Why launch rockets from a 20 km tower?

Late in his conversation with Ashlee Vance on Core Memory, Neal Stephenson was asked which of the strange launch ideas he once studied he would still back. His answer was not a gun or a sling: “the one that I still think somebody ought to build is the tall tower.” A tower 20 kilometres high, with a rocket on top.

▶ Try it yourself in Tall Tower Launch: drag the top of the tower from sea level to 25 km and watch the air fall away below it, the engine bell grow, and the payload crates multiply.

Most of the air is near the ground

Air pressure is the weight of the air above you, so the pressure at a height tells you how much of the atmosphere is still overhead. At the top of Everest about 31% is left. At 20 km, about 5.4% is: 94.6% of the atmosphere is below the launch pad. A rocket that starts there skips the thickest part of the climb, where drag is worst, and it is quieter too. As Stephenson put it, “it’s not as noisy up there,” so the rocket would not have to be built to survive its own roar.

Where the gain really comes from

The surprise in the NASA Glenn study by Geoffrey Landis, Vincent Denis and Valerie Lyons is that drag is the small part. From 20 km, the speed lost to drag falls by only about 1.4%. The big gain is the engine. Near sea level, a rocket nozzle has to be short, or the outside air pushes on the exhaust and the flow separates. In thin air the engine can wear a long vacuum-style bell from liftoff, and in their numbers the average efficiency (specific impulse) rises about 15%. Together, the mass that reaches orbit grows by about 17%.

Why 17% becomes double

That 17% is of everything that reaches orbit, structure and payload together. Landis’s single-stage rocket is 87.5% propellant, so 12.5% of its liftoff mass arrives in orbit, and only 2% of the liftoff mass is payload. The structure does not get any heavier, so all of the extra goes to the payload: 2% + 2.1% = 4.1% of liftoff mass, about twice as much. That is why a small gain doubles it, and why the tool lets you change the payload share: a rocket with a bigger payload share gains far less in proportion.

The catch

Height is not speed. Getting to orbit means going about 7.8 km/s sideways, and 20 km of height is worth only about 0.6% of that energy. The tower helps the engine, not the climb. And the tower itself is the hard part: winds in the jet stream blow at about 9 to 16 km, right through the middle of it. Stephenson studied the idea with Arizona State engineer Keith Hjelmstad in 2013 and said that “in a windless environment making a structure that tall would almost be trivial.” Rockets already start high in other ways: Pegasus drops from an airliner at about 12 km, and Virgin Orbit’s LauncherOne dropped from a 747 at about 10.7 km.

The tool is a model: the air follows the US Standard Atmosphere, and the rocket’s gains are interpolated from the NASA Glenn table for one single-stage design. The tower’s structure, winds, cost and how a fuelled rocket would get to the top are not modeled. The film opens on Neal Stephenson on Core Memory (episode 94) and shows a SpaceX Raptor engine photo by @SpaceX; the sky, tower, air and numbers are drawn from the model. More launch ideas from the same conversation: Neal Stephenson’s launch ideas.

Sources: G. A. Landis, V. Denis and V. Lyons, High Altitude Launch for a Practical SSTO (NASA Glenn, 2003); Core Memory episode 94 with Neal Stephenson, video and episode page; Wikipedia, U.S. Standard Atmosphere and Jet stream; Motherboard, The twelve-mile-high skyscraper is on pause thanks to the jet stream (2013).