The Perfect Machine

by Ronald Florence

  This time the annealing schedule was not an accelerated test but the real thing. After an initial stabilizing period at 500°C, McCauley began the actual cool down on January 21, 1935. The optimum annealing schedule required that the disk cool by exactly 0.8°C each day for ten months. A crew of attendants was assigned to work eight-hour shifts, twenty-four hours a day, seven days a week, manning the controllers for the nichrome heaters in the kiln. Every three hours one of the ten controllers had to be reduced by 1°C, an operation that took approximately ten seconds. The attendants had nothing else to do. McCauley lectured the young men who were hired that if they missed even one adjustment of the controllers—each attendant faced a maximum of three ten-second operations in an eight-hour shift—they would ruin the entire disk. Knowing they might guess that a single missed adjustment would not imperil a twenty-ton mass of glass for the entire annealing period, McCauley drove or walked down to the 3A tank every morning to check the dial settings and thermocoupler readings on the control panel. On the weekends he would take his young children with him, stopping off before or after church on Sundays. He never missed a single day. He often would leave notes on the dials, reminders to the attendants of the importance of the precious object in their care.

  The melting tank and ladling crew stayed busy pouring auxiliary mirrors for the two-hundred-inch telescope, including some tricky oval mirrors for the Coudé focus and a special disk for the Bell Telephone Labs. By midsummer the telescope mirrors had all been poured, and the fires in the 3A tank were extinguished. McCauley was pleased that they had received an order from a “blue chip” company like Bell Telephone in addition to the telescope orders, and that the procedure had become so routine that they could confidently pour mirror after mirror. But no matter how routine the operation became, he couldn’t stop worrying that something would still go wrong, that some accident, some unpredicted disaster, would befall the twenty-ton mass of glass sealed away in the dome-shaped annealing oven. Checking the controllers each morning was his way of fighting off the anxiety. He was a wine master, making daily visits to his cave, reassuring himself that his slowly maturing, priceless treasure would emerge intact and perfect enough for the telescope.

  Left alone the astronomers and engineers might have gone on refining their designs for the telescope forever. By late summer of 1935, with work on the site underway, George Hale concluded that it was time to start building a telescope. From the beginning, he had assumed that Warner & Swasey, with its vast experience, would build the telescope. The company had kept in close touch with each step of the design, and had worked on the models that were exhibited at the National Academy of Sciences. But as the initial sketches began turning into engineering drawings, it became clear that Warner & Swasey did not have the facilities to build the huge telescope mounting. Indeed, few companies did.

  Hale had turned to a shipbuilding yard for the mounting of the one-hundred-inch telescope. He tried again for the two-hundred-inch, asking Homer Ferguson, the president of Newport News Shipbuilding and Dry Dock Company and a member of the Board of Trustees of the Carnegie Institution, if he would be interested in the job. Newport News had pioneered welded construction for ships. The project Hale had in mind would be the largest precision-machined welded structure ever built, but Ferguson declined, citing the backload of the shipyards and observing that the welded steel structures Hale wanted were so large that if they were built in the West, the only place with facilities big enough to anneal them was at Boulder Dam. Ferguson couldn’t recommend a company that could undertake the work.

  Hale was too ill to manage the construction of the telescope himself. He also knew that there was no one on the staff with the experience and time to get it built. John Anderson would have his hands full supervising the figuring and polishing in the optics lab, when the mirror disks arrived. Francis Pease and Russell Porter had no experience supervising large-scale construction. And Walter Adams had a full-time job running Mount Wilson. The engineers on the project, like Mark Serrurier, were too young and inexperienced to undertake the overall management. Hale turned to Max Mason, at the Rockefeller Foundation, for advice.

  Without hesitation Mason said he knew the best man in the world for the job, although he doubted that they could recruit him. Mason had met Clyde S. McDowell at the University of Wisconsin, where McDowell, already a Navy officer, had gotten his doctor of science degree. During World War I, McDowell had been in command of the Naval Experiment Station in New London and had recruited Mason to do research on acoustic devices for antisubmarine warfare. After the war McDowell had gone on to supervise the navy yard in Samoa and then become yard manager at Pearl Harbor, working his way up to the rank of captain. During the postwar disarmament, he had been a strong advocate for both applied and theoretical research, arguing that the war had been “won by research.”

