Chapter 10: Orbital Maneuvers

"Once you are in Earth orbit, you're halfway to anywhere in the solar system." — Robert A. Heinlein

Overview

You are in low Earth orbit. Now what?

That question is the whole of this chapter, and it is where orbital mechanics stops being a description of where things are and becomes a toolkit for putting them where you want. Reaching orbit, the triumph of Part I, only delivers you to a parking spot — a circle a few hundred kilometers up, going nowhere in particular except around. Every mission that matters begins from there and has to move: a communications satellite must climb from its drop-off orbit to the geostationary belt; a cargo vehicle must chase down and dock with a space station; a probe must leave Earth's orbit entirely and fall toward Mars. Each of those is an orbital maneuver — a deliberate change of orbit, bought with delta-v, planned with vis-viva.

Here is the beautiful part, and it is the theme of this whole book made concrete: we already have every tool we need. In Chapter 6 we derived the vis-viva equation, which gives the speed at any point on any orbit from just two numbers. That single equation, applied at the right two points, tells us exactly how much delta-v it costs to hop from one orbit to another. Orbital mechanics is beautiful precisely because so much falls out of so little: the same $v^2 = \mu(2/r - 1/a)$ that told us how fast the ISS moves will now tell us how to fly to the Moon, and the arithmetic is honest — every maneuver is a difference of two vis-viva speeds, and every difference is a bill your rocket has to pay.

But the tools also expose a hard truth, and it is worth stating up front because it governs mission design more than anything else in Part II: changing an orbit is expensive, and changing the plane of an orbit is brutally expensive. The delta-v map of Chapter 3 listed prices like "LEO to GEO, 2.5 km/s" as if they were bus fares; in this chapter we derive those fares from first principles and discover why they are what they are — and why a mission planner will bend over backwards to avoid turning the orbital plane even a few degrees. Every one of those kilometers per second, through the tyranny of the rocket equation, becomes propellant, and propellant becomes mass, and mass is the enemy.

In this chapter, you will learn to:

• Model a burn as an impulsive maneuver — an instantaneous kick — and read off the new orbit it creates.
• Derive the Hohmann transfer, the minimum-fuel way between two circular orbits, and compute the real LEO-to-GEO transfer that every comsat flies.
• Decide when a three-burn bi-elliptic transfer can beat it, and appreciate the price in time.
• Put a number on why plane changes hurt: $\Delta v = 2v\sin(\Delta i/2)$.
• Understand phasing and rendezvous — how two spacecraft actually find each other in the dark.
• Combine maneuvers to save fuel, the trick behind every efficient trajectory ever flown.

Learning Paths

🚀 Space Enthusiast: Read 10.2 (the Hohmann transfer) and 10.4 (plane changes) for the two ideas that will change how you watch every satellite launch. You can take the derivations on faith and focus on the worked LEO-to-GEO example and the "why plane changes are so expensive" punchline.

📐 Engineering Student: Read everything and do the Hohmann derivation in 10.2 yourself before reading ours — it is nothing but vis-viva applied twice, and once you can reproduce it you own transfers forever. The ⭐⭐/⭐⭐⭐ exercises here are the direct foundation for the interplanetary trajectories of Chapter 11.

🎮 KSP Player: You have flown every maneuver in this chapter, probably by trial and error. Sections 10.2, 10.5, and 10.6 will turn your muscle memory into numbers — especially why the game's maneuver-node delta-v readout says what it says, and why plane changes eat your fuel budget alive.

🛰️ Industry Prep: This is the chapter where the delta-v budget of your Mission Design Review gets its real numbers. Section 10.6 (combined maneuvers) and the Checkpoint's maneuvers.py are exactly how launch and satellite operators price a mission. Do the design exercises.

10.1 Impulsive maneuvers

Every orbit change starts with a rocket burn, and a real burn is a messy, drawn-out thing: an engine fires for seconds or minutes, thrust builds and tails off, the vehicle steers, and the spacecraft traces a complicated powered arc while its mass drops as propellant leaves. Modeling all of that in detail is the work of a trajectory-optimization program. For understanding — and for a first, remarkably accurate cut at almost any maneuver — we make one liberating simplification.

Strategy first. If the burn lasts a short time compared with the orbital period — seconds or a few minutes against an orbit that takes ninety minutes or a day — then the spacecraft barely moves along its orbit while the engine is firing. So we pretend the burn is instantaneous: it happens at a single point, changing the spacecraft's velocity vector all at once while leaving its position untouched. This is the impulsive approximation, and it turns every maneuver into a clean before-and-after problem in vis-viva.

Definition (impulsive maneuver). An impulsive maneuver is an idealized orbit change in which the engine burn is treated as instantaneous: the spacecraft's velocity changes by a vector $\Delta\mathbf{v}$ at a single point, while its position $\mathbf{r}$ is unchanged. The magnitude $\Delta v = |\Delta\mathbf{v}|$ is the delta-v the burn costs, priced through the rocket equation of Chapter 3; the direction of $\Delta\mathbf{v}$ decides how the orbit changes.

The power of this idea is that it connects two orbits at a shared point. Before the burn, the spacecraft is on some orbit passing through the point $\mathbf{r}$ at velocity $\mathbf{v}_1$. You apply $\Delta\mathbf{v}$. Now it is at the same point $\mathbf{r}$ but with a new velocity $\mathbf{v}_2 = \mathbf{v}_1 + \Delta\mathbf{v}$ — and that new position-and-velocity pair defines a new orbit, which the spacecraft immediately begins to follow. The two orbits always intersect at the burn point, because that is where the switch happened. This single fact is the skeleton of every transfer in the chapter: to get from orbit A to orbit B, you need a burn point they share (or a bridging orbit that touches both), and a $\Delta\mathbf{v}$ that swings your velocity from the A-value to the B-value there.

The direction you point the burn has standard names, worth fixing now because we use them constantly:

• Prograde (along the velocity): raises the orbit on the opposite side. A prograde burn adds energy, so by the rules of Chapter 6 it lifts the point diametrically opposite the burn — fire prograde at perigee and you raise apogee.
• Retrograde (against the velocity): the mirror image — it removes energy and lowers the far side.
• Radial (toward or away from the body): rotates the orbit's orientation (its line of apsides) without changing its energy much — useful for fine-tuning where periapsis points.
• Normal / anti-normal (perpendicular to the orbital plane): tilts the plane itself. This is the plane change of Section 10.4, and it is the expensive one.

