30 March 2015

NASA's 1992 Plan to Land Soyuz Space Station Lifeboats in Australia

Image credit: NASA
Kosmos 133, the first in the long line of Soyuz ("union") spacecraft, lifted off unmanned from Baikonur Cosmodrome in Central Asia on 28 November 1966. Its mission: to dock with Kosmos 134, another unmanned Soyuz scheduled for launch the following day. The Kosmos 133-Kosmos 134 combined mission was meant to put Soyuz through its paces ahead of a nearly identical piloted Soyuz docking mission.

The Soviet Union's new piloted spacecraft was a union of three modules, which together weighed about 7000 kilograms. They were, from aft to fore, the cylindrical Service Module; the 2900-kilogram Descent Module; and, linked to the Descent Module by a hatchway, the spherical Orbital Module. 

The cramped Descent Module was the only part of Soyuz spacecraft meant to withstand reentry. In addition to a heat shield, parachutes, and solid-propellant rockets for soft-landing on land, it included the main control console and three cosmonaut launch and landing couches.

The Service Module carried a pair of solar array "wings" on its sides for making electricity. It included the main engine which enabled it to change its orbit; orbit-changing maneuvers included the deorbit burn performed when time came to return home. The Orbital Module provided extra living and storage space and carried a docking unit. Both the Orbital Module and the Service Module were cast off following the deorbit burn and disintegrated high in the atmosphere.

Any joy flight controllers near Moscow felt as Kosmos 133 soared above the Earth vanished when they found that its attitude control system did not work properly. They called off the Kosmos 134 launch. Several times they tried to orient Kosmos 133 to point its main engine in its direction of orbital motion so that they could slow the spacecraft and begin reentry. On 30 November, they commanded the first Soyuz to self-destruct when it appeared that its Descent Module would land in China.

November 1966: Kosmos 133, the first Soyuz spacecraft, was launched without a crew. This widely used illustration, of unknown provenance, shows its three modules. From left to right they are: the Orbital Module; the Descent Module; and the Service Module.
When reporting on the half decade that followed Kosmos 133, it needs less space to describe Soyuz and Soyuz-derived spacecraft successes than it does to list their failures. Kosmos 186 and Kosmos 188 successfully performed the first automated docking in late October 1967, and Kosmos 212 and Kosmos 213 repeated the feat in April 1968.

In January 1969, the piloted Soyuz 4 and Soyuz 5 spacecraft docked and two cosmonauts spacewalked between them. Zond 7, a prototype manned circumlunar Soyuz variant without an Orbital Module, flew unmanned around the moon and landed as planned in the Soviet Union in August 1969, a month after Apollo 11 became the first mission to land men on the moon. The two-man crew of Soyuz 9 orbited Earth for nearly 18 days in June 1970, breaking the space endurance record Gemini VII had set in 1966.

These scattered successes should not obscure the fact that, of the 16 individual cosmonauts launched on Soyuz between 1967 and 1971, one-quarter lost their lives. Of the more than 30 Soyuz-derived spacecraft launched in the same period, all but nine failed in some significant way.

Following the deaths of the three Soyuz 11 cosmonauts after they undocked from the Salyut 1 space station on 29 June 1971, Soyuz underwent a major redesign. When piloted Soyuz flights resumed in September 1973, the spacecraft could carry no more than two space-suited cosmonauts. Soyuz spacecraft suffered more malfunctions in the 1970s, often failing to reach their Salyut space station targets, but no more cosmonauts died.

The advent in 1977 of the highly reliable Progress variant, an automated cargo ship for resupplying space stations, marked a break from the past for Soyuz. Malfunctions tailed off and, after a dramatic booster explosion on the launch pad on 26 September 1983, no Soyuz failed to reach its space station target. Even the booster explosion could be seen as a sign of design maturity; despite suffering launch-escape system damage, the Soyuz-T-10a spacecraft saved its two-man crew.

Technology upgrades produced first the Soyuz-T and then the Soyuz-TM, which could transport up to three space-suited cosmonauts. By the early 1990s, Soyuz had developed a reputation for sturdy reliability.

Forward view of Soyuz-TM spacecraft. Image credit: NASA
Aft view of Soyuz-TM spacecraft. Image credit: NASA
Even before the Soviet Union collapsed in 1991, officials with the Soviet aerospace enterprise NPO Energia began to peddle its wares, including Soyuz, at major international aerospace meetings. An implied subtext of these promotional efforts was that, if the West would not buy products from the financially strapped Soviet aerospace sector, then its engineers might sell their technical expertise to countries opposed to Western interests.

The threat - and promise - of Soviet space technology soon attracted the attention of the U.S. government. Spaceflight entered the geopolitical arena in a way it had not done since the early 1970s, when the 1975 Apollo-Soyuz Test Project became the poster child for President Richard Nixon's policy of detente.

In December 1991, Congress directed NASA to study the feasibility of using the Soyuz-TM as a low-cost "lifeboat" or "escape pod" for its planned Space Station Freedom. The concept of a space station lifeboat is an old one, dating back at least to the 1960s. NASA had acknowledged the need for such a vehicle soon after the January 1986 Challenger accident killed seven astronauts and grounded the Space Shuttle fleet for almost three years.

NASA foresaw three scenarios in which a Space Station lifeboat might save lives. First, a medical emergency on board Space Station Freedom might require the rapid evacuation of a sick or injured astronaut. Second, a disaster - for example, a fire - might render Freedom uninhabitable. Finally, another Shuttle accident might ground the Orbiter fleet, stranding a crew on board Freedom with no resupply.

A NASA-designed ACRV for eight astronauts shortly after touchdown. Image credit: NASA
By early 1992, NASA had offered up several designs for an Assured Crew Return Vehicle (ACRV), as it called its planned Space Station Freedom lifeboat. Unfortunately, even the simplest design would cost at least $1 billion to develop. It would, after all, constitute a wholly new piloted spacecraft designed to remain docked to Freedom for years, dormant but always ready.

As part of the preliminary Soyuz ACRV feasibility study for Congress, NASA engineers traveled to Moscow in March 1992 to meet with Russian government and NPO Energia officials. The space agency completed its study the following month.

In its study report, NASA portrayed Soyuz-TM as an interim lifeboat for the period when Space Station Freedom's crew numbered no more than three astronauts. Use of Soyuz-TM would, it was hoped, bring nearer the day when Freedom could be staffed continuously. In about the year 2000, as Freedom's population grew to six or eight astronauts, an "optimized" U.S.-built ACRV would take over from Soyuz-TM.

June 1992: U.S. President George H. W. Bush (right) and Russian President Boris Yeltsin. Image credit: National Archives
On 17 June 1992, U.S. President George H. W. Bush and Russian Federation President Boris Yeltsin signed agreements in Moscow providing for broad space cooperation. A Russian cosmonaut would fly on the Space Shuttle, a U.S. astronaut would live on board the Mir space station, and a Shuttle Orbiter would dock with Mir. The following day, NASA and the Russian Space Agency signed a $1-million contract to jointly assess Russian space technology, including Soyuz-TM, for use in NASA programs.

It had, of course, already become evident that Soyuz-TM would need modifications before it could serve as an ACRV for Freedom. Its Russian language control panel labels would, for example, need to be replaced with labels in English. More significantly, its on-orbit endurance would need to be stretched from 180 days to three years and its docking unit would need to be made compatible with Freedom's docking ports. In addition, NPO Energia would need to find a way to squeeze NASA's tallest astronauts into Soyuz's cramped Descent Module.

Even more challenging was Space Station Freedom's planned orbit about the Earth. NASA expected to assemble its station in an orbit inclined 28.5° relative to Earth's equator. It would orbit over an equator-centered, globe-girdling band of Earth's surface spanning from 28.5° north latitude to 28.5° south latitude. Freedom's planned inclination meant that a Shuttle Orbiter launched from Kennedy Space Center, located on Florida's east coast at 28.5° north latitude, would in theory be capable of reaching the station bearing its maximum possible payload.

Space Station Freedom's orbit meant that, if Soyuz-TM were launched from Baikonur Cosmodrome on the usual Soyuz launch vehicle, then it could not reach the U.S. station. The sprawling Central Asian launch site is located in Kazakstan at 46° north. The Soyuz launcher normally propels Soyuz and Progress spacecraft toward an orbit inclined 51.6° relative to the equator to avoid flying over China during ascent. This meant that, upon reaching orbit, the Soyuz-TM ACRV would need to change its orbital plane by a whopping 23.1° to rendezvous with Freedom.

