The increased use of commercial satellites to meet the burgeoning worldwide market for telecommunications has placed increased demands on the launch service industry. The capacity of this industry will not be adequate to meet the needs of all the proposals for new satellites. Even though not all the proposals will get to the marketplace, there appears to be a shortage of launch capacity. In addition, this industry has new challenges to meet. In contrast to the past when most of the commercial satellites were placed into GEO, new satellites will also be placed into LEO and MEO. These latter orbits will be used by constellations of satellites requiring the launch of numerous satellites at a time and the launch of satellites to replace failed satellites, with little lead time. In addition, there is considerable pressure on the launch industry to make a considerable decrease in the price of entrance into space as well as to increase the reliability of the launches, a point that has been watched with considerable interest by the investment banking community.

Considerable change has occurred since the last report. This is no longer an industry where the need for government and military launches exceeds that of the commercial world. Private investment in new or improved launch systems has increased and may now be even greater than that of governments. To meet the increasing demand for launches, U.S. corporations have acquired launch service capability from other nations and booked launches considerably in advance of their need, just to improve their own position in this competitive growth market. In addition, new launch vehicles are being developed to increase the capacity of the industry and to reduce costs. While much has been done to increase the lift capability of the launch vehicles, little progress has occurred in making significant decreases in the cost of launches. The added lift capability is needed to meet the demands for heavier commercial spacecraft and the need to launch more than one satellite at a time. This latter capability is especially important for the launch and subsequent insertion of numerous satellites into the multi-satellite LEO and MEO orbits. The purpose of this section is to highlight the changes in this industry that have occurred in the five years since the last report.


Ten GEO launches per year were adequate to satisfy the satellite communications business a few years ago. It is now up to thirty and appears that it will increase to almost 100 during the next decade. Launching to LEO will soon exceed launches to GEO. Constellations composed of many satellites, in some cases over one hundred, will put pressures on the industry for timely launches. The launch of commercial satellites is no longer the sole province of the United States. Europe (Arianespace), Russia, China, Japan and the Ukraine have entered this business, with Arianespace replacing the United States as the dominant launch provider. No longer do satellite manufacturers and service providers purchase a launch at a time. They purchase blocks of launches from numerous vendors to ensure the availability of launches when needed. The launch industry has then responded positively to these assured future orders by upgrading the capability of existing rockets and by proposing new launch systems. An example of bulk ordering is the 1997 Hughes purchase of 5 launch options aboard the Chinese Long March rocket. Hughes followed this up with the purchase of 10 launches from Japan's Rocket System Corporation. Space Systems Loral followed a similar path and purchased several launches on the Proton from International Launch Services. The inaugural contracts by Hughes have been important factors in establishing the viability of the Boeing (McDonnell Douglas) Delta III and Sea Launch as well as the Japan H-IIA and the upgrading of the Proton launch facilities. The July 1997 Motorola RFP to provide launch services for its new Celestri system (since cancelled), Iridium replacements, Iridium follow-ons and other satellites totaled 516 satellites, quite an impressive number.

Unfortunately, little radically new technology has been developed in the past few years that promises to lower launch costs significantly. Technology improvements, and thus cost decreases, have tended to be incremental. A fairly common question that was directed towards this panel during its visits and discussions with satellite manufacturers and service providers was, "what can be done about decreasing the cost of launches? Who can help solve this problem? We need to decrease the cost of launching satellites to ensure that satellites will continue to be competitive with terrestrial communications." In addition, the cost of insuring the satellite during the launch is high and has changed little in the past five years, being on the order of 15-20% of the cost of the satellite. Launch failures continue to occur with disconcerting frequency. This is a problem that needs to be addressed. Presently, the cost of launching a heavy satellite to GEO approaches the cost of the satellite itself. In addition to the direct costs of the rocket and its fuel, there are significant costs associated with facilities and staffing at the launch site.

The development of a major new launch vehicle is an expensive undertaking. An interesting question is, can a large launch vehicle capable of decreasing launch costs by a factor of 10 be developed without the infusion of appreciable government funding? The major launch vehicles around the world are the direct result of the infusion of government money to develop and/or improve military missiles or to develop commercial rockets and their launch facilities. Commercial companies have invested appreciable sums of their own funds for the modification and improvement of the lift capability of these vehicles and to improve the launch facilities, but not to achieve the price reduction that is needed. It is an expensive proposition to develop and prove-in a major, new launch vehicle, perhaps too expensive for a commercial service provider.

