The Perfectionists: How Precision Engineers Created the Modern World

  Ed Weiler was there, the NASA engineer who had taken that first grim telephone call. Like everyone else in the room, he had his eyes locked on the screen. The next three seconds were the longest three seconds, Weiler said later, he had ever experienced in his life.

  There was a sudden explosion of exuberance, applause, delight, and joy. The image on the screen was now complete, and it showed before everyone a vivid mass of stars, all in focus, with one star in the dead center occupying just a single pixel of the screen. One star, one pixel.

  The image was sharp, perfectly, precisely sharp. No more fuzziness. No marshmallow. No soft edge. All was exact, aligned, impeccable, just as hoped for back when the project was a mere notion in a group of astronomers’ heads. No other optical telescope that had ever been made and established on planet Earth (even those at the summits of great mountains in Hawaii, in Chile, on the Canary Islands, and in other places where the air was at its thinnest and clearest) could ever rival this.

  Because down there was air—even when thin, it was heavy, windy, polluted, dancing with molecules, potent with distortion. Yet up here, nearly four hundred miles up, high above the troposphere, the stratosphere, the mesosphere, in what is now called the exosphere, where there was just the occasional hydrogen molecule drifting through, there was no air, and no distortion—and where now, at last, and thanks to the cleverness and cost of a whole set of new optics, humankind had a clear-eyed new viewing platform, like no other ever before, from which to observe.

  Half a century after it was first conceived, twenty years after it was first designed, fourteen years after a computer in Danbury told the first polishing arm to travel across the great tablet of Corning quartz and begin to grind and shape its surface, and thirteen hundred and some days after that overflattened eight-foot mirror took its first long gulp of light from the universe wrapped around it, a repaired telescope with new precision optics was able to see clearly deep into the distance, and into the distant past of the cosmos.

  The rest of the Hubble story is still being told today. Four further servicing missions have been up to it, as were scheduled long ago, each one tasked with breathing new life into what has become a beloved old silver workhorse, the greatest of NASA’s great observatories. The longevity of the now almost venerable, if still no prettier, bird is greater than ever anticipated, and it is now expected to continue flying at least until 2030, maybe for a decade longer. It is by all accounts the most successful scientific experiment of modern times, maybe even of all times. And the images it has sent back, tens of thousands of them, have captivated all who see them. The eight-foot mirror, imperfect though it may be, has captured a vision of wonder and rapture to scientists and the lay alike, bringing the universe vividly to life.

  Chapter 8

  (TOLERANCE: 0.000 000 000 000 000 01)

  Where Am I, and What Is the Time?

  Each after each, from all the towers of Oxford, clocks struck the quarter-chime, in a tumbling cascade of friendly disagreement.

  —DOROTHY L. SAYERS, GAUDY NIGHT (1935)

  Time is the longest distance between two places.

  —TENNESSEE WILLIAMS, THE GLASS MENAGERIE (1944)

  The offshore oil rig Orion, nine thousand tons of ungainly ironmongery being hauled slowly across the North Sea by a pair of tugs, was looking for a place to settle herself down and drill.

  I was on the bridge of the lead tug, a small but exceptionally powerful craft called the Trailblazer, from Holland. Orion, her four jacked-up legs towering high over the drilling derrick itself and swaying in a dangerous-looking manner on the swells, had just completed a successful natural gas well five or so miles away. Now we were towing her to a place that the geophysicists back in Chicago had chosen, as the undersea geology looked promising for a new attempt.

  It was March 1967, a bitter-cold early-springtime day, with a stiff nor’easterly breeze. I had worked on this rig for just one month. I was not yet twenty-three years old. The rig was worth ten million dollars, and Amoco Petroleum was renting her for eight thousand dollars an hour. Putting her down in the right place, exactly, was now, quite ludicrously, all down to me.

