15.1 — Fundamentals

The gas turbine engine is the powerplant of virtually every modern commercial and military aircraft. Understanding its fundamental principles — the conversion of chemical energy in fuel into useful thrust or shaft power — is the cornerstone of all subsequent engine study. This section covers the physics that make gas turbine engines work, and the four main engine configurations you will encounter in aviation maintenance.

Everything that follows rests on one idea: a gas turbine is an energy converter, and the substance it works on is air. Air is the working fluid — the medium that is drawn in, compressed, heated, expanded and thrown out again. Kerosene is the fuel: it is the source of the energy, but it is burned in the working fluid rather than being the working fluid itself. The distinction is not a quibble, because it explains the proportions inside the engine. Air enters a large civil engine at hundreds of kilograms per second while fuel enters at a few kilograms per second, so the fuel adds less than about two per cent to the mass passing down the gas path. Almost everything the engine handles, accelerates, heats and finally ejects is air.

The conversion runs in a fixed order, and that order is itself examined. Chemical energy is locked in the molecular bonds of the fuel. Burning releases it as heat energy, which raises the temperature — and therefore the internal energy — of the compressed air passing through the combustor. That hot, high-pressure gas is then expanded, which converts part of its heat and pressure energy into kinetic energy of fast-moving gas and part into mechanical work at a turbine shaft. Finally the kinetic energy of the departing jet, or the mechanical work delivered to a propeller or a rotor, appears as propulsive thrust. The chain runs chemical to heat to kinetic to mechanical and thrust, and it never runs the other way: nothing in the engine turns kinetic energy of the intake air back into chemical energy, and nothing drives the compressor from an external electrical supply once the engine is running.

No step in that chain creates energy. The principle of conservation of energy holds throughout: the energy entering the engine as fuel, plus the energy the incoming air already carries, exactly equals the energy leaving it. What leaves is a mixture of useful output — thrust or shaft power — and unavoidable losses: hot exhaust gas carrying heat away, residual kinetic energy in the jet that is never used, noise, and heat rejected into the oil system and the bypass air. A gas turbine is not a perfect machine and cannot be, so the question an engineer actually asks is never whether energy is lost but where it goes, how much of it can be recovered as useful work, and what a change in that balance is telling us about the engine's health.

Energy, Force and Newton's Laws

Potential Energy

Potential energy is the energy a body possesses because of its position or state. In gas turbine theory, we are mainly concerned with two forms:

• Gravitational potential energy — energy due to height above a reference datum. An aircraft at altitude has potential energy \( E_p = mgh \), where \( m \) is mass (kg), \( g \) is gravitational acceleration (9.81 m/s²), and \( h \) is height (m).
• Chemical potential energy — energy stored in the molecular bonds of aviation fuel. When kerosene (Jet A-1) burns, each kilogram releases approximately 43 MJ of heat energy. This is the primary energy input to every gas turbine engine.

A third form matters inside the engine itself, and it is easy to overlook because nothing is being lifted: pressure energy. Air that has been squeezed by the compressor is in a raised state rather than a raised position, but it holds energy in exactly the same sense — it will do work on anything that lets it expand. Potential energy is therefore best defined as energy stored by virtue of position or condition, a definition wide enough to cover height, chemical state and pressure alike. Compressor delivery air on a large engine sits at forty or more times atmospheric pressure and several hundred degrees Celsius, and it is that stored pressure energy, topped up by the heat of combustion, that the turbine converts back into work a short distance further downstream.

A fourth form appears constantly on the flight line without ever being called potential energy: elastic strain energy, stored in a component that has been deliberately deformed. A torqued bolt is a stretched spring, and the clamping force holding a flanged joint together is the bolt trying to return to its original length. That is why a fastener tightened by an angle-of-turn or bolt-stretch method is controlling stored strain energy directly rather than relying on the friction relationship that a torque wrench measures, and why a bolt that has been stretched past its yield point can look perfectly serviceable while storing far less energy than the joint needs.

Gravitational potential energy is easy to dismiss as classroom material, but it is the reason an aircraft can trade height for distance, and putting a number on it is instructive. An aircraft of mass 70,000 kg cruising at 11,000 m holds \( E_p = mgh = 70{,}000 \times 9.81 \times 11{,}000 \), which is approximately 7,554 MJ relative to sea level. Divide that by the 150.5 MW calculated above and the entire height of the aircraft is worth about fifty seconds of one engine at take-off power. Far more energy sits in the fuel tanks than in the aeroplane's altitude, which is why a drift-down after an engine failure buys range but never buys very much of it.

Chemical potential energy deserves one more remark, because its sheer density is the reason aviation looks the way it does. At roughly 43 MJ per kilogram, kerosene stores something like fifty times as much energy per kilogram as a typical rechargeable battery cell. An engine that can convert that energy at high power-to-weight is therefore not merely convenient; it is what makes long-range flight possible at all, and it is the obstacle every electric and hybrid propulsion concept has to confront.

Kinetic Energy

Kinetic energy is the energy a body possesses because of its motion. For a mass \( m \) moving at velocity \( v \):

In a gas turbine engine, kinetic energy appears in the high-velocity exhaust gases leaving the nozzle. A turbojet produces thrust almost entirely by accelerating a relatively small mass of air to a very high velocity. A turbofan, by contrast, accelerates a much larger mass of air to a lower velocity — this is more efficient because kinetic energy increases with the square of velocity, meaning doubling the mass at the same velocity gives twice the momentum (thrust) but only twice the energy, while doubling the velocity at the same mass gives twice the momentum but four times the energy cost.

Kinetic energy is not confined to the jet pipe, because the gas is moving everywhere inside the engine. Air arrives at the compressor face already travelling at well over a hundred metres per second; it is accelerated by every rotor row and decelerated by every stator row; it is slowed sharply in the pre-combustor diffuser so that a flame can be held in it; and it is accelerated hard again through the turbine nozzle guide vanes. What the engine does, over and over, is trade kinetic energy against pressure energy. Useful output only appears at the very end, when the last of those trades leaves the gas travelling rearwards faster than it was travelling when it arrived.

The energy Engine A spends extra is genuinely wasted, and it is worth being precise about where it goes. Kinetic energy left in the departing jet does no work on the aircraft; it is simply handed to the atmosphere and stirred away as turbulence and heat. The standard measure of how much of the energy actually reaches the aircraft is propulsive efficiency, which for a simple jet is:

Put the two engines above on an aircraft flying at 250 m/s. Engine A adds 600 m/s to the flow, so \( V_j = 850 \) m/s and \( \eta_p = 500/1100 = 0.45 \), about 45%. Engine B adds only 100 m/s, so \( V_j = 350 \) m/s and \( \eta_p = 500/600 = 0.83 \), about 83%. Propulsive efficiency therefore rises as the jet velocity is brought closer to the aircraft's own speed. It would reach 100% only if the jet were left at exactly aircraft speed — which would also mean no velocity change and therefore no thrust at all. Every real engine is a compromise between those two ends, and where a designer sits on that compromise is decided by the speed the aircraft is meant to fly at.

Two distinct quantities are being traded against one another here, and naming them makes the trade easier to hold on to. The useful output is a change of momentum, \( mv \); the bill is paid in kinetic energy, \( \tfrac{1}{2}mv^2 \). One of them is linear in velocity and the other quadratic, and that mismatch is the whole of the argument for moving more air gently rather than less air hard. Every change in engine architecture since the 1960s — larger fans, higher bypass ratios, geared fan drives, open rotors — is an attempt to move further along that curve, and the thing that stops it going further each time is not the thermodynamics but the weight, diameter and drag of the fan needed to do it.

Newton's Laws of Motion

All three of Newton's Laws are fundamental to understanding gas turbine thrust production:

Reading the Three Laws Inside a Real Engine

The First Law is usually taught as a statement about air, but on the flight line it bites hardest as a statement about rotating mass. A large fan and its shaft represent a very substantial rotational inertia, and inertia resists a change of motion in both directions equally. That is why a big turbofan takes several seconds to accelerate from flight idle to go-around thrust: the fuel schedule cannot simply be opened, because the gas generator has to be accelerated against its own inertia while keeping enough compressor surge margin to survive doing it. It is why the same engine takes minutes rather than seconds to run down after shutdown. And it is why free rotation of the compressor is confirmed before a start is attempted — a rotor that will not turn freely is telling you something inside the engine has changed, and the starter has no way of telling you gently.

The Second Law is the one that produces the numbers. In its schoolroom form it is \( F = ma \): a force applied to a mass produces an acceleration. An engine, however, does not accelerate one fixed lump of air; it accelerates a continuous stream of it. Written for a steady flow it takes the form already given in the table above, \( F = \dot{m}\,\Delta v \) — mass flow rate multiplied by the change in velocity. The two statements are not rivals and neither is an approximation of the other: \( \dot{m} \) is mass per second and \( \Delta v \) is the velocity change that mass receives, so their product is still mass multiplied by acceleration, merely accounted for per second rather than per lump. Both phrasings appear in examination questions as the correct description of thrust, and a candidate who has understood that they are the same statement will recognise either.

Only two levers therefore exist for producing thrust: raise the mass flow, or raise the velocity change. Every configuration described later in this note is a different answer to the question of how to divide the work between those two. Raising mass flow means a bigger engine or a bigger fan, which costs diameter, weight and nacelle drag. Raising velocity change means burning more fuel per newton, for the reasons the worked example above set out. There is no third lever, and no engine escapes the trade.

