Chapter 7: Atmospheric Re-Entry

"We choose to go to the Moon in this decade and do the other things, not because they are easy, but because they are hard." — John F. Kennedy, Rice University, September 12, 1962

Overview

For six chapters we have thought about one direction: up. We found the $9.4\ \text{km/s}$ price of orbit, the equation that charges it, the losses that inflate it, and the energy that a body in orbit carries. Now we turn the whole book around and ask the question that almost every crewed spacecraft and every returning sample capsule must answer or die: how do you come back down?

It sounds like the easy half. Gravity is free; surely falling is simpler than climbing. But look at what the falling object is carrying. A spacecraft in low Earth orbit moves at about $7.8\ \text{km/s}$, and that speed is not a detail — it is roughly $30\ \text{megajoules}$ of kinetic energy in every single kilogram of the vehicle, an energy density several times that of the same mass of TNT. To land, the spacecraft must get rid of essentially all of it. You cannot wish energy away; the first law of thermodynamics forbids it. The energy has to go somewhere, and there is only one place for it to go: into heat. Re-entry is the problem of converting a locomotive's worth of kinetic energy per kilogram into heat, at several thousand degrees, in about ten minutes, without cooking the payload inside. It is the most concentrated encounter with theme #2 of this book — space is an unforgiving environment — that Part I contains. There is no repair on the way down. The heat shield must work the first time, for the whole time, or there is no crew to debrief.

In Chapter 6 we studied the energy of an orbit — how much you must add to get up and stay up. This chapter is Chapter 6 read backwards: the story of getting that energy back out. The elegant, almost unreasonable trick is that you do not pay for the braking with propellant. A reverse rocket that killed $7.8\ \text{km/s}$ would need another rocket the size of the one that launched you. Instead you spend a tiny deorbit nudge — about $100\ \text{m/s}$ — to dip into the top of the atmosphere, and then let the air itself do the braking, for free, converting your speed into heat that a shield throws away. The atmosphere that was a thief on the way up (Chapter 4) becomes, on the way down, the only brake you can afford.

In this chapter, you will learn to:

• See re-entry as an energy-disposal problem: ~$30\ \text{MJ/kg}$ of orbital kinetic energy turned into heat.
• Explain why the vehicle is heated by compressing the air ahead of it, not by rubbing against it.
• Compare the ways engineers survive that heat — ablative shields, reusable tiles, transpiration.
• Distinguish ballistic from lifting entry, and use the ballistic coefficient to predict how deep, how hot, and how hard a vehicle decelerates.
• Understand the re-entry corridor — too steep and you burn or crush; too shallow and you skip off into space.
• Estimate the g-forces of entry and explain the radio blackout caused by the plasma sheath.

Learning Paths

🚀 Space Enthusiast: Read 7.1 for the headline (re-entry is an energy problem), 7.2 for the single idea that surprises everyone (it's compression, not friction), and 7.3 for the hardware. The blunt-body story in 7.2 and the corridor in 7.5 are the two ideas that will change how you watch a capsule come home.

📐 Engineering Student: Read everything. The stagnation-temperature estimate in 7.2, the heating scaling in 7.3, and the Allen–Eggers peak-g result in 7.6 are the quantitative core; do the ⭐⭐/⭐⭐⭐ exercises and both case studies. This chapter is the prerequisite for the Shuttle's story in Chapter 37.

🎮 KSP Player: You have felt every idea here — the periapsis you drop into the atmosphere, the orange glow, the way a steep entry rips you apart and a shallow one skips you back to space. Sections 7.4 and 7.5 are the physics behind the "re-entry effects" you already fight; the ballistic-coefficient discussion explains why a flat, draggy craft slows so much more kindly than a dense dart.

🛰️ Industry Prep: Thermal protection (7.3) and the corridor (7.5) are where entry, descent, and landing (EDL) design lives. The Mission Design Checkpoint asks whether your mission returns through an atmosphere at all — a question that reshapes the whole vehicle. This connects forward to Chapter 34's Mars EDL.

7.1 The energy problem

Every discussion of re-entry has to start with a number, because the number is the whole problem. A spacecraft in low Earth orbit is moving at about $v \approx 7.8\ \text{km/s}$ — the orbital speed we derived in Chapter 4. Its kinetic energy per kilogram is

$$ \frac{E_k}{m} = \tfrac{1}{2}v^2 = \tfrac{1}{2}\,(7{,}800\ \text{m/s})^2 = 3.04\times10^{7}\ \text{J/kg} \approx 30\ \text{MJ/kg}. $$

Sit with that figure for a moment. Thirty million joules per kilogram. For comparison, detonating a kilogram of TNT releases about $4.2\ \text{MJ}$, so each kilogram returning from orbit carries roughly seven kilograms of TNT worth of kinetic energy — and unlike an explosion, which is over in microseconds, this energy is bound up in motion and must be removed deliberately. A crewed capsule of $10\ \text{tonnes}$ carries $3\times10^{11}\ \text{J}$; the Space Shuttle orbiter, landing at roughly $100\ \text{tonnes}$, came home with about $3\times10^{12}\ \text{J}$ of kinetic energy — the better part of a kiloton of TNT, all of which had to become heat before the wheels touched the runway. (Vehicle masses here are representative, Tier 2.)

🔗 Connection: This is the exact mirror of a fact from Chapter 4. There we found that reaching orbit takes about thirty times the energy of a mere hop to space, because orbit is about sideways speed, and kinetic energy grows as the square of speed. All of that hard-won kinetic energy is still there when you want to come home. What the launch vehicle spent a skyscraper of fuel and $9.4\ \text{km/s}$ of delta-v to put in, the heat shield must take back out. Re-entry is the launch energy problem, run in reverse and on a deadline.

Strictly, the vehicle must also shed the gravitational potential energy it holds by being up high — but for a low orbit that is a smaller correction, and the kinetic term dominates. Using the specific orbital energy of Chapter 6, the total mechanical energy that must be dissipated to bring a $400\ \text{km}$-orbit vehicle to rest at the surface is about $33\ \text{MJ/kg}$ — close to, and dominated by, the $30\ \text{MJ/kg}$ of pure kinetic energy. So to keep the physics clean, we will speak of "the $30\ \text{MJ/kg}$" and remember there is a little more.

