by Duncan Lunan

Recently I came across a one-volume collection of the late James Blish’s ‘Cities in Flight’ tetralogy of novels  (Gollancz, 2002, Fig. 1).   It reprints them in order of the events within, starting with They Shall Have Stars  (aka Year 2018)  and ending with The Triumph of Time  (aka Clash of Cymbals).  I can see why it’s been done that way, but I think there’s more pleasure in reading them in order of publication, starting with Earthman, Come Home, and ending with A Life for the Stars.  That way, Blish closes the loop, ending with the early life of a character who’s just died at the start of Earthman.  I’m doing the same with this sub-series of articles:  having stated two weeks ago that for over 60 years satellites have had to fit as best they could on to vehicles derived from intercontinental ballistic missiles, I’m now going to tell how it happened.  They could be read the other way round if preferred, which is why I haven’t labelled this ‘Part 3’.

For the Russian  (former Soviet)  space programme, and for the European Space Agency, the task can be done in a sentence.  Almost all Russian launches to orbit have been with what’s now called ‘the Soyuz booster’ and began as the R-7 ICBM  (Fig. 2).  It launched the first Sputniks, and with appropriate upper stages the first moonprobes and the one-person Vostoks starting with Yuri Gagarin’s, then the entire series of Soyuz missions beginning in the late 1960s, and it’s still going strong.  The more powerful Proton booster, originally for lunar and planetary probes, began operations at the same time but wasn’t fully shown to the West until 20 years later, while the N-1 Lenin booster for the one-man lunar lander was so secret that until the fall of the USSR the Soviets were still denying its existence.  Europe’s Ariane series of boosters has had an even simpler line of development, starting with Ariane I in 1979 and continuing to the Ariane VI now about to begin operations in 2024.  But in the USA, things were far more complicated.

That story begins with the surrender of Wernher von Braun’s rocket team to US forces at the end of World War 2, bringing them with as much V2 technology as possible to the USA under Operation Crossbow before Soviet troops overran the territory.  (Stalin ordered the execution of the general who failed to secure them, but Soviet designers managed perfectly well without them.  These days we no longer hear ‘Their Germans are better than America’s Germans’, but it was never anything but a myth.)  Over subsequent years the captured V2s were used for high-altitude research at the White Sands rocket range in New Mexico  (Fig. 3), and the Western Test Range at Cape Canaveral in Florida.  Early records were set by ‘Project Bumper’, the first liquid-fuelled two-stage launches, marrying V2’s with WAC Corporal upper stages built by the Jet Propulsion Laboratory in Pasadena  (Fig. 4). 

For the history of that fascinating period see Escape from Earth, by Fraser MacDonald of Edinburgh University  (Postscript, 2019).  What followed was the result of competition between the three US Armed Services and the commercial companies supplying them – something which could never have happened in the monolithic USSR and was much ridiculed there  (and here), until eventually the range of accessible technologies which it had generated gave the USA the lead.

The US Navy entered the game with a single-stage research vehicle called Viking, whose biggest innovation was the use of a gimballed motor for steering instead of the V2’s graphite vanes in the exhaust  (Fig. 5).  Viking 11 established an altitude record for a single-stage rocket of 156 miles, which stood for many years until smashed in 1958 by Britain’s Black Knight, which reached 300 miles in its first launch, back when we had a space programme  (Fig. 6).  The record of 250 miles set by Project Bumper in 1949 was unbroken until 1956, when von Braun’s Jupiter-C reached 400 miles, but its 5000-mile horizontal range caused far more concern by that time.  The history of the Viking programme up to Viking 12 is grippingly described by project head Milton W. Rosen in The Viking Rocket Story  (Faber & Faber, 1956).  Viking was selected for development as the first stage of the Vanguard booster, to launch the USA’s scientific satellites during the International Geophysical Year, 1957-58.  Great emphasis was placed on its background  as a research booster and not a weapon:  in researching The Making of a Moon  (Muller, 1957),  Arthur C. Clarke was repeatedly beseeched “Don’t call it a missile, it’s a vehicle”. 

