Best I_sp; large tanks; embrittlement
LOX / CH4	3 550	380	830	Sub-cool both for densification	Low	Easy reignition; no coking; ISRU on Mars
LOX / RP-1	3 670	340	1 030	Cryo LOX, room-T fuel	Low	Coking; rich-burn for cooling
LOX / C2H5OH	3 250	310	970	Cryo + room-T	Very low	V-2, Redstone heritage
H₂O₂ (90%) / RP-1	2 970	320	1 280	Storable months	Skin contact hazard	Black Arrow, Skyrora
N₂O₄ / MMH	3 400	320	1 200	Years (sealed)	High (cancer)	Crew Dragon Draco/SuperDraco
N₂O₄ / UDMH	3 470	318	1 195	Years	Very high	Proton, Long March 2/3/4, Soyuz aux
N₂O / HTPB (hybrid)	3 000	250	920	Years	Low	SpaceShipOne/Two
AP / Al / HTPB (solid)	3 600	270	1 800	20 yr (sealed)	Inert	All large solid boosters
Hydrazine (N₂H₄) monoprop	1 200	230	1 010	Years	Very high	Spinning bed catalyst (Shell 405)
Green monoprop (LMP-103S, AF-M315E)	1 900	250	1 240	Years	Low	Replaces hydrazine; NASA GPIM

I_sp values: nominal vacuum at ε = 40 expansion ratio with optimum O/F. Real engines vary ±3 %.

10p. Worked examples

Example A — Single-spool turbojet cycle, cruise design point

Problem. A simple turbojet flies at M = 0.85, altitude 11 km (T_∞ = 216.65 K, P_∞ = 22.632 kPa per ICAO Standard Atmosphere). Design parameters:

• Inlet recovery η_d = 0.97
• Compressor pressure ratio π_c = 20, polytropic efficiency e_c = 0.90 (≈ η_c,isen ≈ 0.86)
• Turbine inlet T_t4 = 1 500 K, combustor efficiency η_b = 0.98, LHV = 43 MJ/kg (Jet-A)
• Turbine polytropic e_t = 0.88
• Nozzle exit fully expanded to P_∞
• Ratios of specific heats: γ_c = 1.4 in cold section, γ_t = 1.33 in hot section
• c_p,c = 1.005 kJ/(kg·K), c_p,t = 1.150 kJ/(kg·K)

Step 1 — Stagnation freestream.

T_t0 = T_∞ · (1 + (γ−1)/2 · M²)  =  216.65 · (1 + 0.2 · 0.85²)  =  216.65 · 1.1445  =  248.0 K
P_t0 = P_∞ · (1 + (γ−1)/2 · M²)^(γ/(γ−1))  =  22.632 · 1.1445^3.5  =  22.632 · 1.6038  =  36.30 kPa

Step 2 — Inlet exit (engine face).

T_t2 = T_t0 = 248.0 K   (adiabatic)
P_t2 = η_d · P_t0 = 0.97 · 36.30  =  35.21 kPa

Step 3 — Compressor exit. With polytropic e_c = 0.90:

T_t3 / T_t2 = π_c^((γ−1)/(γ·e_c))  =  20^(0.4/(1.4·0.9))  =  20^0.3175  =  2.567
T_t3 = 248.0 · 2.567  =  636.7 K
P_t3 = π_c · P_t2 = 20 · 35.21  =  704.2 kPa

Step 4 — Combustor. Fuel-air ratio f from energy balance:

(1 + f) · c_p,t · T_t4  =  c_p,c · T_t3  +  f · η_b · LHV
1.150 · 1500 · (1 + f)  =  1.005 · 636.7  +  f · 0.98 · 43000
1725 + 1725·f          =  639.9  +  42140·f
1085.1                 =  40415·f
f  =  0.02685    →    Air-fuel ratio AF = 37.2

Step 5 — Turbine. Shaft balance: w_turb = w_comp (single spool, no fan), and a fraction (1 + f) of the mass flow passes through the turbine.

c_p,t · (T_t4 − T_t5) · (1 + f)  =  c_p,c · (T_t3 − T_t2)
1.150 · (1500 − T_t5) · 1.02685  =  1.005 · (636.7 − 248.0)
1.1809 · (1500 − T_t5)          =  390.6
1500 − T_t5  =  330.8        ⇒    T_t5 = 1169.2 K

