Telstar

Telstar captured the popular imagination in a way that it is hard to believe any satellite, especially a communications satellite, could do today Perhaps it was the name; euphonious enough to make the satellite the eponymous star of its own pop song, released by the British pop group the Tornados. Certainly AT&T’s impressive publicity machine, one that rivaled even that of NASA, aided the process.

Telstar was launched at 8:35 GMT on July 10, 1962. According to AT&T, more than half the population of the U. K. watched its first transat­lantic transmission, a remarkable percentage given that far fewer people than today owned television sets. Only two Telstar satellites were ever launched, because even before the first went into orbit it was clear that Comsat, not AT&T, would be responsible for operating international com­munications satellites.

Instigating the Telstar project was one of the boldest moves ever made by a private company. Fred Kappel, chairman of AT&T, said that the company would spend $170 million—a considerable sum for the time— on an international communications system if the government would either step aside or facilitate development.

Project Telstar had five objectives: to test broadband communication, to test the reliability of electronic components under the stress of launch into space, to measure radiation levels (electronic components fail and data are lost if radiation alters the dopants in semiconductor material), to pro­vide information on tracking, and to provide a test for the ground station equipment.[19]

A. C. Dickieson was appointed to head Telstar development in the fall of 1960. In his unpublished manuscript he says that he was told to go ahead in the shortest possible time and that he had whatever power and authority he needed to do his job. The team immediately started “spend­
ing with gusto.” For a while Dickieson did not know to the nearest $5 million just how much money was pouring into the project. But by early in 1961 he had spending under control.

The project’s starting point was the early work by Pierce and Kompfner.

On January 6, 1959, while Bell and NASA were still discussing whether they would collaborate on a communication experiment with a passive satellite, John Pierce was arguing internally that Bell should assume a leading role in research in satellite communication, because it seemed likely that “satellites will provide very broad band transoceanic communi­cation more cheaply than submarine cables.” He wrote, “Active repeaters in twenty-four-hour equatorial orbits stationary above one point of the earth have many potential advantages but pose severe problems of launch­ing, orientation, and life, whose solutions lie beyond the present state of the art.” On the other hand, passive satellites were within the state of the art. Bell could, argued Pierce, undertake a program of research because more knowledge about low-noise MASERs and horn antennas would be valuable irrespective of the orbit that communications satellites would eventually occupy. There was so much activity in the field, wrote Pierce, and so little was known.

A few months later, Pierce and Kompfner were contemplating medium-altitude-active rather than twenty-four-hour satellites. There seemed to be just too many technical obstructions to developing the latter.

To John Pierce and his colleagues at Bell, the greatest drawback to twenty-four-hour satellites was the six-tenths of a second that would pass between one person speaking and the other person hearing what was said. The listener, thinking that a pause meant the speaker had finished, might then interrupt. Nothing could be done about the delay because that is how long it takes a signal to travel to and from a satellite in a twenty-four – hour orbit. Also, at that time, the equipment that suppressed echoes from the far end of the telephone line was not very effective, and Pierce was concerned that poor echo suppression coupled with the time delay would make communications via twenty-four-hour satellite intolerable. If, that is, one could have placed a satellite in a twenty-four-hour orbit in the first place. The launch vehicles and guidance and control systems needed to place a satellite in such an orbit were only then being developed.

Power was another problem. Imagine that power is distributed over the surface of a sphere and that the greater the distance of the receiver from the antenna, the greater the surface of the sphere over which the same power is distributed. Thus, if the distance increases by a factor of two, the power available at a particular point will be reduced by considerably more than twice—the square of the distance between transmitter and receiver.[20]

The thinking was that if one was to receive a usable radio signal on Earth from a satellite 22,300 miles away, one needed a high-gain, direc­tional antenna on the spacecraft that would not waste power by broadcast­ing needlessly into space (high-gain directional antennas are common today). But if the satellite was to carry a directional antenna, then the satel­lite’s orientation with respect to the earth—its attitude—would have to be controlled so that it remained constant relative to the earth. This called for a more complex, weightier design. Even at the best of times, satellite designers do not like to add weight. When no launch vehicle yet exists that can reach the desired altitude, the potential weight of a satellite is an even greater problem.

