Rocket Development

From V-2 to R-7: The Founding Years

Soviet rocket development began in earnest in May 1945, when Soviet forces reached Peenemünde and the Mittelwerk underground V-2 factory. The Americans had already taken Wernher von Braun and most of his senior team. The Soviets got production tooling, captured V-2s, and a smaller pool of German technicians who were relocated to Soviet design bureaus for the late 1940s. The historical context page covers the postwar geopolitics in detail.

By 1947, Korolev's team had reverse-engineered the V-2 and launched their own copy, the R-1. The R-2, R-5, and R-11 followed through the 1950s, each a larger and longer-range derivative. None of them had the throw weight to carry a Soviet thermonuclear warhead.

That gap drove the R-7. Stalin authorized the program in 1953 with a single requirement: a missile capable of delivering a 5.5-tonne nuclear warhead to the continental United States. The result, after four years of development, was tested successfully from Baikonur on August 21, 1957. The sixth prototype, designated 8K71, flew roughly 6,000 km to the Kura impact zone on Kamchatka.

Six weeks later, a stripped-down R-7 carried Sputnik 1 into orbit. The same rocket family is still flying crews to the ISS in 2026.

The R-7 Family Tree

Share

The R-7 (nickname "Semyorka," meaning "the Seven," after its design number in the OKB-1 series) used an architecture nobody else had tried at scale. A central core, called Block A, was surrounded by four conical strap-on boosters, Blocks B, V, G, and D. All five elements ignited at liftoff in a configuration called parallel staging.

The strap-ons burned out and fell away first, leaving the core to continue to orbital velocity. The flame pattern at takeoff, four cones plus a central jet, has been the visual signature of Soviet and Russian rocketry for 69 years.

Every R-7 variant that followed kept the same core architecture and added an upper stage. Sputnik (1957) used a stripped configuration. Vostok (1961) added a Block E upper stage to launch Yuri Gagarin. Voskhod (1964) added a more powerful Block I. Molniya (1960) added a fourth stage for high elliptical orbits. Soyuz, introduced in 1966, eventually replaced them all.

Soyuz-U, the workhorse variant that flew from 1973 to 2017, ended with 786 launches and a 97.3% success rate. The current Soyuz-2 added digital flight control in 2004 and is the version still flying today. The story of Soyuz in 2026 is on the Legacy page.

N1: The Moon Rocket That Never Worked

Every other major Soviet rocket program was a Korolev project. The N1 was supposed to be his magnum opus. It was the rocket meant to land a Soviet cosmonaut on the Moon before Apollo did. It launched four times. It exploded four times.

The architecture was brutal. Thirty NK-15 engines on the first stage, all firing in parallel, with a control system called KORD that monitored them in real time and could shut down opposing pairs to balance thrust. NK-15 was a kerosene/LOX engine designed by Nikolai Kuznetsov's OKB-276, an aerospace bureau best known for jet turbines. Kuznetsov had been brought in because Glushko, the Soviet Union's premier rocket-engine designer, refused to build LOX/kerosene engines at all (more on that below).

The fatal decision was made early. Soviet logistics could not move a fully-assembled first stage to Baikonur for a static-fire test. Instead, only two of every six engines in a manufacturing batch were ground-tested, and the actual flight engines were never fired before launch day. This is the consensus root cause for all four flight failures.

The four flights and their failure modes:

Korolev did not live to see any of them. He had died in January 1966. His successor at OKB-1, Vasily Mishin, inherited the program and ran it through all four failures. In May 1974, Glushko replaced Mishin, merged OKB-1 with his own Energomash to form NPO Energia, and cancelled N1 within weeks. Total program cost was about 6 billion rubles over 17 years.

About 150 NK-15 engines, mid-upgrade to the improved NK-33 variant, survived the cancellation in storage at Kuznetsov's factory. Decades later, in the 1990s, Aerojet acquired roughly 40 of them, certified them for U.S. flight as the AJ26, and sold them to Orbital Sciences for the Antares rocket. One of those engines, AJ26 E15, failed on Antares Orb-3 on October 28, 2014. The investigation traced it to a manufacturing defect in a LOX-turbopump bearing bore. The Antares program switched to Russian RD-181 engines after that, until 2022.

Proton: Chelomei's Hypergolic Workhorse

While Korolev was struggling with the N1, Vladimir Chelomei was building the rocket that would carry every Salyut and Mir module, every Russian Federal Space Agency communications satellite, and most of the heavy commercial payloads from the post-Soviet era. The Proton has flown 431 times as of February 2026, with 383 successes and an 88.9% success rate.

Proton started life as the UR-500, Chelomei's proposal for a super-heavy intercontinental ballistic missile capable of delivering a 100-megaton thermonuclear warhead. It was never deployed in that role. The first orbital flight was in 1965, carrying the Proton-1 scientific satellite that gave the rocket family its name.

The Proton uses UDMH and N2O4, a hypergolic combination that ignites on contact without spark or igniter. The propellants are storable indefinitely at room temperature, which simplifies pad operations and supports military-style readiness. They are also extremely toxic. The first stage holds about 419 tonnes of these chemicals, and the launch corridor downrange of Baikonur has been off-limits to grazing land for sixty years.