💡 Intuition: a burn is a nudge to one point of the orbit. Picture the orbit as a loop of wire you can only reshape by pinching it at the spot where you happen to be standing. A prograde pinch at perigee pulls the far side of the loop outward, stretching apogee higher while leaving perigee — where you're standing — put. That is why raising an orbit takes two burns: the first, at your current altitude, lifts the opposite side to the target height; the second, once you have coasted around to that opposite side, lifts the side you started from to match. One burn changes one end. Two burns move both.

Worked Example: one prograde nudge in LEO. A spacecraft is in a circular $400\ \text{km}$ orbit, $r = 6{,}771\ \text{km}$, moving at $v_{\text{circ}} = 7.673\ \text{km/s}$ (from Chapter 6). It fires prograde with $\Delta v = 0.100\ \text{km/s}$. Its new speed at that point is $7.773\ \text{km/s}$, still at $r = 6{,}771\ \text{km}$. What orbit is it on now? Use the energy form: the new semi-major axis follows from vis-viva solved for $a$, $$\frac{1}{a} = \frac{2}{r} - \frac{v^2}{\mu} = \frac{2}{6{,}771} - \frac{7.773^2}{3.986\times10^{5}} > = 2.9538\times10^{-4} - 1.5157\times10^{-4} = 1.4381\times10^{-4}\ \text{km}^{-1},$$ so $a = 6{,}954\ \text{km}$. The burn point ($6{,}771\ \text{km}$) is now perigee; the new apogee is $r_a = 2a - r_p = 2(6{,}954) - 6{,}771 = 7{,}137\ \text{km}$ — an altitude of $766\ \text{km}$, up from $400$. A $100\ \text{m/s}$ prograde tap at perigee lifted apogee by about $366\ \text{km}$, and the far side of the orbit rose while the near side stayed put. Sanity check: the burn added energy, so the orbit grew ($a$ went up from $6{,}771$ to $6{,}954\ \text{km}$), exactly as a prograde burn must.

🔧 Engineering Reality: when "impulsive" breaks down. The approximation is excellent for high-thrust chemical engines, whose burns last seconds to a few minutes. It fails for the gentle, months-long push of an electric thruster (Chapter 20), which spirals slowly outward through thousands of revolutions and can never be treated as a kick at a point — a low-thrust spiral to GEO actually costs more total delta-v than the impulsive Hohmann below, though far less propellant, because its enormous exhaust velocity wins on the rocket equation anyway. It also frays for very large burns deep in a gravity well, where the spacecraft really does travel a meaningful arc during the burn, incurring a small "finite-burn loss" that flight dynamicists add back as a correction. For this chapter, every burn is impulsive, and every answer is the ideal the real maneuver is measured against.

🔄 Check Your Understanding 1. You fire a retrograde burn at a point in a circular orbit. Does the orbit's opposite side rise or fall? Where does the burn point become — periapsis or apoapsis? 2. Why does an impulsive burn leave your position unchanged but your velocity changed?

Answers

1. It falls. A retrograde burn removes energy, lowering the diametrically opposite point; the burn point becomes the new apoapsis (the highest point), and the orbit dips below it on the far side. 2. Because we idealize the burn as instantaneous: in zero elapsed time the spacecraft cannot move along its path ($\mathbf{r}$ is fixed), but the engine's impulse changes its momentum, so $\mathbf{v}$ jumps by $\Delta\mathbf{v}$. Real burns take time and move a little; the approximation is good when the burn is short compared with the orbital period.

10.2 The Hohmann transfer

Now the central maneuver of the entire chapter, and the anchor we will build on all the way to Mars. You are in a circular orbit of radius $r_1$ and you want to be in a larger circular orbit of radius $r_2$, in the same plane. What is the cheapest way — the one that spends the least delta-v?

The answer was published in 1925 by Walter Hohmann, a German engineer who worked it out for fun, decades before anyone could fly it, and it is so clean it feels inevitable. Connect the two circles with a single ellipse that just kisses the inner circle at its perigee and the outer circle at its apogee. Two burns fly it: one at $r_1$ to leave the small circle and coast up the ellipse, and one at $r_2$ to stop climbing and settle into the big circle. That tangent half-ellipse is the Hohmann transfer, and for most orbit changes it is the minimum-delta-v solution.

💡 Intuition: why touch each circle tangentially? A burn is most efficient when it points along your velocity, adding pure speed rather than wastefully turning your direction. At the two ends of the transfer ellipse — its perigee and apogee — the spacecraft moves purely sideways, exactly parallel to the circular orbits it is joining. So both burns are pure prograde/retrograde speed changes with no direction wasted. Any transfer that met the circles at an angle would spend delta-v swinging the velocity vector around, and pay for it. Tangent is thrifty.

Deriving the two burns

Strategy first. We need four speeds and we already own the equation for all of them — vis-viva, $v^2 = \mu(2/r - 1/a)$. At $r_1$: the circular speed we are leaving, and the transfer-ellipse's perigee speed we are joining; their difference is the first burn. At $r_2$: the transfer-ellipse's apogee speed we arrive with, and the circular speed we need; their difference is the second burn. Add the two. That is the entire derivation — no calculus, just vis-viva evaluated at two radii on three orbits.

The transfer ellipse runs from perigee $r_1$ to apogee $r_2$, so its semi-major axis is half the long diameter, $$ a_t = \frac{r_1 + r_2}{2}. $$ The first burn, at $r_1$. We leave the inner circle, where the speed is $v_{c1} = \sqrt{\mu/r_1}$, and join the ellipse, whose speed at perigee ($r = r_1$) is $v_{p} = \sqrt{\mu\!\left(\frac{2}{r_1} - \frac{1}{a_t}\right)}$. Substituting $a_t$ and simplifying the algebra (a common denominator turns $\frac{2}{r_1} - \frac{2}{r_1+r_2}$ into $\frac{2r_2}{r_1(r_1+r_2)}$) gives a tidy closed form: $$ \Delta v_1 = v_p - v_{c1} = \sqrt{\frac{\mu}{r_1}}\left(\sqrt{\frac{2r_2}{r_1 + r_2}} - 1\right). $$ Because $r_2 > r_1$, the bracket is positive: the first burn is a speed-up (prograde), as it must be to raise the far side of the orbit to $r_2$.

