Sea Dragon (rocket)
The Sea Dragon was a 1962 conceptualized design study for a reusable two-stage sea-launched orbital super heavy-lift launch vehicle. The project was led by Robert Truax while working at Aerojet, one of a number of designs he created that were to be launched by floating the rocket in the ocean. Although there was some interest at both NASA and Todd Shipyards, the project was not implemented.
With dimensions of long and in diameter, Sea Dragon would have been the largest rocket ever built. As of 2026, Sea Dragon is generally regarded as the tallest launch vehicle ever formally proposed. With a payload of 550 tons, it has mistakenly been regarded as the largest and most powerful vehicle seriously considered. It is comparable to other unbuilt concepts such as Convair's Nexus and Super Nexus,, Phillip Bono's series of reusable single stage launch vehicles, and various NOVA Post-Saturn launch vehicles.
Design
Truax's basic idea was to produce a low-cost heavy launcher, a concept now called "big dumb booster." To lower the cost of operation, the rocket itself was launched from the ocean and constructed in a shipyard with more standard materials. A large ballast tank system attached to the bottom of the first-stage engine bell was used to orient the rocket vertically for launch. In this orientation the payload at the top of the second stage was just above the waterline, making it easy to access. Truax had already experimented with this basic system in the Sea Bee and Sea Horse. To lower the cost of the rocket, he intended it to be built of inexpensive materials, specifically steel sheeting. The rocket would be built at a sea-side shipbuilder and towed to sea for launch. It would use wide engineering margins with strong simple materials to further enhance reliability and reduce cost and complexity. The system would be at least partially reusable with passive reentry and recovery of rocket sections for refurbishment and relaunch.The first stage was to be powered by a single pressure-fed thrust engine burning RP-1 and LOX. The tank pressure was for the RP-1 and for the LOX, providing a chamber pressure of at liftoff. The first stage would also be equipped with an asbestos-based recovery heatshield for reuse of the vehicle. As the vehicle climbed the pressures dropped off, eventually burning out after 81 seconds. The vehicle would be up and downrange, traveling at a speed of before staging. The normal mission profile would see the first stage land in a high-speed splashdown some downrange.
The noise of the first-stage engine, which would have produced a sound pressure level of approximately 184 dB at liftoff, would have created an extremely challenging sonic and vibrational environment for a traditional land-based launch pad. This issue was common point of issue for vehicles around the scale of Sea Dragon. One solution, proposed by Philip Bono, was the “water-filled acoustic limiter,” which consisted of a parabolic dish filled with water installed beneath the launchpad. However, the size and supporting infrastructure required for such a system would have significantly increased launchpad construction costs. Truax’s design team avoided these construction costs by adopting the ocean-launch concept.
The second stage was equipped with a single extremely large pressure-fed hydrolox engine with a thrust of, fed at a constant lower pressure of. At the end of the entire 260s burn, the second stage would be at in altitude and downrange. To improve second stage engine performance, the engine featured an expanding engine bell which covered most of the first stage tankage, acting as an aeroshell for the first stage during ascent. During the events of staging, this expandable nozzle would go from a linear to a more conical shape, improving the second stage's expansion ratio.
To provide attitude stabilization, the second stage was also equipped with four auxiliary hydrolox engines, each producing of thrust. These engines served as the vehicle's primary Thrust Vector Control system; they provided roll control during the first-stage burn and full attitude control for the second stage. To ensure reliability, these auxiliary engines were ignited and monitored on the ocean surface shortly before the first-stage main engine start. They would remain active for a total of 1340 seconds, continuing to burn after the second stage main engine shutdown to provide the final velocity increment required for orbital injection.
A typical launch sequence would start with the rocket being refurbished and mated to its cargo and ballast tanks on shore. The RP-1 would also be loaded at this point. The rocket would then be towed to a launch site, where the LOX and LH2 would be generated on-site using electrolysis; Truax suggested using a nuclear-powered aircraft carrier as a power supply during this phase. The ballast tanks, which also served as a cap and protection for the first-stage engine bell, would then be filled with water, sinking the rocket to vertical with the second stage above the waterline. Last-minute checks could then be carried out and the rocket launched.
The rocket would have been able to carry a payload of up to or into LEO. This is enough to comfortably launch the ISS in a single launch. Payload costs, in 1963, were estimated to be between $59 and $600 per kg TRW conducted a program review and validated the design and its expected costs. However, budget pressures led to the closing of the Future Projects Branch, ending work on the super-heavy launchers for a proposed crewed mission to Mars.