  Hale trusted Max Mason’s advice. In October he and Mason met with Captain McDowell, who was then stationed at the Navy Office of Inspector of Machinery at the New York Shipbuilding Corporation in Camden, New Jersey. McDowell was quick to offer his suggestions for the telescope project. From what Hale and Mason told him, he thought the design was not far enough along to put out to contract. Navy practice, he explained, was not to enter into contract for any portion of construction until the general features were agreed on. Hale, impressed with McDowell’s quick grasp of the problems of building the machine, and his no-nonsense, take-charge manner, asked McDowell if he would be willing to serve as coordinator and planner of the construction if he could be temporarily released from his navy duties. McDowell said he would be interested if the details could be worked out.

  In less than a week McDowell wrote that he was willing to retire from the navy, passing up the chances for promotion to admiral, to take charge of the building of the telescope. He asked for an “honorarium” of $12,000 per year, substantially higher than any salary paid at Caltech. Hale balked, offering $8,000 and traveling expenses. McDowell held firm, and Mason finally had to persuade Hale to consider comparable commercial salaries and the fact that it wasn’t a salaried position but only a short-term assignment for McDowell.

  Within a month McDowell moved into temporary quarters at the Atheneum, the Caltech faculty club. He was introduced around as “Sandy,” but there was no mistaking his background. He immediately began sending out flawlessly typed memorandums, with multiple carbon copies, to everyone working on the project. Oral authorizations, he noted in an early memo, might prove expeditious, but “in all cases such action must be confirmed as soon as possible in writing.” The memorandums were followed by organizational charts. On McDowell’s charts only Hale and Anderson were above him. Everyone else reported to McDowell.

  Sandy McDowell was ambitious. He had had a promising career in the navy, with a good chance for promotion to flag rank, but the Bureau of Ships, in a period when isolationism and minuscule defense budgets reduced American shipbuilding to a standstill, wasn’t enough. In New London during the war he had directed a staff of the brightest physicists and mathematicians he could locate, working with virtually unlimited wartime budgets. He had done a good job in New London, forcing reluctant scientists to meet deadlines, making decisions that others feared, and ultimately getting the job done by producing workable submarine detection devices before the end of the war eliminated the need.

  Building the two-hundred-inch telescope, the biggest scientific project in the world, was a challenge McDowell couldn’t resist. He hit Pasadena like a whirlwind, firing off a barrage of memorandums, asking each person on the project for timetables, status reports, and flowcharts of project progress. He wanted clear lines of reporting, channels of communication, and specified procedures for every eventuality.

  Francis Pease, who was in Corning for the pouring of the second disk when McDowell arrived in Pasadena, was astonished at the changes in the working atmosphere in Pasadena when he returned. Before McDowell came, the various project committees used their meetings, usually once a week, to cat
ch up on formal reports and to prepare minutes for Hale, who missed most of the meetings. The real work took place in informal bull sessions, in an office, a lab, or under a tree with box lunches in the partially landscaped area outside the buildings on California Street—wherever two, three, or more scientists and engineers could get together to bat around ideas before they went back to their calculations and drawing boards. Now, suddenly, McDowell wanted formal minutes at every meeting. He would sometimes ask each person present to report what they had worked on during the week. Pease, who had been through the birth pains of a large telescope and had worked on this one for a decade and a half, didn’t like the forced efficiency of the new regime. He never got over his discomfort with McDowell.

  By January 1935, barely a month after he arrived in Pasadena, McDowell was back on the East Coast. In a whirlwind trip, he saw Max Mason at the Rockefeller Foundation, Vannevar Bush at MIT, Hannibal Ford of the Ford Instrument Company, GE officials in Schenectady, Warner & Swasey in Cleveland, Hostetter at Corning Glass, and Bethlehem Shipbuilding in Quincy, Massachusetts.