The second burn, at $r_2$. We arrive at the ellipse's apogee moving slowly, at $v_{a} = \sqrt{\mu\!\left(\frac{2}{r_2} - \frac{1}{a_t}\right)} = \sqrt{\frac{\mu}{r_2}}\sqrt{\frac{2r_1}{r_1+r_2}}$, and we need the outer circular speed $v_{c2} = \sqrt{\mu/r_2}$, which is faster. So the second burn is also a speed-up: $$ \Delta v_2 = v_{c2} - v_a = \sqrt{\frac{\mu}{r_2}}\left(1 - \sqrt{\frac{2r_1}{r_1 + r_2}}\right). $$ The total Hohmann cost is the sum, $$ \boxed{\ \Delta v_{\text{Hohmann}} = \Delta v_1 + \Delta v_2 = \sqrt{\frac{\mu}{r_1}}\left(\sqrt{\frac{2r_2}{r_1 + r_2}} - 1\right) + \sqrt{\frac{\mu}{r_2}}\left(1 - \sqrt{\frac{2r_1}{r_1 + r_2}}\right)\ } \qquad\blacksquare $$

Definition (Hohmann transfer). A Hohmann transfer is the two-impulse maneuver between two coplanar circular orbits along an ellipse tangent to both — perigee on the inner orbit, apogee on the outer. It uses the least delta-v of any two-burn transfer between the orbits, and (for orbit-radius ratios below about $11.94$) the least of any impulsive transfer. Raising the orbit takes two prograde burns; lowering it takes the same two burns run backward, both retrograde.

The transfer takes half an orbit

The spacecraft coasts from perigee to apogee — exactly half of the transfer ellipse — so the trip time is half the ellipse's period. Using Kepler's third law from Chapter 8, $T = 2\pi\sqrt{a^3/\mu}$, $$ t_{\text{transfer}} = \frac{T_t}{2} = \pi\sqrt{\frac{a_t^{\,3}}{\mu}}. $$ You do not get to choose this time — it is fixed by the two radii. That rigidity matters enormously when the destination is moving (another planet, a space station), which is the whole subject of launch windows in Chapter 11 and of rendezvous in Section 10.5.

Worked Example: LEO to GEO, the transfer every comsat flies. Take the workhorse case — from a $400\ \text{km}$ circular parking orbit ($r_1 = 6{,}771\ \text{km}$) up to geostationary altitude ($r_2 = 42{,}164\ \text{km}$), using $\mu_\oplus = 3.986\times10^{5}\ \text{km}^3/\text{s}^2$.

The circular speeds are $v_{c1} = 7.673\ \text{km/s}$ and $v_{c2} = 3.075\ \text{km/s}$ (both computed in Chapter 6). The transfer ellipse has $$a_t = \frac{6{,}771 + 42{,}164}{2} = 24{,}467.5\ \text{km}.$$ First burn, at $r_1$: the perigee speed of the ellipse is $$v_p = \sqrt{3.986\times10^{5}\!\left(\tfrac{2}{6{,}771} - \tfrac{1}{24{,}467.5}\right)} > = \sqrt{101.45} = 10.072\ \text{km/s},$$ so $\Delta v_1 = 10.072 - 7.673 = 2.399 \approx 2.40\ \text{km/s}$ (prograde). Second burn, at $r_2$: the apogee speed is $$v_a = \sqrt{3.986\times10^{5}\!\left(\tfrac{2}{42{,}164} - \tfrac{1}{24{,}467.5}\right)} > = \sqrt{2.616} = 1.617\ \text{km/s},$$ so $\Delta v_2 = 3.075 - 1.617 = 1.458 \approx 1.46\ \text{km/s}$ (prograde).

Total: $\Delta v_{\text{Hohmann}} = 2.40 + 1.46 = 3.86\ \text{km/s}.$

Sanity check. Chapter 3's delta-v map listed "LEO $\to$ GTO $\approx 2.5$" and "GTO $\to$ GEO $\approx 1.5$" km/s; our derived $2.40$ and $1.46$ match to within the map's rounding (the small differences come from its slightly different reference altitude). The transfer ellipse here is exactly the geostationary transfer orbit (GTO) of Chapter 9 — the same ellipse Chapter 6 used to introduce vis-viva. We have now derived the map's fares from first principles.

Transfer time: $t = \pi\sqrt{24{,}467.5^{3}/3.986\times10^{5}} = \pi\sqrt{3.675\times10^{7}} = \pi\,(6{,}062\ \text{s}) = 19{,}044\ \text{s} \approx 5.3\ \text{hours}.$ A satellite dropped into GTO coasts for a bit over five hours, climbing and slowing, before its apogee-kick engine fires to circularize at GEO — which is exactly the timeline real operators fly.

🚪 Threshold Concept: a transfer is just two vis-viva speeds subtracted. Every orbit change in this book, from a $100\ \text{m}$ station-keeping nudge to a burn that flings a probe out of the solar system, is priced the same way: find your speed on the orbit you're on and your speed on the orbit you want at the point where they meet, and subtract. Delta-v is a difference of speeds; vis-viva supplies the speeds; the rocket equation turns the difference into propellant. Once you internalize that a maneuver is arithmetic on two orbits sharing a point, the entire subject collapses into a procedure you can carry out on the back of an envelope. This is what mission designers mean when they say a trajectory is "just" a delta-v budget — and it is the deepest sense in which orbital mechanics is beautiful: enormous journeys reduce to subtraction.

📜 From History: Walter Hohmann's paper spaceflight. Hohmann published the transfer in 1925 in a slim book, Die Erreichbarkeit der Himmelskörper ("The Attainability of Celestial Bodies"), working entirely on paper — there were no rockets that could reach orbit, and would not be for another thirty-two years. He was not even a rocketeer by trade; he was a municipal building surveyor in Essen who did celestial mechanics in his spare time. When the Space Age finally arrived, his minimum-energy ellipse turned out to be the route real spacecraft actually take to geostationary orbit and to the planets. It is one of the purest examples of the pattern this whole book celebrates: the physics was understood, exactly, long before the engineering existed to use it.

🔄 Check Your Understanding 1. Why does the Hohmann transfer use two burns rather than one? (Think back to 10.1 — what does a single burn do to an orbit?) 2. In the LEO-to-GEO transfer, the first burn ($2.40\ \text{km/s}$) is larger than the second ($1.46\ \text{km/s}$). Why is more of the cost paid down low?