Arianespace is an interesting example of a successful effort to address the commercial launch market. This consortium now has almost half of the large commercial GEO satellite launch business. The Ariane launch vehicles and their facilities at Kourou, French Guiana are the direct result of the European Space Agency (ESA) plan to capture an appreciable part of the commercial launch market by developing new launch vehicles and creating new, modern ground launch facilities. Ariane 4 has been the workhorse of the industry for the past few years. This effort has been very successful from a market share point of view, but it has come with a high price tag. ESA has developed a new rocket, Ariane 5, to provide increased lift capability and to decrease operational cost, but not to decrease the price of a launch. It is estimated that its development cost, including the creation of the extensive test and launch operation at Kourou, was of the order of $9 billion. It is hard to imagine a purely commercial company putting such a large investment into the development of such a new vehicle. To recover the capital investment for the development of this launch vehicle (say for the launch of 160 satellites over a period of 10 years), with a cost of money of 15%, would require a charge of ~$110 million per GEO satellite. This is an optimistically low charge, since it assumes that Ariane 5 initiates service with its maximum launch capacity of eight launches per year and with two satellites per launch. Even at a (low) cost of money of 10%, it would have to charge ~$90 million per satellite to recover those development costs. Thus, if the cost to develop Ariane 5 is a good indication, it must be concluded that it is necessary to depend on government subsidies for the development of a launch vehicle and facilities for large satellites, especially a launch vehicle that will result in a significant decrease in the cost of access to space. A significant, major reduction in the cost of the launches will require a new approach to the technology of launching satellites. There is a great need for long range R&D to investigate new launch technologies. Of course, this is not to claim that the cost of launches will not decrease somewhat in the future. All of the major launch service providers have programs in place to reduce these launch costs. Also, there are some new approaches that promise to decrease these launch costs, as described in subsequent paragraphs.

To date, launch facilities have been established, maintained and operated with the oversight of government agencies. This is certainly true for the principal launch sites at Cape Canaveral Air Station in Florida, Vandenberg Air Force Base in California, the Baikonur facilities in Kazakhstan, the Tanegashima site in Japan, the Xichang facility in China, and to a lesser degree, the Guiana Space Center in French Guiana. Should the involved governments turn over some of the operation of these facilities to commercial entities as the number of commercial launches exceeds those of the governments? The SeaLaunch proposal to have launches in mid ocean and under the control of the commercial operator is very appealing to many satellite service providers, especially for the priority launches of numerous satellites to LEO. This is an issue that is certainly deserving of more attention in the next few years.

Launch Service Providers

Rockets and launch sites of major commercial operating launch service providers are listed in Table 4.2.

Table 4.2
Major Commercial Launch Services

Launch Service Provider


Launch Site


Ariane 4

Kourou, FG

Ariane 5

Kourou, FG



Cape Canaveral AS, FL

Vandenberg AFB, CA

China Great Wall

Long March


International Launch Services




Cape Canaveral AS, FL

Vandenberg AFB, CA


Baikonur, Khazakhstan

Japan, Rocket System Corp.


Tanegashima, Japan

Orbital Sciences


Wallops Island Flight Facility, VA

Vandenberg AFB, CA


Modified Zenit

Pacific Ocean platform

Yuzhnoe (Ukraine)

Zenit 2

Baikonut, Khazakhstan

This list represents an increase in commercial launch capability over the past five years. Five years ago, Arianespace, McDonnell Douglas (Delta) and General Dynamics (Atlas) launched almost all commercial communications satellites. Russia, Japan and China have been added to this list. The arrival of the Russians, Ukrainians and the Chinese into this business has been accompanied by U.S. government quotas that limit the number and prices of their launches, presumably to prevent these launch providers from offering non-economic based pricing that will inhibit U.S. and European organizations from investing their own funds in the development of new or modified launch vehicles. These quotas may be lifted as a result of the impending shortage of launch capacity, and will be discussed below. There have been institutional changes as well. Boeing acquired the Delta family of launchers from McDonnell Douglas and Lockheed Martin acquired the Atlas family from General Dynamics. Lockheed Martin then entered into a partnership with Krunichev, a Russian organization, to form International Launch Services (ILS) to provide launches by the Proton rocket. Arianespace, Boeing, ILS and Lockheed Martin are aggressively increasing their lift capability and hopefully we will eventually see decreased launch costs.

The following is a brief description of some of the present or aspiring commercial communications satellite launch service providers.

Arianespace: Evry, France

Arianespace was incorporated in 1980 as a commercial company with 53 existing corporate stockholders composed of 41 aerospace manufacturers and engineering corporations from 12 European countries. France has a 55.54% share in the corporation, followed by 18.6% from Germany, 8.1% from Italy, 4.2% from Belgium, 3.0% from the U.K., 2.6% from Switzerland, 2.5% from Spain, 2.3% from Sweden, 2.1% from the Netherlands with the remainder from Denmark, Norway and Ireland. It is closely associated with ESA, which funds the development of launch vehicles and the test and launch facilities at Kourou. The Kourou facility is impressive. It is located at five degrees north of the equator off the northeast coast of French Guiana, a few miles from Cayenne, which has a seaport and commercial airport. This location is ideal for the launch of rockets to place satellites into GEO, since minimum fuel is expended in placing the satellites into the equatorial plane. The rockets are assembled vertically in a modern class 1000, air-conditioned facility, similar to the mode used for the Space Shuttle launch complex at the Kennedy Space Center. Then the satellite is fueled up and tested in an adjacent building and installed on top of the rocket. When ready, the rocket containing the satellite(s) is rolled out on a moveable platform on rails to the launch pads. This enables Arianespace to launch ~12 satellites per year. Ariane 4 now has the capability of lifting a maximum of 4,680 kg (10,300 pounds) to geosynchronous transfer orbit (GTO). Figure 4.12 shows the progressive increase in Ariane lift capability over the years.

Fig. 4.12. Increase in lift capability to GTO for Arianespace.

