Category Paving the Way for Apollo 11

FIRMING UP THE PLAN

The spectacular Soviet achievement of photographing the far-side of the Moon in October 1959 prompted NASA to revise its planning. The Air Force was developing the Agena В as a variant of an upper stage which could restart its engine in space. The payload projection for low Earth orbit of an Atlas-Agena В exceeded that projected for the two-stage Atlas-Vega, and almost matched that of the three-stage variant. On 7 November 1959 NASA decided that all satellites scheduled to use the Atlas-Vega should be transferred to the Atlas-Agena В; the Vega stage would be used only for lunar probes. But on 11 December it was decided that the lunar probes would switch too, and the Vega was cancelled. The Atlas-Agena В was seen as the interim launch vehicle for deep-space probes, pending the introduction of the Atlas-Centaur.

On 21 December 1959 Abe Silverstein of the Office of Space Flight Development at NASA headquarters told JPL to prepare five spacecraft to reconnoitre the Moon in 1961-1962. The objectives now included obtaining high-resolution pictures of the lunar surface during the terminal approach, which would require to be transmitted in real-time since the vehicle would be destroyed on impact. This imaging was to make up for the loss of the orbiter – which Silverstein had ordered in June 1959 and now cancelled because a review had judged it to be too complex for this early point in the program. In addition, Silverstein told JPL to consider adding instruments to perform particles and fields investigations during the cruise to the Moon. This project was to be completed within 36 months, in order to pass on to the next project of the lunar exploration program. It was acknowledged to be a high-risk venture on a short-term schedule, but was intended, in addition to studying the Moon, “to seize the initiative in space exploration from the Soviets”.

In January 1960 Silverstein created the Lunar and Planetary Programs Division, with Edgar M. Cortright as its Director. In March Oran W. Nicks was made Chief of Lunar Flight Systems, and given the task of monitoring the project’s progress. After NASA formally approved the project in February 1960, Clifford I. Cummings at

JPL proposed that it be named ‘Ranger’.[11] Silverstein was not keen, but the moniker was made official and on 4 May Cummings announced it to the Los Angeles Herald – Examiner.

At the end of 1959 Keith Glennan reorganised NASA headquarters. The dominant office in terms of budget and activities remained Silverstein’s, now called the Office of Space Flight Programs.

W. H. Pickering reorganised JPL at the end of 1959 by creating the Lunar Program Office under Cummings, with James D. Burke as his deputy. Cummings duly made Burke Ranger Spacecraft Project Manager. JPL managed the new NASA projects by superimposing small ‘project offices’ on the existing functionally arrayed technical (‘line’) divisions that comprised the laboratory’s core expertise. Indeed, the Ranger Project Office initially comprised only two men and a secretary. Burke’s task was to allocate funds, plan, schedule, assign tasks to the divisions and review progress, but he had no direct supervisory authority over the work since each division did its own design and development and divisional chiefs set their own priorities and assigned engineers within their bailiwicks. Initially, Ranger was the focus of activity, but as other projects claimed attention, most notably the Mariner interplanetary missions, engineers were reassigned and Ranger suffered.

As another element of his restructuring, Pickering added a Systems Division to develop, build and test spacecraft, and a Space Sciences Division to install scientific experiments. For Ranger, the Systems Division would be responsible for systems analysis, including launch to orbit, departure to the Moon, and the requirements of midcourse and terminal manoeuvres; the design and integration of the spacecraft’s systems, including qualification and performance testing, and quality assurance; and assembly and checkout for launch. It would call upon other divisions as subcontractors. Harris M. Schurmeier was the Chief of the Systems Division, and as such he became Burke’s main point of contact with the technical side of the laboratory. In February I960 Schurmeier appointed Gordon P. Kautz as the Project Engineer for Ranger in the Systems Division, but in October Kautz was reassigned as Burke’s deputy and Allen E. Wolfe took the vacated post. The Space Sciences Division, led by Albert R. Hibbs, consolidated JPL’s experimenters into a single group. The Guidance Division was headed by Eugene Giberson. The Engineering Mechanics Division was under Charles Cole. The Telecommunications Division was under Eberhardt Rechtin. The Propulsion Division was under Geoffrey Robillard.

In May 1960 NASA directed JPL to start work on the Surveyor project. As it was doing with Ranger and Mariner, JPL sought to maximise commonality of systems between the two forms of the Surveyor – one for orbital reconnaissance of the Moon and the other to soft land and investigate the physical and chemical properties of the surface. It was expected that because a rough landing from a direct approach would be simpler than entering orbit or soft landing, Ranger would be able to be completed while Surveyor was in development.

Management issues 57

On 27 February 1965 the Ranger experimenters met at JPL with representatives of the Surveyor and Apollo projects to consider the target for the final Block III with a window that would open on 19 March. The Moon would be ‘full’ on 17 March and ‘last quarter’ on 25 March. The Surveyor people argued for Oceanus Procellarum at the western end of the Apollo zone, to identify a safe target for their first soft-lander, but this was rejected. Accepting that the maria were probably all much the same, the Apollo people suggested inspecting a blanket of ejecta, to gain an impression of the roughness of such terrain. The Ranger experimenters themselves argued for a target of particular scientific interest, and this was accepted.

The Ranger team met again on 2 March to consider specific features. To obtain unique data, they considered a variety of locations that were unlikely to be visited by either Surveyor or Apollo – three being Copernicus, Kepler, and Schroter’s Valley near Aristarchus. However, Harold Urey and Gerard Kuiper were both in favour of the crater Alphonsus. Its floor was generally flat, but contained irregular rilles and a number of small ‘dark-halo’ craters which some people thought might be of volcanic origin. Following reports by Dinsmore Alter in America in 1956 of a slight ‘‘veiling’’ of the floor of Alphonsus, Nikolai Kozyrev had monitored the crater for any further such ‘transient events’, and on 3 November 1958 obtained a spectrogram of a ‘‘glow’’ obscuring the 1,100-metre-tall central peak using the 48- inch reflector of the Crimean Observatory. The spectrogram was disputed, but Kozyrev interpreted it as a release of gas. Alphonsus was therefore selected as the primary target for Ranger 9.

The experimenters could not agree a target east of the meridian for early in the window, but a launch on 21 March was compatible with Alphonsus, and thereafter Copernicus, Kepler and Aristarchus on successive days. This list was sent to the Office of Space Sciences and Applications on 10 March. Oran Nicks endorsed it, and passed it to Homer Newell, who concurred. However, NASA had scheduled the Gemini 3 manned mission for 22 March, and the Air Force required a clear 24 hours to reconfigure the Eastern Test Range for a different type of launch vehicle. On 15 March Robert Seamans ordered Gemini 3 postponed to 23 March to give Ranger 9 a chance at its primary target. When the countdown began, the Cape was cloudy and the low-altitude winds were gusty. The clock was held to await an improvement in conditions. Although it remained cloudy, when the winds declined it was decided to proceed.

Soon after lifting off at 21:37 GMT on 21 March (just before that day’s window closed) the vehicle penetrated the cloud deck and was lost from sight. But everything went to plan. The midcourse manoeuvre was deferred to enable radio tracking by the Deep Space Network to precisely define the initial trajectory. It was calculated that

the spacecraft would impact some 640 km north of Alphonsus. The 31-second burn at 12:30 on 23 March ensured that it would fall into the 112-km-diameter crater. The aim point was midway between the central peak, the rilles and the dark-halo craters so that they would all be visible during the approach at a resolution better than was possible telescopically; but they would not appear in the final images, which would give a view of the floor of the crater.

As Ranger 9 approached the Moon on 24 March it became the first spacecraft in the project to make a terminal manoeuvre. At 13:31 the vehicle departed its cruise attitude by pitching, yawing, and pitching again whilst holding its high-gain antenna pointing at Earth. By aiming the cameras along the velocity vector, this manoeuvre would optimise the resolution of the final frames. Ray Heacock, the JPL member of the experiment team, provided the commentary in the auditorium. The electronic scan converter made for the Surveyor project had been hastily modified to process the Ranger video for ‘live’ broadcast by the TV networks. In essence, this comprised two sets of vidicon tubes (one for the wide-angle stream and the other one for the narrow-angle stream) and in each case one vidicon viewed the image displayed on its counterpart, in the process converting the 1,132 lines per frame received from the spacecraft into the 500 lines of the commercial

Ranger triumphs

The sites inspected by the successful Ranger missions.

system.[24] Imaging began at an altitude of 2,400 km, lasted 18 minutes, and ended with impact at 14:08:20 when the spacecraft hit the ground at 9,617 km/hour within 6 km of the aim point.[25] The final picture of the 5,814-frame sequence had a resolution of 25 cm. It was a spectacular way to conclude the Ranger project! The transmission impressed not only the public, but also those scientists who had not appreciated the value of imagery.