Now the definition this whole chapter serves.

Definition (re-entry). Re-entry (atmospheric entry) is the process by which a spacecraft descending from orbit or from an interplanetary trajectory passes into a planet's atmosphere and uses aerodynamic forces to decelerate, converting the overwhelming majority of its kinetic (and potential) energy into heat. For a vehicle returning to Earth from low orbit, this means disposing of about $30\ \text{MJ/kg}$ and slowing from roughly $7.8\ \text{km/s}$ to a survivable landing speed.

Why not just use retro-rockets?

The obvious engineer's question is: why involve the atmosphere at all? Why not turn the ship around and fire the engines to brake, the way you would slow a car — a controlled, gentle stop with no fireball?

Because the rocket equation (Chapter 3) makes it unaffordable. To cancel $7.8\ \text{km/s}$ of speed propulsively costs $\Delta v \approx 7.8\ \text{km/s}$, essentially the same as it cost to gain — and with a good chemical exhaust velocity of $v_e = 3\ \text{km/s}$, the mass ratio for that maneuver alone is $e^{7800/3000} = e^{2.6} \approx 13.5$. You would need to arrive in orbit carrying a second fully fueled launch vehicle, purely to slow down. The tyranny of the rocket equation (theme #1) forbids braking from orbit with rockets for any practical payload.

The atmosphere offers a bargain that no rocket can match. You spend a small deorbit burn to lower the low point of your orbit into the upper atmosphere, and then the air removes the rest of your energy at no propellant cost. How small is the burn? From a $400\ \text{km}$ circular orbit (speed $7.67\ \text{km/s}$), lowering the perigee to about $50\ \text{km}$ — deep enough for the atmosphere to grab you — takes a retrograde burn of only

$$ \Delta v_{\text{deorbit}} \approx 7.67 - 7.57 = 0.10\ \text{km/s} = 100\ \text{m/s}. $$

Worked Example: the bargain, priced out. Compare the two ways home from a $400\ \text{km}$ orbit. Propulsive: $\Delta v \approx 7.8\ \text{km/s}$, mass ratio $\approx 13.5$ — a whole extra rocket. Aerodynamic: $\Delta v_{\text{deorbit}} \approx 0.1\ \text{km/s}$ (mass ratio $e^{100/3000} = 1.03$, a few percent of propellant) plus a heat shield massing perhaps $10$–$20\%$ of the vehicle. The atmosphere does the work of $7.7\ \text{km/s}$ of braking for the price of a tank of paint's worth of propellant and a shield. That factor-of-seventy saving in delta-v is why every vehicle that has ever returned from orbit — from Vostok to Dragon — has come home on a heat shield, not on its engines.

🚪 Threshold Concept: the atmosphere is a free brake, paid for in heat. The deep idea of re-entry is that you trade problems. Braking with rockets is a delta-v problem, and delta-v is exponentially expensive, so you refuse to pay it. Instead you convert the braking into a heat problem — dump all $30\ \text{MJ/kg}$ into the air and into a shield — because a heat problem, unlike a delta-v problem, can be solved with a few centimetres of the right material. Re-entry engineering is the art of accepting a thermal catastrophe on purpose, because the thermal catastrophe is survivable and the propulsive alternative is not. Once you see re-entry this way, the fireball stops looking like a hazard the vehicle barely survives and starts looking like the mechanism — the brake pads glowing exactly as intended.

🔄 Check Your Understanding 1. A vehicle returns from a higher orbit, entering the atmosphere at $11\ \text{km/s}$ instead of $7.8\ \text{km/s}$. By what factor does its kinetic energy per kilogram increase? 2. In one sentence, why is a deorbit burn only ~$100\ \text{m/s}$ when the vehicle must lose ~$7.8\ \text{km/s}$ of speed overall?

Answers

1. Kinetic energy scales as $v^2$, so the ratio is $(11/7.8)^2 \approx 1.99$ — about twice the energy per kilogram. This is why entries from the Moon or from interplanetary space are so much more punishing than entries from LEO, and why they demand the toughest heat shields. 2. The deorbit burn only has to lower the orbit's low point into the atmosphere; once the air can reach the vehicle, the atmosphere removes the remaining ~$7.7\ \text{km/s}$ of speed as heat, at no propellant cost. The burn starts the process; the air finishes it.

7.2 Compression heating, not friction

Ask anyone why a re-entering spacecraft glows white-hot and they will almost always say the same thing: friction. The vehicle is rubbing against the air at enormous speed, and friction makes heat, the way rubbing your hands together warms them. It is a reasonable guess. It is also, in the main, wrong — and getting it right is the key that unlocks the entire design of a heat shield.

The dominant source of re-entry heating is not the air sliding along the vehicle's skin. It is the air being violently compressed in front of it. A vehicle moving at Mach 25 slams into the air far faster than the air can get out of the way — faster than a pressure wave (a sound wave) can propagate the news that something is coming. So the air cannot flow smoothly around the nose; it piles up into a thin, almost stationary shock wave standing just ahead of the vehicle, and behind that shock sits a layer of gas that has been compressed and decelerated almost instantly. Compressing a gas heats it — the same physics that warms a bicycle pump — and here the compression is extreme. The gas in that shock layer is where the temperature lives; the vehicle then bakes in the radiant and convective heat of that superheated air.

Definition (compression heating). Compression heating is the heating of a hypersonic vehicle caused by the near-instantaneous compression of the air in the shock layer ahead of it. Air crossing the bow shock is decelerated from the vehicle's speed to nearly rest relative to the vehicle, and its kinetic energy is converted into thermal energy, raising the shock-layer gas to thousands of kelvin. The vehicle is then heated primarily by convection (and, at the highest speeds, radiation) from that gas — not by frictional rubbing of air along its surface.

