rocket history
Konstantin Tsiolkovskiy
Hermann Oberth
Robert H. Goddard
Wernher von Braun
Sergei P. Korolev
principles of rocketry
early U.S. rocketry
Nazi Germany’s Space Bomber
postwar U.S. rocketry
Thor, Agena, and Delta
the Titan Launch Vehicle
upper stages of rockets
solid rocket propellants
Orion Project
Russian launch vehicles
launch vehicles of other nations
the Sputnik triumph
early Soviet spaceflight
Mercury space programme
Gemini space programme
Apollo space programme
Soviet race to the Moon
Soviet space stations
Skylab space station
Apollo-Soyuz test
Space Shuttle history
the Challenger Accident
the Columbia Accident
Shuttle launches
Space Station
automated spacecraft
Lunar robotic missions
Inner planet exploration
outer planet exploration
exploring other bodies
return to Mars
solar-terrestrial physics
astronomy from space
Earth observation satellites
meteorological satellites
remote sensing satellites
early warning satellites
intelligence satellites
ballistic missiles
Energia and Khrunichev
commercial satellites
Comsat and Intelsat
International space agencies
Cape Canaveral
Vandenberg Air Base
astronauts and cosmonauts
Scaled Composites
space flight chronology

Space Shuttle history


Even before the Apollo moon landing in 1969, in October 1968 NASA began early studies of space shuttle designs. The early studies were denoted "Phase A", and in June 1970, "Phase B", which were more detailed and specific.

In 1969 President Richard M. Nixon formed the Space Task Group, chaired by vice president Spiro T. Agnew. They evaluated the shuttle studies to date, and recommended a national space strategy including building a space shuttle.

During early shuttle development there was great debate about the optimal shuttle design that best balanced capability, development cost and operating cost. Ultimately the current design was chosen, using a reusable winged orbiter, solid rocket boosters, and expendable external tank.

The Shuttle program was formally launched on January 5, 1972, when President Nixon announced that NASA would proceed with the development of a reusable Space Shuttle system. The final design was less costly to build and less technically ambitious than earlier fully reusable designs.

The first Space Shuttle, STS-1, waits on the pad before launch, March 1981.

The prime contractor for the program was North American Aviation (later Rockwell International), the same company responsible for the Apollo Command/Service Module. The contractor for the Space Shuttle Solid Rocket Boosters was Morton Thiokol (now part of Alliant Techsystems), for the external tank, Martin Marietta (now Lockheed Martin), and for the Space shuttle main engines, Rocketdyne.

The first complete Orbiter was originally named Constitution, but a massive write-in campaign from fans of the Star Trek television series convinced the White House to change the name to Enterprise. Amid great fanfare, the Enterprise was rolled out on September 17, 1976, and later conducted a successful series of glide-approach and landing tests that were the first real validation of the design.

The first fully functional Shuttle Orbiter was the Columbia, built in Palmdale, California. It was delivered to Kennedy Space Centre on March 25, 1979, and was first launched on April 12, 1981—the 20th anniversary of Yuri Gagarin's space flight—with a crew of two. Challenger was delivered to KSC in July 1982, Discovery in November 1983, and Atlantis in April 1985. Challenger was destroyed when it disintegrated during ascent on January 28, 1986, with the loss of all seven astronauts on board. Endeavour was built to replace her (using spare parts originally intended for the other Orbiters) and delivered in May 1991; it was launched a year later. Seventeen years after Challenger, Columbia was lost, with all seven crew members, during re-entry on February 1, 2003, and has not been replaced.


click to enlarge 

The Shuttle sits atop the Mobile Launcher Platform (MLP). It consists of Orbiter (on top), External Tank (at centre), and Solid Rocket Boosters (to the right and left of External Tank). Two Tail Service Masts (TSMs) to the either side of the Orbiter's tail provide umbilical connections for propellant loading and electrical power. The Shuttle is a partially reusuable launch system composed of three main assemblies: the reusable Orbiter Vehicle (OV), the expendable External Tank (ET), and the two reusable Solid Rocket Boosters (SRBs). The tank and boosters are jettisoned during ascent, so only the orbiter goes into orbit. The vehicle is launched vertically like a conventional rocket, and the orbiter glides to a horizontal landing, after which it is refurbished for reuse.

