Navaho
The Navaho intercontinental cruise missile project
The Navaho intercontinental cruise missile project was begun just after World War II, at a time when the US Army Air Force considered ballistic missiles to be technically impractical. The Navaho required a large liquid propellant rocket engine to get its Mach 3 ramjet up to ignition speed. This engine, derived with German assistance from that of the V-2, provided the basis for the rockets that would later take Americans into space.
It turned out that mastering the guidance and materials technology needed for a Mach 3 cruise air vehicle was actually more difficult than for a Mach 22 ballistic missile. In the end, the Redstone, Thor, Jupiter, and Atlas rockets were flying before their equivalent-range Navaho counterparts. However the Navaho program provided the engine technology that allowed the US to develop these ballistic missiles rapidly and catch up with the Russians. Navaho also developed chem-milling fuel tank fabrication techniques, inertial and stellar navigation, and a host of other technologies used in later space vehicles. It put North American Aviation, and its Rocketdyne Division, in a leading position that allowed them to capture the prime contracts for the X-15, Apollo, and Space Shuttle projects, thereby dominating American manned spaceflight for the next seventy years.
Completely unknown at the time was that the Soviet Union was developing two equivalent missiles, the Buran and the Burya. These were started later than the Navaho, but the Burya began flight test at about the same time. Like the Navaho, the Burya was canceled in favor of the intercontinental ballistic missile. Unlike the Navaho, it went on to a somewhat more successful flight test career.
In August 1945 the US Army Air Force defined requirements for post-war missile development in a classified document. On Halloween 1945, they invited 17 aircraft systems manufacturers to submit proposals for design studies of specific weapons. North American Aviation chief Dutch Kindelberger, facing a reduction in employment from 100,000 to 6,500 employees as wartime aircraft production orders were canceled, determined that missiles were the future and decided to make the necessary investments to win one of the contracts. He founded a corporate research lab, staffed with experts in such new fields as jet propulsion, rockets, gyros, electronics, and automatic control.
William Bollay, the branch chief for development of turbojet engines in the Navy's Bureau of Aeronautics, was identified as an appropriate leader of the lab. He arrived just in time to deal with the AAF request for proposals. Bollay decided to propose American development of the A4b or A9, a winged, boost-glide version of the German V-2, against the Army Air Force mid-range supersonic missile requirement. On 24 March 1946 NAA received letter contract W33-038-ac-1491 for this missile, designated MX-770, with a range of up to 800 km. The payload requirement was increased from the 900 kg of the A9 to 1360 kg three months later.
North American began its rocket research in the company parking lot, using a bulldozer scraper to protect adjacent parked cars from the inevitable exploding engines. They experimented first with a surplus Aerojet 4.5 kN motor. They then tried some small engines of their own devising, with up to 1.3 kN thrust. But as material came in from occupied Germany, it was obvious that the starting point would have to be the more advanced German technology. So in June 1946 Bollay proposed that North American first refurbish and test a complete V-2 engine system, to be provided by the government. This "Mark I" engine would be followed by redesign of the engine to American engineering standards and materials, followed by fabrication and testing of the "Mark II" American version.
By the end of 1946, two government-furnished V-2 engines arrived. In March 1947 North American rented a large tract of land in the western San Fernando Valley north of Los Angeles, in the Santa Susana Mountains, for use in testing of large engines. A rocket test center was built here, using $1 million of corporate funds and $1.5 million from the Air Force.
Meanwhile, North American had settled on a research and development program, based on additional information received on German breakthroughs in supersonic ramjets. Phase 1 consisted of a study of the German A4b and A9 boost-glide missiles; Phase 2 consisted of study of replacing the rocket engines of the design with supersonic ramjets in order to extend the range; Phase 3 would determine the size and type of the booster needed to boost the reengined A9 to ramjet ignition speed. By the end of the year Bollay had a staff of 43 working at his Technical Research Laboratory in Los Angeles, including 12 PhD's and 18 ME's.
As North American engineers tore down and reassembled the V-2 engines, it was clear that this Model 39 18-chamber engine was an engineering kludge, a prototype that was rushed into production because the ultimate planned engine was unavailable. By the end of the war the Germans had been testing a more advanced Model 39a single-chamber engine. So in the spring of 1947, it was decided to proceed immediately to design, construction and test of a new engine, the NA-704 Mark III, based on the German Model 39a. To assist in this, Bollay's team was free to draw on the expertise of the V-2 designers themselves, now working for the US Army - Wernher von Braun's team, including Walther Riedel, Hans Huter, Rudi Beichel, and Konrad Dannenberg. Dannenberg in particular had been intimately involved with the 'shower head' injector plate that was essential for the single-chamber motor. Dieter Huzel, a close associate of von Braun, was hired by North American as a full-time employee in order to better coordinate work with the German team. In September 1947, preliminary design of the Mark III began, aiming at the thrust of the V-2 engine but with a weight reduction of 15 percent.