How hot is the shock layer? Energy conservation gives a quick, revealing estimate. The oncoming air, in the vehicle's frame, carries kinetic energy $\tfrac12 v^2$ per unit mass; when the shock brings it nearly to rest, that energy becomes thermal (raising the gas enthalpy $c_p T$). Setting $c_p \Delta T \approx \tfrac12 v^2$ with air's $c_p \approx 1{,}005\ \text{J/(kg·K)}$ and $v = 7{,}800\ \text{m/s}$:

$$ \Delta T \approx \frac{v^2}{2\,c_p} = \frac{(7{,}800)^2}{2\,(1{,}005)} = \frac{6.08\times10^{7}}{2{,}010} \approx 30{,}000\ \text{K}. $$

That answer — thirty thousand kelvin, five times the surface of the Sun — is physically impossible, and its impossibility is the lesson. Long before the air reaches such a temperature, the energy goes into tearing the molecules apart: at a few thousand kelvin $\mathrm{O_2}$ and $\mathrm{N_2}$ dissociate into atoms, and higher still the atoms ionize, shedding electrons. These reactions soak up enormous energy without a proportional rise in temperature, so the real shock-layer gas settles at a still-ferocious but finite $6{,}000$–$11{,}000\ \text{K}$. The naive calculation overshoots precisely because the energy is so vast that it dismantles the air — and, as we will see, that dismantling is a gift twice over: it absorbs heat that would otherwise reach the vehicle, and the free electrons it produces are the cause of the radio blackout in §7.6.

Here is the flow, sketched:

   hypersonic flow                     detached BOW SHOCK
   (Mach ~25, cold)             ╱──────────────────────
   ───────────────────►        ╱     SHOCK LAYER
   ───────────────────►       ╱   (compressed air,
   ───────────────────►      │      ~6,000–11,000 K,
   ───────────────────►      │       dissociating)
   ───────────────────►      │    ┌───────────┐
   ───────────────────►      │    │▓▓▓▓▓▓▓▓▓▓▓│  ← blunt heat shield
   ───────────────────►      │    │  vehicle   │     (surface held
   ───────────────────►      │    └───────────┘      near ~1,500 K)
   ───────────────────►       ╲     SHOCK LAYER
   ───────────────────►        ╲──────────────────────
                          stand-off  ↑        hot turbulent wake →→→
                          distance

📜 From History: the blunt-body insight. In the early 1950s, designers of the first warheads and capsules assumed that friction was the enemy, and reasoned that a sharp, slender, streamlined shape — like a needle or a dart — would slip through the air and minimize heating. Every such design burned up. The breakthrough came in 1953 from H. Julian Allen and A. J. Eggers at the NACA Ames laboratory, who realized the intuition was backwards. A blunt body creates a strong, detached bow shock that stands off ahead of the vehicle, and most of the compression energy is dumped into that shock-heated air, which is then swept away in the wake — carrying the heat away from the vehicle rather than into it. A sharp body, by contrast, lets the shock cling to its skin and pours the heat straight in. Allen and Eggers' "blunt-body" principle is why every re-entry capsule from Mercury to Orion, and the underside of every spaceplane, presents a broad, rounded face to the flow. The most counterintuitive shape — a shape that maximizes drag — turned out to be the one that survives. It is one of the great examples of physics overturning common sense in this book.

💡 Intuition: the catcher's mitt, not the knife. Think of the blunt heat shield not as a blade cutting the air but as a catcher's mitt held out in front of the vehicle. It slams the air to a halt and holds a thick cushion of shock-heated gas between itself and the ship — a buffer of hot air that keeps the worst heat out at arm's length in the shock layer, where it can radiate and blow away downstream. A knife-edge would have no cushion; the hot gas would ride right on the metal. Bluntness builds the cushion. Drag, the villain of ascent, is the hero of re-entry.

⚠️ Common Misconception: "The heat comes from friction with the air." Friction (skin friction) does contribute some heating, especially on the long, slanted surfaces of a lifting vehicle like the Shuttle. But for a blunt entry body the dominant heating is compression of the gas in the shock layer, which reaches its temperature whether or not it ever touches the skin. The tell is that the hottest region is the stagnation point at the very nose — the one place the air is brought most completely to rest, and therefore most completely compressed — not the flanks where "rubbing" would be greatest. Design a heat shield around friction and you protect the wrong surface. Design it around compression and you armor the nose, exactly where every real vehicle carries its thickest protection.

7.3 Thermal protection: shields, tiles, and sweat

We have $30\ \text{MJ/kg}$ of energy becoming a shock layer at $10{,}000\ \text{K}$, a few centimetres from the structure. The thermal protection system (TPS) is whatever stands between that inferno and the vehicle. There are three fundamentally different strategies, and the differences come straight from the physics of where you make the heat go.

First, though, a scaling law that governs all of them. The convective heat flux into the stagnation point — the watts per square metre arriving at the nose — follows, to good approximation, the Sutton–Graves relation:

$$ \dot q_{s} \;\propto\; \sqrt{\frac{\rho_\infty}{R_n}}\; v^{3}, $$

where $\rho_\infty$ is the local air density, $R_n$ is the nose radius, and $v$ is the speed. Two features of this formula decide the whole game. First, heating grows as the cube of speed — so an entry at $11\ \text{km/s}$ from the Moon is not merely twice as hot as one at $7.8\ \text{km/s}$ from LEO but about $(11/7.8)^3 \approx 2.8$ times worse in convective flux, and even more once radiation is added. Second, heating decreases with a larger nose radius $R_n$ — which is the quantitative reason the blunt body of §7.2 helps: a broad, rounded nose has a large $R_n$ and so a lower stagnation heat flux. Bluntness wins twice, deflecting energy into the shock layer and spreading the flux over a gentle curve.

🔧 Engineering Reality: Notice the tension the scaling law creates. The rate of heating $\dot q$ peaks high up, where $v$ is still enormous (the $v^3$ dominates); the total heat load — the integral of $\dot q$ over the whole descent — depends on how long you spend getting down. A steep entry has a high peak flux but a short duration; a shallow entry has a lower peak but a longer soak, and so a larger total heat load. This is why there is no single "hardest" entry: a ballistic warhead maximizes peak flux, while a shallow lifting entry maximizes total load. The two failure modes ask for different shields, and the corridor of §7.5 is where the trade is struck.