  I had been given precious little by way of either a briefing or equipment to make sure that Orion settled herself properly. I had a two-way radio that let me talk to the tool pusher up on the rig. I had the British Admiralty maritime chart number 1408 (Harwich to Rotterdam and Cromer to Terschelling), which covered this portion of the North Sea. I had a confidential large-scale geophysical chart of the local ocean floor, fashioned by the American undersea survey teams, and on which someone had marked a big red X, as the place where the planners in Chicago now wanted the rig to put down. Written in pencil beside the X were the rig’s coordinates, something in the order of 53°20'45" N, 3°30'45" E, but with the seconds of arc written to one or maybe even two decimal places.

  Crucially, the tug’s master also had a special chart overlaid with the curved lines (colored in red, green, purple) of the then-most-advanced radio navigation system known. This was the Decca Navigator chart, and like most captains of coastal-going vessels of the time, ours used it in conjunction with a large receiver that was mounted on a swivel at head height. This receiver, rented from the Decca company, sported four dials, three of them with what looked like clock hands, painted with luminous paint so that they could be read at night.

  The receiver picked up the powerful radio signals that were being continuously transmitted from coastal radio stations, masters and slaves, that Decca had built on headlands and cliff tops on the English and German coasts of the North Sea. The signals, which were invariably short pulses, would go out from the master station, and then, a short moment later, the same pulse would be repeated by each of the slave stations. The delay between the master and slave pulses’ reception by a receiver would vary depending on how far the receiver was from each of the slaves—and this, in turn, would allow the primitive computer in the receiver to deduce and determine, by taking a fix from the different distances to the various slaves, where on the chart the receiver was. The dials on the receiver would then show how far along which of the various lines of position (red, green, and purple) our little tug was proceeding. And because it showed where we were on three separate intersecting lines, we could draw them out from the Decca chart and—lo!—they would show where on the navigation chart, or on the geophysical chart, we actually were, to within an accuracy, the Decca makers insisted, of about six hundred feet.

  What I had been told was that when I decided that the rig was exactly above the designated spot—and I knew from the Dutch skipper that there was a surface current running to the northwest, at about six knots, so I had to make allowances for that, as it would set the rig drifting a few score feet while it was settling—I had to instruct the tool pusher by radio to “Drop the legs!” He would immediately order the release of four sets of bolts, and the tall iron legs that were now towering above us would instantly plummet downward with four gigantic splashes, and would hurtle unstoppably down onto the seabed two hundred feet below. There they would pin themselves, by skewering themselves into the soft upper layers and thereby, with the addition of a set of anchors sent down later, fix the rig solidly into position for the next many weeks of her exploration.

  We crept closer and closer. The fathometer pinged every few seconds, the depth below our keel showing a steady thirty-two fathoms. The Permian dome, which was to me just a vague pattern of half-inscribed lines on the geophysical chart that had been interpreted by specialists in Chicago to be a dome, crept nearer and nearer. For a few moments it appeared to be directly underneath where Decca told me the rig was, and I nervously fingered the Transmit button on the radio microphone, pressed it, looked up at the drilling platform, and spoke loudly into the microphone. In a tone as stern and as formal as a nearly twenty-three-year-old could muster, I commanded, “Drop the legs!”

  An instant later, I saw the four small gusts of reddish rust smoke, and the enormous towers of tube iron
trelliswork appeared immediately to collapse into themselves, to vanish quickly from sight. There was a fearful noise of screeching metal and a huge froth of roiling water. We ordered the seamen aboard the rig to release our towrope, and the same for the tug astern. The two tugs turned away and headed out to sea, away from the din, and then we stood off a mile or so and watched as the rig master ordered the jacking-up procedure to begin—noisily, once again, like the sound of a construction site jackhammer, and foot by foot, so the rig climbed up its own now stably rooted legs, pulling itself up by its own bootstraps, up and up, until it was a good forty feet above the waves, and then clear of most effects of storm and swell and surge below. Then someone aboard stopped the machinery, and silence fell, aside from the steady low howl of the gathering wind and the poundings of the swell.