In other technical areas, Bollay's engineers invented a Kinetic Double-Integrating Accelerometer (KDIA) design that allowed an inertial platform to also measure distance traveled. This breakthrough would allow long-distance unpiloted high-accuracy navigation. North American aerodynamicists discovered that the A9 swept-wing design was unstable at transonic speeds. The company selected an aft-wing with forward canard layout for the MX-770. So by June 1947 the navigation concept, the propulsion, and the aerodynamic layout of the missile had all been fundamentally altered from the A9 baseline. North American had spent $3.9 million on the project. Development of specific components - the N-1 navigation system, the Mark III engine - was initiated. Construction of the launch site for the Nativ test vehicle was begun at Holloman AFB in New Mexico.
In parallel with development of these new elements, construction of the Mark II copy of the 18-chamber V-2 engine continued, in order to obtain experience. Detailed design had begun in June 1947 with first drawing release and parts fabrication coming in September. Meanwhile test in the parking lot continued of a 14.7 kN test engine on which a variety of shower-head injector patterns could be tested.
In February 1948 the Air Force ordered a complete redesign of the MX-770 missile to increase its range from 800 km to 1600 km. This was necessary purely for political purposes - when the US Air Force was split from the Army the year before, it had been agreed that the Army would handle missiles under 1600 km range, and the Air Force over that. The MX-770 was now an Air Force missile, and therefore had to have a range over 1600 km. To achieve this, North American decided they would indeed have to change the missile from the A9 pure rocket boost-glide vehicle to one that would use ramjet engines for sustained Mach 3 cruise. This approach had also been studied by the Germans for advanced versions of the A9. The rocket engine was now used for initial boost and to get the vehicle up to the supersonic speed necessary for ramjet ignition. But even with the more efficient ramjets, the entire vehicle would have to grow by 33% to meet the new range requirement. Accordingly the design thrust for the Mark III engine increased from the 249 kN of the V-2 to 333 kN.
In February 1948 the missile was designated the XSSM-A-2. This design had reached 100% drawing release. It consisted of an integrated rocket booster, two ramjets for cruise, and XN-1 inertial navigation with a final dive on the target. But the inertial navigator drifted 1.6 km in accuracy for each hour of flight, which meant the missile could not meet the USAF 800 m CEP requirement.
Meanwhile the XSSM-A-2 was redefined by the customer as only the first step of a revised three-phase program for a family of vehicles using a rocket booster and ramjet cruise. Track, air, and vertical pad launch were to be studied. The first phase would produce a missile with a range of 1600 km while carrying a 1350 kg warhead (the XSSM-A-2). Phase two would produce a missile that could carry a 1350 kg warhead to a 3200 to 4800 km range. Phase 3 would be the intercontinental version, carrying a 4500 kg nuclear warhead to an 8000 km range. In order to obtain the necessary accuracy, North American began development of the XN-2 navigation platform, which coupled the XN-1 inertial system with a star tracker to ensure continued accuracy even in long-range flights. In order to achieve the longer ranges, North American began study of a version of the missile, which would use a separate, jettisonable rocket booster. This would allow the cruise stage to be ignited at near-cruise velocity, and to be filled with only ramjet fuel, which would vastly extend the range while the basic cruise missile remained nearly the same size.
On 26 May 1948 North American made the first launch attempt of the Nativ subscale technology demonstrator. Seven rockets were built in total. There is barely any information on what this rocket was like, and even very inconsistent information on the flight series. It is not even known what kind of rocket engine powered the vehicle (perhaps the 14.7 kN test engine used in the parking lot tests). One source speaks of six launches, another of four, and a third of three. It seems that there were six launch attempts, three of which never made it very far off the pad, two made it some distance aloft, and only one that was considered somewhat successful (reaching Mach 2.23 and an altitude of 18 km).