Through the end of 1997, Arianespace had launched 140 satellites on 104 launches. It has had 11 launches of Ariane 4 in 1997, placing 17 non-military satellites in orbit, and have a back order for the launch of 41 satellites ($3.2 billion). Its last two launches of 1997 were accomplished 19 days apart, quite an impressive accomplishment. Arianespace is an international launch service provider, with 43% of the satellites for customers located outside Europe (16% from the United States), 39% from Europe and the remaining 18% from international organizations such as INTELSAT.

The Ariane 5 rocket program was begun in 1987 by ESA. It is designed to increase the lift capability of Arianespace, thereby enabling launch of the heaviest proposed satellites. It is a simpler rocket than Ariane 4, and thus is expected to have lower operational and construction costs. The initial version will have a lift capability of 5,900 kg (13,000 pounds) to GTO and is ideally suited to launch one very large satellite, two moderately large satellites or several small satellites at a time. Plans are in place to increase this lift capability to GTO to 6,800-7,725 kg (15,000-17,000 pounds) by the year 2000. Arianespace and ESA have asked their suppliers to propose additional modifications to Ariane 5 that will increase its lift capability to 11,000 kg (24,000 pounds) by the year 2007. The present capability of the Kourou facilities is for eight launches per year with the possible increase to ten by the year 2000, and then later to fourteen. After an initial failure in June 1996, and a partial success on October 30, 1997, a third qualification launch was planned for October of 1998. The Ariane 5 rocket is expected to be placed into commercial service in late 1999. Ariane 4 will continue to be used until approximately 2002. Ariane 5 will be phased in for the next 5 years and is expected to be the workhorse of Arianespace for many years into the future.


Boeing is now an active player in the commercial launch services business. It acquired the Delta family of launch vehicles from McDonnell Douglas and, as indicated below, is the key corporation in the SeaLaunch venture.

The Delta family of rockets has been used since 1960. Since then, there have been more than 245 Delta launches. Delta rockets can be configured as two- or three-stage vehicles, depending on the mission requirements. They have an impressive record for reliability.

The Delta II can boost 1,875 km (4,120 pounds) into GTO and continues to be Boeing's primary launch vehicle. The newer Delta III is the largest of the Delta family of expendable launch vehicles and is scheduled for its first launch in 1998. Boeing has signed contracts for eighteen Delta III launches through the year 2002, with 13 from Hughes and five from Space Systems/Loral. The Delta III can lift 8,365 kg (18,400 pounds) to LEO and 3,820 kg (8,400 pounds) to GTO, twice the lift capability of Delta II. Boeing has taken advantage of the Delta II design by incorporating the same booster engine, similar avionics systems, launch operations and liquid oxygen tanks into the Delta III. Its new features include a cryogenically powered (liquid oxygen and hydrogen) single engine upper stage, more powerful strap-on solid rocket fuel motors and a larger fairing to house bigger payloads, 13.1 feet in diameter compared to the Delta II fairing of 9.5 feet. Its first launch on August 25, 1998 ended in failure, presumably due to a malfunction in the control system.

Boeing's response to the Air Force Expendable Evolved Launch Vehicle (EELV) program, described below, is the Delta IV family of launch vehicles. It will consist of three classes of rockets: "small," to launch 4,800 pounds to GTO; "medium," to launch 10,000 pounds to GTO; and "heavy," to launch 33,000 pounds to GTO. The designs for all three of these rockets incorporate a new liquid hydrogen and liquid oxygen burning 650,000 pound thrust booster engine that is 30% more efficient than the conventional liquid oxygen/kerosene engines. The Delta IV Heavy includes the Delta III upper stage engine with modified tanks and the 16.67 foot diameter fairing that Boeing manufactures for the Titan IV.

Boeing has contracts to launch the majority of the Iridium satellites using the Delta II, the first eight Globalstar satellites for Space Systems/Loral, as well as satellites for Matra Marconi Space. Starting in February 1989, Delta launched all 24 of the global positioning systems (GPS) satellites and holds contracts for additional GPS launches through the year 2002. It serves its commercial customers from two launch pads at Cape Canaveral Air Station in Florida and one pad at Vandenberg Air Force Base.

China Great Wall Industry Corporation

This is a state owned entity engaged primarily in industry that utilizes technology developed in China. It has the exclusive authorization from the Chinese government to provide commercial launch services to foreign customers. It utilizes the Long March family of launch vehicles for launches from the Xichang Satellite Launch Center, located at 28.2║N, in a mountainous region about 40 miles northwest Xichang City. Unfortunately, numerous launch failures in the past few years have plagued this operation, resulting in the reluctance of the insurance companies to insure these launches, together with a hesitation by satellite manufacturers and service providers to commit to China Great Wall launches. The successful launch of a commercial satellite on the Long March 3-B, on August 19, 1997, followed by the three-stage Long March rocket launch of two Iridium satellites in December 1997 should allay concerns about using this vehicle for commercial launches.

The China Great Wall Company has launch contracts from several Western commercial satellite manufacturers. The Long March 3-A can launch a 5,500 pound payload to GTO. With the addition of four strap-ons to this rocket, wherein it is called the LM 3-B, it will have a capability of placing a ~29,000 pound payload to LEO and a 10,000 pound payload into GTO (Launchspace Magazine).

