The word “merely” troubled Anderson. Eager to see the promised progress, he visited the West Lynn laboratory. Ellis put on a good show but admitted that there were still enough problems with the spray process that he thought they should hold off on building a larger furnace and the associated spraying equipment, and assigning the additional men to the project, until all the problems were worked out with smaller mirrors.
In the midst of the bad news, Professor Thomson proposed a new idea. Instead of fabricating massive, solid mirror disks for the telescope, he suggested that they could rib the backs of the mirrors, like giant waffles. The ribbing would reduce the weight of the mirror while still maintaining the rigidity, and the pockets in the back would provide a means for supporting the mirror in the telescope. The arrangement, he assured Anderson, would be easy with quartz, though “practically out of the question for glass.” Anderson liked the idea, and Thomson agreed that as soon as the spraying experiments were under control, the laboratory would produce a proposal for ribbed backs on the mirrors.
Thomson’s ribs were an appealing idea. The whole project was filled with appealing ideas. Back in Pasadena they were beginning to worry when they would see a mirror for the telescope. The pessimists wondered if they would ever see one.
In Europe, Hale was supposed to be resting, away from the fray and the demons. But he was obsessed with the telescope. Against the orders of his physician, and from seven thousand miles away, he insisted on regular progress reports. When Anderson and Pease scheduled a trip to the East Coast, Hale sent a list of people they should see, obscure German publications on optics they should research in the New York Public Library, and questions they should pose to experts, consultants, possible subcontractors—anyone who might be useful to the project. He wanted them to ask the Zeiss people about the counterweight schemes they had used on some of their recent telescopes. He had questions for Gano Dunn and Elmer Sperry, of gyroscope fame, about bearings and mounting designs. H. H. Timken, the famed roller-bearing builder, was to be asked whether roller bearings could support the enormous weight of a two-hundred-inch telescope, and Hale wanted them to research the new low-heat-coefficient alloy Invar for possible use in the telescope tube.
Hale had appointed committees for every aspect of the telescope project. Anderson chaired the Committees on Site, Optical Design and Mirror Discs, Bolometric Apparatus, Design of 200-inch Telescope Mounting, Laboratory Design, and Design of Instrument and Optical Shops. The members included astronomers, engineers, and other scientists from Caltech, the Mount Wilson Observatory, and outside organizations like Warner & Swasey and GE. Dozens of astronomers, engineers, opticians, and others—including Harlow Shapley—were listed as official consultants to the project. Each committee theoretically had the authority to consider all options and to make recommendations. On paper the project was reaching everywhere, soliciting and combining opinions from the widest range of sources.
In fact many of the members of the committees served in name only, and many committees existed mostly on paper. The Committee on Optical Design and Mirror Discs was responsible for the big decisions about the mirror, the heart of the telescope. Their charter from George Hale requested that they
begin immediately a theoretical and experimental study of the various possible forms of mirror discs (solid, superposed plates fused or cemented together, two plates fused to an intervening cellular structure, ribbed, etc.), and the efficiency of existing systems and new systems of supporting them in all positions they must take in the telescope.
It was an expansive charter, but the order for the mirrors had already gone out to GE a year before. Anderson briefly explored the idea of mirrors of metal coated with glass, and Sir Charles Parsons in England received a query about his plans for hollow disks, but neither investigation went beyond a few letters and a sample of the proposed disk material. The design criteria Hale sent to the Committee on Design of 200-inch Telescope Mounting were based on “the assumption that the weight of the mirror disc will be that of solid fused silica.”
The committees met from time to time, but most of the decisions on the telescope emerged in notes and memoranda from George Hale. Despite his health problems, he kept his fingers in every pie. Hale asked for answers to his queries in writing, on notebook-sized graph paper that he could insert into the binders he accumulated in his study at the solar laboratory. On Anderson’s own copy of the memo setting up committees, Hale had his secretary, Miss Gianetti, type special requests:
Dr. Anderson:
Please allow for the use of a 60-inch mirror on each side of the 200-inch tube, one for solar work (projecting far enough to permit the 200-inch tube to be completely shielded from sunlight) and one for stellar work. Both to be suitable for photographing in the ultraviolet.