And then there was the all-important question of station keeping, which is essential for a satellite in a twenty-four-hour orbit. The satellite must maintain its position in orbit relative to the subsatellite point despite radiation pressure, lunisolar gravity, and inhomogeneities in the earth’s gravitational field. If the twenty-four-hour satellite was to be constantly accessible to many ground stations and its antennas were to stay pointing toward those ground stations, then some mechanism for station keeping was needed as well as the fuel to operate that mechanism: again, extra weight.

Finally, launch vehicles do not place satellites directly into a twenty – four-hour orbit. The satellite is injected first into an eccentric orbit with its perigee at the altitude at which it separates from the launch vehicle and its apogee at geosynchronous altitude. So a twenty-four-hour satellite would need a motor that could fire at apogee and circularize the orbit: yet more weight.

Pierce and Kompfner were aware of all these problems and the possi­ble poor quality of communication via a twenty-four-hour satellite. Fur­ther, Pierce, who from the beginning had participated in the panels plan­ning military satellites, was aware of and unimpressed by the Advanced Project Research Agency’s (later the Army’s) plans for a twenty-four-hour satellite.

So, in March 1959, when Pierce and Kompfner wrote a paper demonstrating the theoretical feasibility of active broadband satellite com­munication, it was a constellation of medium-altitude satellites that they had in mind, not twenty-four-hour satellites. Bell still intended to work first with passive satellites simply because they would be ready first, but, wrote Pierce, the research department would work towards simple active repeaters. “If we manage to make better components than others and if we make a sensible design, we might take the lead in this field with a compar­atively modest expenditure of money and effort. The tendency of ARPA has been to project elaborate and complicated schemes… our own course has been to get into actual experimental work with NASA and the Jet Propulsion Laboratory as early as possible. In this way we will encounter in an experimental way certain problems such as large antenna, low-noise receivers, tracking, modulation systems, orbit computations and the relia­bility of components which we believe will be important in all satellite communication systems.”

Even though active satellites were at the forefront of Pierce’s mind from the spring of 1959, he was publicly cautious, writing in June of that year to Hugh Dryden, the deputy administrator of NASA, “Right now, we don’t feel we are able to evaluate the merits of active satellite systems as compared with passive systems well enough to propose any concrete steps as the next desirable thing to do.”

Bell’s efforts in the field of active repeaters began to solidify when Leroy Tillotson completed a major memo on active satellite repeaters on August 24, 1959. Much of what he described was similar to what would, after the launch of Echo in August 1960, become an internal Bell project, labeled TSX, which was later renamed Project Telstar.

Central to Tillotson’s plan was the development of a six-gigahertz traveling wave tube. Pierce and Kompfner requested more information and were briefed in October and November 1959. An active-satellite plan­ning committee was formed. Kompfner and Tillotson, but not Pierce, who was more senior, were members. They continued to meet until the re­search effort on active repeaters turned into a full-fledged development project in the fall of 1960.

At the beginning of 1960, the work on Echo was proceeding well and the research staff were considering what to do next. One possibility was a larger passive satellite with enough antenna gain (on the ground), band­width, and transmitter power to relay television signals across the Atlantic. An internal ad hoc group put together an outline of such an experimental system for NASA. As it became apparent what kind of technical pirouettes would be needed for the successful transmission of even the lowest quality TV signal, those working on the proposal became convinced that active satellites were needed.

Thus consensus emerged at Bell that the main effort should be on active satellites—along the lines suggested by Tillotson. A memo from Pierce to Jaffe says that Bell would be putting a major effort into active satellites during the next couple of years. Another letter from Kompfner to Jaffe reviews Bells research on a long-life traveling wave tube and on the effects of radiation damage to solar cells and microwave circuitry. This work continued at a modest level in Bells research department until after the successful launch of Echo on August 12, 1960.

News of Bell’s interest in medium-altitude satellites did not filter out widely until the spring of 1960. When it did, some speculated that the pas­sive scheme had been a smoke screen to cover Bell’s real intentions. If it was, it had not been embarked upon as a smoke screen, though perhaps it was allowed to become one.