At the experimenters’ press conference later in the day, Gerard Kuiper said the ‘dark halo’ craters on the floor of Alphonsus appeared to be volcanic. Harold Urey, no fan of the ‘hot’ Moon hypothesis, allowed that they were probably ‘‘due to some sort of plutonic activity’’. The rilles were revealed not to be clefts but chains of small irregular craters, which some people argued were volcanic. The walls of Alphonsus were very smooth. The central peak was not the harsh edifice depicted by artists. There was no indication that it was a volcano, but neither was there evidence that it was not. There was no clue as to the cause of the ‘transient events’ seen by Alter or Kozyrev.

Mission accomplished 135

In June 1967 the Surveyor Scientific Evaluation Advisory Team considered sending Surveyor 6 to a ‘scientific’ target, with one option being the hummocky Fra Mauro Formation, but NASA headquarters specified Sinus Medii, which would be the ‘first backup’ for an Apollo primary target in the eastern hemisphere. Surveyor 6 would be the project’s third attempt at the meridian – Surveyor 2 had been lost attempting its midcourse manoeuvre, and contact had been lost with Surveyor 4 towards the end of its retro-rocket burn.

Sinus Medii was a relatively small mare plain about 170 km across, bounded to the north and south by highlands. The fact that the northwest-southeast structural trends of the adjacent terrain were radial to Imbrium indicated its origin as sculpture from the creation of that basin. The shapes and trends of the wrinkle ridges, crater chains and small shallow trenches on the plain reflected this structural pattern. The fact that the mare had a larger number of craters with diameters exceeding several hundred metres indicated its surface to be older than most maria. Telescopic studies showed it to have a higher average albedo than most maria. The largest crater on the plain was Bruce, at 7 km in diameter. The centre of the 60-km-diameter target circle was 55 km southwest of Bruce. It was hoped that the lander would set down within sight of a wrinkle ridge.

The launch window for Surveyor 6 was 7-12 November 1967. Although the first unmanned test of the Saturn V launch vehicle was due on 7 November, preparations for the lunar mission went ahead because if it were to become evident that the other mission would not meet its schedule, Surveyor 6 would attempt the first day of its window; otherwise it would be slipped – the Cape’s tracking system required at least 24 hours to reconfigure for different types of vehicle. In the event, the Saturn V was postponed.

Surveyor 6 lifted off from Pad 36B at 07:39:01 GMT on 7 November. Since this was a predawn launch, the Centaur achieved parking orbit in darkness. It flew into sunlight at 07:53:22, initiated the 115-second translunar injection at 08:01:35 and released the spacecraft at 08:04:30. At 02:15:59 on 8 November, once the spacecraft had adopted the attitude for the midcourse manoeuvre, the helium valve was opened to pressurise the vernier propellant tanks. In raising the propellant to 764 psi, the helium fell by 180 psi from its initial 5,182 psia. The burn at 02:20:02 lasted 10.3 seconds, and the 33.1 ft/sec change in velocity moved the aim point 90 km closer to the centre of the target circle. After declining by 208 psi during the burn, the helium regulator maintained the propellant pressures constant throughout the remainder of the cruise.

The pre-retro manoeuvre in which the spacecraft departed from its cruise attitude involved initiating a roll of +82.0 degrees at 00:25:20 on 10 November, a yaw of + 111.8 degrees at 00:29:38 and a final roll of +120.5 degrees at 00:34:56. The initial approach was at 24.3 degrees to the local vertical. The altitude marking radar was enabled at 00:56:16, and issued its 100-km slant-range mark at 00:57:57.038. The delay to the initiation of the braking manoeuvre was specified as 5.875 seconds.

The verniers ignited precisely on time, and the retro-rocket 1.1 seconds later. At that time the vehicle was travelling at 8,460 ft/sec. The RADVS was switched on at 00:58:05.798. The acceleration switch noted the peak thrust of 9,700 pounds fall to 3,500 pounds at 00:58:43.397, indicating a burn duration of 39.4 seconds. After allowing time for the solid rocket thrust to tail off, the verniers were throttled up to their maximum thrust at 00:58:53.297 for a duration of 2 seconds, during which the motor was jettisoned. At burnout, the angle between the vehicle’s thrust vector and velocity vector was 26 degrees. The RADVS-controlled phase of the flight began at 00:58:57.737, when the slant range was 40,574 feet (and because the velocity vector at burnout was offset to vertical, the altitude was 36,625 feet) and the total velocity was 515 ft/sec (and since the vehicle had maintained its thrust along the velocity vector extant at the time of retro ignition, the longitudinal rate was 463 ft/sec). The vehicle immediately aligned the thrust axis along the velocity vector extant at retro burnout and flew with the verniers at 0.9 lunar gravity, very slowly accelerating as it descended. When the altimeter locked on at 00:58:59.892, at a slant range of 35,924 feet, attitude control was switched from inertial to radar and the thrust axis was swung in line with the instantaneous velocity vector to initiate the gravity turn. On intercepting the ‘descent contour’ at 00:59:21.276, the slant range was 24,730 feet and the speed was 552 ft/sec. By the 1,000-foot mark at 01:00:40.534, the vehicle was descending very nearly vertically at 106 ft/sec. The 10-ft/sec mark was issued at 01:00:57.634 at a height of 50 feet.

On receiving the 14-foot mark at 01:01:04.133, the flight control system cut the verniers. At that time the rate of descent was 4.6 ft/sec. After falling freely for 1.3 seconds, the vehicle touched down at 01:01:05.467 with a vertical rate of 11.2 ft/sec. Foot pad no. 1 made contact first, then legs no. 2 and 3 some 25 and 40 milliseconds later, respectively. It rebounded slightly, then settled, with the lateral rate of 1.0 ft/ sec in the direction of leg no. 1 causing each pad to produce a pair of overlapping imprints. The gyroscopes indicated that it was within 1 degree of local vertical. It was a perfect landing!

The verniers had consumed 8.4 pounds of propellant in the midcourse manoeuvre, 41.1 pounds in the retro phase of the descent and 96.8 pounds in the vernier phase – a total of 146.3 pounds of the initial propellant load of 182.6 pounds. The total time spent under RADVS control was 2 minutes 6 seconds, with 1 minute 43 seconds of that flying the descent contour. In contrast, in its improvised descent Surveyor 5 had spent just 62 seconds under RADVS control. In order to have their full functionality available in the event of attempting a ‘lift off and translation’ experiment, it had been decided not to lock the legs as part of the post-landing sequence.

The first 200-line picture was sent at 01:50, and this 24-frame survey of the arc between foot pads no. 2 and 3 continued to 02:35. At an elevation of 3 degrees, the

Sun was barely above the horizon. At 02:55 the solar panel began to scan in azimuth for the Sun, and located it at 03:19. With the landing site at the centre of the Moon’s disk, Earth near the zenith and the vehicle upright, the alignment of the high-gain antenna was simple. It locked on at 03:40. The first 600-line picture was transmitted at 04:02. This camera was the first to have the new box-shaped hood, the mirror of which could seal the aperture to prevent dust or efflux from penetrating the optical system during landing.

The first 360-degree wide-angle panorama was completed by 05:00 and showed a relatively smooth, heavily cratered plain, but there was a feature on the southeastern horizon which, in the low-angle illumination, looked as if it might be a ridge. When this was examined again on Goldstone’s second pass, with the Sun about 13 degrees higher, this identification was confirmed and a series of narrow-angle pictures were taken to record it in detail. The ridge was identified in Lunar Orbiter 2 frame M-113, and when the individual features visible to the lander were located on H-121 by that

orbiter the lander proved to be 10.5 km from the aim point. In high-resolution orbital imagery, the ridge was seen to be 40 km long and to zig-zag generally east-to- west with its individual segments ranging from 300 metres to 2 km in length. It vanished about 1 km southwest of the lander. The base of the nearest section of the ridge was 200 metres from the lander, it was several hundred metres wide and its crest rose about 30 metres above the adjacent plain.

To improve visibility of the surface beneath the vernier engines, Surveyor 6 was provided with three convex mirrors instead of two. Its orientation on the surface was determined by star sightings. Like its predecessor, it had a magnet on foot pad no. 2 to study the concentration of magnetic particles in the surface material. The colour filters had been superseded by polarising filters, and pictures were taken of selected areas during successive Goldstone sessions to build up a dataset in which the Sun’s elevation changed at intervals of about 13 degrees, and thus measure the variation of the polarised component of surface reflection as a function of solar phase angle; the results proved to be insignificant, however.

The fragments displaced and ejected by the foot pads were composed primarily of aggregates of fine-grained material, and in many cases included small bright rock chips. In the immediate vicinity of the lander there were fewer fragments exceeding 2 cm in size than at the other sites, but a greater number smaller than this size. There was also a relative paucity of blocks within 50 metres of the lander – there were only six larger than 20 cm, and the largest was about 50 cm in size. Some were tabular, resembling the layered rocks seen by Surveyor 3 in its medium-sized crater. Most of the fragments within this range were subangular to subrounded, and although many were resting on the surface others were partially buried.