Production of Proton-M ended in 2025. A handful of residual launches are scheduled through 2029 for the Russian Federal Space Agency's Science Power Module-1 and the next Ekspress satellites. Kazakhstan and Russia agreed that Proton launches from Baikonur would end after 2025 in part because of the toxic-propellant footprint. The replacement is Angara A5, which uses kerosene/LOX engines instead.

Energia and Buran: One Flight, Then Shelved

Glushko's revenge for losing the Moon race was Energia. It was the largest rocket the Soviet Union ever flew, designed to launch a space shuttle called Buran. The core stage used four RD-0120 cryogenic engines burning liquid hydrogen and liquid oxygen, the only major use of a hydrolox engine in Soviet rocketry. Four strap-on boosters, each powered by a single Glushko RD-170, surrounded the core.

Energia first launched on May 15, 1987, carrying a classified payload called Polyus. The payload failed to reach orbit because of a guidance error on its own inertial upper stage, but Energia itself worked perfectly. The second and final Energia flight, on November 15, 1988, carried Buran on an uncrewed orbital test.

Buran completed two orbits and then performed something NASA's Space Shuttle was never designed to do. It landed itself. Fully autonomous, no pilot on board, on a runway at Baikonur. As of 2026, no other crewed-capable spaceplane has done this, although the unmanned X-37B has logged autonomous landings since 2010.

Then the Soviet Union dissolved. The Buran program was officially cancelled in 1993. Buran's only flight-tested airframe was destroyed in 2002 when the hangar roof at Baikonur collapsed under snow. A second airframe, OK-1.02, never flew and sits in a museum at the Baikonur Cosmodrome.

Energia is the most accomplished rocket the Soviets ever built that flew exactly twice. The RD-170 engine it left behind, however, did not stop flying.

The Engine Genealogy

The single most exported piece of Soviet technology of the past 40 years was not a hardware platform. It was a rocket engine cycle.

Soviet engineers invented the staged-combustion cycle in 1960, first demonstrated on the S1.5400 engine for the Molniya upper stage. The first large-scale staged-combustion engine to fly was Glushko's RD-253, the first stage of the Proton (1965). The crucial variant, oxidizer-rich staged combustion, was long believed impossible by U.S. propulsion engineers because the hot oxygen-rich gas downstream of the preburner is corrosive enough to dissolve most metals on contact.

When Pratt & Whitney engineers first inspected NK-33 engines in 1995, they assumed the Russian technical documentation was wrong about how the engines actually ran. They were not. Soviet metallurgists had developed alloys that withstood the conditions. NASA confirmed the chamber pressures: 245 to 257 bar in the RD-170 family, compared to roughly 206 bar in the Space Shuttle Main Engine and 70 bar in the F-1 that flew Apollo. The Soviets had been running hotter and harder for 30 years.

The engines that came out of this lineage:

The Cryogenic vs Hypergolic Debate

Two engineers built the Soviet space program. They did not speak to each other for most of the 1960s.

Korolev preferred liquid oxygen and kerosene. Higher specific impulse than hypergolics, cleaner exhaust, less corrosive on hardware. He argued that a crewed Moon rocket had to use cryogenic propellants because nothing else gave the necessary performance per kilogram. He cited the 1960 Nedelin catastrophe, in which an R-16 missile loaded with UDMH/N2O4 exploded on the pad and killed somewhere between 78 and 126 engineers and officers, as evidence that hypergolics were too dangerous for human spaceflight.

Glushko preferred UDMH and N2O4. Hypergolics ignite on contact without an igniter, eliminating one major failure mode. They are storable indefinitely at room temperature, which suits ICBMs designed to sit in silos for years. He also argued that building a LOX/kerosene engine large enough for the N1 was not technically feasible at the time.

When Korolev refused to build the N1 with hypergolic engines, Glushko refused to develop a LOX/kerosene engine for it. Korolev brought in Nikolai Kuznetsov, a jet-engine designer with no large rocket experience, to design the NK-15 instead. The N1 first stage carried 30 of those engines. None of them was ever static-fired before flight.

The rift between Korolev and Glushko had started in 1938. Korolev believed Glushko had denounced him to the NKVD, which sent Korolev to a labor camp in Kolyma and broke his health permanently. A formal commission in 1962 sided with Korolev, but Glushko refused to comply with its findings. The two men ran competing programs until Korolev died in 1966. Glushko outlived him by 23 years.

Modern Russian Rockets in 2026

The Russian rocket fleet in 2026 is smaller, narrower, and increasingly tied to a single supplier (Russia itself). The Ukraine war effectively ended the export market. What remains is a domestic-launch program with three active families.

Soyuz-2, the modernized digital-control variant, handles all Russian crewed and most cargo launches. It is the direct descendant of the R-7 and is still flying weekly. Its profile is on the Legacy page.

Angara A5 is the heavy-lift replacement for Proton. First flight December 23, 2014, fewer than 10 flights in its first 11 years. Each Angara A5 uses an RD-191 in its URM-1 core and four more on the strap-ons. It launches from Plesetsk and from the new Vostochny Cosmodrome in the Russian Far East.

Soyuz-5 (Irtysh) flew its first successful suborbital test from Baikonur Site 45 on April 30, 2026. It is designed to replace Zenit after the Ukraine partnership collapsed in 2022, and to give the Russian program a 16-tonne-to-LEO medium lifter. The first stage uses a single RD-171MV, a modernized variant of the original RD-170.

Frequently Asked Questions