The Orbiter resembles an airplane with double-delta wings, swept 81° at the inner leading edge and 45° at the outer leading edge. Its vertical stabilizer's leading edge is swept back at a 45° angle. The four elevons, mounted at the trailing edge of the wings, and the rudder/speed brake, attached at the trailing edge of the stabilizer, with the body flap, control the Orbiter during descent and landing.

The Orbiter's crew cabin consists of three levels: the flight deck, the mid-deck, and the utility area. The highest flight deck seats the commander and pilot, two mission specialists in the back. The mid-deck has three more seats for the rest of the crew members. Galley, toilet, sleep locations, storage lockers, and the side hatch for entering/exiting the vehicle is also located there, as is the airlock hatch. The airlock has another hatch into the payload bay. It allows two astronauts, wearing their Extravehicular Mobility Unit (EMU) space suits, to depressurize before a space walk.

The Orbiter has a large 60 by 15 ft (18 m by 4.6 m) payload bay, filling most of the fuselage. The payload bay doors have heat radiators mounted on their inner surfaces, and so are kept open for thermal control while the Shuttle is in orbit. Thermal control is also maintained by adjusting the orientation of the Shuttle relative to Earth and Sun. Inside the payload bay is the Remote Manipulator System, also known as the Canadarm, a robot arm used to retrieve and deploy payloads. Until the loss of Columbia, the Canadarm had been used only on those missions where it was needed. Since the arm is a crucial part of the Thermal Protection Inspection procedures now required for Shuttle flights, it will probably be included on all future flights.

Orbital Vehicle Three Space Shuttle Main Engines (SSMEs) are mounted on the Orbiter's aft fuselage in a triangular pattern. The three engines can swivel 10.5 degrees up and down and 8.5 degrees from side to side during ascent to change the direction of their thrust and steer the Shuttle as well as push.

The Orbital Manoeuvring System (OMS) provides orbital manoeuvres, including insertion, circularization, transfer, rendezvous, abort to orbit, and abort once around.

The Reaction Control System (RCS) provides attitude control and translation along the pitch, roll, and yaw axes during the flight phases of orbit insertion, orbit, and re-entry.

The Thermal Protection System (TPS) covers the outside of the Orbiter, protecting it from the cold soak of -121 °C (-250 °F) in space to the 1649 °C (3000 °F) heat of re-entry

The orbiter structure is made primarily from aluminium alloy, although the engine thrust structure is made from titanium.

The External Tank (ET) provides 2.025 million litres (535,000 gallons) of liquid hydrogen and liquid oxygen propellant to the SSMEs. It is discarded 8.5 minutes after launch at an altitude of 60 nautical miles (111 km) then burns up on re-entry. The ET is constructed mostly of aluminium-lithium alloy about 1/8 inch thick.

Two Solid Rocket Boosters (SRBs) provide about 83% of the vehicle's thrust at liftoff and during the first stage ascent. They are jettisoned two minutes after launch at a height of about 150,000 feet (45.7 km), then deploy parachutes and land in the ocean to be recovered. The SRB cases are made of steel about 1/2 inch (1.27 cm) thick.

Computerized fly-by-wire digital flight control
The shuttle was one of the earliest craft to use a computerized fly-by-wire digital flight control system. This means no mechanical or hydraulic linkages connect the pilot's control stick to the control surfaces or reaction control system thrusters.

A primary concern with digital fly-by-wire systems is reliability. Much research went into the shuttle computer system. The shuttle uses five identical redundant IBM 32-bit general purpose computers (GPCs), model AP-101, constituting a type of embedded system. Four computers run specialized software called the Primary Avionics Software System (PASS). A fifth backup computer runs separate software called the Backup Flight System (BFS). Collectively they are called the shuttle Data Processing System (DPS).