While Bell’s scientists and engineers pursued theoretical calculations, AT&T’s management in New York had been working on policy issues. In July 1960, AT&T argued before the Federal Communications Commis­sion that frequencies should be reserved for satellite communication because the company was convinced that satellites would be more eco­nomical than submarine cables for transoceanic communication. A filing on July 8 disclosed AT&T’s plans for a global communication satellite sys­tem costing $170 million. This plan called for fifty satellites without atti­tude control in a three-thousand-mile polar orbit. During the next twelve months the number of satellites, their altitude, and their design would change, but the idea of medium-altitude orbits remained.

In the midst of its political and technical preparations for an active satellite system, on August 9, 1960, the Department of Defense finally released NASA from its agreement to develop only passive satellites.

AT&T now focussed on persuading NASA to select its ideas for an agency led project to develop active communication satellites. On August 11, Pierce and colleagues were briefing senior NASA staff at headquarters in Washington on Bell’s medium-altitude active repeater work. They told NASA of AT&T’s discussions with the communication administrations of Britain, France, and Germany, and of those countries’ interest in joining AT&T in satellite communication experiments. Bell’s idea was that the Bell System would pay all costs, except those of general interest to the space community, such as investigating radiation effects. There was some discussion, too, about the technical difficulties of providing two-way chan­nels (with Echo, voice went via satellite one way and came back over a ter­restrial link).

The day after that meeting, Echo was launched. It was the brightest object in the night sky. In Ceylon, as Sri Lanka was then called, Arthur C. Clarke looked upwards and followed its passage with wonder. In the United States, AT&T’s highly efficient publicity machine ensured that when the public gazed upwards, it was AT&T’s name rather than NASA’s that sprang to mind.

AT&T’s publicity capsized discussions with NASA about launching a satellite based on AT&T’s ideas. A little over two months after Echo’s launch, in a meeting about Bell’s plans for transoceanic communication via active satellite, T. Keith Glennan told senior staff from Bell and AT&T that the company’s methods of publicizing its plans, including making mislead­ing statements, had created difficulties for NASA. Glennan’s remarks about publicity were only a small portion of the criticisms he made. He said that AT&T was not taking account of the “facts of life” and acknowledging through its planning the limited availability of launch vehicles, the prob­lems of scheduling launches, and the aims of NASA’s research and devel­opment program.

By the time of this meeting, on October 27, 1960, NASA’s plans were known publicly to include medium-altitude active repeaters, and AT&T was now one of several companies waiting for the agency’s formal announcement of a competition. Whichever company won, NASA also intended to further the development of communications satellites through “cost reimbursable launch support for private industry.”

So, despite Glennan’s stern admonishments in October 1960, the auguries were not entirely unfavorable for AT&T. NASA, under the lead­ership of T. Keith Glennan, favored the involvement of private industry in the development of communication satellites. AT&T had a vibrant research effort in the supporting technology for medium-altitude active repeaters, and Echo had been a spectacular success.

The days of the Eisenhower administration, however, were numbered. Transition to the Kennedy administration would soon be underway. Though both administrations were concerned about the antitrust implications of policies that favored AT&T, some in the Kennedy administration, notably James Webb, were additionally concerned about the strategic implications of communications satellites and thought that government should retain con­trol over their development. The difference between the two administrations was apparent in a small but telling difference in the budgets that each sub­mitted to Congress for FY62. Eisenhower called for private industry to con­tribute $10 million toward NASA’s communication satellite program. Kennedy’s budget allocated that $10 million from the public coffers.

This was the first faint stirring of the policy upheavals that would exclude AT&T from providing international satellite communications, a role that the company saw as its own as a matter of public trust. In the end, AT&T would have to be content with being one of the common carriers owning Comsat stock.