Strategy 1 — Ablative heat shields: carry the heat away with mass

The oldest and most robust strategy is to build the shield out of a material designed to destroy itself in a controlled way.

Definition (ablative heat shield). An ablative heat shield is a layer of material that protects a vehicle by absorbing heat and progressively eroding — heating, charring, melting, and vaporizing — so that the energy is carried away by the departing material rather than conducted into the structure. The vaporized gas also blows outward into the boundary layer, thickening the cushion of the shock layer and blocking some of the incoming convective heat, an effect called transpiration or blockage.

An ablator is a heat sink you throw away. As its outer surface chars and gasses off, three things happen at once: the phase changes (solid → char → gas) soak up energy endothermically; the hot char layer radiates heat back outward; and the outgassing pushes the searing shock-layer gas away from the surface. The result is that a few centimetres of a low-density resin can hold a $10{,}000\ \text{K}$ flow off a structure that must stay below $200\,^\circ\text{C}$. Apollo used an ablator called AVCOAT — an epoxy-novolac resin packed into a fiberglass honeycomb — and returned from the Moon at $11\ \text{km/s}$ with the shield charred but the crew at room temperature. SpaceX's Dragon uses PICA-X, an in-house descendant of NASA's PICA (Phenolic-Impregnated Carbon Ablator); the same PICA family shielded the Stardust sample capsule through the fastest human-made re-entry on record, about $12.9\ \text{km/s}$. Ablators are the choice when the entry is fastest and hottest, because throwing mass at the problem is the surest defence. Their cost is that the shield is largely single-use — it comes home smaller than it left.

Strategy 2 — Reusable insulating tiles: reject the heat by radiating it

The Space Shuttle wanted to fly again next month, so it could not char away its skin on every flight. Its strategy was the opposite of ablation: don't erode, insulate — hold a hot outer surface that radiates the heat back to the atmosphere while conducting almost none of it inward.

Definition (thermal tile). A thermal tile is a rigid, reusable block of extremely low-density refractory insulation — on the Space Shuttle, a silica-fiber ceramic that was roughly 90% empty air by volume — that survives re-entry without ablating. It protects by tolerating a very hot outer face (which re-radiates heat away) while conducting so little heat through its thickness that the structure beneath stays cool. Unlike an ablator, a tile is intended to fly many times.

The Shuttle's tiles were a marvel of materials engineering. The black high-temperature tiles could sit at $1{,}260\,^\circ\text{C}$ on their glassy coated surface while the aluminium airframe a few centimetres below stayed near room temperature. The famous demonstration was to pull a glowing tile from a furnace and hold it by the edges seconds later with bare fingers — the silica conducts heat so poorly, and radiates so efficiently, that the interior cools almost instantly. The hottest zones, the wing leading edges and the nose cap, where a tile would not survive, used reinforced carbon-carbon (RCC) panels good to about $1{,}650\,^\circ\text{C}$.

🔗 Connection: this is where Columbia was lost. The Shuttle's reusable TPS was brilliant and brittle. On the ascent of STS-107 in 2003, a piece of foam insulation shed from the external tank struck a reinforced carbon-carbon panel on Columbia's left wing leading edge and punched a hole in it. During re-entry, the $10{,}000\ \text{K}$ shock-layer gas of §7.2 poured through that hole into the wing structure and destroyed the vehicle. It is the starkest possible illustration of theme #2 — space is unforgiving; everything must work the first time — because there is no repair, no abort, and no second chance on the way down. We tell the Shuttle's full story, and the organizational failures behind both of its losses, in Chapter 37, and we draw the reliability lessons in Chapter 32.

Strategy 3 — Transpiration: sweat the heat away

The third strategy is the most elegant in principle and the rarest in practice: push a coolant out through a porous surface so that it carries heat away as it vaporizes and forms an insulating film between the hot gas and the wall — the vehicle, in effect, sweats. Transpiration (or "active" film) cooling is standard inside rocket engines, where fuel films protect the chamber wall (a story for Chapter 17), and it has been studied for re-entry for decades. SpaceX at one point proposed transpiration-cooling Starship by bleeding methane or water through tiny pores in its steel skin, before settling on ceramic tiles. The appeal is reusability without erosion; the difficulty is the plumbing, the coolant mass, and the certainty that every pore must stay clear. It remains mostly a promising idea rather than an operational workhorse.

🔧 Engineering Reality: the shield is not free — mass is still the enemy. A thermal protection system is heavy. A capsule's ablative shield and back-shell can be $10$–$20\%$ of the entry mass; the Shuttle's tiles and RCC together massed several tonnes. That is mass which is not payload and which had to be lifted all the way to orbit at the exponential cost of Chapter 3. Theme #4 — mass is the enemy — does not relent for re-entry. Every design is a fight between a shield thick enough to survive and thin enough to afford, which is exactly why engineers care so much about the total heat load (it sets shield thickness) and not only the peak temperature.

🔄 Check Your Understanding 1. Using the $\dot q \propto \sqrt{\rho/R_n}\,v^3$ scaling, give two independent reasons a blunt nose lowers the peak heat flux at the stagnation point compared with a sharp one. 2. Apollo (ablative) and the Space Shuttle (tiles) chose opposite TPS strategies. State the single mission requirement that most drove each choice.

Answers

1. A blunt nose has a large nose radius $R_n$, and the flux falls as $1/\sqrt{R_n}$, so a bigger radius directly lowers $\dot q$. Separately (from §7.2), a blunt body pushes the shock off the surface and dumps compression energy into the shock layer that blows away in the wake, so less of the total energy reaches the skin in the first place. 2. Apollo needed to survive the fastest, hottest possible entry (a one-time return from the Moon at $11\ \text{km/s}$), and ablation is the most robust defence against extreme flux — reuse was irrelevant for a capsule flown once. The Shuttle needed to be reused dozens of times without rebuilding its skin, which ruled out ablation and demanded reusable insulating tiles.