The Shuttle deploys landing gear before landing on a selected runway just like a common aircraft. The design goal of the shuttle DPS is fail operational/fail safe reliability. After a single failure the shuttle can continue the mission. After two failures it can land safely.

The four general-purpose computers operate essentially in lockstep, checking each other. If one computer fails the three functioning computers "vote" it out of the system. This isolates it from vehicle control. If a second computer of the three remaining fails, the two functioning computers vote it out. In the rare case of two out of four computers simultaneously failing (a two-two split), one group is picked at random.

The Backup Flight System (BFS) is separately developed software running on the fifth computer, used only if the entire four-computer primary system fails. The BFS was created because although the four primary computers are hardware redundant, they all run the same software, so a generic software problem could crash all of them. This should never happen, as embedded system avionic software is developed under totally different conditions than commercial software. For example the number of code lines is tiny relative to a commercial operating system, changes are only made infrequently and with extensive testing, and many programming and test personnel work on the small amount of computer code. However in theory it can fail, so the BFS exists for that contingency.

The software for the shuttle computers are written in a high-level language called HAL/S, somewhat similar to PL/I. It is specifically designed for a real time embedded system environment.

The IBM AP-101 computers originally had about 424 kilobytes of magnetic core memory each. The CPU could process about 400,000 instructions per second. They have no hard disk drive, but load software from tape cartridges.

In 1990 the original computers were replaced with an upgraded model AP-101S, which has about 2.5 times the memory capacity (about 1 megabyte) and three times the processor speed (about 1.2 million instructions per second). The memory was changed from magnetic core to semiconductor with battery backup.

STS-26 Discovery lifts off from its platform at Kennedy Space Centre on September 29, 1988. Exhaust plumes billow from the two solid rocket boosters and covers launch pad as the Discovery, atop of the orange external tank clears the launch tower and heads for Earth orbit. STS-26 marks NASA's first human spaceflight mission since the 51L Challenger accident, January 28, 1986.

Other improvements

During STS-101 Atlantis was the first Shuttle to fly with glass cockpit. Internally the Shuttle remains largely similar to the original design, with the exception of the improved avionics computers. In addition to the computer upgrades, the original vector graphics monochrome cockpit displays were replaced with modern full-colour, flat-panel display screens, similar to contemporary airliners like the Airbus A320. This is called a "glass cockpit". In the Apollo-Soyuz Test Project tradition, programmable calculators are carried as well (originally the HP-41C). With the coming of the ISS, the Orbiter's internal airlocks are being replaced with external docking systems to allow for a greater amount of cargo to be stored on the Shuttle's mid-deck during Station re-supply missions.

Shuttle Orbiter, showing Shuttle main engines. The Space Shuttle Main Engines have had several improvements to enhance reliability and power. This explains phrases such as "Main engines throttling up to 104%." This does not mean the engines are being run over a safe limit. The 100% figure is the original specified power level. During the lengthy development program, Rocketdyne determined the engine was capable of safe reliable operation at 104% of the originally specified thrust. They could have rescaled the output number, saying in essence 104% is now 100%. However this would have required revising much previous documentation and software, so the 104% number was retained. SSME upgrades are denoted as "block numbers", such as block I, block II, and block IIA. The upgrades have improved engine reliability, maintainability and performance. The 109% thrust level was finally reached in flight hardware with the Block II engines in 2001. The normal maximum throttle is
104%, with 106% and 109% available for abort emergencies.

For STS-1 and STS-2 the external tank was painted white to protect the insulation that covers much of the tank, but improvements and testing showed that it was not required. The 600 lb saved by not painting the tank results in an almost 600 lb increase in payload capability to orbit. Additional weight was saved by removing some of the internal "stringers" in the hydrogen tank that proved unnecessary. The resulting "light-weight external tank" has been used on the vast majority of Shuttle missions. STS-91 saw the first flight of the "super light-weight external tank". This version of the tank is made of the 2195 aluminium-lithium alloy. It weighs 7,500 lb (3.4 t) less than the last run of lightweight tanks. As the Shuttle cannot fly unmanned, each of these improvements has been "tested" on operational flights.

