Bi-Elliptic Transfer: The Maneuver That Beats Hohmann

A bi-elliptic transfer beats the famously optimal Hohmann transfer whenever your target orbit's radius is more than about 11.94 times your starting radius (and always above about 15.58). It does this by flinging the spacecraft far past the destination on a tall ellipse, then dropping back in. Above that threshold, three burns cost less total delta-v than two.

That single fact is genuinely surprising. The Hohmann transfer is taught as the minimum-energy way to move between two circular orbits, so "well, actually, a three-burn detour can be cheaper" feels like a contradiction. It is not. The Hohmann transfer is only optimal in a bounded regime. Push the size difference between your orbits far enough and the extra burn pays for itself. You can explore the exact crossover for any pair of orbits with the bi-elliptic transfer calculator.

What the maneuver actually does

A Hohmann transfer uses two engine burns. The first lifts you from your circular start orbit onto an ellipse whose far end just touches the target orbit. The second, at that far end, circularizes you. Simple, elegant, and for moderate orbit changes, unbeatable. Walter Hohmann published it in his 1925 book Die Erreichbarkeit der Himmelskoerper ("The Attainability of Celestial Bodies").

A bi-elliptic transfer uses three burns and two half-ellipses:

• Burn 1 — at your starting orbit, fire hard to reach an apoapsis far beyond the target, often many times farther out.
• Burn 2 — coast all the way out to that distant apoapsis, then nudge the velocity so the return ellipse's low point (periapsis) lands exactly on the target radius.
• Burn 3 — coast back down and circularize at the destination.

The trick is that the second burn, performed where the spacecraft is crawling along at its slowest, is astonishingly cheap. The idea was first published by Soviet-Polish rocketry pioneer Ary Sternfeld in 1934, though the now-famous efficiency thresholds were worked out in detail later.

The lever and the Oberth effect

Why does going the long way around save fuel? Think of the distant apoapsis as the end of a very long lever. To raise or lower the periapsis on the far side of an orbit, you change your speed at apoapsis. The farther out apoapsis is, the slower you are moving there, and the less velocity change is needed to swing the opposite end of the orbit by a given amount. In the mathematical limit where apoapsis goes to infinity, the periapsis-adjusting burn costs zero delta-v. You get the orbit change almost for free.

The flip side is the Oberth effect, described by Hermann Oberth in his 1929 book Wege zur Raumschiffahrt ("Ways to Spaceflight"): a burn buys more kinetic energy when you are already moving fast, deep in the gravity well. The bi-elliptic transfer front-loads its energy into that powerful first burn near the central body, where it is most effective, instead of splitting it more evenly the way Hohmann does. Everything follows from the vis-viva equation, which gives speed at any point in an orbit:

v = sqrt( mu * (2/r - 1/a) )

Here mu is the body's gravitational parameter, r your current radius, and a the orbit's semi-major axis. As r grows toward apoapsis, v shrinks, which is exactly why the mid-course correction out there is so cheap.

The 11.94 and 15.58 thresholds

Let R = r_final / r_initial, the ratio of the two circular-orbit radii. There are three regimes:

Below 11.94 there is no apoapsis tall enough to make three burns beat two. Above 15.58, even a modest bi-elliptic path wins. In the middle band, the answer hinges on how far out you push apoapsis. A worked example from the standard literature makes it concrete. Going from a 6,700 km low orbit to a 93,800 km orbit (R = 14, inside the gray zone) costs a Hohmann transfer about 4,133.72 m/s. A bi-elliptic path swinging out to 268,000 km does it in about 4,117.53 m/s, saving roughly 16 m/s, or about 0.4%. Small, but real, and at scale even fractions of a percent of delta-v translate into meaningful propellant mass via the exponential rocket equation.

The price: patience, lots of it

Nothing is free. The bi-elliptic transfer trades fuel for time, and the exchange rate is brutal. A Hohmann transfer rides one half-ellipse; the bi-elliptic rides two, and the journey out to a distant apoapsis can dominate the clock. For the kind of orbit change where a Hohmann transfer finishes in about 15 hours and 34 minutes, real numbers look like this:

• An apoapsis at about 508,000 km (past the Moon) yields roughly a 1% delta-v saving, but takes about 17 days.
• An apoapsis near 11.8 million km yields about a 2% saving, and takes about 4.5 years, by which point other bodies' gravity would wreck the trajectory anyway.

That is why pure bi-elliptic transfers are rare in practice. The far more common use is the cheap plane change. Changing an orbit's inclination is one of the most expensive things you can do, because the cost scales with your speed at the moment you turn. Out at a high apoapsis you are barely moving, so a plane change folded into that second burn costs a fraction of what it would near the planet. Mission designers exploit this to reach high-inclination targets, sometimes raising apoapsis, rotating the plane up high, and dropping back. The energy bookkeeping shares deep roots with interplanetary trajectory design; the same accounting underlies a Hohmann transfer, which remains the right tool for the vast majority of orbit changes.

The bi-elliptic transfer is a beautiful reminder that "optimal" always comes with fine print. Hohmann is optimal among two-burn transfers, and within its radius regime. Loosen those assumptions and a stranger, slower, three-burn path quietly wins.

Frequently Asked Questions