V

… unperturbed by outside influence or occurrence …

            Less than two months after Deer Park’s first Atlas ICBM assumed its station, then President John F. Kennedy, in a speech before congress, stunned the world with 29 now famous words.  In that speech, he made the space race official and set America on a course that would consume it for the next ten year.  The President said, “I believe this nation should commit itself to achieving the goal, before this decade is out, of landing a man on the moon and returning him safely to Earth.”

            With the same type of rocket that had sent Allen Shepard into space resting in a bunker just a few miles away, Deer Park felt it was already part of that race — already part of the new world of rockets and space travel.

            True enough, a small part of the future was resting just a few miles away.  But it wasn’t to be found in the fierce blast of rocket engines hurling men into space.  Rather, the shape of things to come resided in a small, two-foot cube of transistors snugged into the left side equipment pod of the reclining Atlas rocket.  The technology embedded in that cluster, technology specifically developed for the Atlas ICBM program, would change the world more than anyone could imagine.

            Computer hardware in the early 1950s consisted of giant, room size analog devices like the Univac.  Using tubes, consuming prodigious amounts of electricity, and producing vast amounts of waste heat, the only way these computers could be used for guiding ballistic missiles was to leave the computer on the ground and beam instructions to the ascending rocket — as was done with the first of the operational Atlas series — the Atlas D.  From the military point of view, the vulnerability of this radio uplink was unacceptable.

            Wen Tsing Chow, a scientist for the ARMA Division of American Bosch,  proposed his own solid-state digital guidance computer design for the Atlas project.  The result was the eight cubic foot ARMA computer nestled in the missile’s B-2 pod.  In 1954, the same Mister Chow had developed the basic design for the missile’s inertial guidance system.  These two devices, working in tandem, allowed the Atlas E missile complete autonomy once it had risen just one inch above the floor of the launch bay, and created the most accurate ballistic missile the world had yet seen.

            The missileers are quick to point out that the computing power of the onboard unit, plus all the logic units in the bunker, still didn’t add up to the machine intelligence of a modern, hand held calculator.  But when the computer was given the coordinates of a target, and the inertial guidance system tuned so the computer knew precisely where the missile was at blastoff, that primitive brain could carry out its mission with cold efficiency.

            As a matter of policy, crews never knew where their missile was going.

            “On the launch console, in the upper left hand corner, were two buttons marked target A and target B,” Jim Geoghegan said.  “During the launch sequence we were to be told which of those to push.  When pushed, an indicator above the selected button would light.  Since none of us ever recall hearing anything official, it was always a matter of speculation as to what those two buttons actually did.”

            Analyst Technician Dick Mellor recalled, “There was only one programmed target for each Atlas missile.  It wasn’t fed into the guidance system the way it’s done now — as a pre-recorded digital program.  The targeting information was delivered to the site on what were called ‘target constants boards’.  These were plug in circuit boards, very common today, but at the time they were the first of their kind.  Those two boards installed together programmed for one target.”

            “The constants board was a matrix of diodes mounted in a metal framework”, Bob Lemley added.  “The board wasn’t very big.”

            “This was an extremely primitive programming system.  Selected diodes were burned out of the board, leaving the board with a readout sequence that held the target coordinates.”

            “Both boards were plainly visible when installed in the guidance alignment unit which stood close to the countdown logic units at the north end of the launch building’s equipment room.”

            Jim Geoghegan continued by saying, “Rumor, not fact, said that target A was a ‘clean’ burst and target B was a ‘dirty’ burst.  Clean meant an air burst that just blew everything away.  Dirty meant a ground burst intended to pick up millions of tons of contaminated debris and scatter it downwind.  That’s just rumor.  We never really knew.”

            The mating of an onboard inertial platform with an onboard digital computer resulted in a system unperturbed by outside influence and occurrence.  Once the rocket became independent of the launch complex, it would proceed toward its target, and only some type of system malfunction, or some type of damage to the missile, would prevent it from completing that mission.  The point of irretrievability was the moment the missile became airborne.