7.4 Ballistic vs. lifting re-entry

Two vehicles can leave the same orbit at the same speed and have wildly different rides home — one a brutal, brief, high-g plunge, the other a long, shallow, gentle glide. The difference is whether the vehicle makes lift, and how much drag it carries per unit mass. To reason about the second, we reach back for a tool from ascent.

🔗 Connection (ballistic coefficient, from Chapter 5). Recall the ballistic coefficient $\beta = m / (C_d A)$ — a body's mass divided by its drag area ($C_d$ its drag coefficient, $A$ its frontal area) — introduced in Chapter 5 as a measure of how strongly the atmosphere can decelerate a body. We do not redefine it here; we apply it to coming down. The drag deceleration a vehicle feels is $$\frac{D}{m} = \frac{\tfrac12 \rho v^2 C_d A}{m} = \frac{\rho v^2}{2\beta},$$ so for a given air density and speed the deceleration is inversely proportional to $\beta$. A low-$\beta$ body (light, broad, draggy — a capsule with a wide heat shield) decelerates readily even in thin, high air; a high-$\beta$ body (dense, compact — a warhead) knifes deep into the thick lower atmosphere before it slows.

That single fact organizes re-entry. Because a low-$\beta$ vehicle slows high up, where the air is thin, it decelerates gently and encounters its peak heating where the density is low. A high-$\beta$ vehicle punches down into dense air still moving fast, so it decelerates violently and heats fiercely at low altitude. This is why crewed capsules are deliberately built low-$\beta$: a broad blunt shield and a modest mass mean the atmosphere catches them softly and early. It is also why an intercontinental warhead, which wants to arrive fast and give defences no time, is built high-$\beta$ and simply endures the punishing heat with a heavy ablator. Same physics, opposite designs, because they want opposite things.

Now the second lever: lift.

Ballistic re-entry is the simplest: the vehicle makes essentially no lift and falls along a path set by gravity and drag alone, like a thrown stone. Mercury and the earliest capsules flew this way, and a Soyuz reverts to a ballistic entry if its guidance fails. It is robust and needs no active control — but because all the deceleration is crammed into a short, steep plunge, the g-forces and peak heating are high.

Lifting re-entry gives the vehicle an aerodynamic lift force it can point, and that changes everything. Even a small amount of lift lets the vehicle fly a shallower, longer trajectory, spreading the same energy dissipation over more time and more distance. The payoff is a lower peak deceleration and a lower peak heat flux (though, because the descent lasts longer, a larger total heat load — the trade the previous section warned about). Lift also buys cross-range: the ability to steer to a chosen landing site hundreds or thousands of kilometres to the side of the orbital ground track.

Vehicles generate this lift in two ways. A capsule like Apollo, Dragon, or Orion has no wings, but by placing its centre of mass slightly off-axis it flies at a small angle, producing a lift-to-drag ratio of about $L/D \approx 0.3$–$0.4$ — enough, when the capsule rolls to point that lift vector up, down, or sideways, to steer the entry and roughly halve the peak g. A spaceplane like the Space Shuttle used its whole hypersonic shape, flying at about a $40^\circ$ angle of attack (belly forward, presenting the blunt underside per §7.2) to reach $L/D \approx 1$ during entry, banking into long S-turns to bleed energy and reach a runway.

Worked Example: how lift tames the g-load. A ballistic capsule entering steeply from LEO can pull around $8$–$9\ \text{g}$; a lifting guided entry of the same capsule pulls about $4\ \text{g}$. The mechanism is simple to state in terms of §7.6's result: peak deceleration scales with how fast you lose speed, and lift lets you lose it slowly by holding the vehicle up in the thin upper air longer, stretching the deceleration over more seconds. Apollo, returning from the Moon at $11\ \text{km/s}$, would have suffered over $30\ \text{g}$ on a purely ballistic path — enough to injure the crew and strain the structure — so it flew a lifting entry that held the peak near $6$–$7\ \text{g}$. Lift did not reduce the total energy to dissipate; it reduced the rate, which is what the crew and the structure actually feel.

🧩 Productive Struggle. A lifting entry has a lower peak heat flux but a higher total heat load than a ballistic entry of the same vehicle. Before reading on, reason out which of those two quantities sets (a) the maximum surface temperature the shield must survive, and (b) the thickness (and mass) of the shield. Why might a designer choose the ballistic profile for a small robotic capsule and the lifting profile for a crew?

A way to think about it

Peak flux sets the maximum surface temperature (how hot it gets at the worst instant); total load (flux integrated over time) sets how much heat must be absorbed or insulated against, and therefore the shield's thickness and mass. A small robot can tolerate high g and high peak temperature, so a simple ballistic capsule with a robust ablator is cheapest and most reliable. A crew cannot tolerate high g, so a lifting entry is used to cap the deceleration — accepting a longer, heavier thermal soak in exchange for a survivable ride. The choice is a negotiation between what the payload can survive and what the shield can afford.

7.5 The re-entry corridor

A returning spacecraft does not get to enter the atmosphere any old way. There is a narrow band of entry angles that lead to a survivable descent, and missing it in either direction is fatal. That band is the re-entry corridor, and threading it is one of the most demanding pieces of trajectory design in all of spaceflight.

Definition (re-entry corridor). The re-entry corridor is the narrow range of entry conditions — chiefly the flight-path angle $\gamma_E$ at which the vehicle crosses the entry interface (about $120\ \text{km}$ altitude) — that produce a survivable re-entry. It is bounded on the steep side by limits on deceleration and heating (enter too steeply and the vehicle is crushed or burned) and on the shallow side by the skip-out limit (enter too shallowly and the vehicle fails to decelerate and bounces back out of the atmosphere).

