When they are not hurling themselves out of planes, Jean Potvin and Gary Peek ponder wisps of air, rippling fabric and the way falling objects drift in a 170-mph breeze.
It might seem like a lofty subject, but the Army pays the skydivers to study the behavior of parachutes.
Parachutes are notoriously difficult objects for aerospace engineers to get their computer simulations around. Jets are rigid and their trajectories at high speeds are relatively unaffected by wind. Parachutes bend and ripple as they fill up, and the airflow behind inflated chutes is turbulent and chaotic.
"Parachutes being fabric, there's a lot of randomness," said Peek, a consultant with Industrologic Inc., an electronics firm in St. Charles, Mo. Potvin, his partner, is a physicist at St. Louis University's Parks College. He calls parachute design "some kind of a black art."
The pair recently were conducting their research in a burgundy Cessna Caravan as it soared over the Vandalia, Ill., airport. Packed into two milk crates and a beat-up wooden box were instruments designed to record how fast the attached parachutes fell. Sensors along the rigging measured how much of a yank the parachutes delivered as they opened. The goal: to better predict how chutes of different shapes tug on their loads when they open.
A pink helium balloon had given them a sense of the wind before the drops. After takeoff, Potvin threw a hand-held probe with a small chute through the plane's open door as a final wind test. The plane was flying at 140 mph at an altitude of 500 feet.
One by one, Potvin yelled "go" and Peek shoved each box into the wind. The first probe, a yellow and red cone, opened cleanly. The third, a half-moon, inflated neatly as well. The second fell like a stone, its green military paratrooper chute fouled in the wind.
Back in the hangar, Peek reviewed the data on a laptop. The foam-packed boxes had landed hard -- one of the milk crates was cracked -- but the data recorders had worked.
Next the pair will compare the data to sophisticated models they created using the numbers from previous drops. Those models anchor their Parachute Inflation Modeling Suite, or PIMS, a computer program they are developing for the Army.
"The same parachute, flying with the same payload, deployed under the same speeds may give you variations in the opening force from jump to jump," Potvin said. "The (parachute) designer needs to be able to predict how a given system will inflate."