Each degree of plane change would demand hundreds of kilograms of propellants. If the Soyuz ACRV were launched to Space Station Freedom from Baikonur, then the larger, more powerful, and more costly four-stage Proton launcher would have to do the job. Its entire fourth stage, suitable for boosting spacecraft out of Earth orbit toward the moon and planets, would be expended to accomplish the required plane change.

NASA envisioned that a Shuttle Orbiter launched from Kennedy Space Center would deliver the unmanned Soyuz ACRV to Space Station Freedom. Once there, Orbiter or Station robot arms would pluck the Soyuz ACRV from the Orbiter's payload bay and berth it at a waiting Freedom docking port. Alternately, the Soyuz ACRV might be launched unmanned from Florida on a U.S. expendable rocket such as Atlas and perform an automated rendezvous and docking with Freedom.

Space Station Freedom's 28.5° orbital inclination would limit where the Soyuz ACRV's Descent Module could land after it evacuated a crew. The normal Soyuz landing area is located at about 50° north, far beyond the range of a Soyuz ACRV returning from Freedom.

In a June 1993 report, the ACRV Project Office at NASA Johnson Space Center in Houston, Texas, summed up a study of potential Soyuz ACRV landing zones. It noted that, because of Space Station Freedom's orbital inclination, a Soyuz ACRV could land on U.S. soil only in south Texas and south Florida. (The report made no mention of Hawaii, the southernmost U.S. state, over which Freedom would pass regularly.)

The ACRV Project Office then looked abroad to friendly countries with wide-open spaces. Australia appeared ideal. The northern two-thirds of the country lies between 28.5° and 10° south latitude and much of its interior is flat, arid, and sparsely populated.

As part of the June 1992 contract activities, NASA engineers and officials, a U.S. State Department representative, and NPO Energia engineer Valentin Ovciannikov traveled to Australia in November 1992 to conduct a preliminary assessment of four candidate Soyuz ACRV landing zones. The Australian Space Office (ASO), working with the Australian Geological Survey Organization and the National Resource Information Center, chose the zones based on NPO Energia and NASA selection criteria.

November 1992: The NASA ACRV Project Office team's route through Australia during its tour of prospective Soyuz ACRV landing zones. Image credit: David S. F. Portree/NASA
The landing zone survey team stopped first in Australia's capital, Canberra, to meet with government officials. NASA expected that Australia, a signatory of the 1967 United Nations "Agreement on the Rescue of Astronauts, the Return of Astronauts, and the Return of Objects Launched into Space," would stand ready to assist space travelers forced to land on its territory. The team found tentative support for its plans, though the Australians made it clear that they would approve nothing until the U.S. and Australia signed a nation-to-nation treaty covering responsibility for costs and damages.

On 11 November the team began a whirlwind eight-day, 9800-kilometer tour of the proposed landing zones. Team members flew first to Adelaide, the capital of South Australia. There they met with state police to describe the Soyuz ACRV mission and learn about Search and Rescue (SAR) capabilities in the Coober Pedy-Oodnadatta region. Coober-Pedy, "the Opal Capital of the World," is a town of about 2000 people in the Australian Outback north of Adelaide. 

The team learned that the police were responsible for SAR operations throughout Australia, and that Australian SAR personnel and equipment were concentrated in capital cities, not scattered among small Outback communities. In South Australia, the state police had four elite rescue teams and three small airplanes that could reach Coober Pedy's 1400-meter-long asphalt runway in two and a half hours. They leased a single helicopter that could reach the area in four hours.

The next day (12 November), the team flew to Coober Pedy in a small chartered plane. There they learned that the local police and mine rescue service had at their disposal several four-wheel-drive vehicles and an ambulance. They found that much of the area was dry and flat with red, gravel-covered soul of good bearing strength. The hard surface would enable four-wheel-drive vehicles to reach points throughout the area and would help to ensure that the Soyuz ACRV land-landing system would operate properly.

As an aside, the team noted in its report that NASA could learn a great deal by participating in a Soyuz-TM landing. NASA engineers subsequently observed the Soyuz-TM 16 landing in Kazakstan on 22 July 1993. It was an appropriate landing for them to observe, for the spacecraft had been used to test a Russian-built APAS-89 universal docking unit of the type U.S. Shuttle Orbiters would use to dock with the Mir station during Shuttle-Mir missions (1994-1998). The APAS-89 system, which was based on the U.S.-Soviet APAS-75 system developed jointly for ASTP, had been built originally to enable the Soviet Buran shuttle to dock with Mir and its planned successor, Mir-2.

In the south part of the Coober Pedy zone, the survey team gathered data on the "moon plain," a large area where trees - gidgee and acacia - grew along dry watercourses and the soil had "fair to poor" bearing strength. They also noted a field of small sand dunes. NPO Energia's Ovciannikov worried that the Soyuz ACRV Descent Module might roll between two dunes and become stuck with its top-mounted crew hatch buried in the sand. Using a hand-held anemometer and historical weather data from the Australian Bureau of Meteorology, the team determined that wind speeds near Coober Pedy would be acceptable for Soyuz ACRV landings.

The team spent the night in Coober Pedy listening to the distant howls and barks of dingoes, then flew on to Perth, the capital of Western Australia. On 13 November they discussed with state police the SAR capabilities in the area of Meekatharra, about 1240 kilometers to the northeast. 

They also learned of the Royal Flying Doctor Service (RFDS), which had one of its 14 bases in Perth. RFDS provided rapid medical response to two-thirds of the Australian continent, including all four of the candidate Soyuz ACRV landing zones. In their report, the team suggested that NASA doctors should begin to coordinate with the RFDS as soon as possible.

The police in Perth made it clear that current local needs had priority over any future NASA needs. They asked to be alerted 24 hours before an expected Soyuz ACRV landing. In its report, the team noted that this would not be possible for a medical or emergency evacuation, though it would be possible for a crew returning from Freedom during a prolonged Shuttle stand-down.

The team flew to Meekatharra on 14 November. Of great interest was a 2180-meter-long, 45-meter-wide asphalt runway at the Meekatherra Airport. In their report, the team suggested that the runway, built originally for emergency Boeing 707 landings, might be used to land cargo planes bearing rescue equipment, four-wheel-drive vehicles, and helicopters. 

The team judged that Meekatherra's soil had "excellent" bearing strength. Acacia and munga trees stood over less than 10% of the area, which was very flat. There were, however, scattered bedrock outcrops protruding from the windswept plain. In addition to presenting a minor impact hazard, the outcrops included naturally radioactive "uraniferous" deposits. Ovciannikov expressed concern that these might interfere with the Descent Module's altimeter, which relied on a radioactive source.

Meekatherra is only about 500 kilometers from Australia's west coast, a fact that had both pluses and minuses for Soyuz ACRV landings. On the one hand, it meant that debris from discarded Orbital Modules and Service Modules would not fall on land. On the other hand, the Descent Module might fall short of land if it followed a ballistic reentry path; that is, if it failed to rotate about its center of gravity to generate lift. Following a ballistic reentry, quick crew recovery might be crucial; a ballistic reentry would subject the astronauts, who might be weak after a long stay in weightlessness, to deceleration equal to 10 times Earth's surface gravity.

The team flew on to Darwin, capital of the Northern Territory, on 15 November. There territorial police described their 30-member Police Task Force, which was trained to deal with situations diverse as riot control, bomb disposal, and cliff rescue. 

The proposed Soyuz ACRV landing zone in the Northern Territory, the largest of the four candidates, was centered on the town of Tennant Creek (population 3200). The territorial police explained that their SAR resources were based both in Darwin, 970 kilometers from Tennant Creek, and in Alice Springs, 480 kilometers away. 

The team visited the Tennant Creek zone on 16 November. They learned that the Tennant Creek police force included 25 officers but only one four-wheel drive vehicle. The police worried that the Soyuz ACRV soft-landing rockets might start brush fires. Ovciannikov assured them through an interpreter that they would not.

The team noted that the proposed landing zone was in the sprawling Barkley Tableland, a region of black-earth raised plains covered with gold-colored Mitchell grass. Ovciannikov observed that the area resembled the Soyuz-TM "landing grounds" around Dzhezkazgan, Kazakstan. 