G.E.H.
The requests and suggestions of the man who had almost single-handedly shepherded the three largest telescopes in the world into existence, and had gotten the unparalleled grant for this one, could not be ignored. Sometimes Hale’s questions were ahead of anyone else’s on the project. Other times he was off on a tangent, and his questions wasted valuable time. He trusted the men he knew, members of his club, heads of institutions and corporations he had met personally, or who had been recommended to him by one of the old boys. Even if they pointed to the wrong man, to people who knew nothing about the telescope, and took up valuable time that was needed elsewhere, George Hale’s suggestions couldn’t be ignored.
When he returned from the extended trip to Europe, in the late spring of 1929 Hale checked into the sanatorium in Maine for more rest, woodcutting, and forced isolation. It was late spring before he finally went back to Pasadena and his solar laboratory. He joined the regular weekly meetings of the Observatory Committee, but his participation, like his occasional solar research on the spectrohelioscope at the laboratory, was increasingly frequently postponed or interrupted by his old bugaboos: the excruciating headaches, depression, and the demons. The reports from West Lynn, bringing more bad news about progress on the mirrors, aggravated the attacks. The project seemed rudderless.
In Pasadena, Francis Pease tried to refine his sketches into working plans. So many decisions depended on the mirror that it was impossible to produce working drawings. Everyone pretended confidence—of course Thomson and Ellis would work out the problems in the fused-quartz technology and produce the mirror blank—but until they knew for sure that there would be a usable mirror, the basic engineering questions for a large telescope—How do you support a two-hundred-inch-diameter mirror so that it won’t change shape as the telescope moves? Where can observers get access to the light gathered by the great mirror? How do you point the machine with the precision that astronomers require and keep it tracking faint celestial objects as they move with the sidereal rotation of the heavens? How do you move a machine that large with no perceptible vibration?—were on hold.
The design questions were complicated by the fact that the designers still hadn’t agreed on the basic optics of the new telescope. Should it be a fast telescope, with a relatively large ratio between the diameter of the primary mirror and the distance of the primary focus from the mirror? For astronomers who were trying to photograph distant objects at the limits of detection of the telescope and photographic emulsions, the faster the telescope, the fainter the objects it would record with a short exposure. Those who had spent a whole night, or sometimes three whole nights, cramped and cold, with a bursting bladder, while they guided a telescope to keep the pinpoint of a faint star in the crosshair of an eyepiece, knew the advantages of a fast telescope.
Spectroscopists usually favored longer focal lengths and used their devices at the Cassegrain or Coudé foci. But even for them a short focal length could have advantages by permitting a more compact telescope, which would ease the engineering requirements for the mountings and the dome to enclose the instrument and allow some flexibility in the siting of the alternate foci.
But the light-gathering power of
a fast telescope comes at a price. Fast telescopes are more sensitive to light pollution. When Heber Curtis moved from the Lick Observatory to the Allegheny Observatory, in Pittsburgh, he discovered that he could only use slow, long-focal-length telescopes because of the light pollution. A fast telescope also requires a mirror of deep curvature, which is harder to grind and polish to shape. The deep curvature requires a thick mirror blank so that enough material will be left after the grinding to maintain the shape of the mirror. A thicker blank would also be more massive and take longer to adjust to changes in the ambient temperature in the observatory. For a mirror with a deposited surface, like the fused-quartz mirrors, a deep curvature would require either that the deposited layer be thick enough to accommodate the shape ground into the mirror, or that the molded quartz backing be ground into a rough spherical shape before they began spraying on the clear quartz layers. Either way it meant more complications for the already troubled work at GE.