Answers

1. Because one burn only moves the point of the orbit opposite the burn. The first burn (at $r_1$) raises apogee to $r_2$ but leaves perigee at $r_1$ — you are now on an ellipse, not the target circle. The second burn (at $r_2$) raises the other end, lifting perigee up to $r_2$ and rounding the ellipse into the outer circle. One burn per end. 2. The first burn does the heavy lifting of injecting all the extra orbital energy while deep in the well; the second only tops up the speed at apogee, where the orbit is already large and slow. Loosely, most of the energy of the bigger orbit is bought at the bottom, where a little delta-v buys a lot of energy (the Oberth effect: a burn is most energy-efficient where you move fastest).

10.3 When Hohmann is optimal, and the bi-elliptic transfer

The Hohmann transfer is a minimum, but not always the minimum. For a surprising range of orbit changes there is a stranger, three-burn maneuver that costs slightly less delta-v — at the price of a much longer trip. It is worth knowing, both because it occasionally earns its keep and because it sharpens what "optimal" really means.

The idea: instead of transferring directly out to $r_2$, deliberately overshoot. Burn at $r_1$ onto a tall ellipse whose apogee $r_b$ is far beyond $r_2$; coast all the way out to $r_b$, where you are barely moving; give a small prograde nudge there to raise your perigee up to $r_2$; then coast back down to $r_2$ and burn retrograde to circularize. Three burns, two ellipses, one enormous detour. This is the bi-elliptic transfer.

Definition (bi-elliptic transfer). A bi-elliptic transfer moves between two coplanar circular orbits ($r_1 \to r_2$) using two half-ellipses joined at a high intermediate apoapsis $r_b > r_2$, and three burns: a prograde burn at $r_1$ to raise apoapsis to $r_b$; a small prograde burn at $r_b$ to raise periapsis to $r_2$; and a retrograde burn at $r_2$ to circularize. For large ratios $r_2/r_1$ it can cost slightly less total delta-v than the Hohmann transfer, because the middle burn is performed where the spacecraft moves so slowly that changing the orbit is nearly free.

Why would a longer path ever be cheaper? Because of where the work is done. Way out at $r_b$ the spacecraft crawls (vis-viva: large $r$ means small $v$), and out there it takes only a whisper of delta-v to lever the perigee up to $r_2$. You trade the expensive direct climb for a cheap adjustment made at extreme altitude. Whether that trade pays off depends entirely on how big the jump is.

Worked Example: a bi-elliptic that beats Hohmann. Consider a large orbit change: $r_1 = 7{,}000\ \text{km}$ to $r_2 = 105{,}000\ \text{km}$, a ratio of $15$. Circular speeds: $v_{c1} = 7.546\ \text{km/s}$, $v_{c2} = 1.948\ \text{km/s}$.

Hohmann ($a_t = 56{,}000\ \text{km}$): $\Delta v_1 = 7.546\left(\sqrt{\tfrac{2(105{,}000)}{112{,}000}} - 1\right) = 7.546(\sqrt{1.875} - 1) = 2.787\ \text{km/s}$, and $\Delta v_2 = 1.948\left(1 - \sqrt{\tfrac{2(7{,}000)}{112{,}000}}\right) = 1.948(1 - 0.3536) = 1.260\ \text{km/s}$. Total $\Delta v_{\text{H}} = 4.046\ \text{km/s}$; trip time $\pi\sqrt{56{,}000^3/\mu} \approx 18.3\ \text{hours}$.

Bi-elliptic, overshooting to $r_b = 210{,}000\ \text{km}$ (first ellipse $a_1 = 108{,}500$, second $a_2 = 157{,}500\ \text{km}$):

Burn	Where	From speed	To speed	$\Delta v$
1 (prograde)	$r_1 = 7{,}000$	$7.546$	$10.498$	$2.952$
2 (prograde)	$r_b = 210{,}000$	$0.350$	$1.125$	$0.775$
3 (retrograde)	$r_2 = 105{,}000$	$2.250$	$1.948$	$0.301$

Total $\Delta v_{\text{BE}} = 2.952 + 0.775 + 0.301 = 4.028\ \text{km/s}$ — about $18\ \text{m/s}$ less than the Hohmann's $4.046$. But the trip is two tall half-ellipses: $t = \pi\sqrt{108{,}500^3/\mu} + \pi\sqrt{157{,}500^3/\mu} \approx 4.9 + 8.6 = 13.5\ \text{days}$ — against the Hohmann's $18.3$ hours. You save four-tenths of one percent of the fuel and spend eighteen times as long getting there. (FLAG: the $18\ \text{m/s}$ margin is genuine but thin — it depends on the chosen $r_b$; a lower $r_b$ would erase it.)

The break-even is a classic result. Comparing the two costs as the ratio $R = r_2/r_1$ grows:

• For $R < 11.94$, the Hohmann transfer always wins — no bi-elliptic can beat it. This covers almost every Earth-orbit maneuver you will ever plan (LEO-to-GEO is only $R \approx 6.2$).
• For $R > 15.58$, a bi-elliptic with a high enough $r_b$ always beats the Hohmann.
• Between $11.94$ and $15.58$, it depends on how far out you are willing to fly $r_b$ (our $R = 15$ example lives here, and needed a distant $r_b$ to eke out its win).

🔧 Engineering Reality: why almost everyone flies Hohmann anyway. The bi-elliptic's savings are tiny, its transfer times are days to weeks, and its extra burn at $r_b$ demands a precise maneuver far from Earth, where tracking is weaker and a missed burn is costlier. For the overwhelming majority of missions — every trip to GEO, every low-orbit adjustment — the ratio is well under $11.94$ and Hohmann is provably optimal, so the question never arises. Bi-elliptic (and its cousins) surface only for extreme changes, such as a large plane change done cheaply way out at apoapsis (Section 10.4 explains why altitude makes turning cheap), or repositioning far out toward the edge of a planet's gravitational reach. Know it exists; reach for Hohmann.

🔄 Check Your Understanding 1. In the bi-elliptic transfer, why is the middle burn (at $r_b$) so cheap? 2. Your mission needs to raise a circular orbit by a factor of $6$ in radius. Should you even consider a bi-elliptic transfer? Why or why not?

Answers

1. Because at the very high apoapsis $r_b$ the spacecraft is moving extremely slowly (vis-viva: large $r$ forces small $v$), so a small $\Delta v$ there produces a large change in the orbit — it takes almost nothing to lever perigee up to $r_2$. 2. No. A ratio of $6$ is far below the $11.94$ threshold, so the Hohmann transfer is guaranteed to be cheaper and far faster. The bi-elliptic only becomes competitive above a ratio of roughly $12$.