            The device that turned the system autonomous was the inertial platform.  This platform was stabilized in space by two, fluid suspended gyroscopes.  Once the gyros were spun to speed, and the inertial platform erected, leveled and oriented to true north, no matter what attitude the rocket assumed, the platform remained in the same position it had assumed prior to launch.  In a mechanical sense, it remembered where it started.

            The onboard computer needed two things to do its job.  First, it needed a mathematical model of its trajectory from point of takeoff to point of warhead detachment.  The flight program provided this.  The second thing it needed was a constant feed indicating any real world deviation from that mathematical model.  Comparing the two, the computer could manipulate the missile’s flight controls —  directing the two vernier steering rockets, and turning the gimbal mounted thrust chambers of the three main engines — to bring the rocket back on trajectory.  The inertial platform provided this data.

                  Mounted on the inertial platform were three vibrating reed accelerometers.

            “We called them piano wire accelerometers,” Bob Lemley explained.  “Our old Atlas flew in a four dimensional universe.  One dimension was time.  Since time only goes in one direction, all you needed to measure that was a clock.  The other dimensions were forward and back, up and down, left and right.  To measure those you needed three accelerometers — one oriented to measure movement in each direction, each dimension.”

            “Try to visualize a weight.  To each side of this weight I attach a thin piano wire.  Then I stretch the wires tight, so that the weight is suspended between.  If the wires are the same length, when I ‘pluck’ them with an electromagnet, they should both vibrate at the same frequency.”

            This apparatus in motion can be visualized by standing with arms forward and apart with the wire/weight apparatus stretched between the hands so that the central weight is positioned directly in front of the eyes.  From the weight, one piano wire would stretch to the left hand, the other to the right hand.  If both hands were quickly moved in either direction, while maintaining the exact distance between them, then, according to Newton’s first law, the weight in the center of the stretched wire would tend to stay at rest — would tend to lag behind.  If the wire/weight apparatus were moved quickly left, the wire on the left side of the lagging weight would become tighter, and the frequency of its vibration would increase.  The frequency of vibration in the wire to the right, which would now be under less tension because the central weight is lagging in its direction, would decrease.  This simulates the underlying principle used by the accelerometers fixed to the gyro stabilized inertial platform of the missile.

            “The difference in vibration frequency between the two wires can easily be measured,” Bob continued, “and, through experimentation, a value can be derived for the rate of acceleration for any given difference in frequency.”

            “As the missile moved — pitched and rolled, accelerated and decelerated — the masses in the three wire/weight accelerometers — each accelerometer oriented in one of the three dimensions — would try to stay where they were.  And that resistance to motion, changing the vibration frequency in the wires, would be what was being measured."

            After the rocket had been lifted upright, delicate pendulums sensed whether the inertial platform was standing at its zero point.  An optical reader, referencing a star sighting stored in a collimator permanently mounted in a pit in the launch bay floor, read the direction of celestial north.  Using these measurements, a set of servo motors moved the platform to its erect and level position, and turn it to the direction of the target.

            These same servos, acting on directions from inertial platform’s gyroscopes, would keep the platform stable during flight.

            “The North Star, Polaris, was used to align the inertial platform,” Dick Mellor said.  “Since you couldn’t just run out and take a quick reading whenever you received launch orders, the alignment had to be mechanically stored in a collimator. The collimator’s alignment was checked every six months — or whenever there was an earthquake strong enough to change the level of the launch bay floor.”

            Bob Lemley explained, “I’ve seen this done with a theodolite, a high quality surveyor’s transit, though I’ve never done it myself.  So I can’t be too terribly specific.”

            “First, the big door at the bottom of the launch bay ramp was cranked open.  Then, the metal cover over the collimator pit in the launch bay floor just to the west of the missile was opened.  The person using the theodolite needed to be able to shoot Polaris, then vertically flip the theodolite so that the instrument’s sight tube focused directly at the collimator in the pit.”