International Launch Services (ILS): San Diego, CA

ILS is a joint venture established in 1995 to market two of the widely used launch vehicles, the Lockheed Martin Atlas family of rockets and the Russian built Proton. ILS is owned by Lockheed Martin's Commercial Launch Services Company and the Lockheed-Khrunichev-Energia International joint venture with Khrunichev Enterprise and RSC Energia in Russia. It had a backlog of launches at year end 1997 for both the Atlas and Proton worth more than $2.5 billion.

The Atlas family of four launch configurations presently offers launches in the range of 4,000 pounds to 8,200 pounds to GTO. Proton offers two configurations, with three and four stages respectively, and is capable of launching over 10,600 pounds to 27║ inclined GTO. Proton is launched from the Baikonur Cosmodrome in the Republic of Kazakhstan.

The Atlas I commercial payload launch program was initiated in June 1987. It consists of the Atlas booster, a Centaur upper stage and the payload fairing of either 11 or 14 feet in diameter. The Atlas II program was initiated in 1988 to meet the U.S. Air Force needs for launching medium weight payloads and has been adapted to the launch of commercial satellites. The Atlas II stage and a half booster/sustainer engine has been stretched 9 feet and the Centaur by 3 feet over the Atlas I configuration. Four solid rocket strap-ons have been added to the Atlas II to form the Atlas IIAS, achieving an increased lift capability of 3,725 kg (8,200 pounds) to GTO. This has been 100% successful since its first launch in December 1993. Normally flown from Cape Canaveral for launches to GEO, a new pad has been installed at Vandenberg to accommodate the Atlas vehicle launches into polar orbits.

A new launch vehicle, the Atlas IIIA is being developed by Lockheed Martin and is scheduled for its first flight in late 1998. It will be simplified compared to the Atlas IIAS and will feature a more robust single stage booster having a new propulsion system and a single engine upper stage Centaur. This Atlas IIIA booster uses a high performance RD-180 propulsion system produced by the joint venture of Pratt & Whitney and NPO Energomash (Russia) and will burn liquid oxygen and RTP-1 propellant. It will offer lower cost launches. It will be upgraded with the addition of two solid rocket motor strap-ons, and named the Atlas IIIB. Figure 4.13 shows the Atlas launch of Inmarsat III. Table 4.3 indicates the evolution of the Atlas family.

Fig. 4.13. The launch of Inmarsat III, Flight 1 onboard Atlas-Centaur AC-122.

By year-end 1997, the Atlas had flown over 500 times. The Centaur had flown over 100 times as the Atlas upper stage and about 15 times with the Titan.

Table 4.3
Evolution of the Atlas family of Launch Vehicles
Atlas  I2,230 kg,   4900 lbs. to GTO

Atlas II 2,950 kg, 6500 "

Atlas IIA 3,075 kg, 6760 "

Atlas IIAS 3,725 kg, 8200 "

Atlas IIIA 4,055 kg, 8900 " (4Q98)

Atlas IIIB 4,500 kg, 9900 " (2Q00)

The Proton has served as the primary heavy lift vehicle for Russian unmanned space systems since the early 1960s. It has an impressive reliability record with over 200 launches. The first Western commercial mission, the launch of Astra-1F, occurred on April 9, 1996.

The Proton K can be configured as a three or four stage vehicle. The three stage version is used primarily to launch large space station type payloads into LEO.

The Proton M or Proton-M/Breeze-M is a modernized version of the Proton and is capable of placing 22,000 kg (49,000 pounds) into LEO at an inclined orbit of 51.6║. It can place 5,100 kg satellites into GTO. The Proton-M/Breeze-M is a series staged vehicle consisting of four stages, each burning nitrogen tetroxide and unsymmetrical dimethylhydrazine as propellants. The first stage consists of a central tank containing the oxidizer surrounded by six outboard fuel tanks.

The Proton facilities in Baikonur, located at 47.5║ N, have been upgraded by ILS. The rocket is assembled horizontally and the satellite is installed in a building on the complex. It is then rolled out of the building, raised to the vertical and moved to the launch pad. This enables ILS to perform many of the critical tasks in the controlled environment of a permanent building and to use the launch pads for numerous launches per year. These facilities have been used to launch several Western manufactured and owned satellites. Figure 4.14 shows preparation for an Inmarsat III launch on a Proton.

Fig. 4.14. Picture of the Proton containing the Inmarsat III, Flight 2, being readied for launch.

Japan, Rocket System Corp. (RSC)

RSC is the commercial launch service provider for Japan and works closely with the National Space Development Agency of Japan (NASDA), which develops the launch vehicles and the launch facilities at Tanegashima, a small island located at approximately 31║ N and about 100 miles south of Kyushu. After several years of mixed success at launching both commercial and scientific satellites, this provider has had successful commercial launches and is aggressively pursuing new business. It has secured commercial contracts from Hughes and Space Systems Loral for launches of satellites starting in the second half of the year 2000 onboard the new H2A rocket, which is now being developed by NASDA. This rocket is capable of launching 3,000 kg (6,600 lbs) to GTO. NASDA is building a second launch pad at Tanegashima for the H-2A. It plans to develop a capability of eight launches per year. NASDA needs to negotiate an agreement with the fisherman's union to increase the number of days per year that can be used for launching of satellites. The target launch cost is ~ $67.5 million. Unfortunately, the launch of the experimental COMETS satellite on February 27, 1998 was not a success, indicating that more work will be needed to demonstrate the reliability of the H-2.