10.4 Plane changes: why turning is so expensive

Everything so far has kept the spacecraft in one plane — raising and lowering orbits that all lie flat in the same disk. But orbits are oriented in three dimensions, tilted by their inclination $i$ (one of the six orbital elements from Chapter 8), and missions constantly need to change that tilt: a satellite launched from Cape Canaveral enters orbit inclined about $28.5^\circ$ to the equator, but a geostationary satellite must sit in the equatorial plane, $0^\circ$. Bridging that gap is a plane change, and it is one of the most expensive things you can do in space.

Definition (plane change). A plane change is a maneuver that rotates the plane of an orbit — its inclination $i$, the orientation of its ascending node, or both — without necessarily changing the orbit's size or shape. A pure plane change turns the velocity vector through an angle $\Delta i$ while keeping its magnitude, the speed $v$, the same.

Here is why it hurts. At the burn point the spacecraft has velocity of magnitude $v$ pointing one way, and we want velocity of the same magnitude $v$ pointing in a direction rotated by $\Delta i$ out of the plane. The required $\Delta\mathbf{v}$ is the vector connecting the tip of the old velocity to the tip of the new one — the base of an isosceles triangle with two equal sides of length $v$ and apex angle $\Delta i$.

Strategy first. Two velocity vectors, both of length $v$, with angle $\Delta i$ between them. Their difference is the third side of an isosceles triangle. Drop the triangle's bisector and each half is a right triangle with opposite side $\tfrac12\Delta v$ and hypotenuse $v$, so $\sin(\Delta i/2) = (\Delta v/2)/v$. Solve for $\Delta v$.

$$ \boxed{\ \Delta v_{\text{plane}} = 2\,v \sin\!\left(\frac{\Delta i}{2}\right)\ } $$

Read what this equation is telling you, because it is bad news. The cost is proportional to your orbital speed $v$. You are not fighting gravity or distance — you are fighting your own enormous sideways velocity, and you have to redirect all of it. In low Earth orbit, where $v \approx 7.7\ \text{km/s}$, even a small turn is ruinous.

Worked Example: a $28.5^\circ$ plane change, and why altitude is everything. A satellite must swing its orbital plane by $\Delta i = 28.5^\circ$ (the Cape Canaveral case). With $\sin(14.25^\circ) = 0.2462$:

• Done in LEO, at $v = 7.673\ \text{km/s}$: $\Delta v = 2(7.673)(0.2462) = 3.78\ \text{km/s}$.
• Done at GEO, at $v = 3.075\ \text{km/s}$: $\Delta v = 2(3.075)(0.2462) = 1.51\ \text{km/s}$.

The same $28.5^\circ$ turn costs $3.78\ \text{km/s}$ down low but only $1.51\ \text{km/s}$ up high — because at GEO you are crawling at less than half the LEO speed, so there is less velocity to redirect. Note the sizes: a plane change in LEO costs nearly as much as reaching orbit did in the first place, and even the GEO version rivals the entire apogee-kick burn ($1.46\ \text{km/s}$) that circularizes the orbit. This is the rule that governs the maneuver: change your plane where you are moving slowest — high up, at apoapsis. It is the single most important fact about plane changes, and Section 10.6 turns it into real fuel savings.

To feel just how punishing the proportionality is, tabulate the LEO cost per degree and for a few angles (at $v = 7.673\ \text{km/s}$):

That $60^\circ$ row is worth staring at: turning your orbit by sixty degrees costs $2v\sin 30^\circ = v$, the whole orbital velocity — as much delta-v as it took to get into orbit at all. A right-angle plane change costs more than reaching orbit.

🚪 Threshold Concept: you cannot afford to turn, so launch aimed right. Because plane changes scale with your full orbital speed, they are the most expensive routine maneuver in spaceflight, and the entire discipline reorganizes to avoid them. This is why rockets launch on a precise azimuth timed to the second: to be born into the correct orbital plane, because fixing the plane afterward is unaffordable. It is why a satellite headed for the equatorial GEO belt is launched as close to the equator as possible (Kourou, at $5^\circ$ latitude, beats Cape Canaveral's $28.5^\circ$ handily). And it is why interplanetary missions agonize over launch windows: the target's orbital plane is fixed in the sky, and you must leave when Earth's motion lines you up with it, because you cannot buy your way into the right plane later. "Change your orbit" is merely expensive; "change your plane" is very nearly forbidden.

⚠️ Common Misconception: "Just do a little plane change at launch to save the big one later." It sounds efficient to nudge the plane early and often. In fact the opposite is true — plane changes are cheapest where you are slowest, so you want to defer them to the highest, slowest point of the mission, not do them in fast low orbit. The professional move for a GEO satellite is to carry the plane change all the way out to the GTO apogee (at nearly GEO altitude, moving at barely $1.6\ \text{km/s}$) and do it there, folded into the circularization burn — which is exactly the combined maneuver of Section 10.6. Doing that same turn back down in LEO would cost more than twice as much.

🔄 Check Your Understanding 1. Why is a plane change at GEO so much cheaper than the same plane change in LEO? 2. Estimate the delta-v for a $5^\circ$ plane change in LEO ($v = 7.673\ \text{km/s}$). Is it cheap?

Answers

1. Because the cost is $2v\sin(\Delta i/2)$, proportional to the orbital speed $v$, and $v$ at GEO ($3.075\ \text{km/s}$) is less than half of $v$ at LEO ($7.673\ \text{km/s}$). Same angle, less speed to redirect, less than half the delta-v. 2. $\Delta v = 2(7.673)\sin(2.5^\circ) = 15.346 \times 0.0436 = 0.669\ \text{km/s} \approx 670\ \text{m/s}$. That is not cheap — a mere five-degree tilt costs almost $0.7\ \text{km/s}$, more than a quarter of the entire LEO-to-GTO injection burn. Plane changes are never cheap in LEO.

10.5 Phasing and rendezvous

So far we have changed the shape and plane of orbits. But sometimes the orbit is already right and only the timing is wrong: you and your target are on the same orbit, but you are at different places along it, and you need to close the gap. This is the problem of phasing, and its ultimate application is rendezvous — bringing two spacecraft together, the maneuver that makes space stations, crewed docking, and sample return possible.

The engine of all phasing is the single most counterintuitive fact in orbital mechanics, which Chapter 6 proved and which you must now trust with your fuel: a higher orbit is a slower orbit — not just slower in speed, but longer in period. To change where you are along an orbit relative to a neighbor, you temporarily move to a different orbit with a different period, let the mismatch in period open or close the gap, then return.