The Third Law is the one that repays careful reading. Thrust is the reaction to accelerating a mass of gas rearwards. It is not the exhaust pushing against the outside air, and it is not the exhaust pushing against the ground: a rocket works perfectly in a vacuum, and an engine on a test stand produces its rated thrust with nothing behind it but open air. The reaction force appears on the engine's own internal surfaces — on the fan and compressor blades, on the combustor casing, on the turbine nozzle guide vanes, on the internal walls of the intake — as the pressure differences that accelerate the gas act back on the hardware creating them. Add every one of those internal loads together and the resultant is the forward force we call thrust. The same correction applies to a propeller: a turboprop's thrust is the reaction to the prop-wash being accelerated rearwards, not the prop-wash striking the air behind it. In examination wording, impingement on the surrounding air is always the wrong answer.

Because that resultant is distributed over the engine's internal structure, it has to be collected and delivered to the airframe. It travels through the engine casings into the thrust links or thrust-carrying mount, and from there into the pylon and the wing or fuselage structure. That is the practical reason engine mount hardware is inspected for cracking, torqued to specification and, on many types, life-limited: the mounts are the only path by which the reaction force reaches the aircraft. A turboprop or turboshaft installation adds a second reaction to restrain — the torque reaction from driving the propeller or rotor, which tries to rotate the engine and its gearbox in the direction opposite to the driven shaft, and which the mounts and torque struts have to hold.

Rotating mass produces one further reaction that has to be designed for, and it is a direct consequence of the First Law applied to rotation. A spinning rotor resists any attempt to change the direction of its axis, and if it is forced to change direction it responds with a couple at right angles to the applied one. Every engine rotor is therefore a gyroscope, and when the aircraft pitches or yaws, the mounts have to react gyroscopic loads on top of thrust and weight. This is one of the reasons an engine mounting system is a more complicated structure than it looks, why manoeuvre loads appear in mount fatigue calculations, and why the certification case for an engine includes the loads produced by a windmilling rotor that has become unbalanced after a blade release, when both the out-of-balance force and the gyroscopic couple are at their worst.

Reading the Thrust Equation Term by Term

That equation repays being taken apart slowly, because almost everything in engine performance is a consequence of one of its four quantities. The first part of the first term, \( \dot{m}V_j \), is the momentum thrust: the rate at which momentum leaves the nozzle. The subtracted part, \( \dot{m}V_0 \), is the momentum drag: the air already possessed momentum before it reached the engine, and the engine only earns credit for the momentum it adds. The final term, \( A_e(P_e - P_0) \), is the pressure thrust: if the gas is still above ambient pressure as it crosses the exit plane, that excess pressure acting over the exit area contributes force as well.

Momentum drag is the whole reason thrust and airspeed are connected. On a static ground run \( V_0 \) is zero, none of the momentum thrust is given back, and the reading on the test-bed load cell is the largest number the engine will ever produce at that setting. Push the same engine forward through the air and \( V_0 \) grows, the subtraction grows with it, and net thrust falls even though nothing inside the engine has changed.

The pressure term needs a word of explanation because it is often quietly ignored. A simple convergent nozzle can only accelerate gas up to the local speed of sound at its throat; once that happens the nozzle is choked and no further increase in velocity is possible there, however much pressure is available upstream. Surplus pressure therefore has nowhere to go and simply arrives at the exit plane still above ambient, where it acts over the exit area and contributes force. Most civil engines run choked at take-off and climb power, so the pressure term is producing real thrust rather than being a mathematical curiosity.

Air, the Gas Laws and Bernoulli's Theorem

Newton's laws explain what thrust is. The gas laws explain what the engine has to do to the air in order to produce it. Every pressure rise across a compressor stage, every temperature reading on a flight-deck gauge, and every performance shortfall on a hot day at a high-altitude airfield is a consequence of the behaviour set out below. It is also the material that turns a memorised order of the gas path into an explanation of why it has to be in that order.

Air as the Working Fluid

Dry air is approximately 78% nitrogen and 21% oxygen by volume, with argon, carbon dioxide and traces of other gases making up the remainder, plus a variable quantity of water vapour. Only the oxygen takes part in combustion — about a fifth of the air by volume — and even that is only partly consumed, because far more air is drawn in than is needed to burn the fuel. The surplus oxygen and all of the nitrogen are not passengers. They are the mass that carries the heat away from the flame, the mass that is expanded through the turbine, and the mass that is finally accelerated to make thrust. That is why the correct answer to "what is the working fluid of a gas turbine?" is air, and why kerosene, which is merely the energy source, is not.

Over the range of temperatures and pressures a turbine engine works in, air behaves closely enough to an ideal gas for the ideal-gas relationships to be used with confidence. Two of its properties appear in almost every calculation and are worth fixing in mind: the specific gas constant for air, \( R = 287 \) J per kilogram per kelvin, and the ratio of specific heats, \( \gamma \), which is about 1.4 for cold air and falls to roughly 1.33 for hot combustion gas. That change is not a detail. It is one of the reasons the compressor and the turbine cannot be treated as mirror images of one another, and it is why a turbine handling hot gas extracts a different amount of work per unit of pressure ratio than a compressor of the same pressure ratio absorbs.

Mass, Weight and Why the Words Get Mixed Up

Mass is the quantity of matter in a body, measured in kilograms, and it does not change with location. Weight is the force that gravity exerts on that mass, \( W = mg \), measured in newtons, and it does change with location because \( g \) does. A 10 kg component has a mass of 10 kg on the ground, in the air and in orbit; its weight is about 98 N on the ground and effectively nothing in orbit. In Newton's second law and in every gas law it is mass that appears, never weight.

Engineering practice nonetheless uses the two words loosely, and the usage is so entrenched that examinations reproduce it — density, for instance, is very commonly defined as "weight per unit volume" when what is strictly meant is mass per unit volume. The two definitions differ only by the constant \( g \), so at any one location they are proportional and no practical error results. Recognise the loose usage rather than fighting it: if a question offers "weight divided by volume" as the definition of density, that is the answer it wants, and the alternatives on offer will be specific volume or something that is not a density at all.

Density, Pressure and Temperature

Density is mass per unit volume — weight divided by volume — written \( \rho = m/V \), with the kilogram per cubic metre as its SI unit. Its inverse, volume divided by mass, is specific volume; offering volume over mass as the definition of density is a standard distractor and offering pressure over volume is not a density relationship at all. At International Standard Atmosphere sea level conditions, 15°C and 1013.25 hPa, the density of air is 1.225 kg per cubic metre.

Density is the atmospheric property that matters most to an engine, because a compressor is essentially a volumetric machine: at a given rotational speed it swallows a roughly fixed volume of air per second, so the mass it passes is proportional to the density of what it is swallowing. Since thrust is proportional to mass flow, anything that reduces density reduces thrust — and both altitude and ambient temperature reduce it, for reasons that come out of a single relationship:

Two absolute temperature scales are in use and both appear in engine documentation. The kelvin is the SI scale and is obtained by adding 273.15 to a Celsius temperature. The degree Rankine is its imperial counterpart, obtained by adding 459.67 to a Fahrenheit temperature, and it still turns up in North American engine manuals and performance charts. Both start at absolute zero, which is what makes them valid in the gas laws; Celsius and Fahrenheit do not, and substituting either into a gas law gives a wrong answer.

The Standard Atmosphere and Why Altitude Costs Thrust

Because engine performance depends on the state of the air presented to it, that state has to be defined by agreement before any two engines can be compared. The International Standard Atmosphere (ISA) does that. At sea level it specifies 15°C, 1013.25 hPa and a density of 1.225 kg per cubic metre. Temperature then falls at a lapse rate of about 1.98°C per 1,000 ft, or 6.5°C per kilometre, up to the tropopause at 11,000 m — approximately 36,000 ft — above which the standard temperature stays constant at about −56.5°C. Pressure falls continuously throughout, and much faster than temperature does.

The same reasoning explains the effect of a hot day, but with a caution about how much of the effect the gas laws alone account for. At sea level on an ISA+30 day the temperature is 45°C, or 318.15 K, so \( \rho = 101{,}325/(287 \times 318.15) = 1.110 \) kg per cubic metre. That is about 9% below the standard value, and by itself it would cost roughly 9% of the mass flow and therefore of the thrust. The thrust actually lost on such a day is considerably greater than that, because two further effects pile on top of the density loss: hotter air requires more work to compress for the same pressure ratio, and the raised compressor delivery temperature brings the engine up against its turbine temperature limit sooner. Density is the first of three contributors, not the whole story, and the full accounting belongs with engine performance.

Boyle's Law, Charles's Law and the General Gas Law

The ideal gas equation is the general case. The three named laws are simply what it reduces to when one of the three properties is held constant, and each of them describes something real that happens somewhere in the engine.

Specific Heat and the Heat Added in the Combustor

One more property is needed before the cycle can be worked in numbers: the specific heat capacity, which is the quantity of heat required to raise one kilogram of a substance by one kelvin. A gas has two of them, because it can be heated either while it is free to expand or while it is held at fixed volume. Heating at constant pressure requires more heat, because some of the energy goes into the work of pushing the surroundings back as the gas expands, and its value is written \( c_p \). Heating at constant volume requires less, because no expansion work is done, and its value is written \( c_v \). The difference between them is exactly the gas constant, \( c_p - c_v = R \), and their ratio is the \( \gamma \) already met above, \( \gamma = c_p / c_v \). For air \( c_p \) is about 1.005 kJ per kilogram per kelvin, rising to roughly 1.15 for hot combustion gas.