            At that point an assistant in the pit, following directions from the theodolite operator, would adjust the collimator until it matched the sighting taken by the theodolite.

            Rising out of the launch bay floor, just behind the collimator pit cover, was the inertial guidance sight tube assembly.  The upper end of this telescoping tube attach to the missile’s B-2 pod at the level of the inertial guidance platform.  As the missile was erected, this sixteen inches diameter tube extended.  During countdown, as the inertial guidance system was brought online, the collimator in the pit would send a thin beam of light to the missile through the darkened interior of the sight tube assembly.  This beam would reflect off a target on the inertial guidance platform and return to the collimator.  If the returning beam didn’t fall within certain limits, within a certain target on the collimator, the collimator sensed the error and transmitted a signal to the guidance system to adjust the inertial platform.

            In essence, the inertial platform was taking a sighting of the pole star Polaris — a sighting frozen in the adjustments the theodolite team had made to the collimator some time before.  It was a solution derived from the inventiveness of old school engineers — men who used a sensibility drawn from experience to find something workable.

            Having dispersed Fairchild’s nine missiles around a rough 40 by 75 mile oval, the engineers faced the problem of providing secure, dependable, and redundant lines of communication between those far-flung bunkers and Fairchild.  The most jam and snoop proof system available at the time was line-of-sight microwave transmission.  The only problem, such a system required exactly what made it so secure — a direct line-of-sight between transmitter and receiver.

            In 1958, when the military was first looking at placing three missile bases, each with three Atlas D missiles, near the towns of Davenport and Deer Park, and in the Long Lake area, the logical point for transmission between those sites and Fairchild may have been 3,575-foot high Dunns Mountain — southwest of Deer Park.  When the configuration changed to nine single missile Atlas E sites, Lookout Mountain, about six miles southeast, and 475 feet lower than Dunns Mountain, was selected as the relay point for Deer Park.  Possibly this change was due to the necessity of the relay point fitting into an enlarged network of line-of-sight relay points — those points extending not only to the south, to Fairchild, but also to the east and west.

           Robin Feil, a civilian radio communications technician and broadcast engineer stated, “Most of the missile sites were hardwired into their respective communication transceivers.  Because those transceivers needed to be within line-of-sight of either another transceiver, or of a relay point, sometimes the hardwire lines were many miles long.  I seem to recall that Deer Park was one of the few sites with its microwave uplink actually on the grounds.  Deer Park’s transmitter/receiver was located just a few hundred feet southeast of the launch bunker.  It would have looked like a concrete pillbox with a white dome on the side facing its distant transceiver.”

“Fairchild’s Atlas bunker microwave communications system formed a very rough figure 8 — two loops with Fairchild at, again very roughly, its center.  The loops were circular and redundant, so that a failure of any one site’s relay station would not break the communication link between the other sites.”

            “Microwave communications were the primary means for the Wing Commander, located at Fairchild, to give orders to the Crew Commanders on site.  It also allowed a system for site data such as readiness and security situations to be reported to headquarters.”              
            Deer Park’s link to Lookout Mountain split into three routes once it reached the mountain.  South was the connection into the communication center at Fairchild.  The western spur connected into the Gray’s Butte transceiver, which was also the uplink from Reardon’s site 9.  A link east from Lookout to 3,375-foot Antoine Peak, just to the north of Trentwood, connected into the uplink from site 2 at Newman Lake.  Both the Gray’s Butte and Antoine Peak’s transceivers sent microwave beams on to the next site in the loop.

            Since there was no telephone landline into the site, and the only other communication system available was the two-way radio, the microwave system provided a dedicated pathway for all the secure, high priority communications.

            With the bunker, personnel, missile, and communications all at ready, Deer Park’s Atlas was ready to add its weight to America’s mass of nuclear deterrent.