Viking 13 and 14 were launched from the Western Test Range with dummy Vanguard upper stages  (Fig. 7), and by 1957 Vanguard first stages were being launched with them.  But everything changed with the unexpected launch of Sputnik 1 in October 1957, and the Vanguard team was pressured into attempting an ‘all-up’ test with upper stages and payload, only to suffer a catastrophic failure  (Fig. 8).  Vanguard went on to a successful career as a satellite launcher  (Fig. 9), but was completely overshadowed by other events, and further development of the booster was turned over to the US Air Force  (see below).

Wernher von Braun’s Redstone Arsenal, Huntsville, had been ready to launch a satellite since their long-distance ballistic flight in 1956. but repeatedly were refused permission because of the military background.  After Sputnik von Braun promised a satellite launch in 60 days and actually achieved it in 90, with Jupiter-C as the launcher  (Fig. 10). 

The first stage of the stack was the Redstone Intermediate Range Ballistic Missile  (Fig. 11), one of the last rockets to use graphite vanes for steering, which performed so well in service with the US Army that it was nicknamed ‘Old Reliable’, after the Yellowstone Park geyser.  That made it a natural for the first US manned launches into space, with Alan Shepard and Virgin Grissom, in 1961  (Fig. 12).  The remaining Redstones were turned over to Australia for high-altitude research from Woomera, where one of them launched Australia’s first satellite, WRESAT-1  (Fig. 13).  In the early 2000s that booster and others were recovered from the Bush by a team from Sydney University headed by Dr. Kerrie Dougherty, and it was found that in atmosphere entry or weathering the white paint had flaked off to reveal the distinctive US Army olive-green colouring below.

Von Braun’s team had also produced the Jupiter IRBM, which was deployed in Italy and Turkey;  the removal of Jupiter missiles from Turkey was part of the settlement with the USSR in the aftermath of the Cuban missile crisis.  Jupiters carried several payloads for biological research, including the monkeys Able and Baker in 1959  (Fig. 14).  The press kept getting confused between the Jupiter IRBM and the Redstone-based Jupiter-C which had demonstrated the feasibility of ICBMs, so the Jupiter-based launcher for lunar flights was designated Juno II  (Fig. 15)  and the Jupiter-C was retrospectively renamed Juno I.  Media confusion then became total.

Juno II launched the USA’s first successful moonprobe, Pioneer 4.  But von Braun had much bigger aims in mind.  Clustering Redstone and Jupiter tanks above 8 more powerful engines, generating 1.5 million pounds of thrust, he created the Saturn I.  It attracted the derision of a contributor to New Scientist  (‘a rocket that everyone talks about it as if it were in the last stages of proving’), predicting at least 10 launch failures before success.  In the end, Saturn worked perfectly at its first launch  (Fig. 16)  and every time thereafter, a success rate matched only by Britain’s Blue Streak – just a reminder that we were still in the game, with not so many fewer launches, until around the end of Project Apollo.  The hoped-for Saturn missions to Mars and Venus, the original Voyagers, never materialised.  But the man-rated Saturn I-B carried Apollo 7  (Fig. 17), the three crewed missions to the Skylab space station, and finally the Apollo-Soyuz rendezvous mission  (‘Apollo through Binoculars’, ON, July 24^th 2022.) 

Rocketdyne’s development of the F-1 engine, generating as much thrust as all 8 of Saturn I’s, provided the power needed to take Apollo to the Moon along with the Lunar Module for the landing.  Five F-1s were clustered for Saturn V;  other variants, which never flew, included the Saturn C-3, to assemble the lunar mission by Earth Orbit Rendezvous, and the Nova, which would have had 8 F-1 engines and been capable of landing the whole Command and Service Module on the lunar surface  (Fig. 18). 

But in the end only Lunar Orbit Rendezvous provided a fast enough way to fulfil Kennedy’s challenge to land men on the Moon ‘before this decade is out’.  Saturn V worked perfectly on every launch, including the Skylab space station  (Fig. 19), and could have been the heavy lift workhorse for manned spaceflight through to the end of the century, building bigger space stations to orbit both Earth and Moon, build lunar surface bases and mount the first expeditions to Mars.  But the entire post-Apollo programme was cancelled by the Nixon administration, with the exception of a beefed-up version of the Space Shuttle to meet the requirements of the US military – who nevertheless went back to using expendable boosters after the 1986 loss of the Challenger.