Definition (phasing orbit). A phasing orbit is a temporary orbit, entered and later left, whose period differs from a reference orbit so as to change a spacecraft's angular position (its phase) relative to a target on that reference orbit. Dropping to a lower, shorter-period orbit makes you gain ground on the target ahead; rising to a higher, longer-period orbit makes you fall back.

Definition (rendezvous). A rendezvous is the maneuver sequence that brings two spacecraft to the same position and the same velocity at the same time — matching orbits and closing the distance to zero relative speed — so they can dock, berth, capture, or fly in close formation. It combines phasing (to close the along-track gap) with fine orbit-matching and proximity operations.

Worked Example: a $30^\circ$ phasing maneuver. Two spacecraft share a circular $400\ \text{km}$ orbit ($r = 6{,}771\ \text{km}$), period $T = 2\pi\sqrt{r^3/\mu} = 5{,}545\ \text{s} = 92.4\ \text{min}$ (from Chapter 6). Your spacecraft is $30^\circ$ ahead of the target and wants to meet it. You must let the target catch up — so you rise to a higher, slower orbit and deliberately fall behind by exactly one lap's worth of the gap.

Fire prograde so your burn point becomes the perigee of a phasing ellipse whose period is longer by the right amount. You want the target, moving at $360^\circ/T$, to travel the extra $30^\circ$ and arrive at your burn point just as you complete one phasing orbit, so the phasing period must be $$T_{\text{ph}} = \frac{360^\circ + 30^\circ}{360^\circ}\,T = \frac{390}{360}(5{,}545) = 6{,}007\ \text{s}.$$ Kepler's third law inverts to the phasing semi-major axis, $a_{\text{ph}} = \left(\mu\,(T_{\text{ph}}/2\pi)^2\right)^{1/3} = 7{,}142\ \text{km}$ — a modest ellipse whose apogee reaches $2a_{\text{ph}} - r = 7{,}513\ \text{km}$ (altitude $1{,}142\ \text{km}$). The burn to enter it, at perigee $r = 6{,}771\ \text{km}$: $$v_{\text{ph}} = \sqrt{\mu\!\left(\tfrac{2}{6{,}771} - \tfrac{1}{7{,}142}\right)} = \sqrt{61.93} > = 7.8693\ \text{km/s}, \qquad \Delta v = 7.8693 - 7.6726 = 0.1967\ \text{km/s}\ (\approx 197\ \text{m/s}).$$ After one lap you are back at the burn point, now beside the target; a second, equal burn of $0.1967\ \text{km/s}$ retrograde circularizes you back onto the shared orbit. Total phasing cost: $2 \times 0.1967 = 0.393 \approx 0.39\ \text{km/s}$ for a single-revolution $30^\circ$ catch-up.

💡 Intuition: slow down to catch up. The result feels backwards until you hold onto "higher is slower." To catch a target ahead of you, the ground instinct is to speed up toward it — but a prograde burn raises your orbit, slows you, and drops you further behind. The correct move to catch something ahead is to burn retrograde, dropping into a lower, faster, shorter-period orbit that laps around quicker and reels the target in; then raise back up to meet it. (In the worked example above we were ahead and wanted to fall back, so we did the opposite — rose to a slower orbit and let the target come around.) Everything about rendezvous is this dance: down is fast, up is slow, and you gain or lose ground by choosing your altitude.

🔧 Engineering Reality: why big catch-ups take many orbits. Notice that catching up a large angle in a single revolution demands a big period change, which means a big orbit change, which means diving deep or climbing high. A spacecraft trailing its target by $30^\circ$ that tried to make it up in one lap would have to drop into a phasing orbit whose perigee grazes — or dips below — the atmosphere. So real rendezvous phasing is done gently, a few degrees per revolution over many orbits: the ISS-bound Dragon and Soyuz vehicles spend hours to a day or two nudging their orbit slightly, letting the small period difference accumulate the phase, before closing in. Patience is cheaper — and safer — than a violent single-lap dash.

The full rendezvous is a choreography built on phasing, flown thousands of times to the ISS:

1. Launch and phasing. The chaser launches into the target's orbital plane (remember Section 10.4 — you cannot afford to fix the plane later) and into a lower, faster parking orbit. Over several revolutions, small phasing burns close the along-track gap.
2. Far-field approach. A series of Hohmann-like transfers raises the chaser toward the target's altitude, arriving a few kilometers "behind and below," where the orbital-mechanics drift is naturally toward the target.
3. Proximity operations. Within a kilometer, the chaser switches to relative-motion guidance (the strange, looping relative trajectories governed by the Clohessy–Wiltshire equations, an advanced topic), creeping in along carefully chosen approach corridors — the "V-bar" (along-track) or "R-bar" (radial) lines.
4. Docking. The last meters are flown at centimeters per second to a soft capture, matching position and velocity to nearly zero relative motion.

📜 From History: Buzz Aldrin, "Dr. Rendezvous." The first space rendezvous was almost a disaster of intuition. On Gemini 4 in 1965, astronaut Jim McDivitt tried to close on the spent booster by thrusting straight at it — and watched it drift away, because every prograde nudge raised his orbit and slowed him down. The fix came from theory: Edwin "Buzz" Aldrin had written his 1963 MIT doctoral thesis on orbital rendezvous, working out the counterintuitive burns, and earned the nickname "Dr. Rendezvous." His mathematics flew on later Gemini missions, which docked successfully, and then on Apollo, where lunar-orbit rendezvous — two vehicles finding each other in orbit around the Moon, with no rescue possible if they failed — was the linchpin of the entire mission architecture. The paradox of Section 6.6 was not academic; astronauts had to master it before anyone could go to the Moon.

🔗 Connection: finding the transfer that hits a moving target. Phasing answers "how do I change my timing on a shared orbit." The more general question — "what single transfer orbit takes me from here now to there at a specified later time" — is Lambert's problem, the workhorse of real rendezvous and interplanetary targeting, which we meet in Chapter 13. The Hohmann transfer is the special, minimum-energy case of Lambert's problem; rendezvous and Mars-injection targeting are the general case.

🔄 Check Your Understanding 1. You are trailing a target by $10^\circ$ in the same circular orbit. Which way do you burn to catch up, and what happens to your orbital period? 2. Why can't you simply thrust straight toward a spacecraft floating a few kilometers ahead of you on the same orbit?