Lockheed Martin

Lockheed Martin is a major force in the global launch business. It acquired the Atlas launch vehicle from General Dynamics and then formed a partnership with International Launch Services and the Russian entities Khrunichev Enterprises and RSC Energia, to make the Proton available for commercial launches. It also manages the heavy lift Titan 4 program and is also the prime contractor for NASA's X-33 Venture Star, described below.

Orbital Sciences Corporation: Dulles, VA

Orbital Sciences developed the ground launched Taurus rocket to provide a cost effective means of launching satellites weighing up to 1,360 kg (3,000 pounds) into LEO, or up to 365 kg (800 pounds) to GEO. It is fully road-transportable and thus has mission versatility to be launched from a previously unprepared area. Once delivered to a site, it can be launched within eight days from a simple launch pad. This rocket was first launched in March 1994 and is a four stage derivative of the Pegasus(r). It features an upgraded fairing of 92 inches and the powerful first stage Castor 120 motor built by Thiokol Corporation. Its primary use has been to launch scientific, military and commercial satellites into LEO. These rockets are being used to launch the Orbcomm satellites into LEO. Orbcomm is a many satellite constellation that has started offering international two-way data and messaging communications services for mobile customers.

The Pegasus is a novel rocket that was first flown in 1990. It is released at 40,000 feet from Orbital's L-1011 carrier aircraft and is capable of placing a 450 kg (1,000 pound) satellite into LEO. It is a three stage solid propellant rocket and can be launched from virtually anyplace in the world. Its lift capability is used primarily for small satellites placed into LEO, but it can place a 180 kg (400 pound) payload into GTO.

Yuzhnoe; Ukraine

Ukraine has developed Zenit 2 for the launch of commercial satellites, with a lift capability of 13,240 kg to LEO. Unfortunately, a much-publicized launch of 12 Globalstar satellites on September 9, 1998 was a failure, requiring Globalstar to obtain an alternate source of launches for satellites to populate the remainder of its constellation.

New Initiatives

While the above service providers are addressing the near-term-markets for the launch of commercial communications satellites, several new service providers have been organized and new projects have been instituted that promise to address the increased demand for launches of heavy satellites and/or to decrease the cost of the launches. They include:

U. S. Air Force EELV program

The U.S. Air Force has initiated an Evolved Expendable Launch Vehicle Program (EELV) to strengthen the U.S. space launch industry, encourage greater contractor investment, and decrease the Air Force's overall launch vehicle development costs. This effort has the objective to reduce space launch costs up to 50% from today's rate of approximately $12,000 per pound of payload to orbit. The Air Force selected two contractors, Boeing and Lockheed Martin, to participate in Module II, the pre-engineering, manufacturing and development phase of this effort, which is a firm fixed price 17 month contract worth $50 million for each company. The Air Force intends to introduce competition across the lifespan of the EELV program by instituting a dual source program.

Sea Launch, Inc.

Sea Launch is an international consortium led by Boeing that includes Russian, Ukrainian and Norwegian organizations. Boeing is the system's overall integrator and project manager and has 40% ownership of the company. The Russian RSC-Energia holds a 25% share, provides the launch vehicle upper stage and support equipment, and is responsible for launch vehicle integration. KB Yuzhnoye/PO Yuzhmash, a Ukrainian aerospace company, holds a 15% share, has responsibility for launch vehicle processing and operations, and will supply the first two stages of the rocket. Kvaerner Maritime of Oslo, Norway, with a 20% share, has the responsibility of converting a North Sea semi-submersible oil drilling rig into a self-propelled launch platform, constructing an assembly and command ship, and it managing marine operations. The total cost of the project is estimated to be $850 million.

This operation represents a major change in the launch of satellites. The satellites will be launched from a self-propelled platform in the Pacific Ocean, alongside a support facility ship, that can be positioned either on the equator for launches to GEO or to other places in the Pacific that are optimal for launch of satellites into inclined orbits. By being a private, international company, it should avoid launch priority conflicts with government organizations that can interfere with launch schedules. The launch vehicle is based on the two stage highly automated Ukrainian, liquid oxygen/kerosene propellant Zenit rocket, with an upper stage that has been flown over 150 times as the upper stage of the Proton. Satellites will be delivered to the company's spacecraft processing facility in Long Beach, CA, which incidentally is fairly close to the El Segundo, CA satellite manufacturing facility of Hughes and only a few hundred miles away from the Loral and Lockheed Martin manufacturing facilities in Sunnyvale, CA.

Hughes has signed a contract for launches at prices that are "competitive with conventional launch service providers" (Cromer 1997). SeaLaunch can deliver 5,000 kg (11,000 pounds) of payload to GTO. As is the case of launches from Kourou, SeaLaunch provides satellite owners with the cost and fuel benefits of equatorial launches, with the option of utilizing on-board fuel to extend the lifetime of the satellite in excess of what could be achieved by launching at high latitudes, or of launching with heavier communications payloads. Launch from the equator will occur at 152║ W, about 1,400 miles southeast of Hawaii, after a 10-day ocean trip from the base in California. Satellites intended for orbits that are inclined by 45║ or more will be probably be launched from a location off the coast of Baja California. While the initial plans are to have six launches per year, this number can be increased in the future by processing three launch vehicles at a time on the assembly and command ship.