Because a gas turbine adds its heat at constant pressure, it is \( c_p \) that governs the combustor, and the heat needed follows directly:

Continuity — Why the Gas Path Changes Shape

One further relationship governs the physical shape of the engine, and it follows directly from the fact that mass cannot be created or destroyed. In a steady flow, the same mass must pass every cross-section of the duct in every second:

Adiabatic, Isentropic and Isothermal Processes

Three words describe the way a gas is squeezed or allowed to expand, and the examination tests all three against one another.

• Adiabatic — no heat crosses the boundary of the gas. None is added to it from outside and none is taken away.
• Isentropic — adiabatic and reversible, that is, adiabatic with no losses. This is the idealised process used in cycle calculations. A real compressor is close to adiabatic but never truly isentropic, because friction, turbulence, tip leakage and shock losses all add entropy.
• Isothermal — temperature held constant. For a gas being compressed this can only be achieved by continuously removing the heat of compression as it is generated, which requires a cooling arrangement that an aero-engine compressor does not have and could not carry.

Compression in a gas turbine is treated as adiabatic, and the justification is simply speed. The air is through the compressor in a few hundredths of a second, far too quickly for any meaningful quantity of heat to be conducted away through the casings, so essentially all the work done on the air stays in the air. That work raises the internal energy of the gas, and raised internal energy shows itself as raised temperature. Compressor delivery temperature therefore rises because work has been done on the air — not because heat has travelled forward from the combustor through the casings, which is negligible by comparison, and not because the air rubs against the casing wall, which contributes almost nothing.

Bernoulli's Theorem and the Shape of a Duct

Bernoulli's theorem states that the total energy of a flow remains constant along a streamline. It does not say that static and dynamic pressure are equal, and it does not say that one of them is always larger than the other. It says that their sum is fixed, so that whatever one gains the other must give up. Three quantities are involved:

• Static pressure — the pressure the gas exerts on the walls of the duct, and the pressure that an observer drifting along with the flow would measure. This is the pressure a compressor exists to raise.
• Dynamic pressure — the pressure equivalent of the flow's kinetic energy, \( \tfrac{1}{2}\rho v^2 \) at low speeds. It exists only because the gas is moving, and it disappears the instant the gas is brought to rest.
• Total pressure, also called stagnation or pitot pressure — the sum of the two. It is what a probe facing directly into the flow measures, because the flow is brought to rest at the probe mouth and all of the dynamic pressure reappears as static.

The practical consequence is the behaviour of ducts. In subsonic flow — which is where nearly every duct in a gas turbine operates — a convergent duct, whose cross-sectional area reduces, accelerates the gas and its static pressure falls. A divergent duct, whose area increases, decelerates the gas and its static pressure rises. A divergent duct used deliberately to slow a flow down and recover static pressure from it is called a diffuser. Total pressure, meanwhile, never rises in a plain duct: real ducts always surrender a little of it to friction and turbulence, and the quality of a duct is measured by how little total pressure it loses.

Bernoulli in the simple form above assumes constant density, which is a fair approximation at low speed and a poor one inside a compressor or a jet pipe, where density changes a great deal. The principle survives the correction; what changes is the bookkeeping, because the energy account then has to include the internal (thermal) energy of the gas alongside its pressure and velocity. The general statement that survives is the steady flow energy equation: in a duct with no work input and no heat transfer, the total energy of the gas — thermal plus kinetic — is unchanged from one end to the other. Slowing the gas down therefore raises its temperature as well as its pressure. This is why the air arriving at the compressor face of an aircraft in cruise is already measurably warmer than the outside air before a single blade has touched it, and it is the origin of the ram temperature rise that makes an intake total temperature probe read higher than the outside air temperature.

The Speed of Sound and Why It Matters Inside the Engine

Because the subsonic and supersonic rules are opposites, knowing where the local speed of sound lies is not optional. The speed of sound in a gas is \( a = \sqrt{\gamma R T} \) — it depends on the absolute temperature and nothing else for a given gas. It does not depend on pressure, and it does not depend on density independently of temperature. At standard sea level conditions, 288 K, that gives \( \sqrt{1.4 \times 287 \times 288} = 340 \) m/s. At the compressor delivery conditions calculated above, 901 K, it gives \( \sqrt{1.4 \times 287 \times 901} = 602 \) m/s.

That near-doubling has a direct engineering consequence. The rear stages of a compressor run in air where the speed of sound is very much higher than it is at the front, so their blades can be driven at high speeds without the flow over them going supersonic and shocking. It is also why the front fan of a large engine, working in cold air where the speed of sound is at its lowest, is the stage that runs closest to its aerodynamic limit — and why fan tip speed, not core speed, is usually what caps the rotational speed of the low pressure spool. Mach number, the ratio of local flow speed to the local speed of sound, is therefore the parameter that actually matters in a duct, and the same physical velocity can be comfortably subsonic in one part of the engine and supersonic in another.

The Brayton Cycle (Gas Turbine Thermodynamic Cycle)

Every gas turbine engine operates on the Brayton cycle (also called the Joule cycle). It is an open thermodynamic cycle with continuous flow, unlike the Otto cycle (petrol engines) or Diesel cycle which are intermittent. The four stages of the Brayton cycle correspond directly to the four main sections of a gas turbine engine:

What the Horizontal Axis Means

The diagram is plotted against entropy, and it is worth knowing what that quantity is rather than treating it as a label on an axis. Entropy is a measure of how much of a system's energy is no longer available to be turned into useful work. Add heat to a gas and its entropy rises; the amount it rises by depends on the temperature at which the heat was added, because \( \Delta S = Q/T \) for a reversible process — the same quantity of heat added at a high temperature raises entropy less than if it were added at a low one. That is the formal reason why a cycle that adds its heat at the highest practical temperature and rejects it at the lowest practical temperature is the most efficient one, and it is why turbine entry temperature matters as much as pressure ratio.

Two consequences make the diagram readable. First, on temperature-entropy axes the area beneath a process line is the heat transferred during that process, so the area under the 2 to 3 leg is the heat put in by the fuel, the area under the 4 to 1 leg is the heat thrown away in the exhaust, and the enclosed area of the cycle is the net work obtained. A cycle drawn taller or wider encloses more area and does more work. Second, entropy is created — never destroyed — by every real irreversibility: friction, turbulence, mixing streams at different temperatures, and unrestrained expansion. That is what the second law of thermodynamics says, and its practical meaning inside an engine is blunt: every unit of entropy generated is work that has been permanently lost and cannot be recovered downstream by any clever piece of design. Cleaning up a loss is always cheaper than trying to recover the energy it destroyed.

The Four Processes in Detail

The numbered points on the diagram are cycle states, not physical positions in the engine, and each numbered leg between them is a distinct thermodynamic process with a name of its own:

• 1 to 2 — adiabatic, ideally isentropic, compression. The compressor does work on the air. Pressure rises, temperature rises, volume falls, and in the ideal case entropy does not change, which is why the leg is drawn as a vertical line on temperature-entropy axes. No heat is added from outside; the entire temperature rise is the work put in.
• 2 to 3 — constant-pressure heat addition. Fuel is burned continuously in the combustor. Temperature climbs steeply and the gas expands in volume, but the pressure is held essentially constant because the gas is free to flow onwards rather than being sealed in. Entropy rises, because heat has been added, which is why this leg slopes to the right.
• 3 to 4 — adiabatic, ideally isentropic, expansion. The hot, high-pressure gas expands through the turbine and gives up energy as shaft work. Pressure and temperature fall; entropy ideally does not change, so this leg is drawn vertical as well.
• 4 to 1 — constant-pressure heat rejection. The exhaust gas leaves the engine and surrenders its remaining heat to the atmosphere at ambient pressure, closing the cycle so that it can begin again with fresh air at state 1.

One feature of the diagram is worth reading carefully, because it carries a real engineering consequence: the height of point 4 against point 2. The heat carried away along the 4 to 1 leg, everything between point 4 and point 1, is the single largest loss in the cycle, and it is the reason a gas turbine's thermal efficiency is nowhere near 100% however well its components are made. Whether any of it can be reclaimed depends on which of points 4 and 2 is the higher, and that in turn depends on the pressure ratio. The diagram here draws point 4 above point 2, which is the case at the modest pressure ratios of the early jet engines and of most stationary plant: the exhaust leaves hotter than compressor delivery, so a heat exchanger can return part of its heat to the air on its way to the combustor. Raise the pressure ratio at a given turbine entry temperature and point 2 climbs while point 4 falls. For the 1,700 K turbine entry temperature used earlier in this note the two cross at a pressure ratio of about 22:1, and at the 40:1 of a modern engine the ideal figures are 827 K at compressor delivery against 592 K at the fully expanded exhaust, so there is nothing left for a heat exchanger to move. That is why recovery by heat exchanger belongs to low pressure ratio stationary plant. A steam bottoming cycle carries no such restriction, which is how a large modern industrial machine still takes useful work from its exhaust; an aero engine can carry neither installation, so it recovers only what the exhaust can still contribute as jet thrust.

The Same Cycle on Pressure-Volume Axes

The same four processes can be drawn on pressure against volume instead of temperature against entropy, and the two views answer different questions. On pressure-volume axes the compression leg 1 to 2 climbs steeply to the left as pressure rises and volume falls. Heat addition, 2 to 3, runs horizontally to the right — that horizontal line is what "constant pressure" looks like, and it is the single most recognisable feature of the diagram, because on the Otto cycle of a piston engine the equivalent line is vertical: constant volume, with pressure shooting up. Expansion, 3 to 4, falls to the right as the gas gives up pressure and grows in volume, and heat rejection, 4 to 1, returns horizontally at ambient pressure.