  The tool pusher came on the radio. He had just seen the bathymetry reports. “All looks good,” he said. “The current kicked us off a little, maybe. We’re about two hundred feet off the ideal. Pretty good for a beginner. Chicago will be okay with that. It’s good enough. Go get some sleep.”

  They spudded the well later that evening, and then drilled night and day for the next three weeks. We hit gas at six thousand feet, a good and powerful flow of what back in the sixties were the blessings of raw hydrocarbons. A week later we capped the well off, leaving it to be connected to a producing field by a later gang of workers, and Orion and her crew departed with another couple of heavy-haul tugs for further hunting grounds in the sea.

  In due course, I left the rig, then the company, and eventually the profession of petroleum geologist altogether, but the knowledge that I had once helped locate a nine-thousand-ton drilling rig over a Permian salt dome in the heaving middle of an ocean, and had managed to do so with sufficient accuracy to create a flowing gas well, stayed with me for many years.

  We had reached to within two hundred feet of the mark, a figure that seemed to me at the time a very considerable achievement. But being two hundred feet off an X drawn on a chart is, by today’s standards, unimaginably imprecise, a total fail. Places on the surface of the planet can now be located within centimeters (millimeters soon), and they can be so located because of the making of a technology that would eventually replace Decca and LORAN and Geo and Transit and Mosaic and all the other proprietary and radio-based navigation systems of the time, and would indeed also replace the sextant* and the compass and the chronometer and all the various navigation bridge furnishings with which sailors had been determining their positions for centuries.

  It’s called GPS.

  THE BASIC PRINCIPLE of this new technology was unexpectedly born of the development of quite another.

  It was in Baltimore, on Monday, October 7, 1957, when two young scientists, William Guier and George Weiffenbach, arrived at their Applied Physics Laboratory at Johns Hopkins University, enthralled like all American scientists by the fact that, for the first time ever, an artificial moon was currently in orbit around Earth.

  It was Sputnik, a two-hundred-pound, twenty-inch-diameter sphere of polished titanium alloy that the Soviet Union, to the chagrin of the American public, had launched the previous Friday, and which was now orbiting Earth once every ninety-six minutes. The New York Times Sunday edition had reported (on page 193 of its 360-page paper) that the device was continuously emitting radio signals from a tiny transmitter on board. Guier and Weiffenbach (both computer experts, their most recent work being on hydrogen bomb simulations and microwave spectroscopy, respectively) reckoned they could probably determine exactly where the satellite was by recording and then analyzing its radio signals.

  Many years of occasionally partisan bickering led eventually to the acceptance of Vermont-born Roger Easton (third from left) as the inventor of GPS, while he was working at the U.S. Naval Research Laboratory in Washington, DC.

  Photograph courtesy of the U.S. Naval Research Laboratory.

  Accordingly, they used the specialized radio receivers in the lab to tune in to Sputnik’s frequency, and listened intently to the regular heartbeat of its transmissions (a high-pitched beep, sent out a little faster than twice a second) and recorded it on a high-fidelity tape deck. They then analyzed the frequency of the signal and, as they suspected might be the case, heard it alter very slightly as the satellite rose above the horizon, as it then passed directly overhead of them in their Baltimore lab, and finally then set down once again. The frequency change they observed was the Doppler effect—the classic example being the change of perceived frequency of the horn of a passing train—and for the first time ever, it was shown by this pair of physicists to be both detectable and measurable in a satellite signal.

  Shortly thereafter, and by employing as powerful a computer as was available—the Applied Physics Laboratory had a brand-new Remington UNIVAC at hand—the pair was able to digitize the signal and, from the varying frequencies that had now been converted into numbers, to calculate with fair precision how far away Sputnik was on each one of its orbits. The frequency when the satellite was directly above them was the true frequency of the signal; from the variations, the first when it was approaching them and then again when it was moving away from them, gave them the basis for a calculation (as they knew from its circumorbital time that it was moving around the planet at about eighteen thousand miles per hour) of how far away it was.