The first launch was targeted for October 30, 1998, with the launch of PanAmSat's Galaxy 11, the first of the advanced Hughes HS-702 satellites. This date has been postponed, presumably into early 1999, due to government concerns over the export of sensitive technology. The first test launch, with a simulated payload, is scheduled for March 1999. Again, to assure access to launch capacity for its customers' satellites, Hughes has committed to 13 launches and Space Systems/Loral has committed to 5, mostly to GEO. Three of the Hughes satellites will be launched into MEO with inclined orbits of 45║ and 135║, for the ICO Global Communications system.

Project X-33

In its attempt to significantly lower the cost of the launch of satellites into space, NASA concluded several years ago that one of the most promising approaches was to develop a relatively simple and reusable launch vehicle. The agency picked the Lockheed Martin "Skunk Works" to lead a team to build a reduced size, sub-orbital rocket, designated the X-33, to test many of the concepts needed to achieve its goals. This is a ~ $1 billion program. The X-33 would take off vertically, attain orbit with a single stage of engines and fly back to earth, landing horizontally. The first flights of X-33 should start in the early part of 1999.

The next step will be to build Venture Star, a commercial vehicle that is supposed to be capable of placing large (6,800 kg or 15,000 pounds to GTO) commercial satellites into orbit at one tenth the cost of present day launches. This vehicle will be based on "lessons learned" from the X-33 program. The program is expected to cost ~ $10-20 billion to realize a vehicle that can be used for commercial launches. This single stage to orbit vehicle will take advantage of the development of new engines and lightweight and heat resistant materials. All the engines will be functioning before liftoff, and thus they can be checked out in advance of commitment to liftoff. Hopefully, this will improve the reliability of launches, since many recent failures occurred due to poor performance of upper stage engines that are ignited well into the launch. If the present funding continues and no technical roadblocks are encountered, the X-33 will be tested in a sub-orbital mission somewhere around 2004, taking off from Edwards Air Force Base, CA.

Other Reusable Launch Vehicle Projects

Several commercial organizations have initiated reusable launch vehicle projects. These are listed in Table 4.4.

Table 4.4
Reusable Launch Vehicle Projects





Kelly Space and Technology

San Bernardino, CA

Eclipse Astroliner

1,600 kg (3,500 pounds) to LEO

Kistler Aerospace

Seattle, WA


3,600 kg (7,900 pounds) to LEO

Pioneer Rocketplane

Lakewood, CO


18,000 kg (40,000 pounds) to LEO


Daimler-Benz Aerospace (Dasa) of Bremen, Germany and Khrunichev of Moscow, Russia have formed a (51%/49%) partnership to provide launch services under the name "EUROCKOT." The first and second stages of their rocket are components of the SS-19 ICBM and the re-ignitable upper stage is the BREEZE. This configuration has been launched three times, with the last one used to place a small satellite into an almost circular orbit at ~2,000 km with a 64.8║ inclination. Commercial service is expected to begin in mid 1999, with launches from Plesetsk, and will be used for the launch of satellites to LEO. This vehicle has a lift capability of about 1,000 - 1,600 kg, depending on the altitude and orbital inclination.


The Brazil Ministry of Aeronautics has constructed a launch center at Alcantar, which is located in Northern Brazil at 2║ South Latitude. This is an ideal launch site for GEO, or slightly inclined orbit satellites due to its location close to the Equator. Their rocket was developed by the Brazilian Air Force Space Activities Institute and is capable of placing satellites weighing 100 to 350 kg (220 to 770 pounds) into low earth orbit. The first launch of this rocket on November 2, 1997 failed when one of its four solid propellant motors failed to ignite.


The Indian Space Research Organization (ISRO) is developing a new launch system with a cost objective of $40 to $50 million per launch.

Launch Quotas and Government Controls

In the late 1980s, the fledgling U.S. commercial launch industry became concerned about the possibility that the non-economic based launch capability of Russia and China would undermine the U.S. companies' ability to become viable long-term launch service providers for commercial satellites. Agreements with Russia and China to limit their number of launches and their prices were proposed that were intended to help this U.S. industry compete on a worldwide basis. The first agreement was signed in September 1993, limiting Russia to the launch of eight U.S. satellites into GEO through December 2000 and required that these launches be priced within 7.5% of Western prices for similar services. No limit was placed on the launch of satellites to other than GEO orbits. In March 1995, the United States and China agreed to allow eleven Chinese launches to GEO through December 2001, excluding the four launches agreed to in 1989. There is also an escalation provision to permit an additional eleven launches if the average annual number of launches to GEO increases to 20 or more per year. In response to the China agreement the Russian agreement was subsequently modified in January 1996 to increase the number of Russian launches to 15 and increase the potential difference in launch charges to 15%, with the provision that, if the average number of annual international launches to GEO exceeds 24 or more over the time period 1996 to 1999, the Russians could launch an additional four satellites. This was followed by a U.S.-Ukraine agreement in February 1996 to allow five GEO launches from Ukraine and up to eleven more for the use of the Boeing-led SeaLaunch consortium, up to 2001. There were no provisions included in this agreement to increase the number of launches if the above mentioned average annual rate increased. These restrictions are enforced through the approval/denial of export licenses that U.S. operators need to obtain to use non-U.S. launch services.