On the plain, craters up to about 150 metres in diameter generally possessed low subdued rims, but some were rimless. The fact that the smallest craters observed by Surveyor 6 on the plain to possess blocky rims exceeded this size indicated that the fragmental debris layer was up to 20 metres thick. For one crater the rim was not actually visible to the lander, just the associated field of blocks. This was visible in high-resolution Lunar Orbiter pictures, which also showed a bench in the wall of the crater at a depth of about 20 metres that could have marked the contact between the fragmental debris layer and the substrate.

In contrast, the lander observed a crater on the flank of the ridge about 30 metres in diameter and one on the crest of 20 metres diameter with blocky rims, indicating that the fragmental debris layer on the ridge was only 8 to 10 metres in thickness. On the crest to the south of the lander there was a crater 180 metres in diameter that was littered with blocks ranging up to 3 metres in size. In the high-resolution Lunar Orbiter pictures, it was possible to see blocks up to 6 metres in size elsewhere on the ridge. The coarse blocks within strewn fields were angular and faceted, and mostly appeared to be exposed on the surface. In terms of small craters, the size-frequency distribution on the ridge was comparable to that of the plain at the landing site. But a close inspection of the lander’s pictures and the high-resolution orbital imagery indicated there to be many more coarse blocks on the ridge than on the adjacent plain – in this respect the ridge was similar to other examples of wrinkles, suggesting that it was representative. The origin of the ridge was disputed. One idea was that it marked where lava had extruded from a fracture (a dyke) and solidified in place. If

this were the case, then the ridge could have formed at any time since the lava flow that made the plain. But there was no evidence in the cratering to suggest that the ridge was significantly younger than the plain. Another theory was that such ridges were produced when a mare plain was deformed by compressional stress. In this case, the manner in which the ridge zig-zagged suggested that its formation was controlled by regional structures. Such stresses could have been imposed at any time after the formation of the mare. The fact that the fragmental debris layer on the ridge was thinner than on the plain was explicable by the slow but progressive flow of loose material downslope. The profusion of large blocks on the crest was certainly consistent with such ‘mass wastage’. The effect was to smooth the transition between the plain and the ridge. Indeed, in frame H-121 provided by Lunar Orbiter 2 it was difficult to precisely identify the outline of the ridge.

The experiment in which Surveyor 1 pulsed a cold-gas attitude control thruster to study surface erosion had been inconclusive, so Surveyor 6 was to repeat this test by firing a thruster continuously. Since any disturbance of the surface would be subtle, the test was made on 11 November, while the Sun was still low in the east to maximise shadow detail in the impingement area. The downward-aimed thruster on leg no. 2 was fired at 03:23 for 4 seconds, and again at 03:47 for 60 seconds. The ground beneath it was surveyed by the camera prior to, between and after the firings. The nozzle was 10.4 cm above the surface and inclined at 24 degrees to the lander’s vertical axis. Both firings displaced fine grains and individual clumps, and produced

partial erosion of some of the clumps that were too large to be moved. The radius of disturbance was 15 cm for the 4-second firing, and 25 cm for the 60-second firing. The fact that no crater was formed implied that the dimple beneath the mildly pulsed jet on Surveyor 1 had been coincidental.

Surveyor 6 was equipped with an alpha-particle instrument. This was powered up at 05:38 on 10 November. After two 10-minute calibrations of the standard sample between 05:41 and 06:21, activity was suspended for 3.5 hours in order to allow TV surveys to be conducted before the Moon set for Goldstone. Calibration of the alpha-scattering instrument resumed when Canberra took over. A total of 318 minutes had been obtained by 21:00, and at 21:18 Madrid commanded the head to deploy ready to measure the background. The first session began at 21:37, and lasted 33 minutes. Then operations reverted to Goldstone, which undertook TV work. The background measurements resumed at 05:00 on 11 November, and a total of 367 minutes of data had been obtained by 12:07. These operations were allowed more time than in the case of Surveyor 5, whose preliminaries had been abbreviated. The head was finally lowered to the surface at 12:08, some 35 hours into the surface mission. The sample was undisturbed surface. There were few fragments exceeding

several millimetres in size, and the largest in the sampled area was about 1.5 cm in size. Some 7.2 hours of data had been obtained by 23:00, when the instrument was switched off in order to resume TV work. Data collection resumed at 07:48 on 12 November, and a total of 15.7 hours of data had been obtained by 19:55. Activity had to cease at 23:39, when the head exceeded its maximum operating temperature of 50°C. The instrument was off through local noon, but was able to resume sampling at 16:50 on 16 November, when the Sun’s elevation had decreased to 79 degrees and the shadow cast by the solar panel allowed the head to cool. By the time the instrument was switched off at 03:30 on 17 November a total of 30.5 hours had been obtained.

Surveyor 6 was to investigate further how the lunar surface was affected by rocket exhaust. The static vernier firing by Surveyor 5 had produced both viscous erosion and gas diffusion erosion – the latter resulting from the fact that the pressure on the surface was relieved suddenly as the engines were cut off whilst the vehicle was still on the surface. In the case of Surveyor 6, the engines were to deliver a greater thrust and for longer to emphasise viscous erosion, and because such a burn would lift the vehicle off the ground the pressure on the surface would be relieved slowly and thus minimise the disruptive effects of gas diffusion. And since the vehicle was to lift off, it had been decided to impart a horizontal displacement so that upon touchdown the camera would be able to view the original imprints made by the foot pads and the erosional effects of firing the verniers.

This ‘liftoff and translation’ was scheduled for 17 November. As a preliminary, high-resolution pictures were taken to document the state of the area immediately in front of the camera. As the Sun was high in the sky, the solar panel and high-gain antenna were temporarily repositioned to shade and cool the engines to a permissible

pre-ignition temperature. At 08:00 the flight control system was powered up for 35 minutes to verify its status. When the solar panel was stowed in order to prevent its being damaged by the stresses of the manoeuvre, this placed the vehicle on battery power. As the camera installed between legs no. 2 and 3 was on the east side of the vehicle, the flight control system was to fire vernier no. 1 at a lower thrust than the other two engines to make the vehicle lift off inclined at an angle of 7 degrees in the direction of foot pad no. 1, thereby displacing the vehicle to the west whilst causing the material eroded from the surface to be displaced preferentially in the opposite direction. Afterwards, the camera should have a good view of the erosional effects. At 09:46 the flight control system was reactivated, and at 10:32:02 the verniers were ignited and throttled to deliver a total thrust of 150 pounds. The intended period of firing was 2.0 seconds, but the cutoff failed and by the time the backup command took effect a total of 2.5 seconds had elapsed. The manoeuvre consumed 1.5 pounds of propellant. Once the telemetry had been examined to verify the systems, the solar panel and high-gain antenna were redeployed, and within 35 minutes photography had resumed.

The ‘hop’ lasted about 6 seconds, peaked at a height of 12.5 feet, and ended about 8 feet from the initial position in a direction slightly north of west. The vertical rate on making contact with the surface was 12.3 ft/sec, and the horizontal rate was 1.8 ft/sec – which was greater than that of the original landing and caused the foot pads to displace material as ejecta. The post-hop pictures showed the double imprints of pads no. 2 and 3 and the single imprints of the crushable blocks on those legs made at the time of the lander’s arrival. But because the vehicle rolled 5.5 degrees in an anticlockwise direction around its main axis during the hop the imprints of pad no. 1 ended up beneath crushable block no. 3 and thus were not visible for inspection. The imprint of the alpha-scattering head in between legs no. 2 and 3 was obliterated by the blast. At the initial landing, the verniers had been cut off at a height of 12 feet to minimise disturbing the surface, but for the hop they had been fired at even greater

A section of a panorama taken by Surveyor 6 after its ‘hop’, showing the original imprints and the erosional effects of firing the verniers. (Courtesy of Philip J. Stooke, adapted from International Atlas of Lunar Exploration, 2007)

thrust within a foot of the ground. Nevertheless, there was no evidence of explosive cratering – the surface was sufficiently cohesive to resist bearing capacity failure at the imparted gas pressure. Furthermore, although in the case of Surveyor 6 the pressure of the gas on the surface from firing the verniers was thrice that of the static test by Surveyor 5 and the higher pressure would have increased the diffusion into the surface, the rate at which the gas pressure on the surface declined as the vehicle rose was sufficiently slow to inhibit the gas diffused into the surface from escaping violently, with the result that the gas diffusion erosion was no worse than the static test. However, viscous erosion blew dark subsurface material across the undisturbed surface, and there was a striking pattern of fine rays radiating from below where the verniers had been when they ignited. Most of the displaced material was from where the surface had been previously disturbed by the foot pads and crushable blocks. The fact that 1-2-cm fragments left dark trails as they rolled on the undisturbed surface was evidence that the lighter-toned surface material was at most several millimetres thick. Some larger fragments were ejected on ballistic trajectories. One clod of fine­grained material splattered the photometric target on omni-directional antenna boom ‘B’, almost obscuring its pattern.

