Sidebar: The Tupolev OOS


A model of the OOS shuttle, believed to be from a Russian magazine in the 1990s. If you have more information about this picture, please contact the author.

During the 80s the USSR’s space program stayed remarkably focused on Energia/Buran and the Mir space station, especially when compared to the infighting that marred the years 1966-1975. It fended off or adapted to a number of distractions, whether it was Vladimir Chelomei‘s repeated attempts to regain his previous, short-lived position on top of it, or airplane design bureaus suggesting anything from conservative alternatives to the recently discussed Myasishchev M-19 nuclear scram/ramjet.

The OOS was a late Soviet-era shuttle proposal from the Tupolev bureau, an also-ran in that country’s space business despite a strong position in civil aviation and strategic bomber development. Proposed as a fully reusable replacement for Buran sometime around the year 2000, it was about the same size as that craft or the American Space Shuttle, though somewhat heavier at 100 tonnes when fuelled. With a crew of two cosmonauts. it had a payload of 10 tonnes to and from low-Earth orbit.

If you’re a long-time reader of this blog, or just sufficiently into spacecraft, you probably slotted the shuttle pictured above toward the conservative end of that spectrum. Apart from the more-rounded contours, it looks to be much like the Shuttle, particularly in the shape of the underside. There, too, we have the usual ceramic tiles for dissipating the heat of re-entry. The engines are not visible, but I can tell you that there were three, burning LH2 and LOX during the ascent to orbit (though, curiously, switching out the hydrogen with kerosene for orbital maneuvers). Knowing that would likely not change your opinion at all.

Given that it’s was to be fully reusable, the ten-tonne payload mentioned earlier may have got you wondering, though. The actual American and Soviet shuttles had payloads in the 25-30 tonne range, so alright—there’s clearly some sort of tradeoff there. You’d be well-advised to wonder about the rest of the OOS’s configuration. Side boosters but no external tank? Perched on a reusable rocket in some manner, maybe?

Well, no. “OOS” stood for Odnostupenchati Orbitalni Samolyot, ‘one-stage orbital plane’, But a single-stage-to-orbit craft the size of the Orbiter? Surely that’s not possible.

This goes to show that you don’t think like a Soviet aircraft designer circa 1989. The OOS was to have been air-launched, and the other half of the system was the Antonov AKS:


Aerospace aficionados will remember that the An-225, which was used to piggy-back the Buran shuttle around the Soviet Union, was by most measures the largest aircraft ever built. This is two of them, one wing apiece removed and replaced with a sort of aerodynamic bridge, and then 675 tonnes of spacecraft and rocket propellants attached to its underside. It had twelve turbojet engines for when it flew without the orbiter attached (the dark circles in the diagram above, at lower right), with a supplementary ten more being added during launch operations (the white circles). The Aristocrats! With a length of 83 meters (272 feet), a wheelbase of 40m (131 feet) and a wingspan of 153m (502 feet), the combination came in at a whopping 1650 tonnes. By contrast, a fully fueled late-model 747 has a maximum takeoff weight of just under 440 tonnes.

There has been only one successful air-launching system in the world to date, Orbital ATK’s Pegasus. It weighs 23.1 tonnes and can put 0.44 tonnes in orbit; it’s launched from a Lockheed L-1011, already getting into the neighborhood of large airplanes. So start with some skepticism that 20 times this in launch mass and payload are a possibility for the late-era USSR.

Further, I haven’t (unfortunately) been able to find a detailed description of the AKS/OOS’s mission profile. I’d like to see it because I’m having a hard time picturing what the moment of separation would look like. Or rather, I have an image of the support crew aboard the AKS bouncing around like ping-pong balls in a boxcar once the plane, straining to get the orbiter to altitude, suddenly cuts loose 675 tonnes. For that matter, the OOS would have to light its engines pretty quickly thereafter or defeat the purpose of an air launch. As these were in the same class as the RS-25’s on the American Shuttle—the noise aboard the AKS, now presumably not all that far above and behind it, would have been intense.

I’m on record for my begrudging appreciation of the come-what-may technological megalomania that gripped the superpowers post-WWII. The US grew out that uncritical mindset after Love Canal and Three Mile Island, while the Soviets carried on until 1989. That extra time coupled with fossilized technocrats in charge allowed awe-inspiring audacity in technology of it to grow even greater than it did in the West.

Even so, I can’t imagine anyone with the power to make the Tu-OOS happen actually doing so. It would have been an immensely expensive and difficult project right at a time when the Soviet Union was in no position to take one up, and technological limitations would have prevented anything like it at an earlier point in that country’s history. The OOS/AKS was a paper project, and would have remained so.


OOS, la bestia de Tupolev y Antonov

OOS, el sistema espacial de lanzamiento aéreo definitivo

Artist Vadim Lukashevich has numerous renders of the AKS/OOS combination on (screll down to the second half of the page).

Readers will note a lack of primary sources here. I’m convinced of this project’s existence, but any pointers to a source that’s a little more direct than what I’ve relied upon here would be most welcome.