Now that the demand for launches to GEO is increasing significantly, it appears that these restrictions may not be needed, and indeed if they continue, the business plans for some services providers will be seriously impeded by their inability to launch their satellites. The demand for launches may exceed the world's launch capacity, creating a severe problem that can be solved only by opening the business to full competition. In addition, companies such as Lockheed Martin and Boeing have established business partnerships with Russian and Ukrainian entities to launch satellites and to use their rocket technology. The launch of satellites has become a global business activity where the nationality of the companies involved has blurred.

In February 1998, fifteen corporate members of the Satellite Industry Association (SIA) sent a letter to U.S. Vice-president Al Gore requesting the U.S. government to scrap the above number, but not the pricing, restrictions immediately. "The dramatic increase in the demand for satellite launches, international partnerships between U.S. and foreign launch providers and the need for more flexible launch schedules have made these quotas unnecessary." The quota agreements are scheduled to expire in the next few years and new agreements, if any, are required to ensure the availability of launches for planned satellite systems. There is concern that these quota restrictions may be continued. In addition, there is concern over the export of sensitive technology to the launch sites, an issue that has attracted the attention of Congress.

The agreements do not address the launch of satellites into LEO. This is becoming a big market and U.S. companies have plans in place to meet these needs. There is a concern that the numerous stockpiled Russian and Ukrainian ICBMs might be made available to launch commercial satellites into LEO at a price that would significantly undermine the business plans of U.S. commercial launch providers. The United States restricts the use of its ICBMs to government missions. Clearly, this is an issue that needs to be addressed.

Launch Insurance

Insuring launches is a large international business. It typically costs 15-20% of the price of a GEO satellite to insure it against failure during the launch phase of a mission. While this may sound high, the insurance companies have not been awash in big profits from this activity, indeed, they lost money in the early 1990s These high rates are dictated by the number of launch failures, which have been occurring with disconcerting frequency. However, one would expect that the increased frequency and number of launches will improve the reproducibility of the rocket components and improve the success rate of the launches and thus decrease the cost of insurance premiums. Limits on the insurance associated with a single launch have increased also, from ~$100 million in the mid-1980s to ~$600 million now. The increase in the number of launches, the increased use of new unproven launch vehicles and the increase in the number of operating satellites in orbit, which also require insurance, will require the insurance companies to expand their business and increase the risk capital that is necessary to cover all these contingencies. This may be a major problem. It is incumbent on the launch service providers and the satellite manufacturers to improve the reliability of their products to ensure the availability of low cost insurance so that the cost of satellite communications is not burdened by high insurance premiums. In-orbit and launch failures of 1997 and 1998 have further indicated that the cost of insurance will remain high, at least for the next few years.

Orbital Debris

There is concern that debris in space left over from launches, failed launches and failed satellites will destroy or damage communications satellites. There is certainly evidence that the U.S. Space Shuttle has collided with such debris. In 1996 Endeavor had to take evasive action to avoid a collision with a military satellite. Also, there is evidence that the shuttle windshield and some of the tiles may have been damaged by collisions with lightweight debris. This debris consists of large objects, such as spent rocket casings and old satellites, all the way down in size to bolts and flecks of material from exploded, failed rockets and satellites. A one ounce particle at LEO, having a head on collision with a satellite, has the equivalent momentum transfer as a bowling ball hitting a car going at 50 miles per hour. The North American Air Defense Command at Colorado Springs does track debris, most of which is located in the vicinity of the LEO satellite orbits. There certainly is concern that LEO satellites will be damaged by collisions. In 1988, President Reagan signed a National Policy Statement, stating that "all space sectors will seek to minimize the creation of space debris." This policy was enforced as recently as December 1997 when the Federal Aviation Agency (FAA) blocked the launch of eight Orbcomm satellites by the Pegasus XL launch vehicle due to concern about the disposition of unused hydrazine fuel into LEO. Clearly, launch service providers have an obligation to minimize the creation of debris, but little can be done with the debris that is now in orbit or the debris that results from the explosive failure of rockets and their satellites, other than to wait for this debris to gradually de-orbit. Proposals to locate and de-orbit this debris do not appear to be feasible and would certainly be expensive to implement. Should the owners of LEO satellites be concerned about this problem? The answer is "yes," but after all, while there is considerable debris is space, there is still enough distance between these particles that the chance of incurring major damage probably does not justify the additional cost of building a debris resistant satellite. The International Space Station, on the other hand, will have over 200 types of debris shields to protect it against such damage.