The US Navy began to develop the Titan ICBM from Vanguard, but the programme conflicted with development of the Polaris submarine-launched missile, and the Air Force took it over – reluctantly, and being accused of ‘competing with itself’ when it already had the Atlas  (see below). There was no quicker way to fill the ‘missile gap’ with the USSR which filled the headlines.  We now know that was entirely fictitious, concocted by John Foster Dulles out of sabre-rattling speeches by Nikita Khrushchev, taken out of context.  Viewed purely as missiles, there was little to choose between Atlas and Titan I  (Fig. 20):  both were stored below ground in silos, but had to be raised to the surface for fuelling, making them vulnerable to first or second strikes in a nuclear exchange. 

Titan IIs could be launched from within the silos  (Figs. 21 & 22), which had been taken over unaltered from Britain’s Blue Streak programme.  The UK silos and with them the independent missile deterrent had been abandoned when it was found that almost all the suitable strata were within safe Tory seats, except for one in the grounds of Sandringham  (C.N. Hill, A Vertical Empire, The History of the UK Rocket and Space Programme, 1950-1971, Imperial College Press, London, 2001).. 

Another unusual feature of Titan II was ‘fire-in-the-hole staging’, where the second stage engine ignites before separation from the first stage.  It has technical advantages but imposes extra loads on the structure.  Titan IIs used hypergolic propellants, which ignited spontaneously on contact, causing fatal accidents with leaks below ground.  Nevertheless, what led to the missiles’ withdrawal was not the safety of the launch crews, but an in-depth survey which revealed that more than half of them were not willing to fire if ordered to do so.  In the film War Games  (1983)  the military response is to automate the launchers, making them vulnerable to hackers and gamers, but in fact they were pulled out in the mid-1980s and replaced by solid-fuel Minuteman ICBMs.  Nevertheless the image of liquid-fuel missiles in underground silos is deeply embedded in public consciousness and continues to turn up in films and TV dramas.  In Star Trek: First Contact  (1996)  the first starship is launched by a Titan II from a silo, fire-in-the-hole staging and all, and I recently saw another ‘silo’ episode in Scorpion.  Earlier, Titan launches featured in the last episode of The Prisoner and at the end of Dr. No, although the commentary there was from John Glenn’s MA-6 launch in Project Mercury.

In space launches, the Gemini programme used Titan IIs throughout  (Fig. 23), and it would have been used for the X-20 aerospace plane  (Fig. 24), had that not been cancelled along with the USAF plans for the Manned Orbiting Laboratory.  It’s important to realise how advanced NASA’s and the USAF’s future plans were before the big cancellations of the late 60s and early 70s.  An MOL prototype was actually flown on a Titan III booster, topped by a Gemini capsule which had previously flown unmanned  (Fig. 25).  Titan III became the top of the range booster for large payloads in the 1970s, launching the Viking landers to Mars  (Fig. 26), the Voyager probes to the outer planets and beyond  (Fig. 27), and the new generation of large communications satellites led by Intelsat 5  (Fig. 28).  The still larger Titan IV was used mainly for spy satellites, though it did launched the Cassini-Huygens probe to Saturn in 1997, and its last launch in 2005 was a pair of geosynchronous satellites for the National Security Office  (Fig. 29).

Back in the late 1950s, the dispute between the USAF and Army over IRBM policy had led to commissioning their own, Thor, from the Douglas Aircraft Company  (Fig. 20, rear). 

Thors were deployed in East Anglia by the RAF  (John Boyes, Project Emily, Thor IBM and the RAF, Tempus, 2008 – Fig. 30), providing Air Traffic Controllers with amusement as Soviet Tu-104 airliners, equipped with photo-reconnaissance nose-cones like Britain’s Canberra PR-9, wandered ‘lost’ over Norfolk and Suffolk pretending to look for Heathrow.  One Aeroflot pilot, plied with drink by a British counterpart and asked about the nose-cone, would only reply dead-pan “It’s a design feature, the aircraft won’t fly without it”. 