In January 1961, when John Kennedy was inaugurated, and for sev­eral more months, AT&T moved confidently forward with its plans. Bell Labs was working on ways to keep the satellites’ weight down by develop­ing sensitive ground-based antennas that would allow the satellites to transmit at as low a power as possible. That month NASA issued a request for proposals for a medium-altitude satellite to be known as Relay. During the Relay competition, AT&T and NASA suspended discussions about the possibility of the agency launching the telephone company’s satellites. Bell, along with six other competitors, prepared a proposal. The others included RCA, which eventually won, and an outsider with no experience manu­facturing satellites—the Hughes Aircraft Company.

Bell, which did not think much of NASA’s technical specifications, submitted three versions of its proposals. One matched what NASA wanted, including frequencies and a radiation experiment as specified by the agency. The second retained NASA’s radiation experiment but worked at the frequencies that Bell (AT&T was discussing frequency allocation with its overseas partners) thought would eventually be selected for an operational system (AT&T and Bell were correct). The third proposal was Bell’s own design.

On May 18, NASA announced it was awarding the contract to RCA. At the technical debriefing, NASA told Bell that it was the best of the “amateurs” to submit a proposal. Bell’s weaknesses in its bid, according to NASA, were many. The agency awarded the lab poor marks for solar cells made from n-on-p semiconductors rather than p-on-n. Yet Bell’s sci­entists knew that the Evans Signal Laboratory had found that n-on-p semiconductors were more resistant to proton and electron bombardment than p-on-n semiconductors. Bell had fabricated n-on-p solar cells in December 1960 and confirmed the finding (these are what are used now).

The Bell proposal got a particularly low score for the low power of its transmitters, which would call for a low-noise ground antenna, and par­ticularly for further improvement of the MASER that was used for Echo. The agency considered that Bell’s tough specifications for a low-noise ground antenna would make the ground stations that NASA and others might build marginal. Its own course of action, said the agency, did not press the state of the art quite as hard. NASA judged, however, that Bell’s traveling-wave-tube design was excellent and that its radiation experiment was good. The agency asked Bell to design the radiation experiment for Relay. Bell’s view of this critique was that only one criticism—about a VHF antenna—was valid.

Preparing the stack of printed material for its proposal, wrote Dick – ieson, cost several hundred thousand dollars and “chewed up the time of a lot of key people who were sorely needed in designing the company’s own satellite and ground station. But we really had no choice but to bid; with­out launch support, all of our work would be wasted, and NASA con­trolled the launching.”

To the engineers at Bell, NASA’s announcement that RCA had won the contract meant that they could revert to their own ideas for what would be known as Project Telstar. The announcement also left them with one rather major difficulty—the matter of a launch vehicle. Pierce and Kompfner joked with a visiting Soviet scientist that perhaps his govern­ment would launch the satellite.

A more practical discussion was going on between James Webb and Fred Kappel. Webb had called Kappel on May 18 to notify him that RCA had won the contract for Relay and to say that NASA was prepared to launch AT&T’s satellite. Kappel responded that it was important for AT&T to go ahead with its satellite plans. Hard negotiations, led by Webb and Dryden, followed. NASA insisted on being reimbursed for launching the two satellites and that AT&T should assign any patents resulting from the development to NASA. AT&T agreed, and Telstar, which Wilbur L. Pritchard later said was superbly engineered, was underway.

An important aspect of the Telstar project was the two ground sta­tions that Bell Laboratories built. The sophistication and sensitivity of these stations enabled Bell to put less power on the satellite. One station was in Andover, Maine, and the other in France. Britain used its own exist­ing antenna but was unable to receive signals on the first night because of an unfortunate misunderstanding about the polarization of the signal.

The ground station in Maine was 170 feet long and three stories high, and rotated to follow the satellite from horizon to horizon. It was intended to be part of the eventual operational system and had to work in any weather, so Bell contracted for a radome (a radio transparent, domelike shell) to protect the antenna. The wall to which the radome was to be fixed was a massive structure. Someone quipped that in a thousand years, scientists would debate why the entrance door was in precisely that place. Excavation for the antenna foundations began in May 1961, and in January 1962, the antenna was complete. A temporary radome was erected until the special air-supported fabric of the final structure could be delivered. It sagged under the New England snow. Efforts to dislodge the snow failed until someone took a shotgun from the trunk of his car and shot holes in the temporary cover.