When Surveyor 6 arrived, the magnet on foot pad no. 2 had made no contact with the surface material. No changes were observed after firing the cold-gas thruster on that leg. But when pad no. 2 came into contact with the surface following the hop it penetrated to a depth of 10 cm, bounced and came to rest about 12 cm away, thereby not only leaving an overlapping imprint for the soil mechanics team to study but also finally giving the magnet scientists a coating of material to examine. The horizontal displacement from the hop also provided the camera with a baseline for stereoscopic

imaging. Later photogrammetric analysis enabled an extremely detailed topographic map to be produced extending out about 50 metres from the lander.

After the hop, the sensor head was observed to have come to rest upside down! It was switched on at 12:48 on 17 November and found to be too hot, so was turned off again at 12:52 and allowed to cool before undergoing a test to determine whether it could provide any worthwhile data in this orientation – it could monitor solar wind protons bathing the lunar surface, and was operated in this manner for a total of 13 hours between 18 to 20 November and 22 to 24 November, with this experiment concluding at sunset. The alpha-scattering instrument operated for a total of 108.3 hours during which it provided 59 hours of science data, but only 30.5 hours of this was of the surface material and 10 per cent of the data was rejected because it had a low signal to noise ratio – which left 27 hours of surface data for integration.

In the case of Surveyor 5, whose ad hoc descent had required the retro-rocket to operate to within 4,200 feet of the ground, it was conceivable that the aluminium abundance measured by the alpha-scattering instrument was inflated by efflux from the solid-propellant rocket motor. But Surveyor 6 jettisoned its motor at a height of 35,000 feet and measured essentially the same abundances, and this implied that the Surveyor 5 data was valid. The analyses at the two sites suggested that the elements in the lunar surface material were in the form of oxides, and formed compounds and minerals that were familiar on Earth. It was not pristine material condensed from the solar nebula. As in the case of Earth, the Moon has undergone significant chemical differentiation. Although it was concluded that the maria were of a basaltic composition, the data was insufficient to identify the particular type of basalt. The observations of the magnets on these landers were consistent with the fine-grained material being pulverised basalt with little (if any) admixed meteoritic iron.

On 19 November the oxidiser part of the vernier propulsion system developed a leak, possibly owing to the degradation of a rubber o-ring seal. This automatically opened the helium regulator to top up the pressure, which was impossible – with the result that by 25 November both the oxidiser and helium had been completely lost. This leak pre-empted a tentative plan to perform a second hop.

With sunset imminent, the lander recharged its battery to sustain itself through the lunar night. At 16:08 on 22 November the shock absorbers of the legs were locked in order to preclude the deflections suffered by Surveyor 5 when its unlocked legs relaxed upon being chilled.

Sunset was at 13:53 on 24 November. Over the next 6 hours, pictures were taken using the polarising filters to study the solar corona. Between 16:23 and 16:50, and between 19:05 and 19:28, pictures were also taken of foot pad no. 2 illuminated by Earthshine. At 19:03, at the start of the final 10-minute corona exposure, the upper limb of the Sun was about 10 solar radii below the horizon. Camera activity ended at 20:04. It sent some 14,500 pictures prior to the liftoff and translation experiment and by the time it was switched off it had provided a total of 29,952 pictures. The final data from the alpha-scattering instrument on the protons impinging on the Moon was obtained 4 hours after sunset. Temperature monitoring was concluded at 06:41 on 26 November, after 41 hours – it had been hoped to obtain 130 hours of such

A picture of the ‘horizon glow’ phenomenon taken by Surveyor 6 at 14:25 GMT on 24 November 1967, about half an hour after sunset. The sketch shows the position of the solar disk in relation to the horizon and the ‘beads’ at that time. The position of the Sun was determined in relation to the marked stars, the magnitudes of which are indicated beside the circles. The grid coordinates are relative to the digital frame. The diffuse glow is the solar corona.

data, but a problem involving the bimetallically activated switches in the thermally controlled compartments obliged the lander to hibernate early.

Surveyor 6 provided the first measurements of the polarisation of the solar corona out to 30 solar radii, which was several times further than was attainable for a solar eclipse seen from Earth. Pictures taken during the first hour after sunset revealed a surprising ‘horizon glow’. This consisted of a number of glowing segments along a 5- degree arc due west. As the Sun passed progressively further beneath the horizon, these disappeared in groups. Whilst the later exposures were longer than the initial ones in which the band of light was prominent, it had completely disappeared by the time the centre of the solar disk (which spans about half a degree) was 1.2 degrees below the horizon. In fact, this phenomenon had been photographed by Surveyor 1, but it was not recognised until after the Surveyor 6 discovery. One speculation was that the glow was the diffraction of sunlight by fine-grained material on the surface at the horizon. Another idea was forward scattering by particles possessing a mean size of less than 10 microns that were electrostatically levitated a fraction of a metre above the ground at the horizon. ft was impossible to draw a firm conclusion on the data available.

An attempt to awaken Surveyor 6 on 13 December was unsuccessful. Contact was re-established at 16:41 on 14 December, but was lost 3 hours later. Efforts to revive the lander continued until 21 December, and were then abandoned since sunset was once again imminent.

For millennia human beings have peered at the Moon in the sky and wondered what it might be. Within months of its establishment on 1 October 1958, the National Aeronautics and Space Administration set out to develop a program of robotic lunar exploration. In 1961 President John F. Kennedy raised the stakes by challenging his nation to land a man on the Moon within that decade. The resulting Apollo program dominated the agency’s activities throughout the 1960s and into the early 1970s.

It is impractical to cover all the strands of this effort in a single volume in equal detail. Nor can any given strand be properly appreciated in isolation. My approach is therefore to write a series of books, each of which applies a magnifying glass to a certain number of strands and glosses over others. This book focuses on what was known about the Moon at the dawn of the space age and details the robotic projects that paved the way for the first Apollo lunar landing, in particular the Surveyors that soft-landed to investigate the physical and chemical nature of the lunar surface and the Lunar Orbiters sent to reconnoitre possible landing sites.

As such, this book complements: Apollo – The Definitive Sourcebook, which was compiled with Richard W. Orloff and supplements an account of how the Apollo program was organised with the minutiae of each flight; How NASA Learned to Fly in Space – An Exciting Account of the Gemini Missions, which explains the key contribution that the Gemini crews made to the success of Apollo; and The First Men on the Moon – The Story of Apollo 11, which covers that mission from start to finish. In Exploring the Moon – The Apollo Expeditions, which I recently reissued in enlarged format, I detailed what the astronauts of each mission did whilst on the lunar surface. It also complements the excellent To a Rocky Moon – A Geologist’s History of Lunar Exploration by Donald E. Wilhelms, and the International Atlas of Lunar Exploration by Philip J. Stooke.

I used the mission reports as my primary source of information – there are many thousands of pages available on the NASA Technical Report Server. Millions of dollars were spent developing and flying the vehicles used to take close-up pictures of the Moon and, like the mission reports, until recently they remained in archives. I have assembled some of the contiguous photographic sequences taken by the Lunar Orbiters to illustrate the process by which the site for the first Apollo landing was selected. To my knowledge, they have never previously been made available to the public in this form. I have also freely intermixed units of measure, largely following the choice of the appropriate mission reports. Unless stated otherwise, all times are GMT in 24-hour format. Launch, parking orbit, midcourse and terminal phase times are usually specified to the nearest second, but for a Surveyor spacecraft’s powered descent the event times are specified to several decimal places.

In the 1960s NASA was a young and aggressive agency which embodied the ‘can do’ spirit of America at that time in tackling audacious engineering challenges with a tremendous sense of urgency – motivated by the desire to be the first to explore a new world. This is an account of a strand of that story that is often reduced to a few paragraphs in popular histories.

David M. Harland Kelvinbridge, Glasgow January 2009

As Silverstein at NASA headquarters had arranged things, JPL reported to him for the Ranger spacecraft, deep-space tracking and control, in-flight operations and data processing. The procurement of the launch vehicle would be managed by the Office of Launch Vehicle Programs. This was directed by Donald R. Ostrander, who, as a Major General in the Air Force assigned to NASA, was well qualified to liaise with the military and its contractors. Ostrander delegated the task of procuring Agena and Centaur stages to Wernher von Braun in Huntsville. By dividing the spacecraft (JPL and the Office of Space Flight Programs) from the launchers (Huntsville and the Office of Launch Vehicle Programs), this arrangement provided considerable scope for confusion and conflict.

