At the Langley laboratory, proponents and would-be managers of a manned space flight program studied the nonlifting approach to orbital circumnavigation, refined this concept, tested it, restudied it, and invented new ways to prove hardware feasibility and reliability. Floyd L. Thompson, Associate Director of Langley and Acting Director most of the time, gave Robert Gilruth the go-ahead for manned satellite work. In turn, Gilruth gave a free hand to PARD Chief Joseph A. Shortal, Faget, Purser, Charles Mathews, Alan B. Kehlet, Willard S. Blanchard, Jr., Carl A. Sandahl, and others at the Virginia laboratory.61
The search for better experimental methods in manned satellite research produced a concept by Purser and Faget for a new test rocket which would employ a cluster of four solid-propellant Sergeant rockets to provide a high initial thrust. Fired almost vertically and unguided except for large stabilizing aerodynamic fins, the rocket would be an inexpensive means of testing full-scale models of spacecraft in the most critical phases of an orbital mission - launch, abort, and escape at different speeds and under different stresses, parachute deployment, and recovery. Such a vehicle could also "toss" a man in a ballistic capsule to an altitude of perhaps 100 miles. Late in February, Purser and Faget received a job order and authorization to proceed with design work on the test rocket, which at that time they called "High Ride."62
Another experimental technique devised by the PARD engineers was a full-scale "capsule simulator." It was designed to test the practicability of controlling the attitude of a ballistic vehicle manually by activating air jets mounted on its body, similar to the method that would be used to control the X-15 at the peak of its trajectory. In March, Purser and several others in PARD put into operation a crude simulator rig featuring a small bed covered by a tent and attached to a pendulum. The pendulum permitted an oscillation period of two to four seconds, during which the "pilot" attempted to realign the simulator by firing the air jets. Throughout the spring Langley test pilot Robert A. Champine, Purser, and others took turns riding the simulator. Frequently modified and improved, it provided useful data on spacecraft reaction controls.63
Meanwhile Faget and his coworkers were steadily modifying the manned ballistic satellite design itself. Almost from the beginning of their design studies and tests, late in 1957, they had assumed that a ballistic vehicle should enter the atmosphere at an attitude 180 degrees from that of launch, so the g forces would be imposed on the front of the body under both acceleration and deceleration. The "tail" of the capsule when it went into orbit would become its "nose" during reentry. Their original capsule configuration - a squat, domed body with a nearly  flat heatshield - resembled the Mark II missile warhead. The body was recessed slightly from the perimeter of the heatshield, leaving a narrow lip that theoretically would deflect the airflow in such a way as to minimize heat transfer to the after portion. But models of this configuration tested in the Langley free-spinning tunnel proved dynamically unstable at subsonic speeds. The Faget group then lengthened the capsule fuselage and eliminated the heatshield lip. By March 1958, the Langley ballistic vehicle, as described by Faget, Garland, and Buglia at the Ames Conference on High-Speed Aerodynamics, was an elongated cone. This design contrasted sharply with the configuration sketched earlier that month at the ARDC working conference in Los Angeles - a rather deep dome, the rounded front end of which was the heatshield.64
The elongated cone provided dynamic stability during the blazing period of reentry, but tests in the 11-inch hypersonic tunnel and other tunnels at Langley showed that too much heat would be transferred by turbulent convection to its afterbody. Besides thermodynamic considerations, the NACA planners could not figure out how to fit into the top part of the cone the two parachutes necessary for its recovery. The Virginia designers next tried a conical nose shape, then a rounded one with a short cylinder attached to it, but the problems of heat transfer from the heatshield and insufficient space for parachute packaging remained for both of these configurations. It was late summer 1958 before the Langley-PARD researchers had settled on a capsule design combining the advantages of maximum stability in a nonlifting body, relatively low afterbody heating, and a suitable parachute compartment. This was the shape that became the basis of the Mercury spacecraft - a blunt face, a frustum, or truncated cone, and a cylinder mounted atop the frustum. The completely flat heatshield had been discarded because it trapped too much heat, while a rounded face only increased heat transfer. The design ultimately chosen featured a heatshield with a diameter of 80 inches, a radius of curvature of 120 inches, and a ratio of 1.5 between the radius of the curve and the diameter of the shield.65 This heatshield design, as worked out by William E. Stoney, Jr., of PARD at Langley, and confirmed by Alvin Seiff, Thomas N. Canning, and other members of the Vehicle Environment Division at Ames, got rid of a maximum amount of heat during reentry.66
Materials research continued at Langley throughout the spring and summer. In their man-in-space development plans, the Air Force experts initially had favored an ablation heatshield, but their NACA advisers generally felt that the ablation technique was not yet reliable enough for manned reentry. In March, two of the most respected engineers in the NACA establishment, Gilruth and Soulé of Langley, assisted by Clotaire Wood of Headquarters, had presented to the Air Research and Development Command NACA's design concepts for manned orbital flight, including use of the heat sink on a blunt body as the best thermal protection procedure. The question remained open, however. In June, the Wright Air Development Center, the Ballistic Missile Division, and NACA agreed to undertake joint investigation of heatshield materials, the objective being  to compile a sufficient quantity of data for ARDC to make a decision between heat sink and ablation methods within three months.67