On these axes the enclosed area is again the net work of the cycle, but now it is the work done by the gas on the machine rather than the heat balance. Between them the two diagrams make the same point from two directions: the temperature-entropy plot shows where the heat goes, and the pressure-volume plot shows where the work is done. If only one of them is remembered, the useful one for identifying a cycle from a picture is pressure against volume, because the orientation of the combustion line — horizontal for Brayton, vertical for Otto — settles the identification immediately.

Open, Continuous and Constant-Pressure — Three Different Claims

Three descriptions of the Brayton cycle appear repeatedly in examination questions, and each is answering a different question about it. Confusing them is easy, because all three are true at once.

• Open describes the working fluid. It is not recirculated: fresh air is taken from the atmosphere at one end and exhaust gas is discharged to the atmosphere at the other, and the atmosphere itself provides the heat-rejection leg from 4 back to 1. A closed cycle would need a heat exchanger to cool the same fluid back down and return it to the compressor inlet, which is how a closed-cycle industrial or nuclear gas turbine works and is not how any aero engine works.
• Continuous describes the flow. Nothing is trapped, valved or timed. There is not a single valve anywhere in a gas turbine's main gas path, which is one of the reasons it has so few moving parts compared with a piston engine of the same power.
• Constant pressure describes the heat addition, and it is the property that gives the cycle its other name, the constant pressure cycle. It is the single feature that most sharply distinguishes it from the piston engine's constant volume Otto cycle, in which the charge is sealed inside a cylinder by closed valves while it burns, so that pressure rockets up while volume barely changes at all.

Describing the heat addition as taking place at constant pressure carries a qualification that should not be glossed over. A practical combustor loses a small fraction of the compressor delivery pressure to friction, to the pressure drop across the swirlers and dilution holes, and to the momentum change involved in heating the gas. The pressure at the combustor exit is therefore slightly lower than at its inlet, typically by a few per cent, not identical to it. The cycle is still called constant-pressure because that small drop is negligible beside the forty-fold rise the compressor has just produced — and because the alternative label, constant-volume, would be spectacularly wrong.

Where the Turbine's Work Actually Goes

A figure worth pausing on is how much of the turbine's output never reaches the aircraft at all. The turbine's first duty is to drive the compressor, and compressing air is expensive. On a simple turbojet the turbine is sized to extract little more than the compressor and accessories need, and on a modern high pressure ratio core the power the compressor demands is itself of the order of two thirds of the rate at which the fuel releases heat — a figure derived from first principles in the worked example below. That power circulates rather than disappears, because the turbine takes it out of the gas and the compressor puts it straight back into the same gas stream, so it is a measure of how hard the turbine has to work and not of how much of the fuel's energy has been used up. A further small slice is taken through the accessory gearbox for the fuel pump, the oil pumps, the electrical generators and the hydraulic pumps. Only what survives all of that is available as useful output.

What "useful output" means is the single choice that separates the four engine configurations described later in this note, and it is worth stating as a rule. If the turbine is sized to extract only what the compressor and the accessories need, then the surplus energy stays in the gas as pressure and heat, and the propelling nozzle converts it into a fast jet — that is a turbojet. If extra turbine stages are added so that more work is taken out than the compressor needs, the surplus appears at a shaft, where it can drive a fan, a propeller or a rotor, and correspondingly less energy is left in the exhaust to make a jet. There is no third possibility, and no way to have both: every joule the turbine takes out of the gas is a joule the nozzle cannot use, and every joule left in the gas is a joule the shaft does not get.

Pressure Ratio and Thermal Efficiency

The thermal efficiency of an ideal Brayton cycle turns out to depend on one thing only — the pressure ratio the compressor achieves:

Pressure ratio cannot simply be raised without limit, and the reason is a competing effect that governs the whole of modern engine design. Raising the pressure ratio raises the compressor delivery temperature along with it; the worked example earlier in this note put a 40:1 machine at around 900 K at delivery. That matters twice over. First, compressor delivery air is the coolant for the turbine, so as it becomes hotter it becomes less effective as a coolant, and a greater quantity of it has to be bled off to do the same job — air that is then unavailable to make thrust. Second, the temperature window available for heat addition, which is the gap between the compressor delivery temperature and the highest temperature the turbine can survive, is squeezed from below. Higher pressure ratio improves the cycle; higher compressor delivery temperature penalises the components. The pressure ratio actually selected is the point at which those two lines cross for the materials and cooling technology of the day, which is why it has crept upwards only as turbine materials and blade cooling have improved.

The ceiling at the top of the cycle is the turbine entry temperature — the gas temperature at state 3, where the flow first meets the turbine. On modern civil engines it works across a band of roughly 1,200–1,600°C, sitting near the lower end in the cruise and rising to the upper end at take-off power, and the most highly rated designs exceed 1,600°C at take-off. The top of that band is above the melting point of the alloys the turbine is made from, which is only possible because the blades are internally cooled and film-cooled with compressor air and protected by ceramic thermal barrier coatings. Turbine entry temperature is therefore not a thermodynamic limit at all. It is a materials and cooling limit, and it is the number that has risen furthest as engines have improved.

The Absolute Ceiling: the Carnot Limit

Before leaving efficiency, it is worth knowing what the ultimate ceiling is, because it explains why engineers push turbine entry temperature so hard. No heat engine of any design can convert more of its heat input into work than:

Take a turbine entry temperature of 1,700 K and an ambient of 288 K: \( \eta_{carnot} = 1 - 288/1{,}700 = 0.83 \), so 83%. No engine working between those two temperatures can beat that, whatever its pressure ratio and however perfect its components. The ideal Brayton cycle at 40:1 reached about 65% of the heat input, and a real engine achieves appreciably less again. Each step down from 83% is paid for by a specific, identifiable cause: the Brayton cycle does not add all its heat at the peak temperature the way the ideal Carnot cycle does, and then component losses, cooling air and pressure drops take their own share.

The useful lesson is in the shape of the formula. \( T_{cold} \) is fixed by the atmosphere and cannot be changed. The only variable available is \( T_{hot} \), which is turbine entry temperature — so every kelvin gained there raises the ceiling for everything downstream. That is why turbine cooling technology, single-crystal blade casting and thermal barrier coatings have received so much development effort: they do not improve the engine directly, they raise the temperature at which the whole cycle is permitted to operate.

Why Small Efficiency Changes Have Large Effects

A gas turbine has an unusual and somewhat uncomfortable property, and it follows directly from how much of the energy the compressor consumes: wherever the turbine is sized to produce surplus shaft work, that surplus is the small difference between two large numbers. The turbine produces a great deal of work and the compressor consumes most of it, so anything that changes either one is magnified in what is left over. Engineers call the fraction of turbine work consumed by the compressor the back work ratio, and for a gas turbine it is very high — a steam plant, by comparison, spends only a per cent or two of its turbine output on the feed pump, because pumping a liquid is vastly cheaper than compressing a gas.

Efficiencies, and What Each One Measures

Several different efficiencies are used about a gas turbine and they are not interchangeable. Each answers a different question, and quoting the wrong one is a common source of confusion.

Following the Fuel Energy Through the Engine

Putting the pieces together, it is possible to say roughly what becomes of the 150.5 MW of chemical energy released by the fuel flow calculated earlier in this note. The exact split depends on the engine, the power setting and the flight condition, so the ordering below is the one that holds at the take-off condition that fuel flow belongs to, and the proportions should be read as indicative only.

• The largest single share leaves as hot exhaust gas. This is the 4 to 1 leg of the cycle: heat rejected to the atmosphere because the cycle has no way to use gas that has already expanded to ambient pressure. It is the loss the Carnot limit is really describing.
• The next largest share is kinetic energy left in the jet that never acts on the aircraft, which is exactly what propulsive efficiency measures and exactly what a high bypass ratio is designed to reduce.
• Useful propulsive work — thrust multiplied by flight speed — is what remains after those two, and it is a minority of the input in every flight condition.
• Smaller identifiable losses make up the rest: heat carried away in the oil from bearings and gearboxes, power taken by the accessory drives for the aircraft's electrical and hydraulic systems, air bled from the compressor for cabin conditioning and anti-icing, pressure losses in the intake, ducts and combustor, and acoustic energy radiated as noise.

The middle two items are the pair that moves. They sit in that order at take-off, where the aircraft is barely moving and propulsive efficiency is therefore near zero, so almost all of the kinetic energy put into the jet is wasted. In the cruise they change places: the propulsive efficiency worked example earlier in this note puts Engine B at 83% at 250 m/s, which is five parts of useful propulsive work for every one part left behind in the jet. The first item holds its position at every condition, and useful propulsive work stays a minority of the fuel energy at all of them.

Two things follow from that ordering. The first is that anything the airframe takes from the engine — bleed air, generator load, hydraulic demand — is paid for in fuel, which is why bleed and electrical loads are managed rather than ignored on a modern aircraft. The second is that the two largest losses are both thermodynamic rather than mechanical, which is why engine development has been dominated for sixty years by pressure ratio, turbine entry temperature and bypass ratio, and not by better bearings.

The Real Cycle Against the Ideal One

The diagram above shows the ideal cycle, in which the compression and expansion legs are exactly vertical. A real cycle drawn on the same axes leans to the right on both of them, because every irreversibility — blade friction, tip leakage, shock losses, turbulence, the mixing of cooling air into the mainstream — adds entropy. Once that leaning is seen, the consequences are entirely predictable:

• A real compressor delivering a given pressure ratio arrives at a higher exit temperature than the ideal one, so it absorbs more work than the ideal calculation predicts.
• A real turbine expanding through a given pressure ratio also arrives at a higher exit temperature than the ideal one, so it delivers less work than the ideal calculation predicts.
• Both errors act in the same unhelpful direction, so the real cycle's net work output and thermal efficiency are always below the ideal figures, never above.