Turbine pressure ratio (polytropic e_t = 0.88, γ_t = 1.33):

T_t5 / T_t4 = (P_t5 / P_t4)^((γ−1)·e_t / γ)
1169.2 / 1500 = (P_t5 / 704.2)^(0.33·0.88/1.33)
0.7795 = (P_t5/704.2)^0.2183
P_t5/704.2 = 0.7795^(1/0.2183) = 0.7795^4.581 = 0.331
P_t5 = 233.0 kPa

Step 6 — Nozzle. Fully expanded, P_9 = P_∞ = 22.632 kPa, T_t9 = T_t5 = 1169.2 K (adiabatic).

V_9² / 2  =  c_p,t · T_t5 · [1 − (P_9 / P_t5)^((γ−1)/γ)]
         =  1150 · 1169.2 · [1 − (22.632/233.0)^(0.33/1.33)]
         =  1.344e6 · [1 − 0.0971^0.2481]
         =  1.344e6 · [1 − 0.5572]
         =  1.344e6 · 0.4428
         =  5.953e5 J/kg
V_9  =  1091 m/s

Step 7 — Thrust and TSFC.

V_∞ = M · √(γ·R·T_∞) = 0.85 · √(1.4·287·216.65) = 0.85 · 295.1 = 250.8 m/s
F / ṁ_air  =  (1 + f) · V_9 − V_∞  =  1.02685·1091 − 250.8  =  1120.3 − 250.8  =  869.5 N/(kg/s)
TSFC  =  f / (F/ṁ_a)  =  0.02685 / 869.5  =  3.088e−5 kg/(N·s)  =  1.111 lb/(lbf·h)

A real turbojet TSFC at cruise: ≈ 1.0 – 1.2 lb/(lbf·h). Hand calc matches. For a high-BPR turbofan at the same flight condition, TSFC falls to ≈ 0.55 lb/(lbf·h) — the bypass air dominates thrust, and most of that air is accelerated only by the fan, at low pressure ratio (≈ 1.5) and high η_p.

Example B — Saturn V S-IC first-stage Δv

Problem. Saturn V first stage: 5 × F-1 engines, ignition mass m₀ = 2.97 × 10⁶ kg, burnout mass m_f = 753 × 10³ kg (after S-IC sep), I_sp_SL = 263 s, burn time t_b = 161 s.

V_eq_SL  =  I_sp_SL · g₀  =  263 · 9.80665  =  2 579 m/s
Δv_ideal  =  V_eq_SL · ln(m₀/m_f)  =  2579 · ln(2.97e6 / 753e3)  =  2579 · ln(3.945)
         =  2579 · 1.3725  =  3 540 m/s

Real losses (Saturn V S-IC documented in NASA SP-4029 + Apollo flight evaluations):

• Gravity loss (∫ g sin γ dt over the burn) ≈ 1 220 m/s
• Drag loss (∫ D/m dt) ≈ 46 m/s
• Steering loss (cosine of gimbal angle) ≈ 5 m/s
• I_sp increases with altitude (vacuum-side correction) gains back ≈ 460 m/s

Net real Δv at S-IC sep: ≈ 2 760 m/s, altitude ≈ 67 km, velocity ≈ 2 750 m/s. Matches the achieved value within 1 %.

Example C — Hall thruster station-keeping budget

Problem. A 2 000 kg GEO communications satellite needs 50 m/s/yr of north-south station-keeping (15 m/s east-west on top) for a 15-year design life — total Δv ≈ 975 m/s. Compare a Hall-effect thruster (I_sp = 1 750 s, xenon) vs. a hypergolic chemical (I_sp = 230 s, N₂O₄ / MMH).