Fast telescopes with traditional paraboloid mirrors also have small fields of sharp, coma-free focus. The f-ratio of the one-hundred-inch Hooker Telescope was f/5, which was typical for reflectors used for deep-space research. The uncorrected field of sharpness at the primary focus was less than an inch in diameter, because of an aberration introduced when the light from a parabolic mirror was focused on a plate. If they were to make the new telescope even faster, say f/73.3, the field of sharpness at the prime focus would be even smaller. A circle of film half an inch in diameter is a small area in which to concentrate the images of the heavens.
Hale asked Anderson, Adams, Pease, and Frederick Seares, Shapley’s teacher at Missouri and now an astronomer on the staff at Mount Wilson, to explore the diameter of the “good field” at the prime focus of a two-hundred-inch telescope at focal ratios of 1:3.3, 1:4, and 1:5, and to calculate the sharp field that could be used if they tried curved plates instead of a normal flat glass photographic plate. “Considering the great investment in the telescope, and the value of short periods of the best seeing,” he wrote in his memo, “It might easily pay to use such plates for several classes of work.” Graduate students were recruited to do the calculations.
A curved plate, they discovered, would not increase the useful field of a fast telescope. In France, George Ritchey and Henri Chrétien had experimented with a new telescope design that bears their name. By using a hyperboloid shape in the secondary mirror, and a deep, fast primary mirror, the Ritchey-Chrétien design provides an image at the Cassegrain focus free from the coma, or distortions outside the central field of focus, that plague telescopes based on paraboloid mirrors. But the mirrors of the Ritchey-Chrétien design are difficult to figure, it requires curved photographic plates to realize the full promise of the design, and despite the promise on paper, no working telescope had ever been built to the design.
The alternative to increase the useful “good field” of a fast telescope was to use an auxiliary corrective lens to compensate for the aberrations. No one had ever designed a lens that could correct the field of a large f/3.3 telescope. Frank E. Ross, at Yerkes, thought he could come up with a corrector lens if the project could support him and his “computer,” a woman named Margaret Johnston, who got fifty dollars for a half month of work. Ross planned to use the sixty-inch and one-hundred-inch telescopes as test beds for the corrector lens design. But his work was another experiment, with no promise of success. Every stage of design of the telescope, it seemed, called for research and engineering that had never been attempted before.
Shortly after the grant was awarded, Hale had written to his friends at Warner & Swasey, who had built the mounts for almost every large telescope since the first big Lick telescope, asking if their chief designer/engineer, E. P. Burrell, could come to Pasadena to assist in the design. Burrell had recently designed and supervised the construction of the seventy-two-inch Victoria telescope, the newest large telescope, second only to the Hooker.
Pease and Porter had already sketched different designs for a mounting for the telescope. Pease’s was a refinement of the drawings and model he had been working on for almost ten years. It was a conservative approach, a blown-up version of the fork mount of the sixty-inch telescope, relying on massive girders and huge roller bearings for rigidity and smooth motion.
Porter’s telescope design experience was with small amateur telescopes, many of radical and innovative design, like his garden telescope. He had never designed a large telescope. In his early sketches of possible mountings, he tried to combine the rigidity of the English-style mounting of the one-hundred-inch with the versatility of a fork mount. His designs evolved into a split-ring mounting, so different from any other telescope that had been built that the design was relegated to a curiosity.
Hale favored Pease’s design, which drew heavily from features and solutions that had been worked out on the sixty-inch telescope on Mount Wilson and the seventy-two-inch Victoria telescope. The Pease design, which everyone had looked at and talked about, in one form or another, for more than eight years, seemed a safe, simple solution. “It simply remains to adapt the best of these, in the light of recent progress, to the needs of the 200-inch telescope,” Hale wrote. “We now know beyond question that a tube and mirrors having a combined weight of 150 tons, involving a total weight for the moving parts of 500 tons, can be mounted equitorially and without troublesome flexure so as to afford access to the entire available sky.”