On 29 December 1959 Associate Administrator Richard Horner created the Space Exploration Program Council with himself in the chair. It was to seek to improve the management of space flight projects, and to reconcile the inevitable differences that would arise between headquarters and the centres managing individual projects. Its members were Abe Silverstein, Donald Ostrander, and Wernher von Braun (launch vehicles), Harry Goett of the Goddard Space Flight Center (satellites in Earth orbit) and W. H. Pickering (deep-space missions). At its inaugural meeting on 10 February I960 it discussed an internal review sent to Ostrander on 15 January that warned of potential difficulties in the procurement of the Agena B, and how this might affect Ranger. It was decided that Silverstein’s technical assistant, William A. Fleming, should chair a steering committee. This Agena B Coordination Board was formed on 19 February, and drew its membership from von Braun’s team in Huntsville (which was in the process of transferring to NASA, and in July would become the Marshall Space Flight Center), the Goddard Space Flight Center and JPL. The Council also decided that a NASA project engineer should be assigned to the plant in Sunnyvale, California, where the Missile and Space Division of Lockheed manufactured the Agena. In Huntsville, Hans Heuter was made head the Light and Medium Vehicle Office, which was to manage procurement of the Agena B and Centaur stages, and Friedrich Duerr became its Agena Systems Manager. However, whilst von Braun’s team would plan and supervise procurement, the fact that the Air Force did not want an independent line of authority leading to its supplier meant the Ballistic Missile Division would implement procurement from the contractor. The Air Force Space Systems Division in Inglewood, California, of which the Ballistic Missile Division was a part, was commanded by Major General Osmond J. Ritland. In April 1960 Major John E. Albert was assigned to assist NASA in procuring the Agena B, which meant he had responsibility for all Air Force technical matters relating to Ranger. He would work with Duerr in Huntsville for the launch vehicle and Burke at JPL for the interface between the Agena and the Ranger spacecraft. Duerr sent Robert Pace to Sunnyvale as the resident project engineer. Lockheed appointed Harold T. Luskin to work with Albert and Pace. The final contract, which was agreed only on 6 February 1961, was for nine Agena B vehicles. In terms of a production line that was making Agenas for several Air Force programs, this was a small order – and it was treated as such by the company until NASA complained. As Huntsville was

responsible for NASA’s launch operations, in addition to procuring the Atlas-Agena В it had to obtain ground support equipment and the systems required to track the vehicle in its ascent to orbit.[12]

In March I960, as the Army Ballistic Missile Agency was being incorporated into NASA, Donald Ostrander’s Office of Launch Vehicle Programs created the Launch Operations Directorate to manage NASA launches in Florida. In essence it was an expansion of the Army’s Missile Firing Laboratory, and being based at Huntsville it answered to von Braun. Kurt H. Debus, Director of Launch Operations, was keenly aware that he was responsible for activities he could not actually control, because in reality NASA was merely a tenant at the Cape and as such was limited to monitoring the preparation and launch of the vehicles by the 6555th Aerospace Test Wing. His counterpart on the Air Force side was Major General Leighton I. Davis. That same month, the Office of Space Flight Programs set up its own office at the Cape to coordinate the on-site activities of the flight project teams.

On 1 September I960 Richard Horner resigned from NASA. He was succeeded as Associate Administrator by Robert C. Seamans. Noting criticism that the Agena В Coordination Board had proved ineffective at resolving disputes, Seamans ordered a review. On 19 October, Albert Siepert of the Office of Business Administration submitted A NASA Structure for Project Management. On 19 January 1961, the day before he left office with the other Eisenhower political appointees, Keith Glennan endorsed the recommendations. In this new scheme, Silverstein’s office would set budgets for flight projects, establish objectives and review progress. The Marshall Space Flight Center, reporting to Ostrander’s office, would provide launch vehicles and launch operations in support of a project manager at a field centre. In the event of disputes, Seamans would personally decide the issue. The Agena В Coordination Board was dissolved. This revision gave JPL direct authority and responsibility for Ranger. NASA named Burke as its Ranger Project Manager, thereby giving him greater authority than he had when he was simply JPL’s Ranger Spacecraft Project Manager.

When James E. Webb became NASA Administrator in February 1961, he argued that although the Air Force might procure the rockets for NASA, the agency should be wholly responsible for preparing and launching them. On 17 July the Air Force conceded that in due course NASA could install its own launch groups to supersede the 6555th Aerospace Test Wing.[13]

The three successful Rangers satisfied the objective set for the Block III series in terms of supporting Apollo. The maria proved to be cratered on all scales, but with a smoothly undulating surface of generally shallow slopes. And the presence of large blocks of rock lying on the surface suggested sufficient bearing strength to support a lander. In addition, radio tracking had enabled the estimate of the mass of the Moon to be much improved. It also established the axis that is aligned towards Earth to be about 1 km longer, with the centre of mass being offset several kilometres from the geometric centre in a direction away from Earth.

When some 200 scientists gathered at the Goddard Space Flight Center in April 1965 to discuss the combined results of the Ranger project, Harold Urey and Gerard Kuiper still disagreed about whether the Moon was thermally differentiated. Thomas Gold insightfully noted that the pictures represented a mirror in which each person saw evidence to support his own hypothesis.

One lesson of Ranger was that lunar geological units were so severely blurred by impact ‘tilling’ at the fine scale that in undertaking photogeological mapping it was better to use a medium scale of 1:1,000,000, as was used by the Air Force Chart and Information Center in St Louis for its Lunar Astronautical Chart series.

Ranger had provided a close look at several sites, but what was required next was for an orbiter to provide a broader view at better-than-telescopic resolution and for a soft-lander to provide ‘ground truth’.

With four mare sites in the Apollo zone visited, distributed more or less uniformly in longitude from 23°E to 43°W, it was possible to draw some generalisations.

Surveyor 1 inspected a level plain in an ancient 100-km-diameter crater known as the Flamsteed Ring that had been ‘inundated’ in some way by Oceanus Procellarum; Surveyor 3 landed in a subdued medium-sized crater situated on the open plain of Oceanus Procellarum; Surveyor 5 provided a detailed inspection of a very small irregularly shaped crater in Mare Tranquillitatis; and Surveyor 6 inspected the plain of Sinus Medii within sight of a mare ridge. All four sites were very similar in terms of topography, and in terms of the structure of the surface layer and its mechanical, thermal and electrical properties; and the surfaces at the latter two sites were similar in terms of elemental composition and the content of magnetic material. ft seemed unlikely that terrestrial sites situated thousands of kilometres apart and selected in a manner similar to that by which the lunar targets were chosen would prove to be so similar.

At all sites, the undisturbed fine-grained surface material was lighter toned than the subsurface. This difference was as much as one-third for Surveyors 1, 3 and 6, but less for Surveyor 5. The fact that the albedo of the subsurface was the same at all sites meant the exceptional case of Surveyor 5 was due to the surface material being less bright. Observations of the erosion of the fine-grained surficial material by the vernier efflux during the ‘hop’ performed by Surveyor 6 and of the tracks left by the

fragments that were rolled across the surface, indicated that the bright surficial layer was limited to the uppermost few millimetres. The existence of such a well-defined ‘contact’ in a nominally undisturbed surface at four widely spaced sites on the maria implied the action of a process (or combination of processes) which had the effect of increasing the albedo of the material at the surface, for otherwise such a fine layer would be destroyed by the gardening of meteoritic bombardment. Furthermore, the fact that the material at all depths below the surface was uniformly dark, as opposed to there being a gradation in albedo, indicated that whenever an impact mixed the lightened surficial material into the subsurface, it became dark. Perhaps the process which altered exposed material had not had long to act on the material in the small fresh-looking crater in which Surveyor 5 landed. At all sites, the bright rounded rock fragments visible on the surface had textures featuring knobs and pits, whereas these were absent on the highly angular faceted blocks. This hinted that the process which produced the rounding – undoubtedly the relentless meteoroid bombardment – also gave rise to the ‘worn’ texture.

At all sites the fine-grained material was cohesive, and whilst the surficial layer was mildly compressible, its bearing strength increased rapidly with depth. But there was no observable variation in grain size with depth – evidently it was simply a case of the porosity decreasing with depth. It was estimated that the bulk density of the upper centimetre of undisturbed material was in the range 0.7 to 1.2 g/cm3, and that by a depth of several centimetres this had increased to 1.6 g/cm3.

The size-frequency distribution of small craters at all sites matched that expected for a steady-state population resulting from the protracted bombardment of primary meteoroids and the fall of ejecta from such impacts. Furthermore, this distribution was independent of individual differences in the mare surfaces and of the population of craters larger than several hundred metres in size.

