Landing the Space Shuttle – A Flying Machine and the Thrill of a Lifetime

Charlie Precourt describes his first landing at the controls of Atlantis, the culmination of a 10-day mission to the Russian Mir station to deliver supplies and a fresh crew. In a dead-stick descent with no go-around option, Atlantis was slowing through Mach 1 at 50,000 feet, just 3 minutes and 25 seconds from touchdown. A CAPCOM call confirmed winds for Runway 33 at 090 at 8 gusting 15, clearance for nominal aimpoint and nominal drag chute deploy, and a predicted touchdown 2,600 feet from the threshold at 195 knots; Precourt acknowledged, and the stiff crosswind meant he would have to adjust.

The HAC was the heading alignment cone, an imaginary spiral and curved glide path from 40,000 feet to the ground. Nominal conditions meant no abnormal landing procedures were required, and the touchdown airspeed and distance predictions matched expectations. Precourt disconnected the autopilot at 0.95 Mach and about 45,000 feet, after which mission pilot Eileen Collins called advisory information for the rest of the descent. All calls over the next three minutes were highly scripted and rehearsed to keep the crew coordinated through wheel stop at Kennedy Space Center, and hundreds of hours of training kicked in as he reflected on descending from Mach 1 and 50,000 feet to a 195-knot touchdown in roughly the time a typical homebuilt aircraft takes on a 5-mile final.

The key training was in a modified Gulfstream G-II business jet that flew with the shuttle’s performance and feel. It could fly the same 20-degree dive glide path and maintain 300 knots down final by approaching with landing gear down and modulating thrust reversers in flight, whereas the shuttle used a split rudder as a speed brake. The left seat replicated the shuttle commander’s displays and controls, while the right seat remained standard Gulfstream so a safety pilot/instructor could switch the aircraft into and out of shuttle simulation mode; Precourt calls it a literal flying simulator. He also notes a minimum requirement of 500 practice approaches.