Unlike the other landing zones, Tennant Creek had distinct wet and dry seasons, with the former occurring in the southern-hemisphere summer/early autumn months (December through March). Located just 19.5° south of the equator, it was also the hottest of the four zones, with an average of 22 days per year above 40° Celsius (104° Fahrenheit). Flooding from seasonal rains would not interfere with a Soyuz ACRV landing, Ovciannikov explained, though it might impede surface vehicles dispatched to recover the astronauts.

The team flew to Charleville in Queensland on 17 November without stopping in Brisbane, the state's capital. They found that the local airport included two asphalt runways, the largest of which was more than 1500 meters long and 30 meters wide. Though they met with local police, the team's report on the Charleville zone included no SAR data.

Charleville's rolling plains, or downs, differed from the other zones the team surveyed in that they included many large trees (briglow and sandalwood) interspersed with "square" and "circle" treeless areas used for grazing and farming. Charleville police told the team that local ranchers knocked down and burned the trees to create grazing land; if left alone, however, the trees grew back within a few years.

Ovciannikov compared Charleville to the "wooded steppe" on the north edge of the Soyuz-TM landing zone near Arkalyk, Kazakstan. The open areas would make acceptable landing sites, he judged, though the bearing strength of the black and brown loamy soils could be rated only "fair."

The team returned to Canberra late on 18 November. After another meeting with Australian government officials, during which they signed a document that summarized what the parties had learned and what had been agreed, its members departed Australia on 20 November 1992.

Soyuz-TM Descent Module on the treeless steppe in Kazakstan. Image: NASA
Shortly before the team began its Australian tour, U.S. voters went to the polls, where they favored Democrat William Clinton for President over Republican incumbent George H. W. Bush. Many in NASA feared that, after he took office in January 1993, Clinton would not support Space Station Freedom. With no station, their reasoning went, the Space Shuttle would lose its purpose, and U.S. piloted spaceflight would cease.

Clinton did not in fact support Space Station Freedom; that did not mean, however, that he failed to find value in a space station. On 9 March 1993, he ordered NASA to produce three low-cost station designs in 90 days. Aided by an advisory committee, he would select one design for development.

The new President then handed off supervision of NASA to his Vice President, Al Gore. On 25 March, Gore appointed members to the Advisory Committee on the Redesign of the Space Station. MIT's Charles Vest became its chair.

That same month, in a letter to NASA Administrator Daniel Goldin, Russian Space Agency director Yuri Koptev and NPO Energia director Yuri Semenov proposed what would become the NASA station program's salvation: a merger of the financially strapped, politically troubled Freedom and Mir-2 programs. They proposed that the joint station be assembled in an orbit inclined more than 50° relative to Earth's equator. The following month, the Russians provided NASA with a straw-man assembly sequence for the joint station.

On 11 May 1993, Vest advised the White House that, regardless of the design selected, the U.S. station should be built in what he called a "world orbit" inclined between 45.6° and 51.6° so that Russian - and Chinese and Japanese - rockets could easily reach it. This would, he explained, ensure that redundant means of reaching and returning from the station would exist. He added that "the shuttle will likely be grounded during the operational life of the station."

Vest presented the Advisory Committee's report to the Clinton White House on 10 June 1993. Barely two weeks later, on 23 June 1993, the U.S. station program had a near-death experience: the U.S. House of Representatives approved Fiscal Year 1994 station funding by a margin of a single vote (215-216). The close vote, which showed how politically vulnerable Space Station Freedom had become, clearly conveyed to many in NASA that station program reform was essential.

President Clinton soon approved Option A, or Alpha, the redesign option most like Space Station Freedom. Meanwhile, the proposal to merge the U.S. and Russian station programs gained momentum. Engineers and managers in Moscow, Washington, and Houston began to refer to "Ralpha," which was short for "Russian Alpha."

On 2 September 1993, Vice President Gore and Russian Prime Minister Viktor Chernomyrdin released a joint statement on U.S.-Russian space cooperation. In it, they announced a dramatic expansion of the space cooperation outlined in the June 1992 Bush-Yeltsin agreement. Russia became a full partner in the space station; minus Russian participation, it simply would not fly. 

At the same time, however, NASA would pay Russia for its involvement, which put the Russian Space Agency (and through it, NPO Energia) in the role of a NASA contractor. Though ambiguous and controversial in some quarters, the expanded Russian role reinforced the station's geopolitical justification, helping to ensure that the U.S. Congress would support it.

In November 1993, NASA and the Russian Space Agency completed an addendum to NASA's August 1993 Alpha Station Program Plan. It amounted to a blueprint for merging the Alpha and Mir-2 programs. The resulting International Space Station would be assembled in a 51.6° orbit, which meant that Soyuz spacecraft returning from it could land in their long-established recovery zones in central Asia. Soyuz landing zones in Australia were no longer needed.


Mir Hardware Heritage, NASA RP-1357, David S. F. Portree, NASA Lyndon B. Johnson Space Center, March 1995

Alpha Station Addendum to Program Implementation Plan, RSA/NASA, 1 November 1993

Australian Landing Sites Evaluation and Survey, JSC-34045, Assured Crew Return Vehicle (ACRV) Project Office, NASA Lyndon B. Johnson Space Center, 22 June 1993

Assured Crew Return Vehicle (ACRV): Technical Feasibility Study on Use of the Soyuz TM for the Assured Crew Return Vehicle Missions, JSC-34048, Assured Crew Return Vehicle (ACRV) Project Office, NASA Lyndon B. Johnson Space Center, June 1993

Letter with Attachment, C. Vest to J. Gibbons, 11 May 1993

Mir-Freedom Assembly Sequence, NPO Energia, April 1993

Letter, Y. Koptev and Y. Semenov to D. Goldin, 16 March 1993

Assured Crew Return Vehicle (ACRV): Preliminary Feasibility Analysis of Using Soyuz TM for Assured Crew Return Vehicle Missions* *Includes Evaluation of Automated Rendezvous and Docking System, JSC-34023, Assured Crew Return Vehicle Project Office, NASA Lyndon B. Johnson Space Center, April 1992 

26 March 2015

Cometary Explorer (1973)

The world's longest-running spacecraft series is the Explorer series, which began with the launch of Explorer 1, the first U.S. Earth satellite, on 31 January 1958. The U.S. Army carried out Explorer missions until NASA opened its doors on 1 October 1958. NASA Headquarters tapped Goddard Space Flight Center (GSFC) in Greenbelt, Maryland, to manage the Explorer Program. Explorer 6, launched 7 August 1959, was the first in the series to reach Earth orbit under NASA auspices.

The NASA Explorers were envisioned as low-cost science satellites. Explorer 6, a 142-pound spheroid with four paddle-like solar panels, carried lightweight, relatively simple radiation and micrometeoroid detectors. Simple did not, however, mean insignificant: Explorer 6 conducted the first detailed survey of the Van Allen radiation belts, which contain solar radiation particles trapped by Earth's magnetic field. Subsequent Explorers took many forms, but Sun-Earth interactions and the interplanetary environment remained major Explorer Program areas of interest.

Explorer 50/Interplanetary Monitoring Platform-J (IMP-J) satellite. Image credit: NASA
NASA launched Explorer 50/Interplanetary Monitoring Platform-J (IMP-J) on 26 October 1973. The following month, a 35-member GSFC team, the Cometary Explorer Study Group, completed a report for the Greenbelt center's Cometary Study Office which laid out a design for a low-cost dual-comet Explorer mission. One goal of the proposed mission, which aimed to carry out ballistic (unpowered) intercepts of Comet Grigg-Skjellerup and Comet Giacobini-Zinner in 1977 and 1979, respectively, was to gain experience ahead of a Comet Halley mission. Halley, well known to the public and of significant scientific interest, was due to return to the inner Solar System in 1985-1986.

That GSFC should seek a leading role in comet exploration is not surprising. Comets interact profoundly with the Sun and the interplanetary environment. The Greenbelt center staked its claim to comets as early as March/April 1970, when the GSFC-managed Orbiting Astronomical Observatory-2 (OAO-2) spacecraft turned its ultraviolet telescopes toward Comet Bennett, a long-period comet discovered in December 1969.

OAO-2 revealed a large "halo" of hydrogen gas surrounding the comet, which implied that it had a nucleus made up at least partly of water ice. This helped to lend support to astronomer Fred Whipple's "dirty snowball" comet model.