  Their sums (which they then also applied successfully to predicting the orbits of Explorer I, once America had entered the space race) involved many weeks of computer time, and would have profound consequences. For the following March, the chairman of the Applied Physics Laboratory, Frank McClure, realized that his two young colleagues had unwittingly stumbled upon the makings of an application that could have worldwide use.

  As he told the pair when he hauled them into his office and demanded that they close the door, if an observer on the ground could establish with precision the position of a satellite in space, then the opposite, the numerical reciprocal, could be true as well. From the position of the satellite, one could compute the exact position back on Earth of the person or machine that observed it.

  Guier and Weiffenbach had never noticed what in retrospect was blindingly obvious, nor did they immediately appreciate the corollary: that a satellite navigation system based on this simple Doppler principle could do for ships and trucks and trains and even for ordinary civilians, mobile or stationary, what the sextant, the compass, and the chronometer had done for centuries past for mariners, and what LORAN and Decca and Gee were doing at that very moment. It could tell them where they were; moreover, it could tell them what direction they should take if they wanted to go somewhere else. “It occurred to me,” wrote McClure, in a famous memo that claimed the prize for Guier and Weiffenbach, “that their work provides a basis for a relatively simple and perhaps quite accurate navigation system.”

  Quite accurate indeed: the U.S. Navy, which paid for much of the APL work in Baltimore, did some back-of-envelope calculations and came up with the notion that with a good number of satellites, the location of someone’s or something’s position (that of a ship or a submarine) could be achieved within perhaps a half mile. And while that may not be as precise as the six hundred feet guaranteed to Decca, there was a further significant difference, an advantage that was especially relevant in these times of the gathering problems that related to the Cold War. The radio-based Decca-like systems then employed by ships, and by oil rig location tugs such as Trailblazer, were hardly secure, as their transmitters were all based on land, and could easily be put out of service by a canny foe. A system that involved satellites out in space, however, was by its very nature much more protected from outside tampering and interference, from surveillance and from sabotage. Moscow, the enemy du jour, would find it difficult to mess with it or find out anything from its use.

  The U.S. Navy, at the time, was looking for a foolproof, secure, and accurate means of locating its fleet of Polaris-armed nuclear submarines, and thus was born the Doppler satellite navigatio
n system known as Transit. A prototype satellite was successfully put into orbit in 1960, and no more than six years after McClure’s memo (seven years after the launch of Sputnik), a flotilla of U.S. Navy Transit satellites was in orbit around Earth, and the first true satellite navigation system was declared to be fully operational.

  One of the early Transit-system satellites. Launched for the U.S. Navy in the 1950s and ’60s, it used Doppler-based navigation to establish to within three hundred feet the position of American strategic submarines. Transit is seen as the first operational use of a system that eventually led to the modern Global Positioning System, GPS.

  Photograph courtesy of the National Air and Space Museum, Smithsonian Institution.

  Fifteen satellites were built, rather inelegant-looking and insect-like creatures with four solar-panel wings and a long torso attached to a transmitter that acted as a boom to keep the antennas pointing down to Earth. At least three devices at a time were kept up in polar orbit, six hundred miles high. As the world turned beneath them, they swept the land and sea masses below, rising and setting like the sun and sending out signals to receivers on the ground that would be Doppler-affected as they moved toward, over, and away from the receivers. Earth stations, equipped with enormous computers with tape drums whirling back and forth, would predict the true orbit of each satellite as it appeared in the sky, and would radio these data to the ships and submarines that needed to know where in the world they were. And cumbersome and slow though it might have been, and available only once every few hours in its early days, the system most certainly did allow U.S. Navy ships anywhere in the world, at any time of the day or night, and in any weather, to learn their more-or-less-exact location.