That asymmetry — losses cost you work at the compressor and then cost you work again at the turbine — is why component efficiencies of a percentage point or two are pursued so hard by manufacturers, and it is why an engine whose compressor is fouled with dust and insect debris, or whose turbine blade profiles have eroded, shows up as raised fuel burn and raised exhaust gas temperature long before it shows up as any loss of thrust the crew would notice.

Cycle States Against Engine Station Numbers

One source of confusion is worth clearing up before it takes root: the numbers 1 to 4 used for the cycle states above are not the same numbers used to label physical positions along the engine. Engine documentation uses a station numbering convention in which station 0 is the undisturbed free stream ahead of the aircraft, station 1 is the intake entry, station 2 is the fan or compressor face, station 3 is compressor delivery at the combustor inlet, station 4 is turbine entry at the combustor outlet, station 5 is turbine exit, and higher numbers run aft through the jet pipe to the nozzle. Manufacturers differ in the sub-numbers they insert between these — an intermediate compressor exit might be station 2.5 on one engine and something else on another — so read the numbering from the manual for the engine in front of you rather than assuming it.

The practical value of the convention is that it makes flight-deck and maintenance parameters unambiguous. A temperature labelled T3 is compressor delivery temperature and nothing else; a pressure labelled P3 is compressor delivery pressure measured at that one plane. When a troubleshooting chart asks for a P3 reading it is asking for the pressure at a specific station, and taking it from a convenient nearby tapping instead will produce a plausible number and a wrong diagnosis.

The last three rows of the efficiency table earlier in this section explain something counter-intuitive about the pure turbojet. Its cycle is perfectly respectable; what lets it down at subsonic speed is the second multiplier, its propulsive efficiency. An engine can be excellent at one and poor at the other, and when the high bypass turbofan displaced the turbojet on transport aircraft it did so without improving the cycle at all — it improved only the way the cycle's output was delivered to the aircraft.

Force, Work, Power and Energy

The table gives the definitions. The distinctions between them are what the examination actually tests.

Force Is Not Work, and Work Is Not Power

A force that moves nothing does no work. Push against a parked aircraft with all your strength and, if it does not move, the work done on it is exactly zero however tired you become. Work appears only when the force acts through a distance, which is why its formula is \( W = Fd \) and why its unit, the joule, is identical to the unit of energy — work is simply energy transferred by a force. Power then adds time to the account: it is the rate at which work is done, measured in watts, where one watt is one joule per second. The three quantities cannot be interchanged, and their units are a reliable check on any answer: newtons for force, joules for work and energy, watts for power. The horsepower is still met alongside the watt in engine work, and one horsepower is 745.7 watts.

Shaft Power, Torque and Rotational Speed

For any rotating shaft — a turboprop output shaft, a helicopter rotor drive, an accessory drive — power is delivered as the product of torque and rotational speed. The same power can therefore be delivered as a large torque at low speed or a small torque at high speed, and that equivalence is the entire purpose of a reduction gearbox: it converts a turbine's high-speed, low-torque output into the low-speed, high-torque input that a propeller or rotor needs, at very nearly constant power, losing only what friction takes. It is also why torque is the parameter used to set power on a shaft-power engine. With the propeller governed at a constant speed, torque is directly proportional to power, so a torque gauge is a power gauge — which is not true of a thrust-producing engine, where no single mechanical parameter stands in for power in the same way.

Velocity, Speed and Acceleration

Speed is a scalar: a magnitude with no direction attached. Velocity is a vector: a magnitude and a direction. The distinction is not pedantry inside an engine, because a change of direction with no change of speed is still a change of velocity and therefore still demands a force. Gas turned through a large angle by a row of turbine nozzle guide vanes has been accelerated even if it leaves at exactly the speed it entered, and the force that turned it is felt by the vanes as a load their attachments and locating features have to carry. This is one reason nozzle guide vanes are among the most heavily loaded static components in the engine despite doing no shaft work at all.

Acceleration is the rate of change of velocity with respect to time, measured in m/s². It is neither a force nor an average speed — a force is what produces an acceleration, through Newton's second law, and the two must never be swapped in an answer. The accelerations inside a propelling nozzle are extraordinary by everyday standards, and it is worth putting a number on one.

One more quantity completes the set. Momentum is mass multiplied by velocity, measured in kilogram metres per second, and it is what Newton's second law is really about: force is the rate at which momentum changes. For a steady stream of gas, the momentum crossing any plane each second is \( \dot{m}v \), so the thrust equation given earlier is nothing more than the momentum leaving the engine minus the momentum entering it, plus the pressure term. Reading it that way makes it considerably harder to misremember, and it makes clear why raising the mass flow and raising the velocity change are equally valid routes to the same thrust.

Units, Symbols and the Conversions Worth Knowing

Engine work is carried out in a mixture of SI and imperial units, because manufacturers, operators and regulators in different parts of the world settled on different conventions and the documentation has never been fully harmonised. Being fluent in both is a practical requirement rather than an academic nicety: a thrust setting chart, a torque limit and a pressure gauge on the same aircraft may each use a different family of units.

Engine Architecture: Gas Path, Spools and Modules

The cycle described above has to be built out of hardware, and the way that hardware is arranged is as much a part of the fundamentals as the thermodynamics. Two engines can run near-identical cycles and still be very different machines to operate and maintain, depending on how many shafts they have, where the drives are taken from, and whether the major assemblies can be separated from one another without dismantling everything else.

The Gas Path and the Order of the Sections

Air passes through a basic turbojet in one fixed order, front to rear: intake, compressor, combustion chamber, turbine, exhaust nozzle. That order is not a design preference; it is forced by the cycle. The compressor must come before the combustor, because there is no point burning fuel in air that has not been compressed — pressure ratio is what makes the cycle efficient at all. The turbine must come after the combustor, because it is hot gas that carries the energy to drive it. The nozzle must come last, because its job is to convert whatever pressure energy remains into velocity. Any answer that offers a different sequence — the turbine before the combustor, the nozzle near the front, a recirculating loop inside the core — is describing a machine that cannot run.

Of those five, three are described as the main sections of any gas turbine: the compressor, the combustion section and the turbine. Items such as the diffuser, the stator vanes, the nozzle guide vanes and the exhaust cone are components within those sections rather than sections in their own right, and offering one of them as a main section alongside the other three is a standard distractor. The intake and the exhaust are usually treated as part of the installation rather than as core sections, which is why they appear in the gas-path sequence but not in the list of three.

The Cold Section and the Hot Section

The engine is also divided into a cold section and a hot section, and the dividing line is the point at which fuel is burned. Everything upstream of combustion handles air at compressor temperatures; everything downstream handles gas at combustion temperatures.

The pre-combustor diffuser is the item that sits closest to that dividing line. It sits immediately in front of the combustor and its job is to slow the compressor delivery air before it reaches the flame, but it is upstream of ignition and therefore belongs to the cold section. A statement placing the diffuser in the hot section is wrong, and a statement placing the turbine in the cold section is wrong for the same reason in the opposite direction. The test is always the same: has this component seen burning fuel?

The division is far from academic. It drives the choice of materials, as the table shows. It drives inspection intervals and inspection methods, because the failure mechanisms are entirely different — a cold section problem is usually mechanical damage you can see, while a hot section problem is usually time-at-temperature damage you have to look for. And it drives the shape of a shop visit, because hot section refurbishment is a far heavier and more expensive task than cold section work and is very often the event that defines when an engine comes off wing.

The Gas Generator

The combination of compressor, combustor and the turbine stage or stages that drive that compressor is called the gas generator, or simply the core. Its job is exactly what the name says: to produce a supply of hot, high-pressure gas. On its own a gas generator is not a propulsion device at all — it is an energy source, and what is attached behind it decides what kind of engine it becomes. A propelling nozzle makes it a turbojet. A fan, plus the extra turbine stages needed to drive that fan, makes it a turbofan. A power turbine driving a shaft makes it a turboprop or a turboshaft. Recognising the identical core inside all four configurations is the quickest route to understanding why they behave so differently while sharing so much hardware, and it is why manufacturers routinely develop one core into several very different products.

What a Gas Generator Needs Before It Is an Engine

A compressor, a combustor and a turbine will not run on their own, and the supporting systems are not accessories in the dismissive sense — several of them are what makes the machine survivable. Five are always present in some form, whatever the engine type.

• A fuel system to deliver fuel at the pressure the burners need and, far more importantly, to schedule how much is delivered. The fuel control is what prevents the engine over-temperaturing or surging while it accelerates, so it is a protective device as much as a supply system.
• A lubrication system to feed and scavenge the bearing chambers and gearboxes. Its second job matters as much as its first: the oil is a coolant, carrying heat away from bearings running in chambers surrounded by very hot structure.
• An internal air system, taking air from chosen compressor stages to cool turbine discs and blades, to pressurise the bearing chamber seals, and to balance the axial thrust loads on the rotors. This air never makes thrust and every kilogram of it is a cost, which is why it is metered as carefully as fuel is.
• A starting and ignition system, because a gas turbine cannot start itself: something external has to turn the high pressure spool up to a self-sustaining speed while igniters light the fuel.
• An indication system, because almost nothing inside a running engine can be seen. Speeds, temperatures, pressures, fuel flow, oil condition and vibration are the entire evidence base on which the engine is operated and maintained.