The two walls of the corridor are the two failure modes:

• Too steep (dive in). A large downward flight-path angle drives the vehicle into dense air while it is still moving fast. The deceleration spikes — potentially beyond what the crew or structure can take — and the heating rate, which scales as $\rho v^3$, spikes with it. The vehicle burns up or is torn apart. This is the wall that killed the streamlined early designs before the blunt body was understood.
• Too shallow (graze). A small flight-path angle means the vehicle only skims the thin upper air. It decelerates too little, and its trajectory can arc back up and out of the atmosphere — a skip, like a stone bouncing off a pond. For a capsule returning from the Moon or from interplanetary space at or above escape speed, a skip can fling it back onto a long orbit with no guarantee of return, or with the crew's air and power exhausted before it comes back. Even short of a full skip-out, too shallow means overshooting the landing zone and a dangerously prolonged heat soak.

      space  ·  ·  ·  ·  entry interface (~120 km)  ·  ·  ·  ·  ·
                    \            |             /
       too shallow →  \      CORRIDOR        /  ← too steep
       (skip back out) \   (survivable:     /    (excessive g and
        γ too small     \   ~1–2° wide)    /      heating — burn up)
                         \       |        /        γ too large
      ~~~~~~~~~~~~~~~~~~~~~ dense atmosphere ~~~~~~~~~~~~~~~~~~~~~~~~
                                 ↓
                            landing site

How narrow is the corridor? For Apollo's return from the Moon, the survivable band of entry angles was only about one to two degrees wide, centred near $\gamma_E \approx -6.5^\circ$ (measured below the local horizontal at the entry interface). Steeper than roughly $-7.4^\circ$ and the deceleration exceeded what the crew could survive; shallower than about $-5.2^\circ$ and the capsule would skip back toward space. A navigation error of a couple of degrees in the flight-path angle, after a quarter-million-mile coast from the Moon, would have been lethal — which is why so much of Apollo's guidance and navigation effort went into hitting that entry angle exactly. (These corridor figures are widely reported approximations, Tier 2.)

🔗 Connection: lift widens the corridor. A purely ballistic vehicle has only whatever corridor its fixed shape gives it. A lifting vehicle (§7.4) can actively widen its corridor by rolling its lift vector: point lift up to keep from diving too steeply, or point it down to bite into the atmosphere and keep from skipping out. This is the deep reason guided lifting entry is safer than ballistic — the pilot or the computer can steer within the corridor in real time, correcting the entry angle that a ballistic capsule is simply stuck with. It is also how a controlled skip entry can be used on purpose: NASA's Orion capsule, returning from the Moon for Artemis, deliberately skips once to extend its range and reach a Pacific landing zone while capping its deceleration — the failure mode of §7.5, turned into a guidance technique. The corridor is not just a constraint to survive; for a lifting vehicle it is a channel to be flown.

🐛 Find the Error. A student designing a Mars sample-return capsule reasons: "To be safe, I'll enter as shallowly as possible — a grazing entry means the gentlest deceleration and the lowest g-forces, so it's the safest choice for a fragile sample." Identify what is wrong with this reasoning.

Answer

Shallow is only "gentle" up to a point; past the shallow wall of the corridor it becomes catastrophic, not gentle. Enter too shallowly and the capsule skips back out of the atmosphere entirely, failing to land at all — for a sample-return capsule with no propulsion and limited battery, a skip means the sample is lost on an uncontrolled trajectory. And even short of skipping, a very shallow entry stretches the heat load over a long soak, which can demand a heavier shield (§7.3), not a lighter one. The safe choice is not "as shallow as possible" but "inside the corridor" — steep enough to commit to the descent and be captured, shallow enough to stay under the g and peak-heating limits. Safety lives in the middle, not at either wall.

7.6 G-forces and communication blackout

Two experiences define the human ride through re-entry: the crushing weight of deceleration, and the eerie minutes of radio silence. Both come straight from the physics of the preceding sections.

Peak deceleration: a result that ignores the vehicle

How many g's does a re-entering vehicle pull, and what sets the number? There is a beautiful classical answer, due again to Allen and Eggers, for a ballistic entry into an exponential atmosphere. We will state it and read it, and only sketch the derivation.

Strategy first. Model the atmosphere as density falling off exponentially with altitude, $\rho = \rho_s e^{-h/H}$, where $H \approx 7\ \text{km}$ is the scale height (from Chapter 5). For a steep ballistic entry we can treat the flight-path angle $\gamma_E$ as roughly constant and neglect gravity next to the ferocious drag. Writing the deceleration as a function of altitude and finding where it peaks gives a compact closed form.

Carrying that through — integrating the drag equation of motion down through the exponential atmosphere — the speed at any depth turns out to be $v = v_E \exp[-\rho H/(2\beta \sin\gamma_E)]$, and the deceleration $a = \rho v^2 / 2\beta$ reaches its maximum at the altitude where the exponent equals $\tfrac12$. Substitute that back and, remarkably, the ballistic coefficient $\beta$ cancels out:

$$ a_{\max} \;=\; \frac{v_E^2\,\sin\gamma_E}{2\,e\,H}, $$

where $e = 2.718$ is Euler's number (not a speed here). Read what this says: the peak deceleration of a ballistic entry depends only on the entry speed, the entry angle, and the atmosphere's scale height — not on the vehicle's mass, size, drag, or ballistic coefficient. A ping-pong ball and a cannonball entering at the same angle and speed pull the same peak g. The vehicle's $\beta$ decides where (at what altitude) the peak happens, but not how large it is.

Worked Example: peak g for a capsule entry from LEO. Take a shallow guided entry: $v_E = 7.6\ \text{km/s}$, $\gamma_E = 1.5^\circ$ (so $\sin\gamma_E = 0.0262$), and $H = 7{,}200\ \text{m}$. Then $$a_{\max} = \frac{(7{,}600)^2 (0.0262)}{2\,(2.718)\,(7{,}200)} > = \frac{(5.78\times10^{7})(0.0262)}{39{,}140} > = \frac{1.51\times10^{6}}{39{,}140} = 38.6\ \text{m/s}^2 \approx 3.9\ \text{g}.$$ About four g — a heavy but survivable load, matching the ~$4\ \text{g}$ of a nominal Soyuz or Dragon guided entry. Now make it steeper, $\gamma_E = 6^\circ$ ($\sin = 0.105$): the peak jumps to $a_{\max} \approx 154\ \text{m/s}^2 \approx 16\ \text{g}$ — the same speed, four times the entry angle, four times the g. And a ballistic return from the Moon ($v_E = 11\ \text{km/s}$, $\gamma_E = 6.5^\circ$) would give $a_{\max} \approx 350\ \text{m/s}^2 \approx 36\ \text{g}$, which is why Apollo flew a lifting entry (§7.4) to hold the peak near $6$–$7\ \text{g}$. The formula is an idealization — real shallow entries curve rather than hold $\gamma$ constant, and $H$ varies with altitude — so treat these as order-of- magnitude, but the scaling ($a_{\max}\propto v_E^2 \sin\gamma_E$) is exactly why steep and fast entries are the dangerous ones.

