About the time Arnold met Colleen, he also found a new job. His friend Charlie Moore was an engineer for General Mills, and he had a contract with the Navy to develop manned balloons that could stay aloft for up to twelve hours. Charlie had flown one manned flight in 1949, but he needed additional help to continue the program. He really needed three employees – another pilot, a mechanic, and an engineer – but he decided to hire Arnold because he could fill all three roles and thus save the program some money. Despite his lack of degree and his youth, by October Arnold was a General Mills employee and the project engineer for the balloon program. When the Minnesota National Guard activated on December 1, 1950 the Navy, not wanting to lose their prized new project engineer, arranged his to transfer to the Minnesota Air National Guard.
Despite the Navy perception of Arnold’s importance to their program, neither he nor Charlie knew much about manned balloons, and Arnold did not know much about balloons at all. However, if the balloonists-to-be had any doubts, the confidence of youth overcame them.
Photos of these balloons do not inspire confidence that they could safely carry someone even five feet off the ground, let alone 5,000. Except for the use of a single balloon, the setup resembled a “lawn chair balloon” that Californian Larry Walters infamously used in 1982 to ascend to 15,000 feet, astonishing several nearby airline pilots. Federal authorities were not amused.
The GM balloons were legal, although the Navy took great pains to keep the program quiet. Each balloon was made of very thin polyethylene plastic and was twenty feet in diameter, when fully inflated with helium. Instead of the typical basket used on balloons to carry the pilot and passengers, the pilot dangled precariously below the balloon, attached using only the parachute for the instrument and a harness. With this unwieldy get-up, there was obviously no room for passengers, even if any had been willing to go for a ride.
Although Arnold modified later balloons with a plywood swing for the pilot to sit on, there was still nothing between the pilot and the ground, which made for interesting landings. The balloons leaked helium naturally, which caused the balloon to enter a gradual descent once it reached its maximum altitude on any given day. Once the balloon was about one hundred feet above the ground, the pilot would drop a one-inch weighted rope attached to the balloon. When the rope touched the ground, the weight of the balloon was reduced by the rope weight now on the ground, which caused the balloon to temporarily stop descending.
The pilot then brought himself to within five feet of the ground by simply hauling the rope into the basket; at five feet, the pilot pulled a balloon panel designed to rip away and deflate balloon, and the pilot dropped comfortably to the ground. Despite the simplicity of the system, it required a great deal of coordination and effort from the pilot. The landings could also become quite dicey when it was windy, as the rope did nothing to keep the balloon from blowing across the ground.