The Perfect Machine (19 page)

Everyone agreed from the start that the repeated frustrations in the efforts to cast the disk for the one-hundred-inch telescope were convincing evidence that it would be impossible to cast a larger disk of plate glass. Even if a two-hundred-inch mirror could be cast, the best
estimates were that a mass of plate glass that huge might require several decades to anneal. The only answer to the troubles with the one-hundred-inch mirror, which seemed to be caused by the relatively high coefficient of expansion of plate glass, would be a new material. But what? The earliest reflecting telescopes, like Newton’s instruments or the huge telescope Lord Rosse had made, had mirrors made of speculum, a bright metal alloy of copper and tin. Speculum couldn’t be figured to the fine optical surface that modern standards required; it was even more sensitive to changes in temperature than plate glass; after a fresh polish the metal surface reflected only 70 percent of the light that hit the mirror; and every time the mirror was polished enough material was removed to effectively refigure the shape. No recent large telescope mirror had been made from any material other than plate glass.

Early in 1928 Hale had John Anderson draw up a memorandum on the qualities needed for a large astronomical mirror. The material used, Anderson wrote, had to be shapable into the form of a surface of revolution—paraboloidal, hyperboloidal, or plane—the exact geometry to be chosen contingent on the final optical design of the telescope. Because in use the reflecting surface would be exposed to the air, the heat exchange between the air and the mirror had to set up minimal mechanical forces that would tend to distort the mirror from its optical shape. Finally the disk had to be rigid enough to enable it to be supported in all positions without any appreciable change of shape.

The only materials that would meet the first requirement were glass or other “hard transparent substances,” or certain metals or alloys, such as speculum, Magnalite, or chromium. Common metals, such as iron, nickel, aluminum, or silver were impossible to polish to an optical figure, because burnishing metal to a high polish would remove enough material to alter the optical shape. There had been some experiments that had achieved an acceptable optical surface on copper, gold, and tin, but never on pieces larger than two inches square, at exorbitant costs, and with the problem that the materials changed shape dramatically with changes in temperature.

Still the experiments went on. The Philips Lamp Works in Eindhoven, the Netherlands, had fabricated surfaces of glass fused to a chrome-iron backing. The problem with this approach was that if the coefficient of expansion of the backing were different from that of the glass, a change in temperature would introduce strains that would ultimately distort the surface.

That left glass or another transparent material. The surface of a glass mirror would ultimately be coated with a film of silver or another reflective material a few molecules thick, so the actual transparency of the material was not important, although it did have the advantage of allowing the opticians to check for internal strains in the disk. Ordinary plate glass was out; from the experience of the one-hundred-inch
telescope, it was clear that there was no way it could be made to work on a mirror of even greater mass. The search was for an alternate glass or glasslike material.

George Hale knew exactly where to look.

Elihu Thomson was one of those inventor-scientists who found a natural home at the General Electric Company. Next to Thomas Edison, he was the greatest inventor in the company’s history, with close to seven hundred patents for electric welding, transformers, centrifuge cream separators, three-phase AC windings, load regulators, magnetic switches, carbon-brush motors, electric-usage meters, electric refrigeration, and control circuits that made electric traction for trolleys possible. His company, Thomson-Houston, had merged with Edison’s company to form General Electric, and his inventions earned a fortune for GE. In return the company provided generous compensation and built him a substantial research laboratory in West Lynn, near his Swampscott, Massachusetts, home. So much of what he tried ultimately proved successful for the company that GE gave Thomson virtually free rein to pursue his own directions in research.

Among his many interests, Thomson was an avid amateur astronomer. When he was thirteen, his mother had taken him to see a celestial display of meteors and comets; he later experimented with magnifiers that he sold to his friends, and before long he was inventing optical grinding and polishing procedures. Some historians give him credit for discovering that by rotating one flat glass disk over another, with abrasive between the surfaces, a spherical concave shape is produced in the lower disk—the essential procedure for figuring telescope mirrors. His private observatory at his home in Swampscott was as large and well equipped as the facilities at many universities.