Each of these is a subject in its own right and is developed later in this module. The point to carry forward from the fundamentals is that they are all consequences of the cycle rather than additions to it: the internal air system exists because turbine entry temperature exceeds what metal can survive, the oil system exists because the shafts must be supported inside a hot machine, and the fuel schedule exists because the compressor has a surge line that must not be crossed.

Spools: One Shaft, Two or Three

A spool is a compressor, the turbine that drives it, and the shaft joining them, all rotating together as a single assembly at a single speed. The number of spools an engine has is one of its defining architectural features, and the spool arrangement determines which turbine drives which compressor.

The rule governing which turbine drives which compressor is simple and never varies: the turbine immediately behind the combustor drives the highest-pressure compressor, and each successive turbine further downstream drives the next compressor further forward. The reason is energy. Gas leaving the combustor is at its hottest and highest pressure, so the first turbine stage has the greatest energy available to it — and the high pressure compressor is precisely the one that needs the most work per kilogram of air. Working outwards from the flame in both directions keeps the biggest job matched to the biggest energy supply.

Read the speed designations carefully. N1 always means the low pressure spool, which on a front-fan engine is the fan. On a two-spool engine the high pressure spool is N2 — but on a three-spool engine N2 is the intermediate spool and the high pressure spool is N3. Reading N2 as "the high pressure spool" without first establishing how many spools the engine has is a real trap on the flight line, because the same label on two different engine types refers to different hardware.

A direct consequence of the definition of a spool is that everything mounted on the same shaft turns at the same speed. On a conventional dual-spool front-fan engine — one with no fan drive gearbox — the fan, the low pressure compressor and the low pressure turbine form one rotating assembly, so the fan turns at exactly the same speed as the low pressure turbine that drives it. It does not turn at the same speed as the high pressure compressor, which is on a physically separate shaft. The two shafts are concentric, the LP shaft running through the bore of the HP shaft, and they are entirely free to run at different speeds and even, on some designs, in opposite directions. Where a fan drive gearbox is fitted, as it is on a geared turbofan, the fan turns at roughly a third of the low pressure turbine's speed while still being driven by it; that arrangement is described later in this note.

The reason for splitting the machine at all is that different parts of it want to run at different speeds. A large fan is limited by the tip speed of its blades: once the tips approach and pass the local speed of sound, efficiency falls away and noise climbs steeply, so a big fan has to turn slowly. A high pressure compressor is small in diameter and needs high blade speed to do its work, so it has to turn fast. Tying them to one shaft forces a compromise that suits neither. Splitting them lets each run near its own optimum, and it brings a second benefit that is at least as valuable: at part speed the spools automatically settle at different speed ratios from those they hold at full power, which widens the range of conditions over which the compressor stays clear of stall and surge. High overall pressure ratios became practical only once multi-spool architecture was available to make them stable.

Each spool has to be supported on its own bearings, and a concentric arrangement means some of those bearings sit between two rotating shafts rather than between a shaft and static structure. These inter-shaft bearings see the difference between the two shaft speeds rather than either speed by itself, and both their races are moving, which makes their oil supply and their vibration signature more complicated than a conventional bearing's. It is one of the practical costs of a multi-spool engine and one of the reasons a vibration survey has to be interpreted against the engine's actual architecture.

What Happens if a Shaft Shears

Because a spool is a matched pair, breaking the shaft that joins them produces a violently asymmetric result. Consider the low pressure shaft of a turbofan shearing in flight. The LP turbine instantly loses its load — the fan it was driving — while keeping its full supply of hot gas, so the net accelerating torque on it becomes enormous and it accelerates with nothing to restrain it. This is turbine runaway, and it can reach burst speed within a couple of seconds. The fan, meanwhile, has lost its driving torque while still being loaded by the airflow through it, so it slows down. The two halves of one spool therefore move in opposite directions: turbine overspeed at the back, compressor underspeed at the front.

A turbine disc that bursts releases fragments with enough energy to leave the engine and the nacelle entirely, which makes this one of the small number of genuinely uncontained failure modes, so engines are protected against it. Electronic protection watches the rate of change of shaft speed, or watches for a mismatch between speeds that ought to be locked together, and shuts off the fuel. Mechanical protection is also used: some designs allow the unloaded turbine assembly to move rearwards under gas load so that its blades foul static structure and are braked before burst speed can be reached. The maintenance consequence is straightforward — an unexplained speed mismatch between the two ends of a single spool, or an unexplained vibration change accompanying it, must be treated as a possible shaft problem and never written off as an indication fault until that has been eliminated.

The Accessory Gearbox and Its Drive

The engine has to drive its own services and the aircraft's — fuel pumps, oil pressure and scavenge pumps, hydraulic pumps, electrical generators, and usually a dedicated alternator supplying the engine control system. All of them hang on an accessory gearbox, and the power for it is taken from the high pressure spool through a bevel gear and a radial drive shaft, often called a tower shaft, running outwards through a strut in the engine casing.

The drive comes from the HP spool for two reasons and both matter. First, the HP spool has the narrowest operating speed range of any spool in the engine — it never runs slowly the way a fan can — so accessories driven from it see a far more consistent input speed than they would from the fan. A generator or a fuel pump driven off the fan would face a range of input speeds it could not usefully accommodate. Second, the HP spool is the one the starter has to turn to get the engine going, so routing the starter drive and the accessory drives through the same gearbox is the tidiest possible arrangement. On a high bypass engine that gearbox is mounted on the high pressure compressor casing, where the radial drive can reach it directly; on some large engines the drive is carried onwards by a transfer gearbox to a main gearbox slung lower on the core for accessibility, but the power still originates at the HP spool in every case.

Every accessory drive on that gearbox is a potential path along which damage could travel back into the engine, and the standard protection is a waisted drive shaft: a shaft machined with a deliberately reduced-diameter section over part of its length. If the driven unit seizes or is grossly overloaded, the shaft twists off at that waist instead of transmitting the load into the gearbox and the engine behind it. It is a mechanical fuse, and like an electrical fuse its whole purpose is to fail first, cheaply and predictably, so that something expensive and hard to reach does not fail instead. A waisted shaft is not there to save weight and not there to assist dynamic balance, and a sheared one is evidence about the accessory it was driving, not about the shaft itself — fitting a new shaft without establishing why the old one sheared simply arms the same failure again.

Modular Construction

A modular engine is built as a set of self-contained assemblies — typically a fan module, a booster or intermediate module, a core module containing the high pressure compressor, combustor and HP turbine, a low pressure turbine module and an accessory drive module. The defining property is that any one of them can be removed and replaced without disturbing the rest of the engine. The older, non-modular alternative requires the engine to be progressively stripped from one end in order to reach a component in the middle.

The advantage is directly financial. If one module has reached its life limit or has been damaged, only that module goes to the workshop; the rest of the engine remains serviceable, and a replacement module drawn from stores can be fitted and the engine returned to service in a fraction of the time and cost of a full strip and overhaul. It also allows modules of different ages to be combined in one engine, so that remaining life is used up rather than discarded along with the module that happened to fail first.

The price is bookkeeping and balance. Because modules move between engines, the maintenance records have to follow the module and not merely the engine serial number: hours and cycles are tracked per module, and life-limited parts inside a module are tracked individually within it. Rotating assemblies also have to be balanced as a module and, on many engines, checked or trim balanced after installation, because a module built correctly to its own drawing tolerances is not automatically in balance in combination with the modules on either side of it. One thing modular construction is not is a statement about how the engine is physically put together on a stand: building an engine vertically is a workshop convenience adopted by some manufacturers, and it has nothing to do with whether the engine is modular.

Balance, Critical Speeds and Why Vibration Is Watched

A machine whose only moving parts rotate ought to run perfectly smoothly, and the fact that it does not is worth understanding, because vibration is one of the few windows into an engine that stays open while it is running. Any rotor carries some residual out-of-balance: manufacturing tolerances, blade-to-blade weight variation, and later erosion or damage all leave the centre of mass fractionally off the axis of rotation. That offset generates a rotating force proportional to the offset mass, to its radius, and to the square of the rotational speed. The square is the important part — the same tiny imbalance that is undetectable at idle is significant at take-off speed.

Rotors also have natural bending frequencies, and the speed at which a rotor's rotational frequency coincides with one of them is called a critical speed. At that speed the out-of-balance force is applied in step with the rotor's own tendency to flex, so the deflection builds instead of being resisted. Engines are designed so that no critical speed falls inside the normal continuous operating range, but a rotor may have to pass through one during acceleration and deceleration, and this is one of the reasons certain speed bands are marked for continuous operation to be avoided and why an engine is not left dwelling at an arbitrary speed during a ground run.

The practical consequences reach the flight line in three ways. Bearings are supported in squeeze-film oil damping arrangements on many engines, so that a thin film of oil between the bearing housing and the structure absorbs vibration energy rather than passing it straight into the casings. Fan and compressor blades are moved, matched or replaced in balanced sets, and a trim balance is carried out by adding small weights at a defined position after a blade change. And a recorded rise in the vibration level at a particular rotor's own frequency identifies which spool is responsible, because each spool turns at its own speed and therefore signs its own contribution to the signal.