While man-made debris may cause damage to satellites, natural solar debris is certainly a problem, is well documented, and has caused the demise of large GEO located satellites. Solar storms contain a large flux of high energy electrons, protons and heavier particles that do direct damage to components, but worse still, they cause the body of the satellite to charge up to high potentials, resulting in electrical discharges that damage components. Satellite designers go to great lengths to include sound design practices to minimize the occurrence of this problem. Nevertheless, GEO satellites continue to experience abrupt failures or gradual decreases in their usefulness due to electrostatic discharges caused by the normal flux of charged particles from the sun as well as the intense flux of charged particles due to solar storms that occur approximately every eleven years. These electrical discharges are a potential cause of failure or degradation of GEO satellites and are minimized by sound design practices. Nevertheless, these problems continue to occur with disconcerting frequency, resulting in the failure of expensive communications satellites. Radiation damage of components due to energetic electrons and protons has been the subject of considerable study. The problem is minimized by shielding the components and using special fabrication techniques for the silicon integrated circuits. There is a continuing need to study this problem, however. As the dimensions of complex integrated circuits continue to decrease, "single hit" failures due to radiation are of increasing concern. Studies of circuit "lay-out" techniques and revised I/O circuitry are needed to minimize the incidence of such failures. An example of a "single hit" problem is latch-up in VLSI CMOS circuits. The common use of satellites for mobile and data applications in LEO, MEO and inclined orbits may result in the increased incidence of satellite failures due to radiation as these satellites pass through well studied radiation belts. This potential problem needs to be watched very carefully to assure the sound design of these satellites. As mentioned above, the power output of silicon solar cells is degraded by charged solar particle damage over the lifetime of the cells in space, resulting in the need to over-design the array. The increased use of compound materials such as GaAs for solar cells should minimize this problem. The Leonid meteor storms of 1998 and 1999 are forecast to be the worst in 33 years and these particles may cause unexpected failures of satellites. While this storm of small particles will not be as intense as the Leonid meteor shower of 1966, and satellites have not experienced any known problems due to previous meteor storms, this event will be watched very carefully to determine if increased shielding is necessary to protect increasingly complex satellites from future storms.

The Future: Challenges

The international launch industry faces several challenges. The first is to satisfy the demand for launches. It appears that the launch service industry will be hard pressed to launch all the satellites that have been proposed. While all of the proposed satellite systems may not come into existence, the satellite launch industry nevertheless will be challenged to provide all the required launches. This problem is exacerbated by the need to launch on tight schedules. The LEO and MEO constellations containing numerous satellites, in particular, will require timely launches to replace failed satellites and to replace old satellites with upgraded ones that are capable of providing new communications services.

Satellite service providers and manufacturers are most anxious to see reduced launch prices to enable them to be more competitive with their terrestrial counterparts. While the major launch service providers have programs in place to reduce their costs, it is obvious that new technology is required if launch costs are to be reduced significantly in the future. Long-term R&D is needed to investigate radically different ways of placing payloads in orbit. The EELV program of the U.S. Air Force and the X-33 program of NASA are R&D programs that offer the hope of greatly reduced launch costs. These launch vehicle programs are important steps in the struggle to greatly reduce the cost of commercial access.

The geographical location of the launch pad is a factor in the determination of the life cycle costs of the satellite service. Ideally, the most favorable place for a launch to GEO is on the equator. Satellite fuel is a precious resource since it determines the in-orbit life of the satellite. Alternately, reduced onboard fuel requirements can be translated into increased weight or number of satellites that are launched at a time. The Arianespace facility in Kourou has an ideal location, 5║ from the equator. In addition, launches out into the Atlantic Ocean minimize potential catastrophes compared to launching over populated areas. Also, launching east out over the Atlantic provides a 1,000 mph boost to the rocket from the spin of the earth. One just needs to look at a globe to see that there are not many places in the world that offer these advantages and have the political stability to ensure long-term launch operations. It is not uncommon to hear rumors that the Proton, or even the Delta and the Atlas, are trying to secure agreements with the French government to launch from French Guiana. Nevertheless, new launch facilities near the equator are needed to help decrease GEO satellite life cycle costs. Obviously, Brazil is an attractive place for a launch complex. Plans have been proposed to construct launch sites in Australia, Indonesia, an island near Singapore, New Guinea and Korea.

As mentioned above, costs of the operation of satellites can be decreased if launch failures can be reduced, thereby reducing the cost of insurance and improving the certainty of the onset of service, an issue that the venture capital providers typically mention as a reason for staying away from satellite ventures.

Another issue that launch providers as well as satellite manufacturers need to address is the minimization of orbital debris.

The Cape Canaveral and Vandenberg commercial satellite preparation and launch facilities need to be modernized. At present, the rockets are assembled on the pad and then the satellite, which is fueled and tested at an off-base location, is transported to the launch site and mated to the rocket. This not only keeps the frequency of launches per pad at a minimum, but it exposes the satellite to a potentially harsh environment. At Kourou, on the other hand, the rocket is assembled and the satellite is positioned on the rocket inside a clean, air conditioned building and then rolled out to the launch pad. Similar facilities are used at Baikonour. This enables 12 or more launches per year from the same pad, in contrast to 4 or 5 per year from the pads at Cape Canaveral and Vandenberg. Clearly, these last named facilities need to be made state-of-the-art in order to compete with the more current foreign facilities.

Even with these issues, the international launch service plus insurance business is growing at a rate of about 25-35% per year, approaching an estimated $7 to $10 billion in the year 2000.

Published: December 1998; WTEC Hyper-Librarian