Thomson’s own telescope was a refractor he built from glass disks cast by the Paris firm of Mantois, but he understood the problems of mirrors for large telescopes. In 1899 he began experimenting with mirrors in the carriage house of his estate. He started with small concave mirrors, too crude to use in a working telescope. Thomson would focus the image of an artificial star (a point source of light) on two mirrors, one of glass and the other of fused quartz, then compare the focused images as he heated the backs of the mirrors with a flame. The heat would quickly distort a glass mirror, scattering the image. With the quartz mirror it took a considerable period of heating before the image was distorted. Quartz, Thomson concluded, could be an ideal mirror material: If a mirror could hold its figure under the heat of a torch, it would be all but immune to the effects of changes in the ambient temperature at an observatory.

George Hale, who kept his ear to the ground for new technologies that might be applied to astronomy, heard of Thomson’s experiments and sent George Ritchey out to visit Thomson in 1904, offering a modest
grant of three thousand dollars from the Carnegie Institution for additional experiments with fused quartz. At the time Hale was building the Snow solar telescope, one of the first instruments on Mount Wilson. For a solar telescope, in which the mirrors are exposed to the heat of the sun, the low coefficient of expansion of quartz promised a revolution in optical performance. Thomson experimented until the funds were exhausted and Ritchey had to return to the pressing work of grinding and polishing the mirror disk for the sixty-inch telescope. Thomson did not succeed in producing a usable mirror, and with a dozen other projects under way at the same time, some of them big moneymakers for General Electric, he was soon distracted.

Despite his failure to produce a working mirror, Thomson’s experiments were too tempting to ignore. Fused quartz had such a low coefficient of expansion that Thomson calculated that a bar one meter long, raised from room temperature to 1000°C, or near the melting point of gold, would expand by approximately one-half of one millimeter. In the range of temperatures ordinarily encountered at an observatory, the expansion and contraction would be close to negligible. A fused-quartz mirror would be all but immune to the problems that plagued the one-hundred-inch telescope.

A fused-quartz mirror would also be far more efficient to grind and polish than plate glass. Figuring an optical mirror to its final shape is an exquisitely slow process, because the heat generated in polishing affects the optical shape of the surface. In the final stages of the figuring of a mirror, a brief stroke or two with jeweler’s rouge on a tool or a fingertip can heat the mirror enough to produce distortions. The optician then has to wait hours, or even overnight, for the surface to cool enough to test the results of last effort before he can continue. Hours of waiting for minutes of polishing and testing extrapolates into years of work to put the final figure on the mirror of a large telescope. The alternative—grinding or polishing without frequent testing—would risk removing too much material. One slip, a single area polished too deep, could necessitate repolishing or even regrinding the entire disk, a task that could consume months or even years.

Thomson, who had a shrewd business sense as well as the creativity of a polymath inventor, knew the commercial market was too small, even with potential military and industrial applications, to warrant an enormous investment in quartz technology by General Electric. His early experiments seemed promising, but the research was expensive and time consuming. By the mid-1920s all work had stopped while they waited for a customer. At the annual meetings of the American Astronomical Society and the National Academy of Science, Thomson brought pieces of the fused quartz, which he would show to astronomers. Hale would corner Thomson at these meetings to find out the latest progress on fused quartz.

Hale and the Observatory Council made a show of considering a
range of materials for the mirror, but fused silica promised to be so superior to any other material that for Hale the decision of what material to use for the new telescope was already made. This would certainly be George Hale’s last telescope. For a long time it would be the primary research telescope of the world. There was no room for compromise. Fused silica was theoretically the best material for a mirror, and the obvious choice. The only problem was that no one had yet fabricated a functional telescope mirror for an instrument of any size from the material.