The thickness of the fragmental debris layer on the mare plains was clearly related to the abundance of craters with diameters ranging between 1 and 10 km. In the part of Mare Tranquillitatis where Surveyor 5 landed the size-frequency distribution of such craters was twice that of the Oceanus Procellarum inundation of the Flamsteed Ring where Surveyor 1 landed, and the minimum size of the blocky rimmed craters at those sites indicated that the fragmental debris layer was several times thicker in Mare Tranquillitatis than in Oceanus Procellarum. Of all the maria, Sinus Medii had one of the highest size-frequency distributions of craters with diameters larger than several hundred metres, and the fragmental debris layer on the plain near Surveyor 6 was thicker to match. The fact that the cratering indicated the surface of Sinus Medii to be older than the other maria was evidence that the older the surface the thicker its fragmental debris layer. The size distribution of the material on the surface was also related to the thickness of the fragmental debris layer. When the mare lava flow was fresh and its rocky surface was exposed, small impacts were able to excavate it. As a layer of fragmental debris accumulated, it took larger and larger impacts to reach the substrate. Over time, the loose fragments were both reduced in size and increased in number. The trend was therefore towards a thickening layer of ever finer fragments. That is, the regolith ‘matured’.

The implication for Apollo was that an older surface would be a safer landing site.

When viewed from afar, an older surface might look rough by virtue of having large craters with blocky rims, but since only large craters would be able to excavate the substrate this meant that the plethora of small craters (which must be present) would not possess blocky rims. The task for the site selectors was therefore to measure the smallest craters with blocky rims on a mare surface to measure the thickness of the fragmental debris layer, and then seek a flat patch of open ground situated between such craters where it was likely to be relatively free of blocks.

CLASSICAL PHILOSOPHERS

Greek astronomy began with Thales, who was born shortly before 600 BC and lived in Miletus, a city of Ionia, which was a state on the western coast of what is now Turkey. As a philosopher he is regarded as one of the Seven Sages of Greece, and is considered to be the ‘father of science’. He set the seasons of the year and divided the year into 365 days. He also predicted a solar eclipse that occurred in 585 BC. It had been believed that the Moon was self-luminous, but he suggested that it shone by reflecting sunlight. Anaximander, a student of Thales, went to Italy in 518 BC. He opined that Earth floated in space – the prevailing view was that it was in some way supported on pillars through with the Sun passed during the night.

Pythagoras was born about 575 BC on Samos, an island off the coast of Ionia that was a crossroads between Asia, Africa and Europe. In his youth he reputedly visited Thales. Pythagoras considered the Moon to mark a fundamental boundary, in that it and everything ‘above’ was ‘perfect’, while Earth was subject to change and thus to decay. When critics argued that the markings on the face of the Moon indicated that it, too, was imperfect, it was suggested that the Moon was a mirror and the markings it displayed were really on Earth.

Around 450 BC Anaxagoras of Athens decided that Thales was correct in saying the Moon shone by reflecting sunlight. He realised that the Moon was spherical, and used this to explain its monthly cycle of ‘phases’. A generation later, Democritus, who travelled widely in ancient Greece, reasoned that the Moon was a world in its own right with a rugged surface, and he speculated that it might be an abode of life.

In the early fourth century BC, Plato, a student of Socrates, founded the Academy in Athens as the first institution of higher learning. Eudoxus briefly studied under Plato. After learning astronomy, he devised an explanation for the manner in which the constellations on view change with the seasons. He imagined the stars to be on a sphere that was centred on Earth, and the Sun to be on a slightly smaller concentric sphere made of transparent crystal which allowed the stars to be seen through it. The solar sphere turned around Earth on a daily basis, as did that with

the stars, but there was a slight differential in their rates that took a year to complete. Aristotle, another student of Plato, seized on this idea of ‘crystal spheres’ by proposing that there was one for each object that had an independent motion in the sky, and that their rotation was due to the action of angels. Although Eudoxus had envisaged crystal spheres only as a means of exposition, Aristotle believed them to be real and his views would come to dominate natural philosophy.

The points of light in the sky which move against the background of stars were called ‘planets’, meaning ‘wanderers’. In the third century BC Aristarchus of Samos suggested that the Sun might be located at the centre of the ‘planetary system’, with Earth being a sphere, rotating daily on its axis, and travelling around the Sun on an annual basis; but the idea attracted little support and was soon forgotten. Aristarchus also reasoned that because the Moon occults the Sun at a solar eclipse, the Sun must be further away – in fact, much further away. He also inferred that the stars must be considerably further away than the Sun, because they show no parallax when viewed from opposite sides of Earth’s path around the Sun. However, his reasoning on these matters was ignored. He interpreted a lunar eclipse as the Moon’s passage through the shadow cast by Earth, and made a fair estimate of the distance between the Moon and Earth in relation to the diameter of Earth. His contemporary, Eratosthenes of Cyrene, made the first realistic estimate of the Earth’s true diameter, thereby providing a scale to Aristarchus’s calculations.

At the end of the third century BC, Apollonius of Perga on the southern coast of modern Turkey was a Greek geometer with an interest in conic sections, and it was he who introduced the names to the ellipse, parabola and hyperbola. Although it was inconceivable that celestial objects should be less than perfect, detailed observations had shown their motions to be anomalous. Apollonius devised a geometrical scheme in which a celestial body would trace a small circle whose central point travelled in a circle around Earth; the small circle was termed the ‘epicycle’, and its centre was the ‘deferent’. This allowed the Moon to appear at times to lead and at other times to trail its perfect position. Furthermore, this accounted for why the size of the Moon appeared to vary in a cyclical manner. And of course, because the scheme involved only circles it restored purity.

Hipparchus, a Greek living in Alexandria, Egypt, in the second century BC, was the greatest of the classical Greek astronomers. His legacy was a star catalogue, but he also used a solar eclipse to estimate the relative distances of the Sun and Moon to a greater accuracy than had Aristarchus. He reasoned that although the Moon must orbit the Earth’s centre, the location of observers on the Earth’s surface provided the basis for parallax. On scrutinising records of eclipses that had been observed from both Alexandria and Nicaea, which lie on the same meridian but are some distance apart, he used the extents to which the Moon had masked the Sun’s disk to calculate the distance to the Moon relative to the Earth’s diameter. In fact, he calculated the distance of the Moon to within a few thousand kilometres and its diameter to within several hundred kilometres – although obviously he didn’t use kilometres as a unit of measure. Hipparchus also used measurements of the Moon’s orbit to assess Apollonius’s suggestion of deferents and epicycles, found it satisfactory, and provided measurements of the sizes of the epicycles.

In 80 AD the Greek historian Plutarch, who became a citizen of Rome, wrote the philosophical treatise Faces in Orbe Lunare in which he discussed the motion of the Moon across the sky, and how it maintained one face towards Earth as it turned on its axis. He thought that it was a world similar to Earth, and suggested it might be inhabited. A generation later, this latter point led the Greek storyteller Lucien of Samosata to write Vera Historia describing how a whirlwind lifted a ship from the sea and deposited it onto the Moon, where there was a battle in progress between the local inhabitants and invaders from the Sun. The story was a satire on the wars raged by the Greeks.

Claudius Ptolemaeus was born around 85 AD, probably in Alexandria, which was at that time under Hellenistic control. The Royal Library of Alexandria was founded at the start of the third century BC. Over the centuries it had built up an unrivalled catalogue, because whenever a ship docked in the harbour the authorities ordered copies made of any books that were on board. Ptolemy (as he is known in English) used his own observations of the stars and the resources of the library to refine the work of Hipparchus, and wrote up his findings in a book of his own. The library was sacked several times and eventually destroyed, but when this occurred is disputed. Although Ptolemy’s book was lost, an Arabic translation survived as the Almagest. He accepted Earth to be centrally located, celestial objects to be travelling in circles, Aristotle’s belief in the reality of concentric celestial spheres, and also Hipparchus’s endorsement of the deferents and epicycles as the reason for the anomalous motions. The Church of Rome accepted Aristotle’s philosophy, and so, despite its contrived nature, the ‘Ptolemaic system’ – as it became known, even although Ptolemy had not invented it – survived for over 1,000 years.

In March 1958 Major General John B. Medaris of the Army Ballistic Missile Agency requested JPL to compute the payload that could be dispatched into deep space by a configuration of the Jupiter launch vehicle that would later be named the Juno IV.

The result prompted W. H. Pickering to ask Daniel Schneiderman, head of a payload design group that included James Burke, to study preliminary concepts for a spacecraft capable of a flyby of Mars. They outlined a З-axis stabilised vehicle that would face flat arrays of transducer cells to the Sun for power and maintain a high – gain antenna pointing at Earth. It was decided the craft would require a small rocket engine to correct a modest trajectory error inherited from the launcher. These points were accepted by a review in June. When the Juno IV was cancelled in October, JPL proposed that the Atlas-Vega be developed with the capability to dispatch 265 kg to the Moon or 200 kg to either Venus or Mars. To reduce the development costs and improve reliability, JPL decided that all deep-space missions should use a common ‘bus’ to provide not only the main structure but also electrical power, command and control functions, communications, З-axis attitude control and midcourse correc­tion. Peripheral structures would be held against the bus for launch, and be deployed in space.