Answers

1. Burn retrograde. That drops you into a lower orbit with a shorter period, so you go around faster and gain on the target; after you have made up the $10^\circ$ you burn prograde to rise back and match orbits.
2. Because thrusting toward it is a prograde burn, which raises your orbit, lengthens your period, and slows you down — so you fall behind the target instead of catching it. This is the orbital paradox of Chapter 6; the only way to close a gap ahead of you is to first drop lower and faster.

10.6 Combined maneuvers

The maneuvers of this chapter are not meant to be done one at a time. The deepest lesson of mission design is that when a single burn must accomplish two things — say, change speed and change plane — doing them together, as one cleverly aimed $\Delta\mathbf{v}$, is dramatically cheaper than doing them in sequence. The reason is pure geometry: delta-vs are vectors, and vectors add as a triangle, not as a straight sum. Two burns done separately cost $|\Delta\mathbf{v}_1| + |\Delta\mathbf{v}_2|$; done as one combined burn they cost only $|\Delta\mathbf{v}_1 + \Delta\mathbf{v}_2|$, and the triangle inequality guarantees the combined version is never larger, and usually much smaller.

Take the canonical case: the apogee of a geostationary transfer orbit, where a satellite must both speed up to circularize (an in-plane burn) and swing into the equatorial plane (a plane change). Its arrival velocity $v_a$ lies in the inclined transfer plane; the desired velocity $v_{c2}$ lies in the equatorial plane, rotated by $\Delta i$ and larger in magnitude. The single burn that connects them is the third side of a triangle with sides $v_a$ and $v_{c2}$ and included angle $\Delta i$ — the law of cosines:

$$ \boxed{\ \Delta v_{\text{combined}} = \sqrt{v_a^2 + v_{c2}^2 - 2\,v_a\,v_{c2}\cos\Delta i}\ } $$

This one formula contains both special cases we already know. Set $\Delta i = 0$ (no plane change) and it collapses to $\sqrt{(v_{c2}-v_a)^2} = |v_{c2} - v_a|$, the ordinary in-plane burn. Set $v_a = v_{c2} = v$ (pure plane change, no speed change) and, using $1 - \cos\Delta i = 2\sin^2(\Delta i/2)$, it becomes $2v\sin(\Delta i/2)$ — the plane-change formula of Section 10.4. Combined maneuvers are the general case; the earlier sections were its corners.

Worked Example: the real GTO-apogee burn. A satellite arrives at GTO apogee moving at $v_a = 1.617\ \text{km/s}$ in a plane inclined $\Delta i = 28.5^\circ$, and must reach the equatorial GEO circular speed $v_{c2} = 3.075\ \text{km/s}$. With $\cos 28.5^\circ = 0.8788$: $$\Delta v_{\text{combined}} = \sqrt{1.617^2 + 3.075^2 - 2(1.617)(3.075)(0.8788)} > = \sqrt{2.615 + 9.456 - 8.740} = \sqrt{3.331} = 1.825\ \text{km/s}.$$ Compare doing the two jobs separately: circularize first ($\Delta v = 3.075 - 1.617 = 1.458\ \text{km/s}$), then change plane at GEO ($\Delta v = 2(3.075)\sin 14.25^\circ = 1.514\ \text{km/s}$), for a total of $2.972\ \text{km/s}$. The combined burn costs $1.825\ \text{km/s}$ — a saving of $2.972 - 1.825 = 1.15\ \text{km/s}$, almost forty percent, from a single well-aimed burn instead of two. This is not a trick; it is how every geostationary satellite is actually flown.

Now step back and price the whole journey from a Cape-Canaveral LEO to an operational GEO slot, to see how much combining and deferring save:

Deferring the plane change to slow-moving apogee and folding it into the circularization burn saves about $1.15\ \text{km/s}$ — which, through the tyranny of the rocket equation, is a large fraction of the satellite's propellant and therefore of its cost and its usable life. Mass is the enemy, and combined maneuvers are one of the sharpest weapons against it. (A yet-more-optimal design splits a few degrees of the plane change onto the perigee burn as well, shaving a little more; the fully optimal split is a standard result mission designers compute, but the lion's share of the win is simply "do the turn at apogee, combined.")

🔗 Connection: the same trick launches you to Mars. Combining burns is not just a GEO nicety — it is how a probe leaves for another planet. The departure burn from LEO must simultaneously raise the orbit to escape energy and aim the escape asymptote in the right direction, all in one impulse deep in Earth's gravity well where the Oberth effect makes each m/s count most. That combined, energy-adding departure burn is the opening move of the interplanetary Hohmann transfer we build in Chapter 11, and the delta-v it costs is the first line of the Mars mission's budget in Chapter 34.

🔄 Check Your Understanding 1. Why is a combined speed-and-plane-change burn cheaper than doing the two separately? 2. In the GTO-apogee example, would the combined burn save more or less if the plane change were done in LEO instead of at apogee?

Answers

1. Because delta-vs add as vectors (a triangle), not as scalars. The combined cost is the magnitude of the vector sum $|\Delta\mathbf{v}_1 + \Delta\mathbf{v}_2|$, which by the triangle inequality is never more than — and usually much less than — the sum of magnitudes $|\Delta\mathbf{v}_1| + |\Delta\mathbf{v}_2|$ you pay for two separate burns. 2. The combined burn would save far less useful delta-v in LEO, because the plane-change part itself is much more expensive down low (proportional to the higher LEO speed). The whole point is to do the plane change — combined or not — where you move slowest, at apogee. Combining is good; combining at apogee is the real win.

Mission Design Checkpoint: your transfer delta-v and maneuvers.py

You have now derived every tool needed to price the routes of your mission. This checkpoint turns the orbit you selected in Chapter 9 and the energies you recorded in Chapter 6 into concrete transfer delta-vs, and folds them into the delta-v budget you started in Chapter 3.

The design. Open your Mission Design Review (MDR) and add a "Transfers" section to the delta-v budget. For each orbit change your mission needs, compute the delta-v with this chapter's tools and list it as a budget line:

• Track A (GEO comsat): the LEO-to-GEO transfer — perigee burn ($\approx 2.40\ \text{km/s}$) plus the combined apogee circularization-and-plane-change ($\approx 1.83\ \text{km/s}$ from a $28.5^\circ$ GTO), totaling about $4.23\ \text{km/s}$.
• Track B (lunar lander): the trans-lunar injection burn (a transfer ellipse to the Moon's distance), which you will refine in Chapter 11.
• Track C / D (Mars orbiter / asteroid): the Earth-escape departure burn, likewise detailed in Chapter 11.

Add each transfer as a line in your budget, keep a running total, and carry a margin (justified in Chapter 29). This total is what sizes your propellant through the rocket equation.