The SRBs (Solid Rocket Boosters) have undergone improvements as well. Notable is the adding of a third O-ring seal to the joints between the segments, which occurred after the Challenger accident.

Several other SRB improvements were planned in order to improve performance and safety, but never came to be. These culminated in the considerably simpler, lower cost, probably safer and better performing Advanced Solid Rocket Booster which was to have entered production in the early to mid-1990s to support the Space Station, but was later cancelled to save money after the expenditure of $2.2 billion. The loss of the ASRB program forced the development of the Super Light-Weight external Tank (SLWT), which provides some of the increased payload capability, while not providing any of the safety improvements. In addition the Air Force developed their own much lighter single-piece SRB design using a filament-wound system, but this too was cancelled.

A cargo-only, unmanned variant of the Shuttle has been variously proposed and rejected since the 1980s. It is called the Shuttle-C and would trade re-usability for cargo capability with large potential savings from reusing technology developed for the Space Shuttle.

On the first four Shuttle missions, astronauts wore full-pressure Launch Entry Suit (LES) during ascent and descent. The pressured helmet was used from STS-5 until the loss of Challenger. The LES was reinstated when Shuttle flights resumed in 1988. The LES ended its service life in late 1995, replacing by the Advanced Crew Escape Suit (ACES).

Technical data
Orbiter Specifications (for Endeavour, OV-105)

Length: 122.17 ft (37.24 m)
Wingspan: 78.06 ft (23.79 m)
Height: 58.58 ft (17.25 m)
Empty Weight: 151,205 lb (68,586.6 kg)
Gross Lift-off Weight: 240,000 lb (109,000 kg)
Maximum Landing Weight: 230,000 lb (104,000 kg)
Main Engines: Three Rocketdyne Block 2 A SSMEs, each with a sea level thrust of 393,800 lbf (178,624 kgf)
Maximum Payload: 55,250 lb (25,061.4 kg)
Payload Bay dimensions: 15 ft by 60 ft (4.6 m by 18.3 m)
Operational Altitude: 100 to 520 nmi (185 to 1,000 km)
Speed: 25,404 ft/s (7,743 m/s, 27,875 km/h, 17,321 mi/h)
Cross range: 1,085 nautical miles (2,009.4 km)
Crew: Seven (Commander, Pilot, two Mission Specialists, and three Payload Specialists), two for minimum.

Space Shuttle Atlantis transported by a Boeing 747 Shuttle Carrier Aircraft (SCA), 1998 (NASA) External Tank Specifications (for SLWT)

Length: 153.8 ft (46.9 m)
Diameter: 27.6 ft (8.4 m)
Propellant Volume: 535,000 gallon
Empty Weight: 58,500 lb (26,559 kg)
Gross Lift-off Weight: 1.667 million lb (757,000 kg)
Solid Rocket Booster Specifications

Length: 149.6 ft (45.6 m)
Diameter: 12.17 ft (3.71 m)
Empty Weight: 139,490 lb (63,272.7 kg)
Gross Liftoff Weight: 1.3 million lb (590,000 kg)
Thrust (sea level, liftoff): 2.8 million lbf (1,270,058 kgf)
System Stack Specifications

Height: 184.2 ft (56.14 m)
Gross Lift-off Weight: 4.5 million lb (2.04 million kg)
Total Lift-off Thrust: 6.781 million lbf (3.076 million kgf)

At T minus 16 seconds, the sound suppression system begins to release a torrent of water on the Mobile Launcher Platform (MLP) and SRB trenches as the three SSMEs are started at T minus 6.6 seconds, protecting the Orbiter from damage by acoustical energy and rocket exhaust reflected from the flame trench and MLP during lift-off. All SSMEs must reach the required 100% thrust within three seconds. If the onboard computers verify normal thrust buildup, at T minus 0 the SRBs are ignited. At that point the vehicle is committed to takeoff, as the SRBs cannot be turned off once ignited. There are extensive emergency procedures (abort modes) to handle various failure scenarios during ascent. Many of these concern SSME failures, since that is the most complex and highly stressed component. After the Challenger disaster, there were extensive upgrades to the abort modes.