In July 1948, together with the company's Electromechanical Division, Bollay's expanding group was moved into an ex-Consolidated Vultee factory in Downey, east of Los Angeles. It would be here that the Navaho, and later Hound Dog missile, the Apollo command module, and the Space Shuttle would be built. As a foreshadow of the future, the preferred intercontinental Navaho configuration of this time was a delta-winged, recoverable aircraft-type booster looking very much like the B-70 bomber of the 1960's or the reusable boosters of the original space shuttle designs. The cruise stage used a single enormous ramjet engine with a nose intake.
Early in 1949, the first of three Phase II American-built versions of the Model 39 engine was completed. But by then North American was committed to the more advanced design, so this was continued only an interim test article. In May North American settled on a revised design for the intercontinental version of the Navaho. It consisted of a tandem in-line boost stage with a single-ramjet cruise stage atop it. In September, reacting to the surprise explosion by the Soviet Union of an atomic bomb, the Truman administration began pumping more funds into missile projects. In order to proceed to flight test on an accelerated schedule, the Navaho aerodynamic design was frozen so that fabrication of the XSSM-A-2 flight articles could begin.
By late November of 1949, the first version of the Mark III engine was ready for testing at the new Santa Susana facility. Because it lacked turbopumps, propellants were pressure-fed from heavy-walled tanks. The North American team first ran the engine at 10 percent of maximum propellant flow for 11 seconds. However attempts for longer pressure-fed engine runs in December exhibited surges in combustion-chamber pressure (known as "hard starts") that were powerful enough to blow up the engine. Walther Riedel played an important role in introducing design modifications that brought this problem under control. In March 1950, this simplified engine first topped its rated level of 333 kN for four and a half seconds. During May and June, full-thrust runs, exceeding a minute in duration, went well. Meanwhile, Wright had completed design of the ramjets in December 1949 and begun fabrication.
In April, with the first three XSSM-A-2 airframes completed, the Air Force canceled flight test of that 1600-km range version of the Navaho. North American was instructed instead to proceed with development of the 10,200-km range version of the missile using the same aerodynamics, engines, and navigation systems already in development. This was to deliver a 3150-kg nuclear payload - and that was to be achieved by making a separable booster stage with two engines deliver a ramjet-only cruise stage to ignition velocity. The configuration was revised to that of the final Navaho - a twin-ramjet cruise stage launched strapped to the side of a liquid-rocket booster. This arrangement minimized the length of the vehicle, making handling and erecting on the launch pad easier.
Formal redirection of the program came in July 1950, just as the Korean War began. The missile, now designated WS-104A, was to deliver a 3150-kg warhead with a CEP of 450 m over a range of 10,200 km while cruising at Mach 3 at over 18 km altitude. The final missile would be developed in a three-phase program: Phase 1, using a reusable drone dubbed the X-10, would test the aerodynamics, structural concepts, autopilot, and inertial navigation system for the cruise missile using turbojet engines in an aluminum structure to achieve speeds of up to Mach 2. In Phase 2, the G-26 test vehicle would be a 2/3 scale version of the final version, testing the vertical launch booster, and a steel-structure ramjet-powered cruise vehicle that would reach Mach 2.75 and a range of 2300 km. Phase 3 would fly the G-38, the full-sized prototype for the operational system. The payload was sized to match the 20-kiloton W-4 nuclear warhead: 3150 kg in mass, 1.5 m in diameter and 2.3 m long.
Meanwhile, development of the engine for the now-canceled XSSM-A-2 continued. Late in March 1950, the first complete engine, turbopumps included, was assembled. In August, this engine, designated XLR-43-NA-1 by the government, fired successfully for a full minute at 12.3 percent of rated thrust. Late in October, the first full-thrust firing reached 310 kN for less than five seconds. However now a new problem emerged - rough combustion during the build up to full thrust. As a result, of the seven subsequent tests during 1950, only one in mid-November reached the engine's rated thrust level.
This combustion instability in the engine's single large thrust chamber had not been solved by the Germans before the end of the war, and they could not solve it now. However the North American engineers found a solution, and by March 1951 the problem of unstable combustion was under control. This marked an important milestone - the first time, the North American team had encountered and solved an important problem that the Germans could not solve. Combustion instabilities would recur repeatedly during subsequent engine programs, and the work of 1950 and 1951 provided North American engineers with several methods for work the problem.
In just three years of development, the North American team had delivered an engine that weighed less than half as much as the V-2's model 39 (668 kg versus 1126 kg), while delivering 34 percent more thrust. They had formed the corporate technology base for further American development of rocket engines. North American's rocket division, later dubbed Rocketdyne, would go on to be the preeminent American liquid engine rocket builder, building the engines that would take the first American to orbit, the first man to the moon, and power the Space Shuttle.