Hall thruster:

V_eq  =  1750 · 9.80665  =  17 162 m/s
m_p / m₀  =  1 − exp(−Δv / V_eq)  =  1 − exp(−975/17162)  =  1 − exp(−0.05681)
         =  1 − 0.9448  =  0.0552
m_p  =  0.0552 · 2000  =  110 kg propellant (xenon)

Hypergolic:

V_eq  =  230 · 9.80665  =  2 255 m/s
m_p / m₀  =  1 − exp(−975/2255)  =  1 − exp(−0.4324)  =  1 − 0.649  =  0.351
m_p  =  0.351 · 2000  =  702 kg propellant (N₂O₄+MMH)

The Hall option saves 592 kg — enough additional payload for one or two additional transponders. The trade-off: a chemical RCS burn for one station-keep manoeuvre is ≈ 60 s; the equivalent Hall burn at 200 mN thrust on a 2 000 kg sat takes ≈ 7 days of continuous firing (a/m ≈ 1 × 10⁻⁴ m/s²). Most modern GEO comsats (SES, Intelsat, Eutelsat) have flown all-electric platforms since ≈ 2014 (Boeing 702SP, Airbus Eurostar Neo). Starlink uses argon Hall today.

Example D — Rocket chamber regen-cooling heat flux

Problem. A LOX/RP-1 staged-combustion chamber at Pc = 250 bar, T_c = 3 670 K, throat heat flux q” ≈ 80 MW/m². Cooling channels carry RP-1 at 10 kg/s with ΔT_coolant = 200 K. Estimate channel area required and check Bartz-correlation throat heat-transfer coefficient.

Bartz (1957) correlation:

h_g = 0.026/D_t^0.2 · (μ^0.2 c_p / Pr^0.6)_c · (Pc/c*)^0.8 · (D_t/r_c)^0.1 · σ

For typical kerolox: h_g ≈ 25 kW/(m²·K). Driving ΔT (T_c − T_wall_gas) ≈ 3 200 K ⇒ q” ≈ 80 MW/m². The hot-gas wall sees ≈ 700 K with TBC + film cooling; coolant-side wall ≈ 600 K to keep nucleate boiling out of film boiling. Coolant ΔT capacity: ṁ · c_p · ΔT = 10 · 2.0 · 200 = 4 MW total cooling — bounds the chamber throat surface area to roughly 4 MW / 80 MW/m² = 0.05 m² (a 0.13 m diameter throat at 12 cm long). Real engines run two-pass channels and add film cooling at the injector face (≈ 2 – 5 % of fuel flow injected near wall) to share the load.

11p. Edge cases / gotchas

• Compressor stall and surge. When the operating line approaches the surge line (low flow at given speed), boundary layers on the suction surface of the blades separate. Rotating stall is a precursor — a cell of stalled flow propagates around the annulus. Surge is a violent global flow reversal that can destroy the engine in seconds. Mitigations: variable inlet guide vanes (VIGV), variable stators (V2500 has 4 rows), interstage bleed valves, fast-response fuel control.

• Combustor flame-out. Lean blow-out (LBO) at idle / descent, water/ice ingestion, fuel-control transient. Relight envelope (FAR 33.77) is a defined region (altitude, airspeed) within which the engine must demonstrate windmill restart capability. Common cause of dual-engine failure: large hailstone field, volcanic ash (BA 9 Jakarta 1982 — all four engines flame-out from Mt Galunggung ash), bird ingestion (US Airways 1549 / Hudson — geese into both fan stages).

• FOD. Foreign object damage — birds (FAR 33.76 — must ingest 1.85 kg single bird without hazardous engine effect), ice (anti-ice systems on fan blades, spinner), tyre rubber off the runway, hailstones. Fan-blade composite shells (GEnx, GE9X, LEAP) are designed for soft-body impact; titanium or hollow Ti blades on legacy fans.

• Hot-section deterioration. Creep elongation of HP turbine blades (tip rub against shroud), TBC spallation, oxidation/sulfidation in coastal operation. Hot-section life is condition-monitored via EGT (exhaust gas temperature) margin: as deterioration sets in, more fuel is needed to make takeoff power → higher EGT → engine sent for hot-section refurb when EGT margin reaches zero.

• NOₓ at off-design. Lean-premixed combustors run a narrow φ window. At low throttle (taxi, descent) the burner stages out and emissions rise. ICAO LTO (landing-take-off) cycle is the certification metric: NOₓ + CO + HC + smoke integrated over 7 % / 30 % / 85 % / 100 % thrust durations.