Shuttle launch of Atlantis at sunset in 2001. The sun is behind the camera, and the plume's shadow intersects the moon across the sky. Shortly after clearing the tower the Shuttle begins a roll and pitch program so that the vehicle is below the external tank and SRBs. The vehicle climbs in a progressively flattening arc, accelerating as the weight of the SRBs and main tank decrease. To achieve low orbit requires much more horizontal than vertical acceleration. This is not visually obvious since the vehicle rises vertically and is out of sight for most of the horizontal acceleration. Orbital velocity at the 380 km (236 miles) altitude of the International Space Station is 7.68 km per second (27,648 km/h, 17,180 mph), roughly equivalent to Mach 23. For missions towards the International Space Station, the shuttle must reach an azimuth of 51.6 degrees inclination to rendezvous with the station.

Around a point called "max-q", where the aerodynamic forces are at their maximum, the main engines are temporarily throttled back to avoid over speeding and hence overstressing the Shuttle (particularly vulnerable parts such as the wings). At this point, a phenomenon known as the "Prandtl Glauert Singularity" occurs, where condensation clouds form during the vehicle's transition to supersonic speed.

126 seconds after launch, explosive bolts release the SRBs and small separation rockets push them laterally away from the vehicle. The SRBs parachute back to the ocean to be reused. The Shuttle then begins accelerating to orbit on the Space Shuttle Main Engines. The vehicle at that point in the flight has a thrust to weight ratio of less than one — the main engines actually have insufficient thrust to exceed the force of gravity, and the vertical speed given to it by the SRBs temporarily decreases. However, as the burn continues, the weight of the propellant reduces, the ever-lighter vehicle produces more and more acceleration until the thrust to weight ratio exceeds 1 again and the vehicle can hold itself up.

The vehicle continues to climb and takes on a somewhat nose-up angle to the horizon — it uses the main engines to gain and then maintain altitude whilst it accelerates horizontally towards orbit.

Finally, in the last tens of seconds of the main engine burn, the mass of the vehicle is low enough that the engines must be throttled back to limit vehicle acceleration to 3 g, largely for astronaut health and comfort.

Before complete depletion of propellant (running dry would destroy the engines) the main engines are shut down and the external tank is released by firing explosive bolts. The tank then falls to largely burn up in the atmosphere, with some fragments falling into the Indian Ocean.

To keep the shuttle from following the external tank back into the atmosphere, the OMS engines are fired to raise the perigee out of the atmosphere. On some missions (e.g., STS-107 and missions to the ISS), the OMS engines are also used while the Main engines are still firing.

Descent and landing

The outside of the Shuttle heats to over 1,500 °C during re-entry. The vehicle begins re-entry by firing the OMS engines opposite to the orbital motion for about three minutes. The deceleration of the Shuttle lowers its orbit perigee down into the atmosphere. This OMS firing is done roughly halfway around the globe from the landing site. The entire re-entry, except for the lowering the landing gear and deploying the air data probes, is then under complete computer control. However the re-entry can be and has (once) been flown manually. The final landing can be done on autopilot, but is typically hand flown.

The vehicle then starts significantly entering the atmosphere at about 400,000 ft doing around Mach 25. The vehicle is controlled, achieved by a combination of RCS thrusters and control surfaces, to fly at a 40 degrees nose-up attitude producing high drag, not only to slow it down to landing speed, but also to reduce re-entry heating. In addition, the vehicle needs to bleed off extra speed before reaching the landing site. This is achieved by performing s-curves at up to 70 degree bank angle.

Endeavour deploys drag chute after touch-down. In the lower atmosphere the Orbiter flies much like a conventional glider, except for a much higher descent rate, over 10,000 feet per minute (roughly 20 times that of an airliner). It glides to landing with a glide angle of 4:1. At approximately Mach 3, two air data probes, located on the left and right sides of the Orbiter's forward lower fuselage, are deployed to sense air pressures related to vehicle's movement in the atmosphere.