Cometary Explorer spacecraft. A - solid-propellant kick motor; B - hydrazine propellant tank (one of eight); C - science instrument ring; D - solar cell ring (one of three); E - hydrazine thruster (one of six); F - main antenna reflector spin motor; G - main antenna reflector. The right side of the image shows a cutaway of the spacecraft; the left side shows its exterior appearance. Image credit: David S. F. Portree/NASA
The Cometary Explorer Study Group based its 450-kilogram spacecraft on the drum-shaped Explorer 50/IMP-J design, which was very similar to Explorer 43/IMP-H (launched 13 March 1971) and Explorer 47/IMP-I (launched 23 September 1972). The spacecraft's 1.4-meter-wide, 1.8-meter-tall structure would be made up of four stacked 16-sided "rings." Of these, three rings would carry on their outer surfaces solar cells which together would generate 162 watts of electricity.

Within the fourth ring would be mounted most of Cometary Explorer's dozen science instruments. The instrument ring would have attached to it six appendages: four evenly spaced, 61-meter-long cable antennae for measuring interplanetary electric fields and a pair of instrument booms, each about three meters long.

Top view of Cometary Explorer spacecraft. A - cable antenna (one of four); B - instrument boom (one of two); C - counter-rotating high-gain radio antenna. Image credit: David S. F. Portree/NASA
For stability, Cometary Explorer would spin about its long axis at least 15 times per minute. Most of its equipment would not be affected by its spin or would be aided by it; for example, acceleration ("artificial gravity") the spin would create within the spacecraft would help to move hydrazine propellant from eight small pressurized tanks in the lower solar-cell ring to six thrusters spaced around the spacecraft.

Cometary Explorer's top-mounted high-gain antenna, on the other hand, would become useless if it spun with the spacecraft. An electric motor in the antenna base would thus turn the antenna against the spin so that it would remain stationary relative to the rest of the spacecraft. This would help to keep it fixed on Earth throughout Cometary Explorer's two-and-a-half-year mission.

Cometary Explorer would lift off from Cape Canaveral, Florida, on 4 November 1976, at the start of a 10-day launch opportunity. A Delta rocket would place the spacecraft and a solid-propellant upper stage into Earth parking orbit. At the appropriate time, the solid-propellant motor would ignite to place Cometary Explorer on course for Comet Grigg-Skjellerup. Its job complete, the spent upper stage would separate; the spacecraft's small hydrazine thrusters would then tweak its Sun-centered orbital path to ensure a successful comet intercept.

Cometary Explorer inside its 2.44-meter-diameter streamlined launch shroud. A - outline of launch shroud; B - Cometary Explorer spacecraft; C - solid-propellant upper-stage motor. Image credit: David S. F. Portree/NASA
At the time the Cometary Explorer Study Group prepared its report, Comet Grigg-Skjellerup orbited the Sun once every 5.1 years. Its elliptical orbit had a perihelion (point closest to the Sun) of 0.99 Astronomical Units (AU), and an aphelion (point farthest from the Sun) of 4.93 AU. An AU, incidentally, is equal to the mean Earth-Sun distance (149,597,871 kilometers).

Grigg-Skjellerup's orbital elements were the result of a close (0.33 AU) pass by Jupiter in early 1964; prior to that encounter, its orbital period had been 4.9 years and its perihelion distance 0.86 AU. Though Grigg-Skjellerup's orbit had been precisely determined following the Jupiter encounter, the Group advised that observatories on Earth should locate and track the comet before Cometary Explorer's launch to help to ensure a successful intercept.

Cometary Explorer would pass about 1000 kilometers from the Sun-facing side of Grigg-Skjellerup on 11 April 1977, traveling at 15.2 kilometers per second relative to its target. At time of intercept, comet and spacecraft would orbit the Sun only 0.2 AU from Earth. In addition to collecting data on the comet's interactions with the Sun and interplanetary space and the composition of its gas and dust, scientists would attempt to image Grigg-Skjellerup's nucleus.

Departure from Grigg-Skjellerup would mark the start of Cometary Explorer's "extended mission," which would last nearly two years. The spacecraft would for a time follow the initial orbit that had taken it past Grigg-Skjellerup; then, on 26 October 1977, nearly a year after its launch, it would return to Earth to perform a gravity-assist flyby at a distance of about 42,000 kilometers.

Before and after its Earth flyby, Cometary Explorer would pass through and attempt to define the limits of Earth's magnetotail, the part of its magnetosphere pushed outward by the solar wind. During the flyby, the spacecraft would ignite the solid-propellant kick motor embedded in the "thrust tube" at the center of its lower solar-cell ring. This, combined with Earth's gravity, would bend its course toward its second target, Comet Giacobini-Zinner. As Earth grew small behind it, flight controllers would use Cometary Explorer's hydrazine thrusters to refine its trajectory.

The Giacobini-Zinner intercept would take place 1.83 AU from Earth on 19 February 1979. Relative to the comet, the spacecraft would zip along at a speed of 20.8 kilometers per second.

The Cometary Explorer Study Group explained that the Grigg-Skjellerup and Giacobini-Zinner encounters would occur in "the proper order," meaning that the least perilous comet intercept would occur first. Grigg-Skjellerup, Cometary Explorer's primary target, was not a dusty comet, so the group felt that the spacecraft would not suffer crippling damage as it flew past. Giacobini-Zinner, on the other hand, was a dusty comet, so was more likely to damage or destroy Cometary Explorer.

In the foreword to its report, the Cometary Explorer Study Group warned readers that NASA had already rejected its proposed mission. The space agency had cited its rapidly shrinking budget when it turned down the GSFC plan.

The Group argued, however, that its report was still worthy of publication because it had "established the framework for investigating future ballistic intercept missions to comets." In the decade that followed the Cometary Explorer study, several of the Group's members - but most notably Robert Farquhar, Mission Definition Manager for the study - would continue to plan inexpensive, pioneering missions to comets. More often than not, these would aim to prepare NASA to explore Comet Halley in 1985-1986. Several of these proposed missions will be described in future posts.


System Definition for "Cometary Explorer": A Mission to Intercept the Comets Grigg-Skjellerup (1977) and Giacobini-Zinner (1979), NASA TM X-70561, NASA Goddard Space Flight Center, November 1973.

Encyclopedia: Satellites and Sounding Rockets of Goddard Space Flight Center - 1959-1969, NASA, no date (1970).

"NASA Facts: Explorer Satellites," E-10-62, NASA, 1962.

Explorers Program: http://explorers.gsfc.nasa.gov/index.html

More Information

A 1974 Plan for a Slow Flyby of Comet Encke

Missions to Comet d'Arrest and Asteroid Eros in the 1970s (1966)

22 March 2015

The Challenge of the Planets, Part Three: Gravity

Image credit: NASA
It is strange that Lexell's Comet is not better remembered. Discovered by ace comet-hunter Charles Messier on the night of 14 June 1770, it passed Earth just two weeks later at a distance of only 2.3 million kilometers, closer than any other comet in recorded history. On the evening of 1 July 1770, its nucleus shown as brightly as Jupiter at its brightest, and its silvery coma was five times larger than the full moon.

Lexell's Comet then drew close to the Sun - that is, it reached perihelion - and was lost in the glare. Messier saw it next in the pre-dawn sky on 4 August. Having moved away from Earth and the Sun, it had become small and faint. Messier observed the comet with difficulty before dawn on 3 October 1770, then lost sight of it.

Comets are today named for their discoverer or discoverers, but in the 18th century it was the mathematicians who computed their orbits who got all the credit. Comet Halley is, for example, named for Edmond Halley, who computed its orbit and determined that what had seemed like a series of individual comets was in fact a single comet that returned again and again. Partly this was because in Comet Halley's case no one knows who discovered it; records of the comet's apparitions extend back at least to 240 BCE, but it almost certainly was noticed in Earth's skies much earlier.

Lexell's Comet was named for Anders Johan Lexell, who determined that it completed one elliptical orbit around the Sun in 5.6 years. This was for the time a remarkably short period for a comet, raising questions as to why it had not been observed before. Lexell hypothesized that the comet had previously had a large orbit with a perihelion close to Jupiter's orbit, but then had passed Jupiter at a distance of about 3.2 million kilometers in 1767. The giant planet had, he wrote, slowed it and deposited it into its new short-period orbit.