• Inlet distortion. High α (rotation, deep stall), crosswind on static, transonic shock unstart on supersonic engines. Distorted total pressure at the engine face migrates into compressor surge margin loss; severe distortion causes one-side stall. Measured by DC₆₀ index.

• Cooling-air contamination. Particles from compressor erosion (ingested sand, brake dust on military forward airbases, volcanic ash) plug film-cooling holes — and that blade burns through in minutes. Operation in desert environments (CH-47 in Iraq, AH-64 in Afghanistan) shortens hot-section life by 5×.

• POGO instability. Coupled longitudinal vehicle structural mode with propellant feed-line dynamics. Pressure oscillation in fuel line modulates engine thrust, which excites the vehicle’s first axial mode, which modulates feed-line acceleration head, which closes the loop. Saturn V S-II flight 6 (Apollo 6) suffered 5 Hz POGO; fix added helium accumulators to the LOX feed lines on subsequent vehicles to detune the system.

• Combustion instability (“screech” / “chug” / “buzz”). Pressure waves at acoustic frequencies of the chamber grow until they shatter injectors or burn through walls. F-1 development on Saturn V took years to stabilise (baffled injector face with 13 vanes, multiple injector revisions, 2 000+ test firings). RD-170 / RD-180 family use chamber baffles + acoustic resonators.

• Cryogenic feed and chilldown. LH2 boils at 20 K, LOX at 90 K. Before opening the main valves, the entire feed line (tens of metres) must be chilled to suppress vapour pockets that would cavitate the turbopump. Wasted chilldown propellant is a launch-vehicle mass penalty.

• Nozzle flow separation (over-expansion). A vacuum-optimised bell nozzle (large ε = A_e/A_t) at sea-level may have P_e < ~0.4·P_∞ — the boundary layer separates from the wall, generating asymmetric side loads that twist the gimbal. The RS-25 has restart loops to manage start transient flow separation. Vulcain 2 sees side-load > 250 kN at sea level.

• Cavitation in turbopump. Same physics as industrial pumps (see pumps-turbomachinery): inlet P < P_vap of cryo or hypergol → vapour bubbles collapse on impeller. Solved with inducers (axial pre-rotor designed to operate cavitating without performance loss), high tank pressurisation, and high suction-side flow margin.

• Solid-motor nozzle erosion. Al₂O₃ particulate from Al-loaded propellant erodes the graphite/C-C throat — throat area grows during burn, so chamber pressure decays and thrust-vs-time curve droops. Predicted via empirical correlation (Bartz × oxidation correction) and accommodated in trajectory design.

• Restart on orbit / settling burns. Cryogenic upper stages must settle propellant to the tank outlet before main-engine ignition. Done by a small RCS axial burn (≈ 0.01 g) for a few seconds, or by ullage rockets (solid grains). Failure to settle → engine ingests vapour → hard start or no start.

12p. Tools and ecosystem

Cycle-analysis and engine-performance software:

• NPSS (Numerical Propulsion System Simulation) — NASA / Aerospace Industries Association consortium product, the US industry standard for engine cycle simulation. Object-oriented; users assemble engines from elements (compressor, combustor, turbine, nozzle) on a flowsheet.
• GasTurb 14 — Joachim Kurzke, the standard commercial tool for cycle design, off-design and transient performance. Used at every major engine OEM.
• PROOSIS — Empresarios Agrupados (Spain), used in European industry (Safran, RR Deutschland, Avio).
• AEDsys — companion to Mattingly’s textbook (Aircraft Engine Design); free, education-grade.
• GSP — NLR Netherlands Aerospace Centre.
• Cantera — open-source 0-D / 1-D combustion + reacting flow; useful for combustor sub-models, gas-property polynomials.

CFD (specialised for turbomachinery and propulsion):

• ANSYS CFX, ANSYS Fluent — both with turbomachinery extensions; CFX historically stronger on turbomachinery (multi-stage frozen-rotor, mixing-plane).
• NUMECA FINE/Turbo (Cadence) — purpose-built for turbomachines (compressors, turbines, fans).
• Concepts NREC AxCent / TurboAero — 1-D / 2-D throughflow + 3-D blade design.
• STAR-CCM+ Turbomachinery (Siemens) — modern unstructured CFD with turbomachinery template.
• OpenFOAM — open-source; turbomachinery solvers (MRF, AMI sliding mesh) widely used in academic work.
• Loci/CHEM (Mississippi State / NASA) — combustion CFD for rocket chambers.
• Converge CFD — IC engine + rocket combustion specialist (auto-meshing for moving boundaries).