The RAF missiles were eventually returned to the USA and converted for space launches, where they came into their own.  Thors launched the first US moonprobe attempts  (Fig. 31), with an ‘Able’ upper stage which had been the second stage of Vanguard, as well as the first interplanetary medium probe, Pioneer 5, in whose mission tracking by Jodrell Bank played a key rôle..  (The swapping of upper stages between boosters were an example of the versatility which the early rivalries had generated.)  Successful satellite launchers included the 3-stage Thor-Able-Star (Fig. 32, right), but the real success was Thor-Delta  (Fig. 33), which was used in different forms for over 300 launches, far more than all other boosters combined.  The usage ‘Delta’ for the whole vehicle caught on and the later more powerful variants were called Delta II, Delta III, Delta IV  (Fig. 34), and Delta IV Heavy, which launched the prototype Orion capsule intended for the return to the Moon  (Fig. 35).  Delta IV will be replaced by Vulcan-Centaur in 2024.

Along the way the late Prof. Terence Nonweiler’s dictum that aerospace designs should have ‘elegance’ had gone by the board.  When the prototype Space Shuttle Enterprise was rolled out at Kennedy Space Centre on its dummy tank and boosters, the spontaneous comment of the NASA Director was “My God, it looks clumsy”.  The Thor had elegance, particularly in its Thor-Able configuration, as witness the number of film and TV productions it was used in, from The Road to Hong Kong to an ITV drama about an attempt to plant a biological warfare payload on the Moon.  By contrast Delta IV Heavy is monolithic, not to say brutalist, and Vulcan-Centaur simply shapeless – see Part 1, Figs 4 and 15, ON, November 5^th 2023.  

‘Last scene of all, that ends this strange eventful history’ belongs to the Atlas, which was the first of the ICBMs and as it turns out, will be the last of the missile-derived boosters to fly.  The prototype was the MX-774 and pioneering lightweight, pressure-braced construction techniques  (Fig. 36).  The fin shape replicated the V2’s in order to use German wind-tunnel data, and German engineers laughed when they saw it because the shape had been chosen to go through standard German railway tunnels, not for aerodynamic reasons  (John L. Chapman, Atlas, the Story of a Missile, Gollancz, 1960).  One unusual aspect was that under Prof. Krafft A. Ehricke, Atlas was always envisaged as a space launcher as well as a weapon – he even designed a moonbase to be built using it.  Its success was due partly to its lightweight construction and partly to its unique ‘one and a half stage’ design, in which the two outer engines and their supporting structure were jettisoned on reaching the upper atmosphere, leaving a ‘sustainer’ engine to carry the vehicle into orbit – clearly shown in John Glenn’s photo of his Atlas booster, just after his Mercury capsule separated from it  (Fig. 37).  Atlas-Able boosters launched a second set of unsuccessful US Moon probes, but went on to launch all of the later moonprobes and planetary probes of the 1960s and early 70s, with Agena second stages and later the Centaurs with liquid-hydrogen propellant.  Atlas launched all the unmanned and manned spacecraft of Project Mercury  (Fig. 38), and the modified Agena rendezvous and docking targets for the later Gemini missions.

The end of the Cold War gave Atlas a new lease of life.  During the Moon race the most advanced rocket engines in the world, with ‘full regenerative cooling’, were built for the second stage of the Soviet N-1 Lenin Moon rocket  (Fig. 39). 

When the pretence that Russia was never going to the Moon got underway, the makers were ordered to destroy them.  But they couldn’t bring themselves to do it  (Fig. 40), and in the 1990s 60 of them were rediscovered and promptly snapped up by US manufacturers, for use on the Antares and Atlas V boosters.  The New Horizons mission to Pluto and Arrakoth flew on the final version of a missile originally built to destroy the USSR, now flying on Russian engines  (Fig. 41).  Renewed tensions with Russia have put an end to such cooperation, but the United Launch Alliance still has engines in hand to keep Atlas V flying till 2030 – at which point the saga of space launchers built out of military missiles will finally be over.

Current and Recent Launch Vehicles

NB.  The lower entries in this Table prioritise launch systems available or being developed in the UK.  A complete survey of launchers worldwide is beyond the scope of this article.

See also: Rockets, for November 5th and Rockets 2: Launch Costs Per Vehicle