But this Mark III engine was now just a way-station to the more powerful 530 kN engine required for the new intercontinental Navaho. However, the Army had now directed von Braun's team to develop at utmost speed an 800-km range pure ballistic missile. With minor modifications, the Mark III would fit this requirement. So although the Mark III never went into production for the Air Force, its Army derivative boosted various versions of the Redstone tactical missile. And it would be this engine on the Redstone that would place the first American satellite in orbit and boost the first American into space.
North American originally designed the X-10 to be reusable, air-dropped from a B-36, then automatically landed at an airfield. But the Air Force dropped support for air launching in August 1950 - a decision that would have consequences later. By December 1950 North American was completing model tests of the X-10 at its own Santa Susanna wind tunnel facility, taking it up to Mach 2.87.
In June 1951 the Air Force inspected the X-10 full-sized mockup. The highly classified Mach-3 design, with cropped delta wing and canards, was far in advance of contemporary manned aircraft. The same month, the XLR-43-NA-1 engine was run for six seconds at 430 kN thrust, compared to its 333 kN rated thrust. Preliminary design of the new LR-71-NA-1 engine's lightweight tubular-construction 540 kN-thrust combustion chamber was begun. The brazed tubes would carry fuel to cool the chamber more evenly than the simpler double-walled German design of the XLR-43-NA-1. In September 1951 the X-10 detailed design drawings were released to the shop for fabrication.
The course of 1952 saw a wide range of testing activities. A 20-inch diameter ramjet, a subscale version of the 40-inch engine for the Navaho, was flown on Lockheed's X-7 rocket. Of the seven launches, only one was completely successful, and two partially successful. The G-26 version of Navaho reached the mock-up stage; and design of the G-38 was begun. In June a test version of the 540 kN combustion chamber for missile was run. This was the most powerful ever tested, but still used the double-walled combustion chamber design of the LR43 rather than the brazed tubular wall construction planned for the production LR71. But the first complete prototype was first operated on 19 November. By the end of the year the drawings for the G-26 were released to the shop, and the first X-10 had been completed and was in static testing.
On 23 December the USAF released the full-scale development contract for the Navaho G-26, consisting of 10 cruise missiles, 13 boosters, and five N-6 stellar-inertial navigation systems (early flights would be radio-controlled). First delivery was scheduled for 1953 and first launch February 1956.
But early in 1953 another technical improvement was deemed necessary, and this would impact the schedule just agreed. North American's rocket group had begun a Rocket Engine Advancement Program, to identify improved rocket technology. One early conclusion was that the performance and logistics could be significantly improved by shifting from alcohol to kerosene fuel. The decision was made to change the propellants for both the Navaho G-26 test vehicle and the G-38 production booster. The LR71 engine, improved and modified to burn kerosene, would be designated LR83. This meant a delay to Navaho, but would be the basic engine that would power the Thor, Jupiter, Atlas, Saturn I, and Delta rockets into the next century.
The XN-2 navigation system for the Navaho had been flight tested aboard a C-97 aircraft between April 1952 and May 1953. A series of fifteen X-10 flights were conducted at Edwards Air Force Base from October 1953 to March 1955. A lot of time was wasted and four vehicles were lost in proving the autolanding capability for use at the skid strip at Cape Canaveral. In retrospect, an expendable vehicle might have completed the aerodynamic and guidance system test work more quickly.
Meanwhile North American had also begun operations at Cape Canaveral in 1953, building two missile assembly buildings, a vertical launch facility for the XSM-64 (LC9) and a 70 m x 3000 m landing strip (the 'Skid Strip') to recover the reusable X-10 and G-26 test vehicles.
North American tripled its field office staff at the Cape from 22 to 77 people in 1954. It also began installing equipment in the guidance laboratory, the blockhouse and the Navaho flight control building even before construction of those facilities was completed. Meanwhile the XN-2 navigation system was still found lacking for the production missile. The XN-6 introduced paired counterrotating gyroscopes to compensate for the precession errors resulting from the single-gyro-per-axis XN-2 design; and introduced hydrodynamic bearing gyros in place of the XN-2's air bearings. N-6 flight testing began aboard a T-29 in May 1954.
On 11 January 1955 the Air Force expressed its confidence in the missile by placing the second G-26 second production contract. This added 12 more cruise stages, 21 boosters, and 6 N6 navigation systems, bringing the total procured to 22 cruise stages, 34 boosters, and 11 navigation systems. The first X-10 was launched from the Cape on 19 August 1955, and the Navaho quickly replaced the Matador as the range's principal user. Support facilities were completed in the last half of 1955, and seven more X-10s were launched from the Cape over the next twelve months.