Communication blackout: the plasma sheath

Recall from §7.2 that the shock-layer gas gets hot enough to ionize — to strip electrons off atoms, producing a soup of free electrons and ions: a plasma. That plasma wraps the vehicle in a conducting shell, and a conducting shell does to radio waves what a metal mirror does to light: it reflects them. For several minutes near peak heating, the spacecraft is sealed inside a cocoon of ionized air that its radio signals cannot escape. Ground controllers hear nothing.

Definition (plasma blackout). Plasma blackout (communication blackout) is the interruption of radio contact with a re-entering vehicle caused by the layer of ionized gas — the plasma sheath — that forms in and around the shock-heated flow. Free electrons in the plasma reflect and absorb radio waves below a critical frequency (the plasma frequency), cutting off transmission and reception until the vehicle slows and the sheath cools and recombines.

The physics is quantitative. A plasma transmits radio waves above its plasma frequency and reflects them below it; that frequency depends only on the free-electron density $n_e$:

$$ f_p \approx 8.98\,\sqrt{n_e}\ \ \text{Hz} \quad (n_e\ \text{in electrons per m}^3). $$

Worked Example: what gets blocked. Near peak heating the electron density in the sheath is on the order of $n_e \sim 10^{17}$–$10^{18}\ \text{m}^{-3}$. Take $n_e = 10^{17}\ \text{m}^{-3}$: $$f_p = 8.98\,\sqrt{10^{17}} = 8.98\times(3.16\times10^{8}) \approx 2.8\times10^{9}\ \text{Hz} = 2.8\ \text{GHz}.$$ At the denser $10^{18}\ \text{m}^{-3}$, $f_p \approx 9\ \text{GHz}$. Any radio link below that frequency — which includes the VHF, UHF, and S-band ($\sim 2.2\ \text{GHz}$) frequencies most spacecraft use — is reflected by the sheath and cannot get through. That is the blackout. The way out is to slow down (so the sheath cools and its electrons recombine, dropping $n_e$ and $f_p$) or to use a much higher frequency that rides above $f_p$.

For Mercury, Gemini, and Apollo, blackout meant three to four minutes of total silence — the tensest interval of any mission, when controllers could only wait to hear whether the capsule had survived. The Space Shuttle softened it with a trick of geometry: the plasma is thinnest and coolest on the vehicle's leeward (upper) side, so the orbiter relayed its signals upward through the Tracking and Data Relay Satellites in high orbit, punching through the thin part of the sheath and cutting the blackout dramatically. The physics never went away; the engineering routed around it.

📜 From History: four minutes of silence. During Apollo re-entries, flight controllers in Houston knew the exact clock time at which the plasma sheath would swallow the capsule's signal and the exact time it should reappear. In between there was nothing to do but watch the clock — a self-imposed helpless interval built into every lunar return, a few minutes when the most advanced machine humanity had ever built was simply gone, its fate already decided by a heat shield poured months earlier. It is the purest expression of theme #2 in the crewed program: on the way down, everything must already have been done right, because there is nothing left to do. We return to the human side of these minutes — the mission-control room and the wait — in Chapter 31.

🔄 Check Your Understanding 1. Two capsules enter at the same speed and angle, but one is twice as heavy (higher $\beta$). Which pulls the higher peak g, and at what altitude does each peak? 2. Why does communication blackout end as the vehicle slows down?

Answers

1. Neither pulls a higher peak g — the Allen–Eggers result $a_{\max} = v_E^2\sin\gamma_E/(2eH)$ is independent of $\beta$, so both reach the same peak deceleration. But the heavier, higher-$\beta$ capsule reaches that peak lower in the atmosphere (it penetrates deeper before decelerating), where the air is denser. The lighter capsule peaks higher up. 2. As the vehicle slows, the shock layer cools; below the ionization threshold the free electrons recombine with ions, so the electron density $n_e$ falls, the plasma frequency $f_p = 8.98\sqrt{n_e}$ drops below the radio link's frequency, and signals pass through again. Blackout is a high-speed, high-heating phenomenon; it ends when the heating does.

Mission Design Checkpoint: does your mission come home?

Across this book you are developing one mission — a comsat to GEO (Track A), a lunar lander (Track B), a Mars orbiter (Track C), or an asteroid rendezvous (Track D). This chapter adds a question that reshapes the vehicle for some tracks and not at all for others: does any part of your mission enter an atmosphere, and if so, how fast and how hot?

The design. Open your Mission Design Review (MDR) and add a short Entry / Disposal note:

• Track A (GEO comsat): does not re-enter. Geostationary satellites are boosted at end of life to a graveyard orbit (Chapter 35), never facing an atmosphere. Record "no atmospheric entry; disposal to graveyard orbit." (A satellite left in low orbit is the opposite case — it will deorbit and demise in the atmosphere, by design or by drag.)
• Track B (lunar lander): the Moon has no atmosphere, so there is no free brake — every bit of descent deceleration must be done propulsively, at full delta-v cost. This is the flip side of §7.1: with no air to convert speed to heat, a lunar landing pays the rocket-equation price in propellant that Earth entry avoids. Note "fully propulsive descent; no TPS; ~$2\ \text{km/s}$ of landing delta-v."
• Track C (Mars orbiter): Mars has a thin atmosphere — enough to heat you, not enough to stop you easily ("too thin to land on comfortably, too thick to ignore"). Aerocapture or aerobraking can save propellant on arrival (Chapter 11, Chapter 34); note the entry speed and whether you will use the atmosphere for braking.
• Track D (asteroid rendezvous): the asteroid has no atmosphere, but a sample-return capsule faces the fastest, hottest Earth entry of all — $12\ \text{km/s}$ or more, like Stardust and OSIRIS-REx. Record the return entry speed and an ablative-TPS note.