Gas Turbine Engine Types

All gas turbine engines share the same core — compressor, combustor, turbine — but differ in how they convert the energy extracted by the turbine into useful output. The four main types encountered in aviation are:

It helps to see those four as points on a single continuum rather than as four unrelated machines. The variable that orders them is how much air is worked on for every kilogram that passes through the core. A pure turbojet works on nothing but its core flow — an effective bypass ratio of zero. A low bypass military turbofan works on perhaps half as much again as its core swallows. A civil high bypass turbofan works on several times more. A turboprop, whose propeller disc is many times the diameter of its intake, works on an air mass equivalent to a bypass ratio of many tens to one. Move along that line and the propulsive device gets larger while the velocity it adds gets smaller, and the flight speed at which the arrangement stops being competitive comes down with it. Nothing else about the four types is as fundamental as that single trade, and the sections below are best read as four different resolutions of it.

Turbojet

The simplest gas turbine configuration. All of the air entering the intake passes through the core (compressor → combustor → turbine). The turbine extracts only enough energy to drive the compressor; all remaining energy is converted to a high-velocity exhaust jet that produces thrust. Turbojets are efficient at very high speeds (Mach 2+) but are noisy and fuel-hungry at subsonic speeds. They are rarely used on modern civil aircraft but are still found on some military aircraft and older designs.

"All of the air passes through the core" is the defining statement, and everything else about the turbojet follows from it. Because the core flow is small — tens of kilograms per second rather than the many hundreds a large turbofan passes — the only way to obtain useful thrust is to give that small flow a very large velocity increase. Jet exit velocities of the order of 600 m/s and above are normal, and at subsonic flight speeds that leaves an enormous velocity difference between the jet and the air it is being thrown into.

The mechanism behind the fuel consumption is the one set out in the worked example earlier in this note: the energy poured into the jet rises with the square of the velocity change while the thrust obtained rises only in proportion to it, so a fast jet is an expensive way to buy thrust at subsonic speed. The mechanism behind the noise is separate and just as unforgiving. Jet mixing noise — the roar of a high-velocity jet shearing into still air — rises with a very high power of the jet velocity, classically the eighth power, so a modest reduction in jet velocity produces a very large reduction in noise. Between them, those two relationships drove the turbojet off civil aircraft: noise legislation and fuel price were attacking the same design feature from two directions.

The picture reverses at high speed. As the aircraft's own velocity climbs, a slow bypass jet runs out of margin over the flight speed and can no longer add momentum usefully, while a turbojet's high jet velocity still leaves plenty of room to work in. Ram compression in the intake also becomes increasingly valuable, doing part of the compressor's job for nothing: the temperature rise it produces grows with the square of the flight Mach number, and the pressure ratio it produces grows faster still. Above roughly Mach 2 the turbojet, or a very low bypass turbofan, is the appropriate architecture. Above about Mach 3 the compressor and turbine become an encumbrance rather than a help, because ram compression alone is sufficient, and that is the point at which the ramjet takes over. Military installations also frequently add reheat downstream of the turbine for short periods of greatly increased thrust.

Maintenance considerations follow the architecture directly. A turbojet's entire mass flow passes through the hot section, so the hot section is worked hard for every newton produced and hot section life tends to dominate the maintenance programme. There is no cool bypass stream washing the outside of the core, so nacelle and jet pipe skin temperatures are higher, fire zone ventilation and fire protection are correspondingly more demanding, and the exhaust hazard area behind a running turbojet is both longer and hotter than behind a high bypass engine of the same thrust.

Turbofan

The dominant engine type in modern commercial aviation. A large fan at the front of the engine is driven by a turbine (usually the low-pressure turbine via a shaft running through the centre of the engine). The fan accelerates a large mass of air, most of which bypasses the core through an annular duct. The bypass ratio (BPR) is the ratio of bypass air mass flow to core air mass flow:

Modern high-bypass turbofans (e.g., CFM LEAP, Rolls-Royce Trent XWB, GEnx) have BPRs of 9:1 to 12:1, meaning 9–12 kg of air pass through the bypass for every 1 kg through the core. Approximately 75–85% of total thrust comes from the bypass (cold) stream. High BPR engines are quieter and more fuel-efficient than turbojets because they produce thrust by moving a large mass of air at moderate velocity rather than a small mass at high velocity.

A fan is a propeller that has been placed inside a duct and given many more, much shorter, much stiffer blades. The duct is what makes it a different machine. It allows the blade tips to run at transonic speeds without the losses an unducted propeller would suffer there, it controls the flow entering and leaving the blades so that the working conditions along the span can be designed rather than accepted, and it lets the bypass stream be exhausted through a nozzle of chosen area. None of those is available to an open propeller, and together they are why a fan can be used at the high subsonic cruise speeds where a propeller cannot.

Turbofans are classified by bypass ratio, and the classification is useful because it predicts almost everything else about the engine. Low bypass engines, with ratios below about 2:1, are found on military aircraft and on older civil types; their jet velocity is high, they are slim, and they tolerate supersonic flight. Medium bypass covers roughly 2:1 to 4:1. High bypass begins around 5:1 and reaches the figures quoted above for current civil engines, with geared designs going higher again. As the ratio rises, the fan diameter grows, the jet velocity falls, fuel consumption and noise both improve, and the engine becomes progressively less suited to high subsonic and supersonic flight because its slow bypass jet has less and less margin over the aircraft's own speed.

The two streams may be exhausted separately or mixed. In a separate-flow or unmixed installation, typical of large high bypass engines, the cold stream leaves through its own annular nozzle around the core while the hot stream leaves through the core nozzle further aft. In a mixed-flow installation, usual on lower bypass engines and on many business-jet engines, the two streams are brought together in a lobed mixer ahead of a single common nozzle. Mixing recovers a small amount of thrust and reduces jet noise by evening out the velocity profile leaving the engine, but it costs length, weight and duct pressure loss, which is why it is not universal.

Driving the fan is a demanding job in its own right. The fan absorbs a large amount of power at a low rotational speed, while a turbine is happiest producing power at high speed, so the low pressure turbine has to be given a large annulus and several stages to extract that power at a speed the fan can accept. On a large conventional turbofan the low pressure turbine may have five or more stages and account for a substantial share of the engine's weight. The alternative is to insert a reduction gearbox between the low pressure turbine and the fan, so that each can run at its own best speed and the turbine can be made smaller, faster and lighter with fewer stages; that arrangement is dealt with among the alternative turbine constructions elsewhere in these notes.

Two further points are worth carrying to the flight line. First, the bypass duct air does considerably more than make thrust: it cools the outside of the core, it ventilates the fire zones, and on most large engines it is the working stream for the thrust reverser. An obstruction or a damaged duct panel is therefore a multi-system problem, not a simple thrust loss. Second, the fan is the largest rotating assembly on the aircraft and the first thing any ingested object meets. Fan blade condition, blend repair limits, blade root lubrication and fan trim balance after a blade change or a blade set rotation are routine and consequential tasks, and an out-of-limits vibration indication traced to the fan is far more common than one traced to the core.

One historical variant is worth knowing because it clarifies what the definition of a turbofan actually requires. Early bypass engines were sometimes built as aft-fan designs, in which the fan was placed behind the turbine rather than in front of the compressor and was driven not by a shaft but by an extra row of turbine blades formed on the outer ends of the fan blades themselves, with the hot gas driving the inner portion and the fan pumping cold air through the outer portion. The arrangement avoided a long shaft and kept the fan out of the way of the core, but it put the fan blade roots in the hot gas stream, which limited both the temperature and the life that could be achieved. Front-fan designs displaced it entirely. The point that survives is the definition: what makes an engine a turbofan is the presence of a bypass flow driven by turbine power, not the position of the fan.

Turboprop

Uses the gas turbine core to drive a propeller through a reduction gearbox (RGB). The turbine extracts almost all available energy from the gas flow, leaving very little for exhaust thrust (typically only ~10%). The propeller generates ~90% of the total thrust by accelerating a very large mass of air at low velocity. Turboprops are very efficient at speeds below 400 knots and at lower altitudes, making them ideal for regional airliners (e.g., ATR 72, Bombardier Dash 8) and utility aircraft.

The reason so little energy is left over for the exhaust is a deliberate design choice, not an accident of layout. Extra turbine stages are fitted behind the ones that drive the compressor, sized specifically to extract shaft power, and the more work the turbine takes out the less pressure and heat remain in the gas to make a jet. The residual jet thrust is nonetheless real, which means a turboprop's output cannot be described honestly by shaft power alone. The industry answer is equivalent shaft horsepower (ESHP), which adds the residual jet thrust — converted into an equivalent power at the aircraft's speed — to the shaft horsepower actually delivered to the propeller. Quoting only the shaft power understates the engine; quoting the engine as thrust alone is meaningless, because thrust and power are different quantities and cannot be exchanged without knowing the speed.

The reduction gearbox is not an optional refinement. A gas turbine's power turbine runs in the tens of thousands of revolutions per minute, while a propeller is limited to roughly one to two thousand by the speed of its blade tips. The tip sees the vector sum of its rotational speed and the aircraft's forward speed, so a propeller turning too fast has supersonic tips, loses efficiency sharply and becomes extremely noisy. The gearbox reconciles the two by trading speed for torque at very nearly constant power. That also makes it a heavily loaded, closely monitored assembly with its own oil supply, its own chip detectors and its own inspection schedule — on a turboprop, gearbox condition is as much a part of powerplant maintenance as the engine itself, and a metal indication from the gearbox is as serious as one from an engine bearing chamber.