  McDowell searched out the contacts he had developed in his navy years. Vannevar Bush had done research on automatic guiding mechanisms for guns, and Hannibal Ford had built servo controls for gun turrets. The problems of mounting and guiding the telescope, for McDowell, were only a variation of the problems of controlling naval guns. John Anderson had to point out the difference: In guiding an astronomical telescope, the guide star might be as faint as the fifteenth magnitude, visible only in a large telescope, and stars, which had to be held on the slit of a spectrograph for hours, really weren’t comparable to the targets of the naval guns.

  McDowell wasn’t surprised by Anderson’s response. He was used to what he thought of as the hair-splitting of scientists. He saw his responsibility as cutting through the fine distinctions, urging, and even forcing mutual comprehension between the scientists and real-world engineers.

  One of McDowell’s earliest moves was to hire a retired army man from Pasadena, Col. M. L. Brett, to serve as resident superintendent at Palomar. Brett, a West Point graduate and former ordnance officer, had been a personal aide to the secretary of war in 1918. He was an asthma sufferer and eagerly accepted the position because he thought the mountaintop would be good for his health. His brother was a general who had served in Southeast Asia, and Brett did his best to carry on the family tradition, setting up his headquarters at the William Beech cabin east of the observatory site with his personal “striker,” a young man who cooked, kept house, and polished the knee-high boots the colonel wore each day.

  The conditions on the mountain were primitive. There was no telephone, an old gasoline-driven generator for power, and all supplies had to come up by the rugged Nate Harrison Grade. Communications were via mail and visitors from Pasadena. Even so, Brett didn’t have to advertise for labor. In 1935 an unconfirmed rumor of work was enough to prompt men eager for jobs, or with goods to sell, to hitch a ride up to the old Nate Harrison Grade to the peak to seek out the colonel. A young man named Ben Traxler, whose radio repair shop had failed in the depression, got a ride up the mountain with a hardware merchant eager for a new account. When Traxler mentioned his radio experience, Brett suggested the possibility of opening an amateur radio station on the mountain for regular communication with Pasadena. Traxler knew that an amateur license couldn’t be used for regular traffic, but he needed a job enough to agree with the colonel’s ideas. Two weeks later he was told to report to the mountain with long underwear, warm bedding, and any tools he would need.

  A few of the newly recruited workmen slept in the back of the colonel’s cabin. The others slept in a bunkhouse nearby, called the “Boar’s Nest,” with an attached kitchen and mess hall. Traxler tried a tent his first night, thinking he would enjoy the privacy, until he discovered the cold of nights on the mountain. By morning he had moved into the bunkhouse.

  Each morning Colonel Brett lined the workmen up in a semblance of an inspection, so he could march back and forth in his freshly shined boots and sharply pressed trousers, barking orders for the day’s work. The orders from Pasadena arrived in a steady stream. McDowell expected the work on the mountain to run like a Navy Sea Bee operation, and he had time for even the smallest procedural details: All mail from Palomar to Pasadena was to go in locked mailbags, to be unlocked at Palomar by Colonel Brett and at Pasadena by Miss Gianetti; spring faucets were to be used in all showers and the washhouse to conserve water; meal expenses were to be held under $1.25 per worker per day. Some of the crew cleared brush from the eventual building site with a small Caterpillar tractor, others started on digging and pipe laying for a well with a holding tank in the lower valley. The plans called for a pump station to lift the water three to four hundred feet to a 1-million-gallon tank and water tower, initially for the concrete work on the observatory and eventually to supply the observatory.

  While Colonel Brett’s crew started on site work at the top, a road crew from San Diego started on the new Highway of the Stars up the mountain. The flurry of New Deal alphabet agencies had provided for road projects like this one, and word soon got out that Basich Brothers, the contractors, would establish a WPA camp to house the road-workers. To an old cattle farmer like George Mendenhall, the whole New Deal smacked of Reds and Communism; the WPA was nothing less than government-funded bands of revolutionaries. “Owing to the general lack of control over these men, I am forced to refuse permission for any of these men to cross the French Valley at any time,” he wrote in protest.