Considering the unreliability common in early ballistic missiles, and especially the widespread lack of confidence in the hard-pushed Atlas, some fast and almost foolproof means of escape would be essential to any launch system for manned space flight. The Air Force man-in-space designs had included an escape mechanism with many moving parts and a degree of complexity unacceptable to the NACA engineers. The Air Force plans envisioned a pusher rocket escape system, meaning that a rocket or rockets would fire at the base of the capsule to hurl it clear of the booster. The PARD rocket experts, again led by Faget, rejected this approach and began working on a solid-fueled tractor escape rocket. This would be mounted above the capsule and would pull it upward and away from a faulty launch vehicle. By the end of August 1958, Willard Blanchard and Sherwood Hoffman of PARD, working on plans and suggestions hurriedly made by Faget and Andre J. Meyer, Jr., had drawn designs for the escape rocket and tower, consisting of a slender rocket case and nozzle and three thin struts fastened to the cylinder of the capsule. The Wallops Island engineers already were planning a series of test firings of the awkward-looking escape mechanism, using "boilerplate" capsules, or full-scale metal models.68
The solid-fueled tractor rocket with a minimum of components reflected the Langley-PARD preoccupation with the easiest, most dependable way to get a manned spacecraft into orbit. There were certain interlocking aspects of the approach. The "bare" Atlas, the regular ICBM without an upper stage, should be the booster. With the ballistic capsule, acceleration forces during launch would be about 5 or 6 g; on a shallow reentry trajectory, deceleration loads should not exceed 8 or 9 g. But an abort and reentry after a launch following the steep trajectory and unbroken acceleration of a single-stage booster could impose as much as 20 times the force of gravity on the capsule passenger. Air Force planners had considered a two-stage booster and a flight profile with a more shallow trajectory, or a variable-drag device like the Avco metal parachute, to lessen the abort-reentry g loads - although by midsummer cost considerations were pushing the Air Force toward the bare Atlas.69 For body support, the Air Force had thought in terms of some kind of rotational apparatus to maintain continuously optimum positioning in relation to the direction of acceleration.70 This procedure, the NACA engineers felt, was too complicated and probably entailed too much weight.
As Man-in-Space-Soonest was taking shape in late spring, featuring a two–stage booster and either a rotatable interior cabin or a rotatable couch, Faget had another idea. Why not build a lightweight, stationary couch that a man would lie not on but in? This was the fundamental principle behind the contour couch designed by Faget, fabricated out of fiber glass at Langley, and tested on the big Navy centrifuge at Johnsville late in July.7l There, in what Faget called "the only technical 'break-through' of the summer," Carter C. Collins and R. Flanagan Gray  of the Navy endured more than 20 g while riding in the contour couch. Then, said Faget, "we were able to disregard the USAF 'ground rule' (and a rather firmly established one in their minds) that 12 g was the reentry design limit." The bare Atlas could be used to hurl a man into orbit, and an abort need not impair his safety.72
60 See "Current NACA Aerodynamic Research Relating to Upper Atmosphere and Space Technology," NACA Hq., March 10, 1958; "NACA Research into Space," 6-15; memo, Purser to Robert R. Gilruth, "Langley Manned Satellite Program," April 11, 1958.
61 Purser interview.
62 Purser log; Grimwood, Mercury Chronology, 14; William M. Bland, Jr., "Project Mercury," in Eugene M. Emme, ed., The History of Rocket Technology: Essays on Research, Development, and Utility (Detroit, 1964), 212.
63 Purser log; Purser memo.
64 "How the Mercury Capsule Design Evolved," Aviation Week, LXX (Sept. 21, 1959), 52-53; letter, Alan B. Kehlet to L.S.S., July 2, 1964; Faget interview.
65 "How Mercury Capsule Design Evolved," 53-54; Christopher C. Kraft, Jr., "A Review of Knowledge Acquired from the First Manned Satellite Program," NASA/MSC fact sheet No. 206.
66 William E. Stoney, Jr., "Aerodynamic Heating of Blunt Nose Shapes at Mach Numbers Up to 14," in "NACA Conference on High-Speed Aerodynamics, Ames Aeronautical Laboratory, Moffett Field, Calif., March 18, 19, and 20, 1958, A Compilation of the Papers Presented," 227-244; Stoney, interview, Houston, Feb. 13, 1964; Alvin Seiff and Thomas N. Canning, interviews, Moffett Field, Calif., April 22, 1964.
67 Wood, interview, Washington, Sept. 1, 1965; memo, Wood, "Background on WADC Letter to NASA of October 22, 1958, Covering 'Ablation/Heat Sink Investigation - Manned Reentry,' " Nov. 7, 1958. The Air Force and NACA investigators completed their tests in October 1958. ARDC had conducted tests at the Chicago Midway Laboratories, while the NACA engineers had worked in the hot jet facilities at Langley. Both groups devoted most of their time to studies of ablation. Memo, Lester J. Charnock, Tech. Dir., Deputy Chief of Staff for Plans and Operations, to Comdr., Air Research and Development Command, "Ablation/Heat Sink Investigation - Manned Reentry," Oct. 21, 1958.
68 Memo, George M. Low to NASA Administrator, "House Committee Staff Report on Project Mercury," Jan. 26, 1960; "How Mercury Capsule Design Evolved," 55; Willard S. Blanchard, interview, Langley Field, Va., Jan. 6, 1964; Bland, "Project Mercury," 215.
69 Besides fear of excessive g on the abort-reentry flight profile, the Air Force was reluctant to use the bare Atlas because of its unproven nature and because it supposedly would place a capsule in a lower orbit than the Thor-fluorine rocket, necessitating more tracking stations to maintain constant voice contact with the human passenger.
70 Stanley C. White, interview, San Antonio, Aug. 18, 1965.
71 See pp. 43-46. 73 Faget, marginal notes on "Outline of History of Man-in-Space Program."