The code. No canonical astrotools module is assigned to this chapter, so treat this as a standalone analysis helper you can keep in your MDR folder — it estimates the peak g (Allen–Eggers) and whether you'll suffer blackout:

import math

def peak_g(v_entry, gamma_deg, scale_height=7200.0):
    """Peak deceleration of a ballistic entry, in Earth g's (Allen-Eggers).
    Independent of the vehicle's ballistic coefficient.
    v_entry in m/s, gamma_deg = entry flight-path angle below horizontal."""
    a_max = v_entry**2 * math.sin(math.radians(gamma_deg)) / (2 * math.e * scale_height)
    return a_max / 9.81

def plasma_frequency_ghz(n_e):
    """Plasma frequency (GHz) for electron density n_e (electrons/m^3).
    Radio links below this frequency are blacked out."""
    return 8.98 * math.sqrt(n_e) / 1e9

print(round(peak_g(7600, 1.5), 1), "g   (shallow LEO entry)")
print(round(peak_g(11000, 6.5), 1), "g   (ballistic lunar return)")
print(round(plasma_frequency_ghz(1e17), 1), "GHz blocked")
# Expected output:
# 3.9 g   (shallow LEO entry)
# 35.7 g   (ballistic lunar return)
# 2.8 GHz blocked

Those three numbers — a survivable ~4 g for a shallow LEO entry, a bone-crushing ~36 g for a ballistic lunar return (hence lifting entry), and a blackout that blocks everything below ~3 GHz — feed the entry section of your MDR and, for the returning tracks, justify the thermal-protection choice you will carry to the capstone. If your mission never comes home, you have just saved yourself a heat shield; write down why, because that is a design decision too.

Summary

Re-entry is the disposal of orbital energy as heat, done deliberately and survived by design. Carry these forward:

Numbers worth memorizing: orbital kinetic energy $\approx 30\ \text{MJ/kg}$; deorbit burn $\sim 100\ \text{m/s}$; shock-layer gas $\sim 10{,}000\ \text{K}$; heating $\propto v^3$; peak g $\propto v_E^2 \sin\gamma_E$; Apollo corridor ~$1$–$2^\circ$; blackout blocks below ~3 GHz.

Spaced Review

Retrieval strengthens memory. Answer from memory before checking, then look back at the cited chapter.

1. (Ch. 6) In terms of specific orbital energy, re-entry is the removal of the energy Chapter 6 taught you to compute. For a circular LEO orbit, how is the kinetic energy per kilogram related to the magnitude of the specific orbital energy, and why does that make ~$30\ \text{MJ/kg}$ the amount you must shed?
2. (Ch. 5) The ballistic coefficient $\beta = m/(C_dA)$ appeared on ascent. In re-entry, does a *high*-$\beta$ or *low*-$\beta$ vehicle decelerate higher in the atmosphere, and which one do you want for a crewed capsule?
3. (Ch. 5) Re-entry heating scales as $\rho v^3$ and ascent drag as $\rho v^2$, both containing the density $\rho$ that falls off with a scale height of about $8\ \text{km}$. Why does the peak of each occur at a middle altitude rather than at the ground or in vacuum?
4. (Ch. 6) A vehicle returning from a high, energetic orbit hits the atmosphere faster than one from LEO. Using $E_k \propto v^2$ and heating $\propto v^3$, explain why interplanetary-return entries are so disproportionately punishing.

Answers

1. For a circular orbit the specific orbital energy is $\varepsilon = -\tfrac12 v^2$ (kinetic $+\tfrac12 v^2$ plus potential $-\mu/r = -v^2$ gives $-\tfrac12 v^2$), so the magnitude of the specific orbital energy equals the kinetic energy per kilogram, $\tfrac12 v^2 \approx 30\ \text{MJ/kg}$. That kinetic energy is exactly what the atmosphere must remove to bring you to rest, which is why the number that defined the orbit in Chapter 6 is the number the heat shield must dispose of here. 2. A low-$\beta$ vehicle decelerates higher (it is draggy per unit mass, so thin high air is enough to slow it); you want low $\beta$ for a crewed capsule so the deceleration is gentle and the peak heating happens where the air is thin. A high-$\beta$ vehicle knifes deep and decelerates violently in dense air. 3. Because both the density $\rho$ and the speed $v$ vary in opposite senses through the descent: high up, $v$ is huge but $\rho$ is nearly zero; low down, $\rho$ is large but the vehicle has already slowed. The product $\rho v^n$ is small at both ends and peaks in the middle, where there is simultaneously enough air and enough speed — the same conspiracy that put max-Q at a middle altitude on the way up. 4. Kinetic energy grows as $v^2$, so a faster entry carries proportionally more energy to dump; but the heating rate grows even faster, as $v^3$ (and radiative heating faster still), so a modest increase in entry speed inflates the peak heat flux dramatically. A lunar return at $11\ \text{km/s}$ has ~2× the energy but nearly 3× the convective flux of a LEO entry — which is why the fastest returns need the most aggressive ablative shields.

What's Next

Part I is complete. We can now get up (the rocket equation and ascent), stay up (orbital speed and energy), and come home (re-entry) — the full arc of a flight, understood from first principles of momentum, gravity, and energy. Along the way one object kept reappearing without ever being examined on its own terms: the orbit itself. We have treated orbits as speeds to reach and energies to carry, but we have not yet asked what shape an orbit is, why it is that shape, how long it takes to go around, or where in its loop a spacecraft will be at a given time. Those questions are the province of orbital mechanics, and they are surprisingly beautiful — the same inverse-square gravity of Chapter 2, followed carefully, produces ellipses, and the ellipses obey laws that Kepler found by staring at data decades before Newton explained them. In Chapter 8 we open Part II by deriving Kepler's laws from Newton's gravity and learning to describe any orbit with six numbers. The physics of getting there is behind us; the geometry of being there begins now.