When the approach and landing phase begins, the Orbiter is at 10,000 ft (3048 m) altitude, 7.5 miles (12.1 km) to the runway. The pilots apply aerodynamic braking to help slow down the vehicle. The Orbiter's speed is reduced from 424 mph (682.3 km/h) to approximately 215 mph (346 km/h), (compared to 160 mph for a jet airliner), at touch-down. The landing gear is deployed while the Orbiter is flying at 267 mph (429.7 km/h). In additional to applying the speed brakes, a 40 ft (12.2 m) drag chute is deployed once the nose gear touches down at about 185 knots. It is jettisoned as the Orbiter slows through 60 knots.

After landing the vehicle stands on the runway to permit the fumes from poisonous hydrazine that was used as propellant for attitude control to dissipate.

Main engine exhaust, solid rocket booster plume and an expanding ball of gas from the external tank is visible seconds after the Space Shuttle Challenger accident on Jan. 28, 1986.

Operations, applications and accidents

From left to right: Columbia, Challenger, Discovery, Atlantis and Endeavour. Not illustrated: Enterprise, Explorer and Pathfinder.


Individual Orbiters are both named, in a manner similar to ships, and numbered, using the NASA Orbiter Vehicle Designation system. Whilst all Orbiters are externally very similar, they have minor internal differences; new equipment is fitted on a rotating basis as they are maintained, and the newer Orbiters tend to be structurally lighter.

  • Handling test article designed with no spaceflight capability:
    • Pathfinder (Orbiter Simulator, no series number)
  • Mockup for display at Kennedy Space Centre visitor complex.
    • Explorer
  • Main propulsion test article, with no spaceflight capability:
    • MPTA-ET (External Tank) which is now attached to Pathfinder
    • MPTA-098 suffered major damage due to engine failure.
  • Structural test article, with no spaceflight capability:
    • STA-099 which became Challenger
  • Test vehicle suitable only for glide/landing tests, with no spaceflight capability without major refit:
  • Lost in accidents (see below):
    • Challenger (OV-099, ex-STA-099) - destroyed after lift-off - January 28, 1986
    • Columbia (OV-102) - destroyed during re-entry February 1, 2003

Crew rotation of the ISS
Manned servicing missions, such as to the Hubble Space Telescope (HST)
Manned experiments in LEO
Carry to LEO:
Large satellites — these have included the HST
Components for the construction of the ISS
Carry satellites with a booster, the Payload Assist Module (PAM-D) or the Inertial Upper Stage (IUS), to the point where the booster sends the satellite to:
A higher Earth orbit; these have included:
Chandra X-ray Observatory
Many TDRS satellites
Two DSCS-III (Defense Satellite Communications System) communications satellites in one mission
A Defence Support Program satellite
An interplanetary orbit; these have included:
Magellan probe
Galileo spacecraft
Ulysses probe

Flight statistics (as of August 25, 2005)
Shuttle Flight days Orbits Distance
-mi- Distance
-km- Flights Longest flight
-days- Crews EVAs Mir/ISS
docking Sat.
dep. †
Columbia 300.74 4,808 125,204,911 201,497,772 28 17.66 160 7 0 / 0 8
Challenger 62.41 995 25,803,940 41,527,416 10 8.23 60 6 0 / 0 10
Discovery 255.84 4,027 104,510,673 168,157,672 31 13.89 192 28 1 / 5 26
Atlantis 220.40 3,468 89,908,732 144,694,078 26 12.89 161 21 7 / 6 14
Endeavour 206.60 3,259 85,072,077 136,910,237 19 13.86 130 29 1 / 6 3
Total 1,045.99 16,557 430,500,333 692,787,174 114 *17.66 703 91 9 / 17 61

† Satellites deployed
* This was flight STS-80, during November 1996.

Two Shuttles have been destroyed in 114 missions, both with the loss of the entire crew:

Challenger — lost 73 seconds after lift-off, January 28, 1986
Further information: STS-51-L
Columbia — lost during re-entry, February 1, 2003
Further information: Space Shuttle Columbia disaster
This gives a 2% death rate per astronaut per flight.

