What it was: Two closely related spaceplanes and their associated launcher, in turn related to a number of other projects, which were studied in the late 80s and early 90s as a way of getting astronauts to and from Space Station Alpha or the ISS. This included rescue missions in the event the space station had to be evacuated quickly.
Details: On January 28, 1986 the Space Shuttle Challenger exploded and the American manned space program entered a period of reflection. In the end the result was to continue flying Shuttles at the cost of them never approaching the rate or affordability pictured for them in the early 80s, let alone as they were originally envisioned.
The field was wide open for other possibilities, though, and into it stepped the Personnel Launch System (PLS) consisting of the HL-20 and either a Titan IIIC rocket or a Titan IV. Together they were intended to address a number of perceived problems with the Shuttle, particularly in conjunction with a space station: its cost and complexity, its lack of a crew escape system, and its inability to stay attached to the station for very long (partly for technical reasons and partly because it was sheer overkill for most reasons why you’d want to leave one attached).
As designed the HL-20 was a small spaceplane about 7.5 meters long, 7.2 meters in width, and 10,884 kilograms; the HL-42, proposed a few years later, was so-named because it was almost exactly the same design but 42% larger in each direction. At the time they were the latest in a long line of NASA lifting bodies going back to the “Flying Bathtub”, the M2-F2 the HL-10. Furthermore they benefited from what was known about the MiG-105 and BOR-4 lifting bodies, respectively a subsonic test article and an unmanned subscale version of the Soviet Spiral spaceplane, the latter of which had been seen in an overflight by an Australian Air Force plane while being recovered after a re-entry test.
The initial proposal was that the HL-20 would have been attached to a Titan IIIC, an Air Force rocket that had just enough strength to lift it into low earth orbit; it had the additional advantage that there was a man-rated version which had been developed in the early 60s for launching the never-flown Air Force spaceplane Dyna-Soar. This plan did change quickly, though, as the USAF had moved on from the Titan IIIC—its last flight was in 1982—and started flying the more powerful Titan IV in 1989 just as the PLS was getting warmed up.
The HL-20 was proposed and championed by NASA’s Langley Research Center, but most of the work on it was done by Rockwell International and Lockheed’s Advanced Development Projects (more famously known as the Skunk Works). Their intention was to develop something relatively simple that could be flown ten to twelve times a year, and deliver from eight to ten astronauts and a minimum of cargo to the space station; as time went by an initially speculative mission to serve as a lifeboat for station crews gained in importance too. To that end they came up with a glider which had no main propulsion engines, unlike the Shuttle, and only orbital maneuvering engines. Once paired up with the powerful Titan IV, some of the systems needed for launch were moved into a fairly heavy adapter (4435kg) at the aft end of the craft which attached the HL-20 to its launcher, a strategy similar to that used on the final versions of its contemporary, the Hermes.
Technically the HL-20 wasn’t a winged craft but a lifting body: its curved shape provided its lift. By eliminating actual wings sticking out from the fuselage of the plane it became easier to build and reduced heating complications at the joints between wing and body. Somewhat surprisingly, though, the complication was added back in by making the “wing” tips of the body hinged so that they would fold up over the top of the craft, the intention being to make it easier to transport by making it smaller in this configuration.
The most important difference between the HL-20 and the Shuttle, though, was its abort systems. Once the Shuttle was launched, or when it was re-entering, there were no provisions for emergency escape—either the Shuttle made it where it was going as a whole or it didn’t make it at all. The HL-20 was to have been equipped with emergency escape rockets to push it away from a failing launch vehicle on the pad, and further up in its trajectory it could be brought back to Earth underneath three parachutes. In those cases it wouldn’t have enough speed to land as it normally would—horizontally on a runway of choice—so it would aim for the water tail first and be supported by inflation devices around its aft end until rescue came. There were times where the crew would have no options during an HL-20 missions (often because the Titan IV used toxic UDMH and N2O4 for propellant, which even the Russians baulk at using for manned missions), but unlike the Shuttle the astronauts at least had a chance in some of the ways a mission could go very wrong.
