August/September 2004 | Volume 55, Issue 4
The main problem blind flight was meant to solve was flying in poor visibility—at night or in bad weather. The airplanes of the time tended to drift, like a car with no one controlling the steering wheel. If visibility was good and the pilot could see the horizon, he would notice if the plane was sliding into a banked turn and could bring it back to straight and level flight. But without that visual information, sensing up from down during a turn was often impossible. The acceleration in three dimensions made it feel as if “down” was toward the cockpit floor even when the plane was severely tilted. The pilot would, however, think the plane was diving during such a turn, and his attempts to counteract the dive would lead to a downward spiral, possibly ending in a crash.
Doolittle’s plane was equipped with the latest instrumentation, including an artificial horizon and a gyrocompass. He took off from Mitchel Field in Garden City, New York, climbed to 1,000 feet, flew an oval course totaling about 15 miles (tracking his position with the help of a radio signal broadcast from the ground), and landed safely. The flight took 15 minutes. Landing was the hardest part, since Doolittle’s altimeter, while extremely sensitive, was accurate to within five feet at best. Without knowing exactly when he would hit ground, he flew the plane at a low angle into a large, grassy field, relying on a set of especially rugged shock absorbers to soften the jolt.
Instruments-only flight allowed aviation to change from a pastime for daredevils into a regular business, running on schedule like a railroad. Later innovations included radio-assisted navigation and communication, automatic pilots (which had been developed in primitive form as early as 1914), and radio-guided landing. Today’s aviation software can essentially fly a plane by itself, with the pilot overseeing the process and taking control when necessary. It’s a lot less exciting than the seat-of-the-pants days, but it’s a lot safer too.