While the technical details of the accidents are quite different, the organizational problems show remarkable similarities. In both cases events happened which were not planned for or anticipated. In both cases, engineers were greatly concerned about possible problems but these concerns were not properly communicated to or understood by senior NASA managers. In both cases the vehicle gave ample warning beforehand of abnormal problems. A heavily layered, procedure-oriented bureaucratic structure inhibited necessary communication and action. In both cases a mind set among senior managers developed that concerns had to be objectively proven rather than simply suspected.

With Challenger an O-ring which should not have eroded at all did erode on earlier shuttle launches. Yet managers felt because it had not previously eroded by more than 30%, that this was not a hazard as there was "a factor of three safety margin". Morton Thiokol designed and manufactured the SRBs, and during a pre-launch conference call with NASA, the Thiokol engineer most experienced with the O-rings pleaded with management repeatedly to cancel or reschedule the launch. He raised concerns that the unusually cold temperatures would stiffen the O-rings, preventing a complete seal, which was exactly what happened on the fatal flight. However, Thiokol senior managers overruled him, dismissing his safety concerns and allowed the launch to proceed. Challenger's O-rings eroded completely through as predicted, resulting in the complete destruction of the spacecraft and the loss of all seven astronauts on board.

Columbia was destroyed because of damaged thermal protection from foam debris that broke off the external tank during ascent. The foam had not been designed or expected to break off, but had been observed in the past to do so without incident. The original shuttle operational specification said the orbiter thermal protection tiles were designed to withstand virtually no debris hits at all. Over time NASA managers gradually accepted more tile damage, similar to how O-ring damage was accepted. The Columbia Accident Investigation Board called this tendency the "normalization of deviance" — a gradual acceptance of events outside the design tolerances of the craft simply because they had not been catastrophic to date.

While the Shuttle has been a reasonably successful launch vehicle, it has not met the goal of greatly reducing launch costs. There are various ways to measure per-launch costs. One way is dividing the total cost over the life of the program (including buildings, facilities, training, salaries, etc) by the number of launches. This method gives about $1.3 billion per launch[1]. Another method is calculating the incremental (or marginal) cost differential to add or subtract one flight — just the immediate resources expended/saved involved in that one flight. This is about $55 million. Neither figure is right or wrong; they are simply different ways to examine the picture.

The total cost of the program has been $145 billion as of early 2005, and is estimated to be $174 billion when the Shuttle retires in 2010. NASA's budget for 2005 allocates 30%, or $5 billion, to Space Shuttle operations.

Original goals of the Shuttle included operating at a fairly high flight rate (roughly 12 flights per year, at low cost, and with high reliability. Improving in these areas over the previous generation of single-use and unmanned launchers was a motivation. Although it did operate as the world's first reusable crew-carrying spacecraft, it did not greatly improve on those parameters, and is considered by some to have failed in its original purpose.

Although the final design differs from the original concept, the project was still supposed to meet USAF goals and be much cheaper to fly in general. One reason behind this apparent failure is inflation. During the 1970s the U.S. suffered from severe inflation. Between when the program began in 1972, and first flight in April 1981, inflation increased prices over 200%. When evaluating shuttle development costs in later-year dollars, this superficially appeared to be a large cost overrun in the program. In fact when discounting inflation, the shuttle development program was within the initial cost estimate given to President Richard M. Nixon in 1971.

The high shuttle operational costs have been much more than anticipated, if counting all associated support resources (total expenditures, including development costs, divided by number of flights). Some of this can be attributed to a lower flight rate, operating beyond the 10-year anticipated lifespan of each Shuttle, and higher than anticipated maintenance costs. The marginal or incremental per launch costs have been about 50% more than early projections.

Some reasons for higher than expected operational costs can be ascribed to:

Maintenance of thermal protection tiles turned out to be very labour intensive, averaging about 1 person·week to replace a tile, with hundreds damaged with each launch.