Burrell drew up a design based on Pease’s sketches and drawings. His drawings were passed on to Professors Paul Epstein and Romeo Martel of the California Institute, for calculations of the flexure in the mounting, and also to Hale’s friend Gano Dunn and his colleague Samuel R. Jones of J. G. White Engineering. The old-boy network was in full swing.
While the engineers calculated, Warner & Swasey built a model of the Pease design for exhibition at the National Academy of Sciences. The model, with a huge fork mount carrying the entire weight of the telescope on oversize roller bearings, and with a filigree box-girder construction for the tube of the telescope, relied on the same Brooklyn Bridge school of over engineering that had produced the one-hundred-inch telescope. Like Pease’s earlier model, which had been brought out for late-night discussions at Mount Wilson, the Smithsonian model was an exhibition piece, to meet public demands and queries. The actual design work on the telescope was in suspension, awaiting progress on the mirror.
While they waited Pease explored options for constructing the telescope. The mounting for the one-hundred-inch telescope had been built on the East Coast, but labor was 15 percent cheaper in California than on the East Coast, and 15 percent of $2 million—Pease’s estimate of the fabrication cost—was $300,000. Los Angeles had more sunlight, cheaper power, cheaper gas, freedom from extreme temperatures, and freedom from strikes in the nonunion shops. G. W. Sherburne, a local machinist, estimated that they could erect their own local plant for the fabrication, with a salvage value of 50 percent. At Mount Wilson’s own shops, overhead was less than 25 percent. By contrast, when they had paid the Fore River Shipyard to build the one-hundred-inch telescope mounting, the overhead had run from 35 to 110 percent, plus a 10 percent profit.
To explore another option, Pease organized a conference at the Llewellyn Iron Works in Los Angeles, a large foundry that assured him and Anderson that they could fabricate anything that could be built on the East Coast at considerable savings.
With the actual design of the telescope on hold, awaiting progress on the mirrors, Porter turned his efforts to designing an astrophysics laboratory for the California Institute campus. The building would provide laboratory space, offices, a library. Construction was scheduled to start in the spring.
Work was already under way on machine and instrument shops, which Porter had also designed, and Sherburne was persuaded to come in and take charge of the machine shop. Hubble was named chairman of the Astrophysical Observatory and Laboratory Advisory Committee. On the recommendation of the committee, the machine shop was equipped with for
ty-inch tools, large enough to do much of the fabrication for instruments, auxiliary telescopes, and some of the precision-drive equipment that would be required for the big telescope.
Porter, who had considerable experience with mirror grinding from his days of writing for amateur telescope makers in Scientific American, and more experience of cold than anyone else, argued that the original idea of grinding and polishing the mirror at the observatory site was less than ideal. The opticians at the Santa Barbara Street optical labs agreed. The laborious grinding and figuring of a large mirror was too delicate a job for a mountaintop. There wasn’t room in the Santa Barbara Street laboratories for a two-hundred-inch mirror, and after the brouhaha over the application, the relationship between the Mount Wilson Observatory and the new project was still so tentative that an optical laboratory for the Caltech campus, large enough to house the mirror-grinding project, became the next item on Porter’s drafting board. While the architectural details of the buildings were being fleshed out by a New York architectural firm, site work began on California Street in Pasadena—the first tangible evidence of the telescope project.
Visitors to the campus were told the purpose of the buildings, and some were even shown the Porter drawings of various designs that had begun to line the halls and offices of temporary buildings. But buildings and drawings were no substitute for a telescope.
12
Depression
The new year came without good news from the GE labs in West Lynn. Ellis and his staff fiddled for months before they had the furnace and auxiliary equipment ready to surface a twenty-two-inch disk, a substantial leap up from their previous efforts, and the last trial disk on their schedule before they began a five-foot auxiliary mirror that would actually be used in the telescope.
The Perfect Machine Page 18