This August, Britain’s Bloodhound car will accelerate at 2 g’s along a stretch of desert in South Africa and, if all goes well, it will push past the land-speed record of 1227.986 kilometers per hour (763 mph). That’s just shy of the sound barrier.
You might almost call that the easy part, compared with the problem of stopping. “The desert we have is 12 miles long (19 km), 2 miles wide, not much room to slow down,” notes Jules Tipler, a spokesman for the project.
And that’s just for this year’s speed trial. Next year, a beefed-up version with three rockets instead of one will made a run for a rounder number: 1000 mph (1609 kph). Then it, too, will have to slow down, fast.
Why the three rockets? You need that monstrous thrust because drag goes up with the cube of speed.
Tipler describes how the dash to 1000 mph will look: “You start up your jet engine, give it time to spool up—it’ll be going from zero to 100 mph in about 12 seconds. Then, you go from 100 to 1000 mph in about 25 seconds. When it hits 350 mph, you start up the Jaguar V-8 engine, pumping concentrated hydrogen peroxide in, which then breaks into water and oxygen, igniting fuel in the rubber. So, the rocket kicks in. It goes at 1000 mph for one mile, the ‘flying mile,’ 3.6 seconds—with 20 tons of thrust, 15 tons of air resistance.”
It gets its motive force from a jet engine and a hybrid rocket motor with a solid-fuel section and a liquid oxidizer. By hybridizing a rubbery solid and a liquid, you get much of the convenience of the former while retaining the controllability inherent in the latter. The liquid is hydrogen peroxide, and it provides the oxygen—cut off the flow from the pump—a V-8 engine--and the rocket shuts down. Besides, hydrogen peroxide is itself something of a fuel, providing a bit of energy just by changing into water and oxygen. It’s a low-heat reaction, and that explains why it has figured in James Bond-style rocket belts.
The rocket engine comes from Nammo, in Norway. Watch this test firing:
After all that comes the harrowing part: braking. It starts when the flow of hydrogen peroxide is turned off, killing the rocket, and the flow of fuel to the jet engine is stopped.
“You’re losing 60 mph per second—in a road car, you’d call it a crash,” Tipler says. “Then come 20 seconds of 3-g deceleration. At 800 mph you can put on the air brakes, which increases the drag on the car, maintaining the 3-g deceleration. Coast down the desert for 5 to 6 miles, and use your wheel brakes at 160 mph. Stop at the support struck, pull a great big U-turn in the desert, with a turning radius 240 meters. Refuel it with another ton of peroxide, half a ton of jet fuel, a splash more unleaded for the jaguar engine. Then, back again.”
That’s if all goes well. If not, two parachute brakes can be deployed, an expedient to be avoided not least because of the time it would take to repack the chutes for a second run. Several runs a day are what’s planned, the better to improve by tiny increments.
But here’s the rub: For each of those runs, at every speed, under all temperatures and wind conditions, the car must maintain neutral lift—neither going up like a plane nor hunkering down like an Indy car. That’s uncharted aerodynamics, the kind that, in the olden days, fighter jockeys would test in Chuck Yeager fashion. The Bloodhound team will do that, certainly, but they will try to minimize the danger and expense by testing as much as possible in silico. Computational fluid dynamics, often used in designing engines, will be the stand-in for pilots as engineers fine-tune the aerodynamics.
“The answers you get are only as good as the model itself,” says Joe Holdsworth, a system engineer in the project and one of 20 or so people actually building the car. “So you need to compare the model with actual aerodynamic data.”
With that in mind, the car—still just in skeletal stage—will sport some 500 sensors, including a bunch of cameras that provide views of interest only to the engineers, as well as three in-cockpit screens to help the driver compensate for narrow views offered by tiny stretched-acrylic windows. (The apertures have to be small to minimize the material’s weakness with respect to the aluminum skin, itself bored with 192 tiny holes that each connects to its own air-pressure gauge.)
The sensors yield two streams of data. A highly-detailed stream gets stored for use in modeling later on, and a stream with less data is used in real time, to help the driver stay within safe bounds.
That real-time data is also relayed to the Web. And that’s not an afterthought; it’s a design feature. The Bloodhound project gets its corporate funding in part on the strength of its educational value as a showcase for science, particularly for children. Indeed, of the 80 people who work on the project, at least a dozen make regular presentations at schools or are involved in public relations generally.
But the team members who manage the fiber optic link are mainly in it for the speed. “It runs 2 to 3 gigabits of data, about 500,000 measurements per second,” Holdsworth says proudly.
Despite this capacity to capture the car’s every move, patient testing still matters. That’s why the team is limiting itself this year to squeaking past the land-speed record set way back in 1997 by Britain’s Andy Green. And Green, 52, will be behind the wheel this time, too. He practices in jet aircraft, pulling hard-g turns that test mind and body. His passengers have been known to pass out, but he always keeps his blood just where it should be by simply clenching his muscles.
It’s the 21st century, but test pilots still need the right stuff.
Philip E. Ross is a senior editor at IEEE Spectrum. His interests include transportation, energy storage, AI, and the economic aspects of technology. He has a master's degree in international affairs from Columbia University and another, in journalism, from the University of Michigan.