Two shaft arrangements exist and it is worth knowing which one you are looking at. In the fixed-shaft, or direct-drive, arrangement drawn in the schematic above, the propeller is geared to the same shaft that carries the compressor and turbine, so propeller speed, compressor speed and turbine speed are locked together in a fixed ratio. In the free-turbine arrangement, a separate power turbine sitting downstream of the gas generator turbine drives the propeller through the gearbox and turns independently of it. The free turbine allows the gas generator to be started and run without turning the propeller, which makes starting easier and allows the propeller to be held stationary on the ground; the fixed shaft is mechanically simpler and gives a more direct relationship between power lever movement and engine response. Both arrangements are in widespread service, and neither is a modern replacement for the other.

The propeller is also what sets the aircraft's practical speed limit, and the mechanism is the same tip-speed problem seen from the other end. Propeller efficiency holds up well while the blade tips remain comfortably subsonic, but the tip Mach number combines rotational and forward speed, so as the aircraft flies faster the tips approach the speed of sound no matter what the propeller itself is doing. Past that point efficiency collapses and noise climbs steeply. This is why the turboprop's advantage is confined to a lower speed band — and why, inside that band, it is genuinely and substantially more economical than a turbofan rather than merely cheaper to buy.

On the flight line the practical consequence is that a turboprop is set up and monitored differently from a turbofan. Power is set on torque rather than on a fan speed or an engine pressure ratio, for the reason given above: with the propeller governing at a constant speed, the torque gauge is reading power directly. The propeller and the engine are controlled as a single system, which means a fault that presents as an engine problem may in fact lie in the propeller control and the reverse is equally true, so troubleshooting has to treat the two together from the outset rather than clearing one before starting on the other.

Turboshaft

Functionally similar to a turboprop but the output shaft drives a rotor system (helicopter) or other machinery (ship, generator, pump) rather than a propeller. In a turboshaft, the engine is designed to produce shaft horsepower (SHP) with virtually zero residual jet thrust. Most turboshaft engines use a free turbine (also called a power turbine) — a separate turbine stage that is mechanically independent of the gas generator spool. This allows the power turbine to rotate at a different speed from the gas generator, which is essential for helicopter rotor speed control.

The mechanical independence of the free turbine is worth spelling out, because it is what makes a turboshaft practical for a helicopter at all. There is no shaft connecting the power turbine to the gas generator; the two are coupled only by the gas flowing from one into the other. That means the gas generator can be accelerated and decelerated to follow the power demand while the power turbine, and therefore the rotor, is held at constant speed by its governor — exactly what a helicopter needs, since rotor speed must stay within a narrow band whatever the collective is doing. It means the engine can be started, run and brought up to idle with the rotor stationary and the rotor brake applied. And in an autorotation the rotor goes on turning while the gas generator idles, but not because it is driving the power turbine round with it. The rotor drive carries a freewheel unit, a one-way clutch between the engine output shaft and the main gearbox, and the moment the rotor tries to turn faster than the engine is driving it that unit releases and disconnects the rotor from the engine altogether. Rotor speed then rises clear of power turbine speed on the gauges — the classic needle split — which is the opposite of the two being carried round together. It is that freewheel unit rather than the free turbine that makes autorotation possible: the same unit is fitted whatever the engine, which is why piston-engined and fixed-shaft turbine helicopters autorotate too. What the free turbine contributes is a light power turbine left carrying nothing once the drive has released, and a gas generator that can go on running at idle without the rotor having to turn it.

Because the two shafts are not mechanically connected, each needs its own speed indication, and the naming convention here differs from the turbofan convention. On a two-spool turbofan, N1 is the low pressure spool and N2 the high pressure spool. On many turboshaft and turboprop engines the gas generator spool is designated N1 or Ng, and the free power turbine is N2, Nf or Np. The labels are the same but the hardware they point at is not, so establish which engine architecture a manual or a question is describing before interpreting any speed designation on it.

"Virtually zero residual jet thrust" is a design requirement rather than a passive observation. On a helicopter, residual thrust from the exhaust would be a force acting in whatever direction the exhaust happened to point, which would have to be trimmed out and which would change with every power setting. Turboshaft exhausts are therefore arranged to minimise it, and where they are angled upwards or to the side the reason is clearance, infra-red signature or the avoidance of hot gas reingestion rather than any attempt to make thrust. The turbine is sized to leave as little energy in the gas as it usefully can, which is the exact opposite of the turbojet's sizing philosophy despite both engines using the same kind of core.

The same architecture is used well beyond rotorcraft. An auxiliary power unit is a small turboshaft whose load is an electrical generator and a bleed-air supply rather than a rotor. Marine propulsion, pipeline pumping and emergency generating sets use larger versions of the same idea. In every case the engine is rated in power rather than thrust, its health is assessed by a power assurance check that compares achieved torque and gas generator speed against the ambient conditions of the day rather than by any thrust measurement, and its operating limits are expressed as torque, gas generator speed and turbine temperature.

Choosing Between Them

Comparing the four is easiest if the comparison avoids the properties already described and concentrates instead on how each one is rated, geared and maintained, because those are the differences an engineer meets first.

The choice between them in service is not a question of one being better than another. It is a question of matching the propulsive device to the mission, and the drivers are usually practical rather than thermodynamic: the length and surface of the runways to be used, the altitude and temperature of the airfields, the sector length, the local noise rules, the acquisition and maintenance cost the operator can carry, and the performance required after an engine failure. A turboprop's low-speed thrust and short field performance can matter far more to a regional operator than the cruise speed it gives up, while a long-haul operator will pay for cruise efficiency and accept a longer take-off run.

One consequence of the common core is worth knowing, because it explains a great deal about parts commonality and about type training. Manufacturers routinely develop a single gas generator into several products: the same core, in different states of tune, may appear in a regional turbofan, an industrial power generation set and a marine engine. Modules, accessories, control philosophies and even maintenance practices can then carry across between apparently unrelated applications, which is why an engineer who understands one core has a substantial head start on any engine built around it.

The Gas Turbine Compared with the Piston Engine

The gas turbine did not replace the piston engine on large aircraft because of the cycle alone. It replaced it because of a set of mechanical and operational consequences that follow from the cycle, and setting the two side by side is the quickest way to see which properties are fundamental and which are incidental.

The most-quoted advantage is smoothness, and the reason for it is purely mechanical: a gas turbine has no reciprocating parts. In a piston engine every piston is brought to a complete standstill and reversed twice per revolution, and the inertia forces required to do that are large, cyclic and impossible to balance out entirely; the firing impulses then arrive at the crankshaft as a series of separate blows. A gas turbine's rotating assemblies simply turn, at a steady speed, with a steady flow of gas through them. What vibration remains comes from residual imbalance in the rotors and from aerodynamic excitation, not from the working cycle itself. That is the root of the very low healthy vibration baseline described earlier, and it is what makes a turbine's vibration signature such a sensitive instrument.

It is worth being explicit about what the smoothness does not come from, because the alternatives are offered as examination answers. It does not come from better lubrication: a turbine's oil system is excellent, but a well-lubricated piston engine still shakes. And it does not come from running at a lower temperature: a gas turbine's hot section runs very much hotter than any piston engine's cylinder head. The absence of reciprocating masses is the principal reason.

Several operational differences follow from the same root. A gas turbine has no mixture control, because the fuel is metered into a continuously burning flame rather than into a measured charge of air, and the combustor runs with a very large excess of air at all times. It has no ignition system running continuously either: the igniters fire only during starting and when continuous ignition is deliberately selected for protection against flame-out in heavy rain, turbulence or icing conditions. And it responds to a power lever movement with a lag rather than immediately, because the fuel control has to accelerate the rotors without pushing the compressor into surge — which is why turbine-powered aircraft are flown with the spool-up time anticipated rather than discovered.

Starting shows the same divergence. A piston engine needs only to be turned fast enough for the first charge to fire, after which it is self-sustaining almost immediately. A gas turbine has to be driven up to a speed at which the compressor is delivering enough air for combustion to be stable and at which the turbine can produce more work than the compressor absorbs; below that speed the engine cannot sustain itself, so the starter must keep assisting well after the fuel has lit and the flame has established. That is why a turbine start is a timed, monitored sequence with a defined light-up point and a defined starter cut-out speed, rather than a single event. It is also why the failure modes at start have their own names and their own distinct causes, and why the fundamental parameters watched during a start are gas generator speed and turbine temperature — the two quantities that say whether the compressor is delivering and whether the combustor is behaving.

The gas turbine's weaknesses are real and they are the reverse side of the same characteristics. It is thermodynamically poor at low power settings, because both the pressure ratio and the turbine entry temperature fall away as the engine is throttled back, and the ideal-cycle relationship set out earlier shows immediately what that costs. It is expensive to manufacture, because its hot section parts are made from costly alloys by costly processes. It is intolerant of the sort of neglect a piston engine will survive, since a small quantity of the wrong debris in the gas path can end an engine's life in seconds. And its efficiency advantage depends on flying fast and high, which is why the piston engine remains entirely sensible in a light aircraft cruising at 120 knots and entirely hopeless in an airliner.

One habit of mind carries forward from all of this into the rest of the module, and it is the practical reason the physics in this note matters. A gas turbine is normally assessed without being opened: its condition is inferred from a handful of parameters read at the flight deck and on a test run. Because the cycle behaves in predictable ways, those inferences are sound. If a compressor becomes fouled or a turbine's tip clearances open up, its component efficiency falls; the cycle must then burn more fuel to reach the turbine work the engine is being asked for, so for the same thrust the exhaust gas temperature rises and the margin to the temperature limit shrinks. Nobody has to see the blades to know that. Every diagnosis of that kind is an application of the four processes, the gas laws and Newton's laws set out above, which is why they are worth understanding rather than merely memorising.