Testing of Telstar began in November 1961 and took 2,300 hours. The satellite was due at Cape Canaveral in May 1962 but was delayed for two weeks until a loose wire could be traced. The launch was set for July 10. In late May Bell heard about Project Starfish, a high-altitude atomic – bomb test. They were worried that if Telstar was in orbit, radiation from the explosion would seriously damage its electronics. They relaxed when they heard that the explosion was set for the day before their launch, believing that by the time Telstar reached orbit, the worst would be over. They later learned, as did APL, which had its TRAAC satellite aloft, that fallout persisted and precipitated along magnetic field lines. Telstar began to falter on November 18, 1962, and failed the following February.

Nonetheless, Telstar allowed an examination of the signal after pas­sage at various angles through the ionosphere, the earths magnetic field, and the atmosphere. Different methods of frequency modulation were tested, and the impact on available bandwidth assessed. These and other results were published in the open literature. Harold Rosen, of the Hughes

Aircraft Company, said, “Telstar showed that there were no propagation anomalies, that it was easy to calculate what the propagation would be like. It was a confidence builder.”

Telstar attained orbit despite antagonism and suspicion between AT&T and Bell on the one hand and NASA on the other. Exactly how these tensions, which also existed under T. Keith Glennan, played out in the decision to select RCA for Project Relay is not easy to discern. Robert Seaman, who was NASA’s associate administrator at the time, said in his exit interview that the message came through loud and clear from the Kennedy administration that the AT&T design was not the one to pick. In 1966, Webb said that the RCA proposal was clearly the best proposal for the research requirements of NASA, even if it was “… not necessarily the best as the first step towards an operational satellite system as desired by AT&T.” Webb’s phrasing is telling.

The disagreements between NASA and AT&T covered everything from choice of frequency to operation of the ground stations and negotia­tions with foreign telecommunications companies. AT&T was particularly jealous of the relationships it had built over the years with common carri­ers in other countries. The transatlantic submarine cable TAT-t, for exam­ple, was a joint venture with the British and Canadians. A British cable ship had laid the cable.

It is also clear from Dickieson’s unpublished manuscript in the AT&T archives that Bell’s technical people did not respect the technical ability of some of those with whom they dealt at NASA and that they thought much of the required paperwork pointless. Dickieson wrote, “the NASA people assigned to receive this paper were interested in the shadow, and not the substance, so we were able to keep them happy without [hav­ing them] interfere with our work.”

The attitude of the Bell engineers comes through best in this follow­ing anecdote related by Dickieson. The National Physics Laboratory in the U. K. wanted to use Telstar in an experiment with the U. S. Naval Observa­tory to synchronize clocks in the two countries (this was before atomic clocks). Dickieson set things up. The experiment was performed and the results published. “At a subsequent meeting, Leonard Jaffe brought up the subject, and made it clear that the approved method was: first, discussions between the state departments of the two countries; second, reference to the technical organizations of the two governments, and finally down to scheduling by the Ground Station Committee.” Says Dickieson, “I did not argue the matter, because I thought that if another useful experiment appeared, we would do it first and argue later.”

Despite all this, the two organizations successfully launched the two Telstar satellites.

From February 1962, when President Kennedy submitted his Com­sat bill to Congress, it was clear to Bell’s engineers that the lab and AT&T were out of international satellite communication. They still had both satellites to launch, and the team worked on, fueled by the need to prove that private enterprise could operate in the field. To the public, the battles behind Telstar were unimportant. To them Telstar was the satellite that first broadcast live transatlantic television and promised a new era of interna­tional communication. Among those watching were a group of engineers at the Hughes Aircraft Company, in Culver City, California. After rejec­tions and ridicule, they had won a contract for their ideas for a twenty- four-hour satellite in August 1961—nearly a year before Telstar went into orbit. The engineers watching Telstars broadcast were envious. They had dispatched a telegram of congratulations to Bell but were eager to see their own satellite in orbit. One of them, Harold Rosen, said “It was interesting in two respects, one was the beautiful picture coming from overseas. And two, it didn’t last very long.” They knew their satellite would be altogether different.