But North American management could see the ballistic missile competition gaining on them. The Thor, Jupiter, Atlas, and Titan development programs were all underway on a crash basis. But these missiles needed guidance systems and engines - and the Navaho's systems were far in advance of anything else available.
On 7 November 1955 North American broke up the Navaho program into three new divisions. The Missile Development Division, in Downey, would handle the Navaho missile and its possible derivatives. The Autonetics Division would handle inertial navigation and other avionics products, and move from Downey to Anaheim. The Rocketdyne division would handle North American's liquid rocket motors, and move to a new facility in the western San Fernando Valley in Canoga Park, close to North American's rocket test facilities in the Santa Susanna mountains. These divisions were now free to market their products for use on other company's missiles.
Rocketdyne liquid rocket engines were selected for the Thor, Jupiter, and Atlas missiles. The Titan was powered by Aerojet engines, purposely done by the Air Force in order to 'second source' the Atlas in case of some unforeseen catastrophic technical problem with the design.
By the middle of 1956, North American had 605 people working on the Navaho program at the Cape. Five more X-10 flights were completed in the last half of 1956, but problems with an auxiliary power unit delayed the Navaho G-26's first launch for six months. When the first vehicle finally made it into the air, it exploded 26 seconds into the flight. Three more Navahos were launched over the next seven months with equally dismal results. In addition to those failures, the first in a series of 2,800 km long auto-navigator test flights was attempted ten times in the first three months of 1957 without a single launch.
More promising were successful static tests of the booster rockets and North American's isolation of problem areas revealed in the first four flights. But while the "never-go-Navaho" sat on the pad or failed in flight, the first Jupiter, Thor, and Atlas ballistic missiles began test in the spring of 1957. They were equally unsuccessful at first, but a Jupiter made a full-range 2100-km flight on 31 May. The Army had already demonstrated the necessary re-entry vehicle technology on launches of the 'Jupiter-A' multi-stage versions of the Redstone. It was clear that these ballistic missiles, each little more complex than the booster stage of Navaho alone, could deliver a nuclear warhead over the same ranges - at seven times the speed.
But it was still a surprise when Air Force Headquarters terminated the Navaho development program on 12 July 1957. 4,705 employees were laid off the day the termination notice was received via an announcement over the public address system to "stop what you are doing, proceed to the nearest exit, and deposit your badge in the bin indicated". By the end of the month the total staff laid off at North American alone amounted to 15,600 employees. At the time the program was canceled full-range G-38 missiles were in fabrication with first flight test planned by the end of 1958.
On 21 August 1957, the Air Force handed North American a consolation prize - it was awarded a contract for the Hound Dog air-launched cruise missile. This was a crash project to equip B-52 bombers with supersonic nuclear missiles that could penetrate heavily-defended Soviet sites. The Hound Dog would use the aerodynamics of the G-38, and navigation and control systems developed for the Navaho, and an off-the-shelf turbojet engine. The program kept the core engineering and production staff together at North American's Downey facility - and this would be the team that won the Apollo contract, and later the Space Shuttle.
Because the auto-navigator showed promise for use on other missiles, the Air Force authorized continuation of flight test of existing assets. Engineering staff was kept on at Cape Canaveral to launch the five completed G-26 missiles, at a total cost of $4.9 million. These were flown through February 1958, and were somewhat more successful than the earlier series, with one missile managing to reach a range of 2000 km before its ramjets failed.
Seven further completed G-26 Navahos were to be tested in support the of North American's B-70 bomber and F-108 long-range interceptor programs. But after two unsuccessful launches in late 1958 that project was completely canceled at the insistence of the B-70 Weapons System Project Office.
North American also attempted to revive the Navaho program with the Air Force. Test vehicle and reconnaissance versions were pitched. After Sputnik was launched, the company proposed the use of G-26 or G-38 boosters to launch a modified manned X-15 into space. None of these ideas were considered favorably by the customer.
As North American closed out the Navaho program, three X-10s were selected as target drones for Bomarc antiaircraft missile tests. Two X-10's launched successfully on 24 September and 13 November 1958, but both burned at the end of their missions after running off the end of the Skid Strip. The last X-10 was launched on 26 January 1959, but self-destructed and crashed approximately 105 km downrange after a power failure. It was the Navaho project's final flight.