Trajectory / mission analysis:

• STK (Ansys / AGI) — commercial flagship for satellite mission design, launch vehicle ascent, interplanetary trajectories.
• GMAT — NASA Goddard, open-source; used for routine GSFC mission design.
• Copernicus — NASA JSC trajectory tool, optimisation-heavy.
• AGI ODTK — orbit determination toolkit.

Suppliers (current 2026 product line):

• GE Aerospace — LEAP-1A/B/C (with Safran, BPR 11), GEnx-1B/2B, GE9X (777X, BPR 9.9, 470 kN), CT7/T700 turboshafts, GE Catalyst turboprop, F110, F414, F404 military.
• CFM International (GE + Safran 50/50) — CFM56-5B/7B (legacy), LEAP-1A (A320neo), LEAP-1B (737 MAX), LEAP-1C (C919). RISE open-rotor demo 2025 – 2028.
• Pratt & Whitney (RTX) — PW1100G/PW1500G/PW1900G geared turbofan family, PW4000, PW6000, F119 (F-22), F135 (F-35), PT6A.
• Rolls-Royce — Trent 1000 (787), Trent XWB (A350), Trent 7000 (A330neo), Trent 900 (A380), AE 3007, EJ200, Pearl business-jet family, UltraFan demonstrator (geared, BPR 15+).
• Safran — M88 (Rafale), M53 (Mirage 2000), Arriel, Arrius, Makila helo engines; CFM partner.
• Honeywell — HTF7000, TPE331, TFE731, APUs; partner on some Garrett legacy.
• MTU Aero Engines — partner on V2500, PW1000G LPT, GP7200; military RB199 (Tornado).
• Aerojet Rocketdyne (L3Harris) — RS-25, RL10, AR1 (proposed), Aerojet legacy.
• Blue Origin — BE-3 (LH2 expander, New Shepard + New Glenn upper), BE-4 (LOX/CH4 ox-rich SC, 2.4 MN SL, on Vulcan and New Glenn).
• SpaceX — Merlin 1D / 1D Vac (kerolox GG), Raptor 1/2/3 (LOX/CH4 FFSC, Starship).
• Rocket Lab — Rutherford (electric-pump-fed kerolox), Archimedes (kerolox ox-rich SC, Neutron).
• Firefly Aerospace — Reaver, Lightning (tap-off cycle).
• Astra, Relativity Space (Aeon R), Stoke Space (Zenith), Impulse Space (Saiph) — newer vehicle OEMs each with proprietary engines.
• Electric propulsion: Busek, Aerojet (XR-5), Safran (PPS), Sitael, ThrustMe, ENPULSION (FEEP), ExoTerra.

13. Cross-references

• thermodynamics — Brayton cycle (jet engines), rocket combustion thermochemistry, real-fluid effects, adiabatic flame temperature.
• heat-transfer — turbine-blade film cooling, regen-cooled rocket chambers, TBC, ablative materials.
• fluid-mechanics — compressible flow in inlets and nozzles, shocks, supersonic diffusers, mixing.
• pumps-turbomachinery — axial compressors / turbines (jets), centrifugal turbopumps (rockets), Euler equation, cavitation in turbopumps, NPSH.
• materials-ceramics — single-crystal Ni superalloys, TBC (8YSZ), SiC/SiC CMC for combustor liners and turbine shrouds.
• aerodynamics — installed engine performance, nacelle drag, inlet–airframe integration, propeller theory.
• electromagnetics-engineering — Hall-effect thruster plasma physics, magnetic-field design.
• electric-motors — electric aviation propulsors, eVTOL motors.
• power-electronics — eVTOL inverters, electric-pump-fed rocket controllers (Rocket Lab Rutherford).
• structural-dynamics — POGO instability, rotor dynamics, gimbal-actuator loads.
• aerospace-defence (planned) — ARINC 429/664, MIL-STD-1553, AS9100, flight-data formats.
• spacecraft-attitude-control (planned) — GNC interaction with propulsion, throttle authority, restart envelopes.