Lexell's Comet was due to reach perihelion again in 1776, but this occurred on the far side of the Sun as viewed from Earth and so was not observed. Astronomers eagerly awaited its next perihelion in 1781 or 1782, but nothing was seen. Again, Lexell offered an explanation: in 1779, as it neared the point in its new orbit where it was farthest from the Sun - its aphelion - the comet had again intersected Jupiter. This time, it had sped up and entered an unknown but probably long-period orbit. It might even have escaped the Sun's gravitational grip entirely. In any case, Lexell's Comet has not been seen since and is officially designated "lost."

The light-show of 1 July 1770 should have ensured that no one forgot Lexell's Comet, but both its close pass by Earth and its orbit changes soon faded from memory. If they had not, then Michael Minovitch's mathematical research in 1961-1964 might not have shaken the interplanetary mission planning world the way they did.

Minovitch, in 1961 a 25-year-old graduate student at the University of California - Los Angeles (UCLA), began his research while working a summer job at the Jet Propulsion Laboratory (JPL) in Pasadena, California. He calculated that a flyby spacecraft which passed behind a planet would be seized by the planet's gravity and in effect towed as the planet orbited the Sun. This would increase the spacecraft's speed. As it departed the planet's vicinity, it would keep the speed it gained. Conversely, a flyby spacecraft that passed ahead of a planet would be slowed. Minovitch viewed this as a new form of propulsion; he called the effect the planet's motion had on the spacecraft "gravity thrust."

Minovitch determined that a spacecraft could use gravity thrust flybys to travel from world to world indefinitely without use of rocket propulsion. It could even return to the vicinity of Earth, enter a close solar orbit, or escape the Solar System entirely. In all, he calculated about 200 different planetary-flyby sequences using charts he devised and computers at JPL and UCLA.

Many engineers who learned of Minovitch's results assumed at first that they violated fundamental physical law. It seemed that the flyby spacecraft would get something for nothing. This was, of course, incorrect: when the spacecraft was slowed, the planet gained a very tiny amount of momentum; when the spacecraft was accelerated, the planet lost a very tiny amount of momentum. Nature thus balanced its books. Minovitch, for his part, was not very skilled at first at explaining his discoveries; he seems to have understood the clean elegance of numbers far better than he did the fuzzy vagaries of human beings.

Nevertheless, he had his champions. The most important was Maxwell Hunter, who met Minovitch at the American Astronautical Society's Symposium on the Exploration of Mars (6-7 June 1963) and quickly recognized the significance of his work. Before joining the professional staff of the National Aeronautics and Space Council (NASC) in January 1962, Hunter had worked at Douglas Aircraft for 18 years. He ended his career there as Chief Engineer for Space Systems. As part of the NASC, he was well placed to promote Minovitch's discoveries; the advisory body, chaired by Vice President Lyndon Baines Johnson, provided advice directly to President John F. Kennedy.

Hunter described Minovitch's "unconventional trajectories" in a report to NASC Executive Secretary Edward Welsh in September 1963. The report became the basis for a prominent article in the May 1964 issue of the important trade publication Astronautics & Aeronautics. Hunter permitted Minovitch to review a draft before the article went to publication.

In June 1964, a month after Hunter's article made the spaceflight world aware of Minovitch's labors, JPL began planning what became Mariner Venus/Mercury 1973, the first planetary mission to employ one of the trajectories Minovitch had calculated. The MVM '73 spacecraft would fly past Venus to slow down and enter a Sun-centered orbit that would take it past Mercury. The flight past Venus was labelled a "gravity-assist flyby" - Minovitch's "gravity thrust" moniker never caught on.

At nearly the same time, high-energy propulsion systems, which had been deemed essential for travel to worlds beyond Venus and Mars, rapidly began to lose support. As described in the previous post in this "Challenge of the Planets" series, the leader among these systems was electric (ion) propulsion.

In 1962, JPL engineers had prepared a preliminary design for an automated 10-ton nuclear-electric "space cruiser" and proudly presented it at a conference attended by about 500 other electric-propulsion engineers. It was received with great enthusiasm. The system was still early in its development, but the JPL engineers expected that, with sufficient funding, they might develop it for interplanetary spaceflights in the 1970s.

By late 1964, however, such brute-force high-energy systems were increasingly seen as needlessly complex and costly (at least as far as the preliminary reconnaissance of the Solar System was concerned). NASA could instead use a relatively small booster rocket to place on an interplanetary trajectory a package comprising a small chemical-propellant propulsion system for course corrections, star-trackers for precise spacecraft position and trajectory determination, a cold-gas thruster system for turning the spacecraft, science instruments, a computer, an electricity-generating isotopic system or solar arrays, and a radio. By 1962 standards, such a package hardly qualified as a spacecraft, yet it remains the basic form of our proudest interplanetary flyby and orbiter spacecraft to this day.

Electric-propulsion supporters were loathe to give up their labors. In addition to developing small station-keeping electric-propulsion systems for Earth-orbiting satellites, they sought planetary exploration niches where electric propulsion could outshine gravity-assist trajectories.

Ironically, given the adventures of Lexell's Comet, the most significant niche they identified was comet rendezvous. Before the end of the 1960s, the 1985-1986 Comet Halley apparition became a particularly important target for electric-propulsion supporters. Their efforts to explore Comet Halley using electric propulsion will be described in forthcoming posts.

In the years that followed Mariner 10 (as MVM '73 came to be known), more of Minovitch's gravity-assist trajectories were put to use. Though often mistakenly attributed to JPL's Gary Flandro, among Minovitch's trajectories was the Jupiter-Saturn-Uranus-Neptune path of Voyager 2. (Flandro's oft-cited "grand tour" paper saw print in mid-1966, nearly five years after Minovitch began his research; in it Flandro gave credit where it was due by citing two of Minovitch's JPL internal reports.)

The Voyager 2 sequence of flybys has been touted as a once-in-176-years opportunity to visit all the outer Solar System planets during a single mission; Minovitch, however, was quick to point out that this claim is spurious. Jupiter, Saturn, Uranus, and Neptune are each massive enough to bend a passing spacecraft's path and accelerate it toward any other point in the Solar System at any time.

Voyager 2, with a mass at launch of about 726 kilograms, left Earth on 20 August 1977 atop a Titan IIIE rocket. It flew within 564,000 kilometers of Jupiter's trailing side on 9 July 1979; within 102,000 kilometers of Saturn's trailing side on 25 August 1981; about 82,000 kilometers from Uranus's trailing side on 24 January 1986; and within 5000 kilometers of Neptune on 25 August 1989. In all, its primary mission spanned just over 12 years.

The intrepid spacecraft then began its Interstellar Mission, which continues to this day. At this writing, Voyager 2 is more than 19 billion kilometers from the Sun; unless humans catch up to it and reverently bring it home, it will in centuries to come depart the Solar System entirely and wander among the stars.

Minovitch calculated Venus-Earth gravity-assist trajectories; these came in handy beginning with the loss of the Space Shuttle Orbiter Challenger (28 January 1986) and subsequent cancellation of the Shuttle-launched Centaur G-prime upper stage. The accident and stage cancellation grounded the Galileo Jupiter Orbiter and Probe mission, which had been set to launch to Earth orbit in May 1986 in a Space Shuttle payload bay then boost directly to Jupiter on a Centaur-G-prime.

The Space Shuttle resumed flights in September 1988. Galileo was launched in the payload bay of the Orbiter Atlantis (18 October 1989) and boosted from Earth orbit using a solid-propellant Inertial Upper Stage that was incapable of sending it directly to Jupiter.

Instead, Galileo flew by Venus (10 February 1990), Earth (8 December 1990), and Earth again (8 December 1992) before it built up enough speed to begin the trek to Jupiter. Galileo reached Jupiter on 7 December 1995. Over the course of 35 Jupiter-centered orbits, it explored the four largest Jovian moons using gravity-assist flybys to speed up and slow down. A final gravity-assist series caused it to orbit nearly 26 million kilometers from Jupiter and then perform a pre-planned death-dive into its atmosphere on 21 September 2003.