  The problems with Mendenhall were smoothed over and by August 1935 everyone was ready to celebrate the beginning of work. George Hale received an invitation to the ceremony but was too ill to attend. Everyone else—Walter Adams, John Anderson, Francis Pease, Sinclair Smith, and Mark Serrurier—from Caltech and Mount Wilson enthusiastically trooped down to Palomar. Mendenhall, Beech, and some of the other property owners on the mountain came up for the festivities, joined by civic officials from San Diego and the reporters they were able to drag along. Poor Hale, bedridden in his darkened room, had to rely on reports from others. “I attended with considerable pleasure and amusement the party yesterday on the top of Palomar Mountain,” Adams wrote. “It was very much of a love feast.”

  20

  Swept Away

  The old A Factory at the Corning Glass Works sat close by the bank of the Chemung River, hemmed in by the tracks of the New York Central Railroad. It was a beautiful site. The river was wide enough to afford a perspective across to the hills in the distance and swift enough to keep itself clean. From the riverbank, a visitor could see the hillsides above the factory where so many of the Corning employees lived, and the Monkey Run that ran down off the slope into the Chemung. The river was part of life in Corning. Its changes signaled the seasons, from the low water of late summer to the wild flow of spring runoff. Old-timers could read the river the way an experienced sailor can read the sea.

  From time to time the river misbehaved. When the spring runoff reached a peak, the Chemung sometimes overflowed its banks. If that weren’t enough, after a sustained period of rain the Monkey Run, which carried runoff from the hills into the river, would surge over its channel, and the flood waters would find their way into A Factory, extinguishing furnace fires in the cave level and leaving behind a layer of mud that would take weeks to clean up. The worst flood anyone could remember had been in 1918. There were still marks on the walls of the factory caves to show the height the waters had reached.

  McCauley took some ribbing for his excessive caution, but when the controllers and transformers for the annealing oven for the two-hundred-inch disk were installed, he had made sure they were on a raised platform several inches above the high-water mark. The entire casting and annealing setup was designed so that the only equipment below the high-water mark were the bases of the four lifting screws, which were shielded with watertight steel cylinders, open at the top, and filled with oil to lubricate the screws and protect them a
gainst the inevitable dust from the factory cave. Even without the protective shields, the hoist screws would only be vulnerable to flooding in the lowered position, when the disk was being moved into or out of the annealing oven.

  For most of June 1935, the carpenters and millwrights had worked on a crate for the mirror. The shipment plans had been debated in Corning and Pasadena. Hale’s initial preference was shipment via the Panama Canal, but track clearances on tunnels and underpasses were too low to move the disk to a dock in New York, Baltimore, or Philadelphia, which left Albany as the only seaport they could reach from Corning. Even if they could get a direct cargo ship from Albany, via the Panama Canal to Los Angeles or San Diego, shipping by sea would involve the extra handling of the disk at Albany and a West Coast port, in addition to the initial loading onto a railcar at Corning and the final unloading in Pasadena.

  The alternative was to ship the disk by rail from Corning to Pasadena. The 120-inch disk had been shipped successfully across the country by rail in a crate, shielded and supported by heavy timbers. A 10-foot-diameter mirror disk was a large cargo, but it required no special routing or handling. The two-hundred-inch disk was another matter. Even if it were suspended vertically in a well car, so that the bottom of the disk would be only inches above the tracks, the crated two-hundred-inch disk would require tunnels and overpasses with a height of eighteen feet. The normal east-west routes through Kansas City didn’t have the clearance. The New York Central Railway authorities thought it might be possible to find a route through Chicago or St. Louis that would accommodate the disk, though it would take months of planning by their schedulers and the schedulers of other rail lines.

  The crate builders, told that their cargo was close to irreplaceable, that it had to fit under tight railroad bridges and tunnels, and that it had to be movable by crane, spent months engineering a crate. Here, too, the Caltech engineers got into the act, sketching a metal drum, constructed of half-inch and quarter-inch boilerplate, reinforced with heavy channel and angle sections. The American Bridge Company, in Elmira, was engaged to fabricate the larger sections from hot steel. Corning millwrights and carpenters fabricated the balance of the crate from sheet and angle steel, with felt, rubber, and cork cushioning designed to hold the disk rigidly. As each section of the complex crate was finished, it was test-fitted to the original two-hundred-inch disk in the temporary steel building on the riverbank.