14. Citations

1. Mattingly, J. D. Elements of Propulsion: Gas Turbines and Rockets, 3rd ed. AIAA Education Series, 2024. ISBN 978-1624106989. The canonical US graduate text; covers cycle analysis, component design, rocket fundamentals end-to-end.
2. Hill, P. G.; Peterson, C. R. Mechanics and Thermodynamics of Propulsion, 2nd ed. Addison-Wesley, 1992. ISBN 978-0201146592. The classic; clear treatment of compressible flow, cycle thermodynamics, and component matching.
3. Kerrebrock, J. L. Aircraft Engines and Gas Turbines, 2nd ed. MIT Press, 1992. ISBN 978-0262111621. Strong on compressor and turbine aerodynamics, off-design behaviour.
4. Cumpsty, N. A. Jet Propulsion: A Simple Guide to the Aerodynamic and Thermodynamic Design and Performance of Jet Engines, 3rd ed. Cambridge University Press, 2015. ISBN 978-1107511224. The single most accessible introduction; written from Rolls-Royce experience.
5. Sutton, G. P.; Biblarz, O. Rocket Propulsion Elements, 9th ed. Wiley, 2017. ISBN 978-1118753651. The canonical rocket text; covers liquid, solid, hybrid, electric, and nuclear in one volume.
6. Huzel, D. K.; Huang, D. H. Modern Engineering for Design of Liquid-Propellant Rocket Engines. AIAA Progress in Astronautics & Aeronautics, Vol. 147, 1992. ISBN 978-1563470134. The engine-designer’s bench reference; based on Rocketdyne design practice.
7. Goebel, D. M.; Katz, I. Fundamentals of Electric Propulsion: Ion and Hall Thrusters, 2nd ed. Wiley, 2024. ISBN 978-1394164547. The canonical EP text; JPL-grounded.
8. Anderson, J. D. Hypersonic and High-Temperature Gas Dynamics, 3rd ed. AIAA Education Series, 2019. ISBN 978-1624105142. Reference for scramjet, re-entry, and high-T gas effects.
9. Mattingly, J. D.; Heiser, W. H.; Pratt, D. T. Aircraft Engine Design, 2nd ed. AIAA Education Series, 2002. ISBN 978-1563475382. Companion to (1); design-oriented with the AEDsys software.
10. Heiser, W. H.; Pratt, D. T. Hypersonic Airbreathing Propulsion. AIAA Education Series, 1994. ISBN 978-1563470356. Foundational scramjet text.
11. Walsh, P. P.; Fletcher, P. Gas Turbine Performance, 2nd ed. Blackwell, 2004. ISBN 978-0632064342. Rolls-Royce-rooted; the standard performance-analyst reference.
12. Kuo, K. K. Principles of Combustion, 2nd ed. Wiley, 2005. ISBN 978-0471046899. Used heavily for combustor and solid-rocket internal ballistics.
13. SAE AS755 — Aircraft Propulsion System Performance Station Designation and Nomenclature. SAE International, latest revision 2017. The station-numbering standard cited throughout.
14. SAE ARP755 — Gas Turbine Engine Performance Presentation and Nomenclature for Digital Computer Programs. SAE International. Companion to AS755 for digital cycle codes.
15. FAR Part 33 — Airworthiness Standards: Aircraft Engines. 14 CFR Part 33, current rev. Type-certification requirements (vibration, bird ingestion, FBO, durability).
16. MIL-STD-5007D — Engines, Aircraft, Turbojet and Turbofan, General Specification For. US DoD, 1973 (still active). The US military engine spec from which all later commercial requirements descended.
17. NASA SP-8089 — Liquid Rocket Engine Turbopumps. NASA Space Vehicle Design Criteria monograph series (SP-8005 to SP-8125, 1968 – 1976). The most concentrated body of practical rocket-engine design knowledge ever published.
18. AIAA Joint Propulsion Conference proceedings, annual (since 1965). The single largest peer-reviewed venue for propulsion R&D.
19. AIAA Journal of Propulsion and Power, ongoing. Primary refereed journal.
20. Marek, S. L. Liquid Propellant Rocket Engines (NASA TM SP-125, 1971). Open NASA reference.