In March 1928, before he submitted the formal proposal for the telescope to Wickliffe Rose, George Hale met A. L. Ellis, Elihu Thomson’s lab assistant, at the Commodore Hotel in New York. Ellis was carrying a beautiful piece of fused clear quartz. Hale asked him how much it would cost to fabricate a mirror for the two-hundred-inch telescope from the material. On the spot, on a scrap of brown paper, Ellis wrote up an estimate of the cost to fabricate a series of mirrors from the eleven-inch blank the laboratory had already produced, up to a two-hundred-inch quartz mirror. Ellis’s rough figure for equipment, labor, and material—without profit or overhead—for a series of mirrors, from a twenty-two-to sixty-, one-hundred-, and a two-hundred-inch mirror that would be fabricated on Mount Wilson, came to $252,000. It seemed a modest sum in a budget of $6 million.

As soon as Rose gave the Observatory Council the go-ahead to begin work on the telescope, Hale had Henry Robinson, of the board of trustees at the California Institute, formally ask Gerard Swope, the president of General Electric, to undertake the fabrication of the mirror blanks. Swope wired back:

GENERAL ELECTRIC COMPANY WILL BE DELIGHTED TO DO THE WORK ON THE FUSED QUARTZ LENS UNDER THE PERSONAL DIRECTION OF PROFESSOR THOMSON WHO IS MUCH INTERESTED IN IT AT MANUFACTURING COST WITHOUT ANY OVERHEAD FOR COMMERCIAL OR ADMINISTRATIVE EXPENSES WHICH I ASSUME IS WHAT YOU HAD IN MIND.

Like almost everyone who had never looked carefully at a large reflecting telescope, Swope called it a
lens
instead of a mirror. No matter. If General Electric and the famed Professor Thomson could produce a two-hundred-inch-diameter disk of fused quartz, Hale and the astronomers would not only have the biggest telescope in the world, but the best.

Hale delayed a public announcement of the telescope as long as he could. He had always been wary of publicity, afraid that tentative, exploratory ideas and the meandering process of scientific research would be pummeled by a press too eager to “expose” science and scientists and to demand “results.” Astronomers, he feared, could be
tempted by the same pressures that had corrupted Sinclair Lewis’s Martin Arrow smith.

From his experiences with the Yerkes and Mount Wilson telescopes, Hale knew that once the grant for the new telescope was announced, the sheer scale and audacity of the project—more than the celebrated discoveries at Mount Wilson, or the famed debate in Washington—would open the process of building the telescope, and the observational astronomy program, to public and media scrutiny. The Observatory Council would be flooded with questions and offers. Developers with a mountain on their land would offer it as the site for the new telescope. States and counties would campaign to get the big telescope in their jurisdictions. Crackpots would come forward with their ideas of how to build a “giant eye.” Shipping companies, foundries, machine shops, construction companies, and self-promoting entrepreneurs would offer their services as they sought some tenuous connection with the prestigious project.

California also had more than its share of fundamentalist and revivalist movements, led by evangelists who were quick to brand science and technology as the work of Satan. A few had already spoken out against the telescopes on Mount Wilson. A larger telescope, designed to reach even deeper into the mysteries of the universe, would be a prime target for their sermons. A campaign by fundamentalists would be an even greater threat than the union strikes and picket lines that increasingly disrupted businesses in Southern California, because the police couldn’t be expected to show the same eagerness for scuffles with men and women of the cloth that they demonstrated against the unions.

Hale also feared that men like Harlow Shapley, at institutions fearful that their own plans would be slighted because of the funds committed to a big telescope, would be quick to join the public doubters. The debate in Washington, and the publicity that surrounded Hubble’s work at Mount Wilson, had turned the attention of many universities to the possibilities of big telescopes. The academic world in 1928 was not exempt from jealousies, rivalries and backstabbing, and many an astronomy department chairman was willing to bad-mouth the new telescope if it would help channel funds or facilities to a local project. Whatever the motives of the critics, the vagueness of the plans for a big telescope would make it difficult to answer questions and challenges in the newspapers or on the radio. There were no working drawings, few calculations, no engineering studies. Hale and his colleagues needed time to experiment, to make mistakes, to explore possibilities that a strict budget analyst would no doubt rate as cost-ineffective.