Every jet engine, rocket, and gas turbine ever built obeys the same fundamental law: Newton's third law applied to a working fluid. This page builds your understanding from the ground up: the physics first, the equations second, then the real engineering that makes it work at 35,000 feet.
Before turbines, combustors, or nozzles, there is one idea. If you accelerate mass in one direction, you get a force in the opposite direction. Everything in propulsion is an application of this.
Every propulsion system, from a child's balloon to a Saturn V, works by throwing mass backward. The reaction force is thrust. The general thrust equation captures this precisely:
F = ṁ(Ve − V∞) + (pe − p∞)Ae
The first term is momentum thrust: the mass flow rate ṁ times the gain in velocity (exhaust Ve minus freestream V∞). The second term is pressure thrust: the pressure difference at the exit plane times the nozzle exit area. In a perfectly expanded nozzle (pe = p∞), all thrust comes from momentum.
Specific impulse (Isp) is the single most important performance number in propulsion. It tells you how much thrust you get per unit weight-flow of propellant consumed:
Isp = F / (ṁ_prop × g₀)
The units are seconds. A higher Isp means you burn less propellant for the same thrust. Every second of Isp you gain translates directly to payload capacity or range.
A turbofan at cruise has an effective Isp of roughly 6,000 to 7,000 seconds, because it generates thrust mostly by accelerating bypass air rather than by burning fuel for momentum. That figure follows directly from its own TSFC: Isp = 1/(g₀·TSFC), so 15 g/kN·s gives about 6,800 s. A kerosene/LOX rocket engine achieves Isp ≈ 311 s at sea level. The difference (ten-fold) explains why rockets need to be 90 to 95% propellant by mass to reach orbit, while aircraft can cruise for 14 hours on a reasonable fuel fraction.
Every jet engine, every gas turbine power plant, and every turboshaft helicopter engine runs on the Brayton cycle. Understanding this cycle is the key to understanding why engines are designed the way they are, and what limits their efficiency.
Air enters the compressor and is compressed adiabatically and reversibly, no heat exchange, no friction losses in the ideal case. Temperature and pressure both rise. The isentropic efficiency η_c of a real compressor (85 to 92%) accounts for irreversibilities. Temperature rise: T₀₂ = T₀₁ · (OPR)^((γ-1)/γ). Note these are stagnation temperatures, not static: compressor work is set by total enthalpy rise, and the intake has already done some of it. At M 0.8 with a static temperature of 250 K, the compressor face sees T₀₁ = 250(1 + 0.2 × 0.8²) = 282 K, and with OPR 45 the exit is T₀₂ ≈ 282 × 2.97 = 837 K. Ignoring ram compression and using the static 250 K would give 742 K, which is both wrong and hides the fact that the intake is part of the cycle.
Fuel is injected and burned at approximately constant pressure. Temperature jumps to the turbine entry temperature (TET), up to about 1,900 K stagnation at the rotor inlet in modern engines. That is well above the maximum usable metal temperature of single-crystal nickel superalloys, roughly 1,350 to 1,400 K, though still below their incipient melting point near 1,600 K. The blade does not melt; it creeps, oxidises and loses strength long before melting, which is the limit that actually governs design. so internal cooling channels, film cooling holes, and thermal barrier coatings are essential. The combustor also reduces NOₓ emissions, the primary driver of modern combustor design.
Hot gas expands through the turbine, doing work. The high-pressure turbine (HPT) extracts just enough work to drive the compressor. These are directly shaft-coupled. The low-pressure turbine (LPT) extracts remaining work to drive the fan (turbofan) or propeller (turboprop). Turbine isentropic efficiency η_t ≈ 88 to 92%. The pressure ratio across the turbine is determined by the work extraction required.
In an open-cycle gas turbine (all jet engines), the exhaust is ejected to atmosphere. The heat rejection is not actually a closed process. The nozzle accelerates the remaining gas to produce the exhaust jet. In an ideal cycle this is isentropic; in practice, nozzle efficiency η_n ≈ 95 to 99%. The exit velocity Ve determines the momentum thrust component. Higher Ve = more thrust but lower propulsive efficiency.
Every air-breathing engine takes the Brayton cycle and adapts it to a specific flight regime. The differences (bypass ratio, compression method, combustion speed) are engineering responses to the same fundamental constraints.
A turbofan adds a large fan in front of the core engine. Most of the air accelerated by this fan bypasses the combustor entirely. It goes around the core and is exhausted directly. The bypass ratio (BPR) is the ratio of bypass airflow to core airflow.
The physics reason this works: propulsive efficiency η_p = 2V∞ / (Ve + V∞). Maximum efficiency occurs when exhaust velocity Ve equals flight speed V∞, meaning you accelerate a lot of air by a small amount, rather than a little air by a large amount. A high-BPR fan does exactly this.