The code. Create astrotools/maneuvers.py, the maneuver module of the package. It uses the orbital speeds from orbits.py (Chapter 6) and returns the delta-vs of this chapter. Keep these signatures stable — later chapters call them.

import math

def hohmann(mu, r1, r2):
    """Two-burn Hohmann transfer between circular orbits r1 -> r2 (same length unit).
    Returns (dv1, dv2, dv_total) in the speed unit implied by mu and r."""
    a_t = (r1 + r2) / 2.0
    v_c1 = math.sqrt(mu / r1)
    v_c2 = math.sqrt(mu / r2)
    v_p = math.sqrt(mu * (2.0 / r1 - 1.0 / a_t))   # transfer-ellipse perigee speed
    v_a = math.sqrt(mu * (2.0 / r2 - 1.0 / a_t))   # transfer-ellipse apogee speed
    dv1 = abs(v_p - v_c1)                           # leave inner circle
    dv2 = abs(v_c2 - v_a)                           # join outer circle
    return dv1, dv2, dv1 + dv2

def plane_change(v, di_deg):
    """Delta-v for a pure plane change of di_deg degrees at orbital speed v:
    dv = 2 v sin(di/2)."""
    return 2.0 * v * math.sin(math.radians(di_deg) / 2.0)

def dv_budget(legs):
    """Roll up a list of (label, dv) legs into a total delta-v budget."""
    return sum(dv for _, dv in legs)

if __name__ == "__main__":
    MU = 3.986e5            # km^3/s^2  (work in km, km/s)
    R_LEO, R_GEO = 6771.0, 42164.0

    dv1, dv2, dvt = hohmann(MU, R_LEO, R_GEO)      # Track A: LEO -> GEO transfer
    pc = plane_change(3.075, 28.5)                 # 28.5 deg plane change at GEO speed
    budget = [("LEO->GTO perigee", dv1),
              ("GTO->GEO apogee",  dv2),
              ("plane change 28.5 at GEO", pc)]
    print(round(dv1, 3), round(dv2, 3), round(dvt, 3))
    print(round(pc, 3))
    print(round(dv_budget(budget), 3))
    # Expected output:
    # 2.399 1.457 3.857
    # 1.514
    # 5.371

The three numbers on the first line are the LEO-to-GEO Hohmann burns and total ($2.40 + 1.46 = 3.86\ \text{km/s}$) we hand-derived in Section 10.2. The `dv_budget` total ($5.37\ \text{km/s}$) is the naive sum — doing the plane change separately; Section 10.6 showed that combining it into the apogee burn cuts the real figure to about $4.23\ \text{km/s}$, which is the number you should carry into your MDR. By the capstone, maneuvers.py will price every leg of your trajectory, and dv_budget will roll them into the single total your rocket must be able to pay.

Summary

Orbital maneuvers reduce the entire problem of moving through space to differences of vis-viva speeds. Carry these forward:

Numbers worth memorizing: LEO→GEO Hohmann $\approx 3.9\ \text{km/s}$ ($\approx 5.3$ h); plane change $\approx 134\ \text{m/s}$ per degree in LEO; a $28.5^\circ$ change costs $3.78\ \text{km/s}$ in LEO but only $1.51\ \text{km/s}$ at GEO; combining circularization and plane change at GTO apogee $\approx 1.83\ \text{km/s}$.

Spaced Review

Retrieval strengthens memory. Answer from memory before checking, then look back at the cited chapter. This chapter revisits Chapter 6 (energy and vis-viva) and Chapter 8 (Kepler's laws and orbital elements).

1. (§10.2, Ch. 6) The entire Hohmann derivation rests on one equation from Chapter 6, used to find four speeds. Which equation, and what are its two inputs?
2. (§10.2, Ch. 8) The Hohmann transfer time is $\pi\sqrt{a_t^3/\mu}$. Which of Kepler's laws gives the period $T = 2\pi\sqrt{a^3/\mu}$ this comes from, and why is the transfer time half of it?
3. (§10.5, Ch. 6) Chapter 6's threshold concept was "a higher orbit is a slower orbit." State how phasing uses it to let a trailing chaser catch a target ahead.
4. (§10.4, Ch. 8) A plane change alters which of the six orbital elements from Chapter 8, and leaves which ones (size and shape) unchanged?
5. (§10.2, Ch. 6) Without computing, is the first Hohmann burn (at the inner orbit) prograde or retrograde when raising an orbit, and what does it do to the spacecraft's specific orbital energy?

Answers

1. The vis-viva equation, $v^2 = \mu(2/r - 1/a)$. Its two inputs are the current distance $r$ and the orbit's semi-major axis $a$; every Hohmann speed (two circular speeds, plus the transfer perigee and apogee speeds) is one evaluation of it. 2. Kepler's third law relates period to semi-major axis ($T^2 \propto a^3$). The transfer covers only perigee-to-apogee, exactly half of the transfer ellipse, so it takes half the full period. 3. To catch a target ahead, the chaser burns retrograde, dropping to a lower orbit that (being lower) is faster and has a shorter period; it laps around more quickly, closes the gap, then raises back up to match. Higher-is-slower means "down = fast = gaining ground." 4. It changes the inclination $i$ (and/or the right ascension of the ascending node $\Omega$) — the orientation of the plane — while leaving the semi-major axis $a$ (size) and eccentricity $e$ (shape) unchanged; a pure plane change keeps the speed and hence the energy the same. 5. Prograde (a speed-up). It raises the specific orbital energy $\varepsilon$, which enlarges the orbit (bigger $a$) and lifts the far side to the target radius — exactly as Chapter 6's energy rules require.

What's Next

We have learned to move between orbits around one body — to climb from LEO to GEO, to turn a plane, to chase down a station. But the most consequential Hohmann transfer of all does not stay near Earth: it leaves. If you treat the Sun as the central body and Earth and Mars as two nearly circular orbits around it, then flying from one planet to another is a Hohmann transfer writ enormous — a half-ellipse around the Sun, with a departure burn at Earth's orbit and an arrival burn at Mars's. Everything you derived in this chapter carries over directly; what changes is that the target is moving, so you can only leave at the rare moments when Earth and Mars are lined up for the crossing — the launch window. In Chapter 11 we scale the Hohmann transfer up to interplanetary size, meet the launch window and the departure energy $C_3$, and watch a gravity assist steal speed from a passing planet for free — the trip to Mars, and beyond, built entirely on the two burns you now understand.