Like the Shuttle the fuselage of the HL-20 was to have been built out of aluminum, but its outer surface was more advanced. Where the Shuttle had ceramic tiles applied to a substrate before being attached to metal, the HL-20 would have had a graphite/polyimide composite outer skin that had the same heat expansion properties as aluminum; this would have let the tiles be directly attached to the hull.
Upon arrival back home the HL-20 had one more potential advantage. If it were returning long-term station crews who were physically weak after being in freefall for months, its hatch was specifically designed to be large enough that astronauts could be retrieved easily from inside. In a joint effort on NASA’s behalf, North Carolina State University and North Carolina A&T University built a full mockup of the craft to test out this and other human factors of the crew area
What happened to make it fail: Both vehicles needed the Titan IV for launching and that meant cutting a deal with the Air Force, which controlled that particular launch vehicle. This was particularly true as the Titan IV wasn’t man-rated, and so would have had to go through an expensive run of upgrades and tests before astronauts could be perched on top of one. The USAF wasn’t interested, perhaps in part because the Titan IV was the result of the Department of Defense wanting to divorce themselves from NASA after the Challenger explosion grounded their satellite launches. Instead they went ahead with the development of the Delta IV and Atlas V rockets.
NASA might have managed to push things forward anyway by offering the Air Force more, or developing their own launch vehicle to tempt the DoD back into bed (a rocket called the NLS was roughed out), but spending more money like that ran into another wall. When Space Station Alpha became the ISS, the hybrid nature of the station opened up another possibility for getting people to and from it: the Soyuz spacecraft and its associated launcher. Simply renting taxi rides on Russian spacecraft and leaving one attached to the ISS as a lifeboat turned out to be a lot cheaper. NASA took this actual route, swallowing their own national pride and (temporarily, at least) leaving the US without a functional manned spacecraft program for the first time since 1981.
What was necessary for it to succeed: This is another one of those projects that needed the USSR to continue for a little while longer, with the added twist that it’s not a Russian program but an American one. Once the Soyuz became available only a considerably more capable American craft would get the go-ahead, and the HL-20 didn’t qualify. The HL-42 was then proposed, the whole reason for its increased size being that it could carry more people than a Soyuz and so allow more crew aboard the station. This approach became pointless when the ISS was upgraded to allow two Soyuzes to be docked at the same time.
If history diverts so that the Russian part of the ISS disappears, though, possibilities open. While it’s conceivable that an all-American station would have gone ahead without any lifeboat capability at all, it’s not likely. And while NASA gave lip service to developing their own ballistic capsule for that purpose, there was much more internal support for a winged craft. This left only two possibilities at the time, either something like the X-24A (an approach that was revived for the X-38 CRV in the late 90s) or something like the HL-20. At that point it comes down to capability: it’s hard to fit fuel tanks into the X-24A fuselage shape, so do you want to be able to maneuver in orbit? If so, you’d be better off with an HL-20 or HL-42.
Given NASA’s inclination to go for more capable (and more complex) projects, it seems likeliest that in the absence of anything other than the Shuttle for station crew return they’d have gone for the full Personnel Launch System rather than a mere return-only lifeboat based on the X-24A.
As it happens, there’s still one more possibility for the HL-20 to fly. It’s a slick little design, and there have been two commercial attempts to revive it. Orbital Sciences Corporation proposed a derivative of it, Prometheus, for the initial phase of NASA’s Commercial Crew Development competition, which is looking for a commercial craft to deliver crew to the ISS; unfortunately for the HL-20 it didn’t make it through to the second round. However, a second HL-20 proposal, this one from Sierra Nevada Corporation, did, and furthermore proceeded on to the third round too. Their Dream Chaser proposal is considered to be the third likeliest to ultimately win the competition out of the three remaining competitors (the other two being the Boeing CST-100 and SpaceX’s Dragon, both ballistic capsules), but stranger things have happened.