The CFM LEAP (A320neo) has BPR ≈ 11. The Rolls-Royce Trent XWB (A350) has BPR ≈ 9.3. The Pratt & Whitney GTF (A220) uses a gearbox to decouple fan and LP turbine speeds, allowing BPR above 12 with further efficiency gains.
The turbojet sends all ingested air through the compressor, combustor, and turbine. It produces a high-velocity, relatively small-mass-flow jet. The opposite of what propulsive efficiency demands at subsonic speeds.
From η_p = 2/(1 + Ve/V∞): at subsonic cruise (V∞ ≈ 240 m/s) with a turbojet exhaust at Ve ≈ 600 m/s, propulsive efficiency is only 2/(1 + 2.5) = 57%. A high-BPR turbofan achieving Ve/V∞ ≈ 1.3 gives η_p ≈ 87%.
The turbojet finds its niche at high supersonic Mach numbers (M 2+) where exhaust velocity is naturally close to flight speed. The Rolls-Royce/Snecma Olympus 593 on Concorde operated at M 2.04, at that speed, the efficiency penalty largely disappears.
A turboprop extracts nearly all the combustion energy through the turbine to drive a propeller via a reduction gearbox. Only a small residual jet contributes to thrust (typically 10 to 15% of total thrust at cruise).
Propellers generate thrust by accelerating a large mass of air by a small velocity increment, exactly what propulsive efficiency demands. At speeds below about 400 knots, a propeller is significantly more efficient than a jet nozzle. This is why turboprops dominate the regional market: ATR 72, De Havilland Dash 8, Bombardier Q400.
The reduction gearbox is essential: the gas generator runs at 20,000 to 40,000 rpm, but the propeller is limited by its helical tip Mach number, √(V_tip² + V∞²)/a, which must stay below roughly 0.85 to 0.90 or compressibility losses and noise rise sharply. At sea level that is a helical tip speed near 290 to 300 m/s, not 200, and because the aircraft's own forward speed contributes, the allowable rotational tip speed falls as the aircraft flies faster. That is the real reason turboprops are capped near M 0.6 to 0.65. It puts the propeller at only 1,000 to 1,500 rpm, so the gearbox ratio is typically 10 to 15:1.
A ramjet has no rotating components whatsoever. Compression comes entirely from the ram effect, the kinetic energy of the incoming supersonic flow converted to pressure through a series of shock waves in the intake. This only works above approximately Mach 2, which is why ramjets must be accelerated to operating speed by another propulsion system first.
A scramjet (supersonic combustion ramjet) goes further: it maintains supersonic flow throughout, including through the combustor. At Mach 5+, slowing the flow to subsonic for combustion would generate unacceptable total pressure losses and heating. Instead, fuel (usually hydrogen) must mix and ignite in a supersonic stream in milliseconds, an extraordinary engineering challenge. X-51A Waverider achieved sustained scramjet flight at Mach 5.1 in 2013.
A rocket carries both fuel and oxidiser. It does not need atmospheric oxygen. It works in vacuum, at any altitude, and at any speed. This is the only propulsion system that can reach orbit.
The rocket nozzle is a convergent-divergent (de Laval) nozzle. The convergent section accelerates subsonic flow to sonic (M = 1) at the throat. The divergent section then further accelerates supersonic flow, counterintuitively, a larger cross-section makes supersonic flow go faster. The nozzle converts thermal energy (high pressure, high temperature combustion products) entirely into kinetic energy.
Propellant combination determines Isp. LH₂/LOX gives highest Isp (450 s vacuum) but requires cryogenic storage and has very low density. Kerosene/LOX (Falcon 9 Merlin, 311 s SL) is denser, easier to handle. Methane/LOX (SpaceX Raptor, ~380 s vacuum) is the emerging choice for reusability, cleaner burning than kerosene, easier to produce on Mars.
How do all five engine types compare across the metrics that actually matter? This is the chart that should be on every propulsion student's wall.
The Tsiolkovsky equation, staging, orbital insertion, Hohmann transfers, and reentry: what happens after the engine fires.
Inlet aerodynamics, nozzle flows, and the thrust/drag balance: propulsion and aerodynamics are inseparable.
Simulate combustor flows, nozzle expansion, and inlet shock structures: CFD is the primary design tool for modern engines.
Thrust levels, fuel burn rates, and how engine performance shapes aircraft trajectory and range.
Bypass ratio optimisation, scramjet combustion analysis, and turbine cooling studies: propulsion project ideas with methodologies.
New to propulsion? Start with the broader aerospace engineering disciplines and how propulsion fits into the full picture.
SheCodes Lab teaches Python and C++ from scratch, side by side, free, no experience needed. Includes an engineering module covering NumPy, pandas, ISA models, cost index, and flight data analysis. The same tools used to build the calculators on this site.
shecodeslab.com →