When in July 1959 NASA told JPL to focus on the Moon in the short-term, the laboratory decided to stick with the bus concept so as to exploit this development to prepare for planetary missions in 1962. This made Ranger much more sophisticated than strictly required for a flight to impact on the Moon. Although recognised to be a high-risk venture in the short term, this strategy was expected to pay off in the long term. The preliminary outline was submitted to NASA on 1 August. The scientific payload for Ranger was to be specified by the Office of Space Flight Programs, and Silverstein delegated this task to Homer Newell. By the autumn, Newell’s working groups and JPL had agreed on a rough priority for experiments to be carried on six Vega-launched lunar flights. The first two spacecraft, designated Block I, would not be sent towards the Moon, but would test the basic spacecraft systems in the deep – space environment by using Earth orbits with apogees significantly beyond the orbit of the Moon. The spacecraft engineers had argued for not carrying any experiments, in order to increase the mass available to provide redundancy in key systems until their reliability could be determined, but Newell insisted that such orbits provided an excellent opportunity to study particles and fields in space. The Block II spacecraft would be devoted exclusively to studying the Moon. However, then the Vega was cancelled and Silverstein directed JPL to prepare a five-flight series for the Atlas – Agena B. On 28 December 1959 Pickering proposed that the bus it had planned for the Vega-launched spacecraft be revised to suit the Agena B, and NASA concurred. Once again, the first two missions would be test flights carrying particles and fields payloads.

Daniel Schneiderman issued the design concept for Ranger on 1 February 1960. Reflecting the fact that the frame of the third stage of the Atlas-Vega was to have been hexagonal, the bus was a hexagonal disk 1.5 metres in diameter. The systems would be contained in rectangular boxes on its sides. For the test flights, a ‘tower’ was to be mounted on top of the hexagon with a platform to support the experiments and a fixed low-gain antenna on the tip. It would not, however, have the midcourse engine. On the underside would be a wider hexagonal frame scaled to mate with the Agena, and this would have a pair of solar panels that would be held against the bus for launch and hinged out in space, and a 1.2-metre-diameter high-gain antenna dish

ELECTRONICS

BAY VI

Details of the Block I Ranger spacecraft.

that would be stowed directly beneath the bus for launch and gimballed out in space. The Systems Division at JPL completed the design of the Block I in May I960, and Burke froze it. The design was then split up in terms of functions and the individual tasks assigned to the various divisions. These missions were to provide data on flight performance in space that could not be gained by testing on the ground – at least not with the facilities then in existence.

The Block II spacecraft would have the same hexagonal frame, solar panels and high-gain antenna, but include the midcourse engine and a structure to accommodate a rough landing package. The operational flights were to activate a TV camera and a gamma-ray spectrometer during the terminal approach and,

OMNI ANTENNA

MAGNETOMETER

ION CHAMBER

LYMAN ALPHA TELESCOPE

COSMIC DUST DETECTOR

ELECTROSTATIC ANALYZER
PITCH & ROLL JETS

VELA HOTEL EXPERIMENT

EARTH SENSOR

ANTENNA GEAR BOX

YAW JETS

ELECTROSTATIC

ANALYZER

FRICTION EXPERIMENT

SUN SENSOR

SOLAR PANEL

HIGH-GAIN ANTENNA

ELECTOSTATIC ANALYZER

The configuration of the Block I Ranger spacecraft.

shortly prior to impact, eject the landing package that contained a single-axis seismometer. Conceived at a time when America’s only experience in deep space was the Pioneer probes, this was an extremely ambitious design.

JPL contracted out the development of the surface package. It was an experiment, and as such represented a distraction to the spacecraft engineers since it would not directly contribute to the exploitation of the bus for planetary missions. In February I960 JPL issued three competitive contracts. It received the proposals on 15 April, and on 25 April selected the one submitted by the Aeronutronic Division of the Ford Motor Company. The design mounted the spherical capsule above a solid-propellant
retro-rocket (later revised to a liquid engine) and at an appropriate height above the Moon a pulse-type radar altimeter dish antenna would command separation from the bus. At burnout, the retro-rocket would be jettisoned. A crushable shell of balsa would protect the fibreglass capsule from the impact, and the scientific payload would be immersed in high-viscosity fluid. Once the capsule had come to rest, the offset centre of mass of the payload would allow it to adopt an upright orientation within the fluid. Including its support structure, the surface package subsystem had a mass budget of 136 kg. The propulsion was contracted to the Hercules Powder Company, and the altimeter to the Ryan Aeronautical Company. The contract for the battery powered single-axis seismometer had been placed in July 1959 with Frank Press at the Seismological Laboratory at Caltech and Maurice Ewing at the Lamont-Doherty Geological Observatory at Columbia University. It was a magnet suspended in a coil by a spring, and restrained radially so that it would respond only to motion parallel to its axis. It weighed 3.6 kg, and by floating it in a viscous fluid inside the ‘survival shell’ of the 44-kg capsule it could withstand the deceleration force of 3,000 times Earth gravity on hitting the ground at 60 m/s. Aeronutronic’s contract called for the first surface subsystem to be delivered by September 1961. The company formed a Lunar Systems group headed by Frank G. Denison specifically for the project, and he was to report to Burke at JPL.

The Ranger spacecraft would require 100 to 150 watts of power when fully active, which was to be generated by a pair of solar panels. Attitude determination would be by photocells and gyroscopes. Attitude control would be by thrusters squirting cold nitrogen gas. After being released by the Agena, the spacecraft would deploy its solar panels and orient itself to face its longitudinal axis towards the Sun for power generation, then aim its high-gain antenna at Earth. Later, it would have to adopt the attitude required for the midcourse manoeuvre, then re-establish its cruise attitude. This burn would be made by a hydrazine monopropellant rocket engine. Because the engine was installed beneath the hexagonal framework, it could not be fired until the high-gain antenna had swung clear. The autopilot would hold the vehicle’s attitude during the manoeuvre using vanes in the rocket exhaust. For the Block I, a simple ‘alarm clock’ would prompt a sequencer to perform the various actions of deploying the solar panels, locking onto the Sun and deploying the high – gain antenna. For the more demanding Block II, a ‘computer’ would drive the sequencer – but the attitude for the midcourse manoeuvre and the magnitude of the burn would be specified by Earth. The low-gain antenna was for use after the craft was released by the Agena and before it adopted cruise attitude, and later for the midcourse manoeuvre during which the high-gain antenna would not attempt to hold its ‘lock’ on Earth. The high-gain antenna would send telemetry while cruising, and scientific data (including TV) during the terminal approach. On the Block II, a boom would swing the low-gain antenna away from its initial position above the surface package shortly prior to the terminal approach – by which time the high-gain antenna would have locked on – in order to clear the way for the separation of the surface package, which would occur when the radar altimeter indicated that the slant range had reduced to 24 km.

The Vega-launched Ranger had been allocated a launch mass of 364 kg in order to match the predicted performance of the three-stage Atlas-Vega. NASA’s decision to switch to the Agena B for deep-space missions was based on the expectation that the performance of the Atlas-Agena B would almost match this. But whereas the Air Force’s Agena payloads operated in Earth orbit, NASA wanted the Agena to remain in ‘parking orbit’ only briefly before reigniting its engine to head for deep space. On 11 July I960 Lockheed announced a cut in its estimate of the Agena B’s capability in this role by 34 kg to 330 kg. This did not affect the 307-kg Block I, but the design of the Block II had started out as 364 kg – including the 136-kg surface package subsystem. The issue of the Block II mass remained in doubt through the remainder of the year, with the Agena B Coordinating Board failing to achieve a resolution. Meanwhile, James Burke told Harris Schurmeier’s Space Division to do all it could to lighten the Block II, short of deleting equipment. But by the end of 1960 the mass of the Block II design still exceeded 330 kg. Worse, Aeronutronic warned that their subsystem may well exceed its mass limit. On 14 December the Space Technology Laboratories were asked to re-evaluate the capacity of the Agena B by combining Lockheed’s data for launch to translunar injection and JPL’s data for the remainder of the mission. The calculation would take several months to perform.