Missions that used gravity-assist flybys include (in no particular order) Voyager 1 (which flew by Jupiter and Saturn), the Cassini Saturn Orbiter (which carried out a Venus-Venus-Earth-Jupiter sequence of gravity-assist flybys and dozens of gravity-assist flybys of Saturn's lone large moon, Titan), the MESSENGER Mercury orbiter (Earth-Venus-Venus-Mercury-Mercury-Mercury), the Rosetta comet-rendezvous spacecraft and Philae lander (Earth-Mars-Earth-Earth), the Juno Jupiter orbiter (Earth), and the New Horizons Pluto flyby spacecraft (Jupiter). Even the Dawn Vesta/Ceres mission, which mostly relied on solar-electric propulsion, used a gravity-assist Mars flyby on 4 February 2009 to gain speed and reach the Asteroid Belt between Mars and Jupiter.


"Gravity Propulsion Research at UCLA and JPL, 1962-1964," R. Dowling, W. Kossmann, M. Minovitch, and T. Ridenmoure, History of Rocketry and Astronautics, AAS History Series Volume 20, J. Hunley, editor, 1997, pp. 27-106

Comets: A Chronological History of Observation, Science, Myth, and Folklore, D. Yeomans, John Wiley & Sons, New York, 1991, pp. 157-160

The Voyager Neptune Travel Guide, C. Kohlhase, editor, NASA JPL, June 1989, pp. 103-106

"Fast Reconnaissance Missions to the Outer Solar System Utilizing Energy Derived from the Gravitational Field of Jupiter," G. Flandro, Astronautica Acta, Volume 12, Number 4, 1966, pp. 329-337

"Utilizing Large Planetary Perturbations for the Design of Deep Space, Solar Probe, and Out-of-Ecliptic Trajectories," JPL Technical Report No. 32-849, M. Minovitch, December 1965

"Future Unmanned Exploration of the Solar System," M. Hunter, Astronautics & Aeronautics, May 1964, pp. 16-26

"Determination and Characteristics of Ballistic Interplanetary Trajectories Under the Influence of Multiple Planetary Attractions," JPL Technical Report No. 32-464, M. Minovitch, October 1963

Future Unmanned Exploration of the Solar System, M. Hunter, Report to the Executive Secretary, National Aeronautics & Space Council, September 1963

More Information

The Challenge of the Planets, Part One: Ports of Call

The Challenge of the Planets, Part Two: High Energy

The Seventh Planet: A Gravity-Assist Tour of the Uranian System (2003)

21 March 2015

Where to Launch and Land the Space Shuttle? (1971-1972)

Image credit: NASA/North American Rockwell/General Dynamics
NASA's ambition in 1971 was to build a fully reusable Space Shuttle which it could operate much as an airline operates its airplanes. The typical fully reusable Shuttle design in play in the first half of 1971 included a large Booster vehicle and a smaller Orbiter, each of which would carry a crew.

The Booster's rocket motors would ignite on the launch pad, drawing liquid hydrogen/liquid oxygen propellants from integral internal tanks. At the edge of space, its propellants depleted, the Booster would release the Orbiter. It would then turn around, reenter the dense part of Earth's atmosphere, deploy air-breathing engines, and fly under power to a runway at its launch site. Because it would return to its launch site, NASA dubbed it the "Flyback" Booster. It would then taxi or be towed to a hangar for minimal refurbishment and preparation for its next launch.

The Space Shuttle Orbiter, meanwhile, would arc up and away from the Booster. After achieving a safe separation distance, it would ignite its rocket motors to place itself and an internally carried cargo into Earth orbit. Like the Booster, it would draw liquid hydrogen/liquid oxygen propellants from internal tanks.

After accomplishing its mission, it would fire its motors to slow down and reenter Earth's atmosphere, where it would deploy jet engines and fly under power to a runway landing. As with the Booster, the Orbiter would need minimal refurbishment before it flew again.

The Space Shuttle Orbiter (upper left) climbs toward low-Earth orbit as the fully reusable Booster begins its turn to fly back to its launch site. Image credit: NASA North American Rockwell/General Dynamics
Unlike an expendable launcher - for example, the Saturn V moon rocket - a fully reusable Space Shuttle would not discard spent parts downrange of its launch site as it climbed to Earth orbit. This meant that, in theory, any place that could host an airport might become a Shuttle launch and landing site.

NASA managers felt no need for a new launch and landing site; they already had two at their disposal. They planned to launch and land the Space Shuttle at Kennedy Space Center (KSC) on Florida's east coast and at Vandenberg Air Force Base (VAFB), California. Nevertheless, for a time in 1971-1972, a NASA board reviewed some 150 candidate Shuttle launch and landing sites in 40 of the 50 U.S. states. A few were NASA-selected candidates, but most were put forward by members of Congress, state and local politicians, and even private individuals.

The Space Shuttle Launch and Recovery Site Review Board, as it was known, was chaired by Floyd Thompson, a former director of NASA's Langley Research Center in Hampton, Virginia. The Board got its start on 26 April 1971, when Dale Myers, NASA Associate Administrator for Manned Space Flight, charged it with determining whether any of the candidate sites could host a single new Shuttle launch and landing site as versatile as KSC and VAFB were together. The consolidation scheme aimed to trim Shuttle Program cost by eliminating redundancy.

Image credit: David S. F. Portree (base map by Daniel Dalet/d-maps.com)

The reusable Booster flies back to its launch and landing site after releasing the Orbiter. Note the triangular flat Orbiter attachment area on top, the crew cabin above the nose, and the air-breathing jet engines just visible under the left wing. Image credit: NASA/North American Rockwell/General Dynamics
The proposed Space Shuttle launch and landing sites were a motley mix. Many were Defense Department air bases of various types (for example, Patuxent Naval Air Station, Maryland), while a few were city airports (for example, the Lincoln, Nebraska Municipal Airport). Texas proposed two sites on the Big Bend of the Rio Grande River and Wyoming offered 11 of its 23 counties. White Sands, New Mexico, was an attractive candidate; it would go on to become an alternate Space Shuttle landing site. KSC and VAFB were on the list, of course, as were NASA's Marshall Space Flight Center in Huntsville, Alabama, and Ellington Air Force Base in Houston, Texas, which had as its chief function to serve NASA's Manned Spacecraft Center.

Texas had the most candidate sites (22) of any state, while Nebraska and Wyoming tied for second place with 12 sites each. Furthest north and east were Presque Isle Air Force Base, Dow Air Force Base, and Loring Air Force Base, all in Maine. Furthest south were sites around Brownsville, Texas. VAFB was the westernmost site considered.

The reusable Booster lands on a runway less than an hour after launch from a nearby launch pad. Image credit: NASA/North America Rockwell/General Dynamics
With the exception of Alaska and Hawaii, the 10 states that contained no candidate Space Shuttle launch and landing sites lacked obvious disqualifying features (or, at least, appeared no less qualified than most of the states that contained candidate sites). Alaska and Hawaii were judged to be too far from established U.S. aerospace industry sites to be considered. The Midwestern states of Iowa, Illinois, Indiana, and Minnesota contained no candidate sites, though sites existed close to their borders in neighboring states. West Virginia alone among states east of the Mississippi River and south of the Ohio River lacked any candidate sites. The east coast states of Rhode Island, Connecticut, and New Jersey rounded out the list of no-shows.

In its efforts to cull unsuitable sites, the Thompson Board focused most of its attention on the effects of sonic booms, the sudden waves of air pressure produced when an aircraft or spacecraft exceeds the speed of sound (that is, "breaks the sound barrier"). Sonic booms, which the Board wrote had "the startling audibility and dynamic characteristics of an explosion," were a bone of contention in the U.S. in the early 1970s; concern at the time over possible injury to people on the ground and damage to structures helped to kill U.S. plans to develop a supersonic passenger liner akin to the Anglo-French Concorde.

The Thompson Board determined that the Space Shuttle would generate its most powerful sonic boom during ascent, while the Booster and Orbiter formed what was in effect a single large vehicle. The Booster's rocket plume would, for purposes of calculating sonic boom effects, make the ascending, accelerating spacecraft stack seem even bigger. The Shuttle stack's flight path characteristics - for example, the pitch-over maneuver that it would perform as it steered toward orbit - would create a roughly 10-square-mile "focal zone" for sonic boom effects about 33 nautical miles downrange of the launch site. Winds could unpredictably shift the focal area by several miles.

"Overpressure" in the focal zone would almost certainly exceed six pounds per square foot and might reach 30 pounds per square foot, which would be powerful enough to damage structures. The Board noted that plaster and windows could suffer damage at an overpressure as low as three pounds per square foot . It urged that "the severe overpressures associated with the focal zone. . .be prevented from occurring in any inhabited areas."

