Being an avid player of Kerbal Space Program, it is intensely frustrating to return to the real world and have to remind myself that rocket stages can’t generally be swapped out between rockets (meaning that we’re unlikely to ever see - for example - a Falcon 9-Centaur). Especially given that, in the past, such swapping of rocket stages was fairly common (usually in the form of taking a bunch of rockets and slapping an upper stage on them):
- The Able, which originated as the second stage of the Vanguard rocket,1 was quickly grafted onto the Thor missile to produce the Thor-Able space launcher. It was also later attached to the Atlas D,2 producing the Atlas-Able.3 (The AJ-10 engine used by the Able would go on much further, its evolved variants seeing a long and productive career in such uses as the Apollo Service Propulsion System and the space shuttle’s OMS engines.)
- The Agena was also mated to the Thor, producing the Thor-Agena, and, later, also to the Atlas, forming the Atlas-Agena, which launched a great number of scientific and military satellites in the 1960s (the Ranger lunar impactors being an example of the former, and the Corona spy satellites an example of the latter) - as well as being used, in modified form, for the Agena Target Vehicle (where the Agena was itself the payload, being used as a target for manned Gemini spacecraft to dock with) - before being superseded by the entry into service of...
- ...the Centaur, the first hydrolox rocket stage ever flown and probably the rocket stage used (or planned to be used) on the greatest variety of launch vehicles. It started out as yet another upper stage for the Atlas, in a configuration known as (surprise surprise) the Atlas-Centaur; once the Centaur’s initial teething troubles were worked out, this quickly became the standard configuration of the Atlas, for which it flies even today, although the Atlas in question is now an Atlas V, and the Centaur is no longer regarded as a separate upper stage on top of the Atlas, but, rather, as an integral component of the Atlas V itself. It was also planned to be used as the Saturn’s final stage, although progressive simplifications of the Saturn design and Wernher von Braun’s intense dislike of the Centaur (mainly the latter) meant that it never actually flew in this role. In the 1970s, a number of heavy interplanetary probes (two each of the Helios, Viking, and Voyager probes) - too heavy for an Atlas-Centaur - were launched by the Titan III-Centaur, also known as the Titan IIIE, before that launcher was cancelled in favour of the space shuttle, which was also planned to carry the Centaur (thus making the Shuttle-Centaur, allowing it to launch interplanetary probes beyond even a Titan III-Centaur’s wildest dreams, as well as gargantuan Earth-orbiting military payloads; when the Shuttle-Centaur programme was cancelled in the wake of the Challenger disaster, the Centaurs meant to ride the shuttle (special wide-bodied versions known as the Centaur G and G-prime) were instead mated to more Titans, to form the Titan IV (which, among other payloads, launched Cassini–Huygens to Saturn).4 Finally, the Centaur is to be used as the upper stage of the United Launch Alliance (ULA)’s under-development Vulcan, intended to replace all of the ULA’s current launch vehicles.
In contrast to this illustrious history of interchangeable rocket stages, each of today’s launch vehicles uses its own, proprietary stages; even the ULA’s two current launchers (the Atlas V and the Delta IV Heavy), despite both being made by ULA members and both needing hydrolox upper stages with essentially the same requirements, still use different designs of upper stage (the Centaur for the Atlas V and the DCSS for the Δ4H), rather than standardising on the Centaur for both.
Why aren’t the rocket stages of today modular and interchangeable, as was common in the past?
1: Yes, that Vanguard.
2: It was originally intended to be used with the slightly-earlier Atlas C, but the first one blew up during a static test, and then the Atlas D was ready.
3: All three launches of which failed, but that’s beside the point here.
4: In fact, if I’m not mistaken, an absolute majority of U.S.-launched interplanetary satellites have ridden some version of a Centaur!5
5: Plus at least three more that would have flown on Shuttle-Centaur flights had the programme not been cancelled post-Challenger.