In early 1961, awaiting this report, Burke faced a dilemma. To match the overall project timescale of 36 months specified by Silverstein, the schedule called for the test

flights in July and October 1961, and the lunar flights in January, April and July 1962. To achieve this, he would have to commit the Block II to a mass heavier than the currently projected capability of the Atlas-Agena B. Alternatively, he could let the schedule slip until he had a definitive mass figure. Part of NASA’s rationale for Ranger was to steal a march on the Soviets. However, on 12 February 1961 they dispatched a 454-kg spacecraft on an interplanetary trajectory towards Venus. It was З-axis stabilised, had solar panels and a high-gain antenna, and utilised the parking orbit technique rather than direct ascent – all of which JPL had hoped Ranger would pioneer.[14] Under pressure, on 16 February Burke told Schurmeier that the Systems Division “must begin removing items’’ from the Block II in order to lighten it. He also told Aeronutronic that if their subsystem exceeded its allotted mass it would not be carried.[15] The deletions that Burke specified would cut the mass of the Block II to ЗЗ2 kg, which was still marginally greater than Lockheed’s revised estimate. The design of the Block II was frozen in April 1961 at this mass. Ironically, in late May 1961 the Space Technology Laboratories reported that the Agena B would be able to place З82 kg on the Block II trajectory, which was even more than the Vega figure! Rather than introduce a further delay by trying to exploit this situation, Burke let the development proceed. If the actual capability of the Atlas-Agena B in this role had been known when JPL drew up the specifications for the Block II in early 1960, a significant degree of redundancy would have been built into the spacecraft. In fact, the mass-saving exercise had deleted those redundant systems that existed, making it even more technologically risky than initially envisaged. In effect, the design of the Block II was predicated on the assumption that all of its systems would work, which in turn put it at risk of loss if a single system were to fail.

In January 1958 Joshua Lederberg, a geneticist at the University of Wisconsin, warned the National Academy of Sciences that if a spacecraft transported terrestrial microbes to another body of the solar system, this would preclude a later experiment to determine whether life independently originated there. The International Council of Scientific Unions established an ad hoc committee to consider the issue, and then urged all nations to sterilise their spacecraft. On 15 October 1959, with the approval of Keith Glennan, Abe Silverstein directed that any NASA “payloads which might impact a celestial body must be sterilised before launching’’.

The case for such a precaution on planetary missions was self-evident, because at that time the seasonal variation of the dark areas on Mars was widely believed to be due to vegetation, and although the atmosphere of Venus was permanently cloudy, one idea was that conditions on the surface might resemble the carboniferous period of Earth’s past. As regards the Moon, one theory (not widely held, but impossible to falsify with the data available) posited there might be water ice at a shallow depth beneath the surface, and this environment might be conducive to microbial life that would not be able to survive on the surface. It was therefore decided that spacecraft destined for the Moon must be sterilised.

JPL soon decided that NASA’s recommendation of immersing the spacecraft in lethal gaseous ethylene oxide would have to be augmented by ‘dry heat’ treatment. In April I960 George L. Hobby, a research biologist in the Space Sciences Division, was assigned to work with James Burke to draw up sterilisation procedures for the Block II spacecraft. The first task was to define the term ‘sterile’ – which basically was an issue of deciding how efficient the process had to be. Ideally all components and subassemblies would be subjected to a temperature of 125°C for a period of 24 hours, but this proved impractical owing to its deleterious effects on the electronics, and one by one waivers were issued to protect particularly sensitive items. Once the spacecraft had been fully assembled (using the sterilised subassemblies) it would be thoroughly cleaned with alcohol. It would be transported to the Cape in a controlled environment. After the spacecraft had been installed on its launcher and passed its final checks, the aerodynamic shroud would be sealed and gaseous ethylene oxide pumped in for a time to complete the sterilisation process. On 26 June 1961 Robert Seamans approved these procedures.

As a matter of policy, the choice of experiments for space missions was made by Homer Newell’s Space Sciences Steering Committee. Abe Silverstein retained the final approval. Experimenters therefore submitted proposals to NASA headquarters, and, if successful, worked with a field centre to implement the experiment. In the case of JPL, this was the Space Sciences Division. While James Burke concentrated on the spacecraft, the launch vehicle and its support systems, Albert Hibbs prepared the scientific experiments to be carried. The particles and fields experiments did not require much development because they were by now being flown widely, but a lot of care was required to integrate them into the spacecraft. The final list of scientific experiments for the Block I was a solar plasma detector, a magnetometer, a trapped – radiation detector, an ion chamber, a cosmic-ray telescope, a Lyman-alpha detector, a micrometeoroid detector and the Vela Hotel package. The Vela Hotel provided by the Atomic Energy Commission was a late addition which Silverstein approved on 29 June 1960. A network of satellites were to carry X-ray and gamma-ray sensors to detect above-ground nuclear tests. But if the Sun issued microsecond-duration bursts of X-rays, then these might cause the satellites to report false detections. The highly elliptical orbits of the Ranger test flights would have their apogees above the van Allen belts, and the 3-axis-stabilised spacecraft would enable the Vela Hotel sensor to ‘stare’ at the Sun to determine whether it produced such emissions.

The in-flight experiments for Block II consisted of a TV camera and a gamma-ray spectrometer. The TV camera was to be activated by command from Earth when the spacecraft was within 4,000 km of the Moon. The Astro-Electronic Division of the Radio Corporation of America in Hightstown, New Jersey, which had supplied the TV system for the Tiros meteorological satellites, was hired to provide a slow-scan vidicon imaging tube and related electronics. The Space Sciences Division at JPL built the optical element, which was essentially a telescope with a focal length of 1

metre and an aperture ratio of f/6. The image scanned off the vidicon tube would have 200 ‘lines’. The plan was for the camera to provide about 100 pictures during the terminal approach, with the transmission terminating when the separation of the surface package perturbed the attitude of the bus and the high-gain antenna lost its lock. In excellent ‘seeing’, the best telescopes had a lunar surface resolution of about 300 metres. An image taken by this camera at an altitude of 50 km was expected to provide a resolution 100 times better. On 16 October 1961 Newell’s Space Sciences Steering Committee named Gerard Kuiper, Gene Shoemaker and Harold Urey as the experimenters who would receive and interpret the pictures transmitted by this camera.6

The gamma-ray spectrometer experiment was led by James R. Arnold, a chemist at the University of California at San Diego. Its scintillation counter was to detect the natural radioactivity originating from the uppermost layer of the lunar surface. In particular, it would detect gamma rays issued by the decay of uranium, thorium and potassium. If these large-ion lithophile elements were widespread, this would imply that the interior of the Moon had undergone significant thermal differentia­tion. The instrument was mounted on an 18-metre-long boom that would be deployed after the midcourse manoeuvre in order to determine the ‘background’ from the spacecraft itself and from celestial sources. It would operate until the high- gain antenna lost its lock. In fact though, such a study would be better done from a polar orbit in order to obtain global coverage.

On 27 March 1961 Walter E. Brown, head of the Data Automation Systems Group of JPL’s Space Sciences Division, pointed out that if the signal from the radar altimeter were to be telemetered to Earth, the radar echo could be correlated with the imagery to gain insight into the density, conductivity and thickness of the material at the surface – in particular its dustiness. This information would be of use to the team planning the soft lander. On 28 April, after it was established that this modification would not adversely affect operations, James Burke authorised the modification. The Space Sciences Steering Committee in Washington duly designated Brown as the investigator for the radar reflectivity experiment.

THE SECOND GENERATION

After the success of Luna 3, the Soviets developed a new spacecraft designed to deliver a capsule to the lunar surface using the rough landing technique. Luna 4 was launched at 08:16 GMT on 2 April 1962, and after cruising in parking orbit it set off for the Moon. After an ineffective midcourse manoeuvre, the 1,422-kg vehicle made a flyby at 13:25 on 5 April at a range of 8,500 km and passed into solar orbit. After several further failures, the Soviet Union conducted a deliberate flyby mission.

FILLING THE GAP

Zond 3 lifted off at 14:38 GMT on 18 July 1965. After parking orbit, it was sent on a trajectory to pass by the illuminated leading limb of the Moon. Imaging began at 01:24 on 20 July at an altitude of 11,570 km and ended at 02:32 at 9,960 km, with the closest point of approach at 9,220 km. It had been intended to launch this probe in 1964 as a companion to Zond 2 on a mission to Mars, but it was held back. The pictures were not transmitted until the narrow-beam of the high-gain antenna was able to lock onto Earth, which occurred on 29 July at a range of 2.2 million km. The objective of this flight was to test deep-space communications for an interplanetary mission, and the Moon was merely a convenient photographic target. It transmitted two dozen pictures of Oceanus Procellarum and around onto the far-side to view the area which had not been visible to Luna 3.

The results indicated that although there were few maria on the far-side, and those were small, there were multiple-ring structures which for some reason had not been flooded by lava. The Orientale basin was seen in its entirety for the first time, since even at the most favourable libration barely half of it was observable in reprojected telescopic pictures. In addition to the concentric rings, there were radial patterns in evidence. There was a small patch of mare material inside the central ring, and small patches between the rings, but otherwise the entire structure was ‘on display’ in its

magnificence. It boggled the mind that Earth must once have been disfigured by such structures!