Image credit: NASA/North American Rockwell/General Dynamics
Based on this and other concerns, the Thompson Board trimmed the list of candidate Space Shuttle lauch and landing sites to just seven. These were: KSC; VAFB; Edwards Air Force Base, California; Las Vegas, Nevada; Matagorda Island, Texas; Michael Army Field/Dugway Proving Ground, Utah; and Mountain Home Air Force Base, Idaho.

As the Thompson Board continued its deliberations, the Space Shuttle design was undergoing rapid and profound changes. At its 22 June 1971 meeting, the Board discussed NASA Administrator James Fletcher's 16 June announcement that the space agency would spread out Shuttle costs by adopting "series development" of the Booster and Orbiter. The Orbiter would be developed first. Until the Booster could be developed, the Orbiter would be coupled with an "interim expendable booster" - possibly a modified Saturn V S-IC first stage - that would separate after depleting its propellants and fall back to Earth downrange of the launch site.

In addition, Fletcher had told reporters that Shuttle contractors would abandon work on the Orbiter's reusable internal liquid propellant tanks in favor of expendable external tanks. The expendable tanks would be less technologically challenging than their reusable internal counterparts and thus would have a lower development cost. The tanks would break up high in the atmosphere after separating from the Orbiter.

Semi-reusable Space Shuttle with reusable Orbiter, twin reusable Boosters, and expendable External Tank. This is the design the Nixon White House favored by late 1971. Image credit: NASA/McDonnell Douglas/TRW
The Thompson Board received a whirlwind series of briefings on the Shuttle design changes at KSC, the Manned Spacecraft Center, and Marshall Space Flight Center in late September 1971, after which Floyd Thompson declared a two-month recess to give the Shuttle design time to firm up. Then, on 5 January 1972, President Richard Nixon announced to members of the press gathered in San Clemente, California, that he would seek new-start funding for the Space Shuttle Program in the Fiscal Year (FY) 1973 NASA budget.

During the press conference, Fletcher displayed a model of the Shuttle configuration the Nixon White House favored. It included a reusable delta-winged Orbiter with twin unmanned reusable booster rockets and a large expendable external tank for Orbiter ascent propellants. Though this put the semi-reusable design in the spotlight, NASA remained hopeful that the reusable manned Booster would eventually become part of the Shuttle system.

On 15 March 1972, as NASA and Nixon's Office of Management and Budget jousted over the Shuttle's development cost, Fletcher announced that the reusable Booster would be abandoned entirely in favor of the White House design. A pair of reusable Solid Rocket Boosters (SRBs) would separate from the Orbiter/External Tank (ET) combination after expending their propellants and descend on parachutes for recovery. NASA's Office of Manned Space Flight subsequently determined that the SRBs could not touch down in "a controlled manner" on land; they would instead need to splash down and be recovered at sea.

Semi-reusable Shuttle drops its twin reusable Solid Rocket Boosters over water downrange of its coastal launch and landing site. The large expendable ET provides liquid oxygen/liquid hydrogen propellants to the reusable Orbiter's rocket motors. Image credit: NASA/McDonnell Douglas/TRW
The Thompson Board met just twice more. At its 27 March 1972 meeting, it discussed the implications of the 15 March booster decision and officially eliminated all non-coastal candidate launch and landing sites from consideration. At its final meeting on 6 April 1972, the Board compared the cost of building and operating a single new Space Shuttle launch and landing facility at Matagorda Island, 65 miles (105 kilometers) south of Houston, Texas, with the cost of modifying and operating both KSC and VAFB.

The Board members assumed that NASA would build five reusable Orbiters, begin Space Shuttle flights in FY 1978, and ramp up to 60 Shuttle flights per year beginning in FY 1985. To launch that many missions from Matagorda Island, the Shuttle fleet would need one Orbiter Thermal Protection System (TPS) maintenance and checkout bay, three vehicle assembly highbays for mating Orbiters with their ET/twin SRB stacks, three Mobile Launcher Platforms for transporting Shuttle Orbiter/ET/twin SRB stacks to their launch pads, three launch pads, three firing rooms, and one Orbiter landing strip.

If NASA opted for the two-site approach, three Orbiters based at KSC would conduct 40 missions per year using one Orbiter TPS bay, two vehicle assembly highbays, two Mobile Launcher Platforms, two pads, two firing rooms, and one landing strip. The two Orbiters based at VAFB would conduct 20 missions per year using one Orbiter TPS bay, one vehicle assembly highbay, two Mobile Launcher Platforms, one pad, one firing room, and one landing strip. The KSC/VAFB plan would thus need one more TPS bay, Mobile Launcher, and landing strip than the Matagorda Island plan.

The single-site plan would, however, incur greater construction costs than the dual-site plan, for the simple reason that Matagorda Island had no spaceflight infrastructure already in place. The Board estimated that Matagorda Island construction and operations would cost $5.365 billion through FY 1990, while KSC/VAFB would together cost $5.137 billion.

The single-site plan would, as had been predicted, lead to reduced Shuttle operations costs, but these savings would amount to only $87.6 million. Constructing the Matagorda Island site would, on the other hand, cost $315 million more than would modifying KSC and VAFB to support Shuttle launches. This meant that the single-site option would cost $228 million more than the two-site option.

In addition to the greater monetary cost, the single-site option would introduce substantial programmatic risk and societal costs. The Texas coastal site was partly privately owned, so construction could not begin there until NASA had negotiated purchase of the private land. Infrastructure such as roads, railways, a harbor, an airport, waste treatment plants, and a potable water system would need to be built new or expanded.

Thousands of workers would need to relocate to the Matagorda Island area over a period of less than five years, placing enormous strain on local housing, schools, and what few amenities existed in the immediate area. At the same time, established communities around KSC, already under pressure as the Apollo Program drew to an end, would suffer economically catastrophic job losses.

The Thompson Board briefed James Fletcher on its results on 10 April 1972. Four days later, Fletcher announced at a press conference at NASA Headquarters that Space Shuttles would launch from KSC beginning in 1978 and that launches from VAFB would be phased in early in the 1980s.


Space Shuttle Launch and Recovery Site Review Board, NASA, 10 April 1972

Astronautics and Aeronautics 1971, NASA SP-4016, 1972, p. 281

Astronautics and Aeronautics 1972, NASA SP-4017, 1974, pp. 139-140

Space Shuttle: The History of the National Space Transportation System, The First 100 Missions, Dennis Jenkins, Third Edition, 2008, p. 155

16 March 2015

The Challenge of the Planets, Part Two: High Energy

JPL's nuclear-electric "Space Cruiser" could in theory reach Pluto from Earth orbit in slightly more than three years. Image credit: NASA/JPL
President John F. Kennedy did not call only for a piloted lunar landing by 1970 in his 25 May 1961 "Urgent National Needs" speech before a joint session of the U.S. Congress. Among other things, he sought new money to expand Federal research into nuclear rocketry, which, he explained, might one day enable Americans to reach "the very ends of the solar system."

Today we know that Americans can reach the "ends" of the Solar System without resorting to nuclear rockets. When President Kennedy gave his speech, however, it was widely assumed that "high-energy" propulsion - which for most researchers meant nuclear rockets - would be desirable for round-trip journeys to Mars and Venus and an outright necessity for voyages beyond those next-door worlds.

In his speech, President Kennedy referred specifically to the joint NASA-Atomic Energy Commission (AEC) ROVER nuclear-thermal rocket program. As the term implies, a nuclear-thermal rocket employs a nuclear reactor to heat a propellant (typically liquid hydrogen) and expel it through a nozzle to generate thrust.

ROVER had begun under U.S. Air Force/AEC auspices in 1955. AEC and the Air Force selected the Kiwi reactor design for nuclear-thermal rocket ground testing in 1957, then the latter relinquished its role in ROVER to the newly created NASA in 1958. As President Kennedy gave his speech, U.S. aerospace companies competed for the contract to build NERVA, the first flight-capable nuclear-thermal rocket engine.

Nuclear-thermal propulsion is not the only form of nuclear-powered high-energy propulsion. Another is nuclear-electric propulsion, which can take many forms. This post examines only the form known widely as ion drive.

An ion thruster electrically charges a propellant and expels it at nearly the speed of light using an electric or magnetic field. Because doing these things requires a large amount of electricity, only a small amount of propellant can be ionized and expelled. This means in turn that an ion thruster permits only very gradual acceleration; one can, however, in theory operate an ion thruster for months or years, enabling it to push a spacecraft to high velocities.