The main engines were highly complex and maintenance intensive, necessitating removal and extensive inspection after each flight. Before the current "Block II" engines, the turbopumps (a primary engine component) had to be removed, dissembled, and totally overhauled after each flight.

Launch rate is significantly lower than initially expected. This does not reduce actual operating costs, but if dividing total program costs by number of launches, more launches per year produces a lower per-launch cost figure. Some early hypothetical studies examined 55 launches per year, but the maximum possible launch rate was limited to 24 per year, based on manufacturing capacity of the external tank. Early in the shuttle development, the expected launch rate was about 12 per year. Launch rates reached 9 per year in 1985 but averaged less thereafter.

Early cost estimates of $118 per pound of payload were based on marginal or incremental launch costs, and based on 1972 dollars and assuming a 65,000 pound payload capacity. Correcting for inflation and other factors, this equates to roughly $36 million incremental costs per launch. Compared to this, today's actual incremental per launch costs are about 50% more, or $55 million per launch.

Shuttle operations

The Shuttle was originally conceived to operate somewhat like an airliner. After landing, the orbiter would be checked out and start "mating" to the rest of the system (the ET and SRBs), and be ready for launch in as little as two weeks. Instead, this turnaround process takes months; Columbia was once launched twice within 56 days. Because loss of crew is unacceptable, the primary focus of the Shuttle program is to return the crew to Earth safely, which can conflict with other goals, namely to launch payloads cheaply. Furthermore, because in some cases there are no survivable abort modes, many pieces of hardware simply must function perfectly and so must be carefully inspected before each flight. The result is high labour cost, with around 25,000 workers in Shuttle operations and labour costs of about $1 billon per year.

During development, shuttle features were primarily chosen based on capability required to service the future space station. Even though the initially planned Space Station Freedom was significantly scaled back, the shuttle was still vital to service it. No other launch vehicle had the shuttle's payload capability or could return large items from the space station to earth.

NASA's plan for using the shuttle to launch all unmanned payloads declined, then was discontinued. Following the Challenger disaster, carrying in the shuttle payload bay the powerful liquid fuelled Centaur upper stages planed for interplanetary probes was ruled out. The Shuttle's history of unexpected delays also makes it liable to miss narrow launch windows. Advances in technology over the last decade have made probes smaller and lighter, and as a result unmanned probes and communications satellites can use cheaper and more reliable expendable rockets, including Delta launcher, and Atlas V.

Looking back and ahead

Opinions differ on the lessons of the Shuttle. While it was developed within the original development cost and time estimates given to President Richard M. Nixon in 1971, the operational costs, flight rate, payload capacity, and reliability have been worse than anticipated.

In general future designers look to less complex, more reliable launch systems with lower maintenance costs. One approach is Single Stage To Orbit (SSTO), which would be 100% reusable and use a single stage. NASA evaluated several concepts in the 1990s, and selected the X-33, which would eventually have been the Venturestar. During design that program increased in complexity and development cost, encountered problems and was finally cancelled.

Another variant of SSTO is a hypersonic, scramjet-powered, air breathing vehicle. This would be launched and landed horizontally like an airliner. It would achieve much of orbital velocity while still within the upper atmosphere. It was originally investigated by the U.S. Department of Defence, but passenger-carrying civilian versions were planned, sometimes called the "New Orient Express". The official name was the Rockwell X-30. Like the X-33, the X-30 encountered major technical difficulties, primarily due to the system complexity and materials required for hypersonic flight, and was also cancelled.

Another approach is lower cost expendable launch vehicles. NASA currently uses these for unmanned launches, and plans to use them for future manned launches. NASA plans on using modified shuttle components to build an expendable Shuttle Derived Launch Vehicle. This technology would be used to develop two separate launchers, one for manned missions and the other for unmanned heavy cargo. This contrasts with the current shuttle where astronauts and heavy cargo are launched in a single vehicle. Unlike the shuttle, this future launcher and associated crew exploration vehicle will have a launch escape system to save the crew in the event of a disaster.