American rocket pioneer Robert Goddard first wrote of electric rocket propulsion in his laboratory notebooks in 1906. By 1916, he conducted experiments with “electrified jets.” He described his work in some detail in a report in 1920.

Interest remained minimal, but picked up in the 1940s. The list of ion-drive experimenters and theorists reads like a "Who's Who" of early space research: L. Shepherd and A. V. Cleaver in Britain, L. Spitzer and H. Tsien in the United States, and E. Sanger in West Germany all contributed to the development of ion drive before 1955.

In 1954, Ernst Stuhlinger, a member of the German rocket team the U.S. Army brought to the United States at the end of the Second World War, began small-scale research into ion-drive spacecraft while developing missiles for the Army Ballistic Missile Agency (ABMA) at Redstone Arsenal in Huntsville, Alabama. His first design, poetically nicknamed the "cosmic butterfly," relied on banks of dish-shaped solar concentrators for electricity, but he soon switched to nuclear-electric designs. These had a reactor heating a working fluid which drove an electricity-generating turbine. The fluid then circulated through a radiator to shed waste heat before returning to the reactor to repeat the cycle.

Stuhlinger became a NASA employee in 1960 when the ABMA team at Redstone Arsenal became the nucleus for Marshall Space Flight Center (MSFC). In March 1962, barely 10 months after Kennedy's speech, the American Rocket Society hosted its second Electric Propulsion Conference in Berkeley, California. Stuhlinger was conference chairman. About 500 engineers heard 74 technical papers on a wide range of electric-propulsion topics, making it perhaps the largest professional gathering ever devoted solely to electric propulsion.

Among the papers were several on ion propulsion research at the Jet Propulsion Laboratory (JPL) in Pasadena, California. JPL had formed its electric-propulsion group in 1959 and commenced in-depth studies the following year.

One JPL study team compared different forms of "high-energy" propulsion to determine which, if any, could perform 15 robotic space missions of interest to scientists. The missions were: flybys of Venus, Mars, Mercury, Jupiter, Saturn, and Pluto; Venus, Mars, Mercury, Jupiter, and Saturn orbiters; a probe in solar orbit at about 10% of the Earth-Sun distance of 93 million miles; and "extra-ecliptic" missions to orbits tilted 15°, 30°, and 45° with respect to the plane of the ecliptic. In keeping with their robotic payloads, all were one-way missions.

The five-person JPL comparison study team found that a three-stage, seven-million-pound chemical-propellant Nova rocket capable of placing 300,000 pounds of hardware - including a hefty chemical-propellant Earth-orbit departure stage - into 300-mile-high Earth orbit with a meaningful scientific instrument payload could achieve just eight of the 15 missions: specifically, the Venus, Mars, Mercury, Jupiter, and Saturn flybys; the Venus and Mars orbiters; and the 15° extra-ecliptic mission. A chemical/nuclear-thermal hybrid comprising a Saturn S-I first stage, a 79,000-pound Kiwi-derived nuclear-thermal second stage, and a 79,000-pound Kiwi-derived nuclear-thermal stage with interplanetary payload could carry out the Nova missions plus the 30° extra-ecliptic mission.

A 1500-kilowatt ion system starting from Earth orbit could achieve all 15 missions. The JPL team told the Berkeley meeting that an unspecified chemical-propellant booster rocket would launch the 45,000-pound ion system into a 300-mile-high orbit as a unit. There the reactor and ion thrusters would activate and the slow-accelerating ion system would begin gradually to gain speed and climb toward Earth-escape and its required interplanetary trajectory.

For several of the missions to more distant targets - for example, the Saturn flyby - the ion system had enough time to accelerate so that it could reach its goal hundreds of days ahead of the Nova and chemical/nuclear-thermal hybrid systems. It could also provide its instrument payload and long-range telecommunications system with ample electricity, boosting data return. A smaller ion system (600-kilowatts, 20,000 pounds) that could be launched atop NASA's planned Saturn C-1 booster rocket could accomplish all but the extra-ecliptic 45° mission.

Missiles & Rockets magazine devoted a two-page article to the JPL comparison study. It headlined its report "Electric Tops for High-Energy Trips," which must have been gratifying for many long-time ion-drive supporters.

Many technical problems remained, however. The five JPL engineers who performed the comparison study optimistically assumed that for every kilowatt of electricity its 1500-kilowatt system applied to generating thrust, only 13 pounds of hardware - reactor, turbo-generator, radiator, structure, wiring - would be required. In 1962, a ratio of about 70 pounds of hardware per kilowatt of thrust with a maximum generating capacity of only 30 kilowatts was considered much more realistic.

They also assumed that its electricity-generating system and its ion-drive system could operate more or less indefinitely despite the presence of moving parts operating at high temperatures. The whirling turbo-generator, for example, would for some missions need to operate non-stop at a temperature of about 2000° Fahrenheit for years. A one-year operating time was considered a bold aspiration in 1962.

The five engineers did not specify the precise form their ion-drive spacecraft would take, but it would probably have resembled the design depicted at the top of this post. A trio of JPL engineers produced it during the 1960-1962 period, while the five-person JPL team conducted its comparison study.

The automated, 20,000-pound "space cruiser," as the three engineers dubbed their creation, would include a radiator surface area of roughly 2000 square feet, making it a large target for micrometeoroid strikes. In 1962, little was yet known of the quantity of micrometeoroids in interplanetary space, so no one could judge accurately the likelihood that such a radiator might be punctured, nor the mass required for effective puncture-resistant radiator tubes, redundant cooling loops, or "make-up" cooling fluid.

The five-person team only briefly mentioned the potentially profound effects of ion-drive power and propulsion systems on other spacecraft systems. The turbo-generator, for example, would impart torque to the spacecraft, creating a requirement for a spin-nulling attitude-control system - for example, a momentum wheel and chemical-propellant thrusters (the momentum wheel is visible near the center of the truss in the image above).

The turbine, flow of coolant through the radiator, and momentum wheel would, it was expected, cause vibration that could interfere with scientific instruments. In addition, ion drive systems would of necessity generate powerful magnetic and electric fields that might make difficult many desirable scientific measurements.

The space cruiser engineers sought to reduce radiation effects by placing its reactor at its front (upper right in the illustration above) and its science instruments at its rear. Unfortunately, this put the instruments among the space cruiser's ion thrusters, where intense electric and magnetic fields would occur.

The space cruiser designers looked at a thermionic power system that would use electrons from its reactor to produce electricity directly and would include neither moving parts nor high-temperature systems. They did not favor it because it was new technology. In addition, the thermionic system's nuclear reactor would need cooling fluid, a circulating pump, and a radiator, so in terms of vibration and micrometeoroid damage would offer only a little improvement over the better-understood turbo-generator design.

Close on the heels of the ARS Electric Propulsion Conference in Berkeley, NASA Headquarters opted to concentrate electric propulsion research at the NASA Lewis Research Center in Cleveland, Ohio. The move was probably intended to eliminate costly redundant research programs and keep JPL and MSFC focused on their Apollo Program-related tasks. Research did not stop entirely at NASA MSFC and JPL, however. Stuhlinger, for example, continued to produce designs for piloted ion-drive spacecraft.

Ironically, while the nearly 500 electric-propulsion engineers met near San Francisco, a young mathematician working alone near Los Angeles was busy eliminating any immediate need for ion drive or any other kind of high-energy propulsion system for planetary exploration. The third part of this three-part series of posts will examine his work and its profound impacts on planetary exploration.


"Electric Tops for High Energy Trips," Missiles & Rockets, 2 April 1962, pp. 34-35

Nuclear Electric Spacecraft for Unmanned Planetary and Interplanetary Missions, JPL Technical Report No. 32-281, D. Spencer, L. Jaffe, J. Lucas, O. Merrill, and J. Shafer, Jet Propulsion Laboratory, 25 April 1962

The Electric Space Cruiser for High-Energy Missions, JPL Technical Report No. 32-404, R. Beale, E. Speiser, and J. Womack, Jet Propulsion Laboratory, 8 June 1963

"Electric Spacecraft – Progress 1962," D. Langmuir, Astronautics, June 1962, pp. 20-25

More Information

The Challenge of the Planets, Part One: Ports of Call

The Challenge of the Planets, Part Three: Gravity