6 thoughts on “Sidebar: The Tupolev OOS

  1. I hadn’t heard of this project, but it doesn’t surprise me. The Soviets had a lot of ideas for space projects, and the idea of an orbiter lobbed into space via a double-An-225 is somehow apt. They didn’t think small.

    • It’s of a piece with the general Soviet tendency to want to bash reality into a useful shape, whatever the cost: drain the Aral to grow cotton, embrace the power of the atom, famines as an instrument of public policy. It’s breathtaking in its own horrible way. Bolting dozens of turbofan engines to a flying brontosaurus in an assault on the sky is child’s play by comparison.

  2. The An-225, which I had the pleasure to fly on when I was seconded on it as “flight engineer” by the Soviet Minister of Defence when I was negotiating its possible use as a heavy-lifter in the West, was also the lifter-of-choice for the HOTOL spaceplane.

    • The Buran program was full of one-offs: one An-225, one orbiter, OK two Energia rockets. I wonder if the capability to build another An-225 for, say, Skylon (since we’re talking about HOTOL) even exists any more. Thirty years on and it feels like that boat must have sailed.

  3. I am very very skeptical of -sub-sonic air launch. Aside from some auxiliary advantages (mainly that launch azimuth enjoys very free choices, and that the dangerous aspects of a launch are conveniently removed far from people except the poor flight crew) the speed that a subsonic air launch contributes to the total delta V to orbit is next to zero; 300 meters per second or so, maybe 10 percent more, out of an orbital speed of 7800 or so and a higher total “mission delta V” due to gravity and air drag losses. Getting it under 9000 m/sec is fantastically good and generally involves high G forces. Now if we could go to supersonic speeds, as Skylon does “internally” as it were, and some older designs from the ’60s and ’70s would do with advanced jets or rockets or some combination thereof, realistically a surge up to Mach 5 or so is possible, 1500-1600 m/sec. As those familiar with the rocket equation know, even small reductions in delta V requirement can integrate to big mass ratio savings; 1500 m/sec (approaching a US statute mile per second) can make a big difference in the size of the spaceplane part. But if making a subsonic turbojet in the 1600 tonne takeoff weight range is a staggering leap forward from developed state of the art, imagine how much more difficult to achieve such weights, or even half of it, with a supersonic design! (Not to mention the layout shown here for this OOS would be insane beyond the sound barrier–the assembly has to be configured to be streamlined suitably both with and without the orbiter attached, lots of luck with that! Surely it can be done, but equally surely would involve adding yet more weight to the overall assembly).

    Of course I know air launch is supposed to offer other advantages–much touted is that by climbing up to altitudes where air density is low, perhaps a tenth that of sea level, the problem of rocket engine design, specifically the nozzle, is simplified to optimizing for vacuum performance. With the sort of ker-lox rocket engines the superpowers used for direct vertical launch from Earth’s surface, the difference between sea-level optimized nozzles firing into sea level atmosphere and the same core engine with a vacuum nozzle in vacuum might be a 20 percent or so penalty for the sea level version at launch, in thrust and Isp, the latter governing the mass ratio for a given desired gross delta-V. Taking the rocket to the stratosphere and then firing it means both simplified engine design and a more efficient engine. This is clearly worth something, but the problem is as this article indicates, air launch ain’t cheap–it is bad enough when the airplane you need is a modification of some commercial model available in large numbers and thus at reasonable prices; a one-off that also pushes the state of the art far beyond former benchmarks is going to cost a lot more than the ratio of its weight to the bigger commercial models. Maybe ten times as much, maybe closer to 100 counting the R&D and the clean sheet design engineering. The alternative to spending all that money for one each megaplane is to instead develop a zeroth stage booster rocket system to bring the system one considered “air launching” to the same speeds and altitudes and then fire the main rocket–and in fact, typical first stage boosters deliver the upper stack to speeds and altitudes considerably greater than subsonic and stratospheric–and that’s good because the faster one separates from the first sage, and higher, the less delta-V the upper stack must deliver, improving the mass ratio. A supersonic air launch system is sort of competitive with a ground launched first stage, but subsonic is completely outclassed.

    It might make sense to invest in air launch, even marginal subsonic versions, if the launch frequency envisioned was really high. If it is possible to make the upper Orbiter stack (here ambitiously a single reused stage) have fast turnaround, and to pay for many payloads and many operations, one is getting good value from the expensive airplane and might even justify making many models of it for lower unit prices. If one envisions being able to pay for only a few launches a year though, a much simpler disposable first stage vertical rocket booster is going to be cheaper overall, and then one might invest in making that rocket stage recoverable and reusing it instead, raising the bar of minimum flight frequency for the airborne version still higher. Since in fact I suspect designing a booster to merely put the upper stack to subsonic speed at say 10 km altitude would be inefficient, one instead would have an upper stack that lights higher and faster with a more normal supersonic, higher altitude boost phase, and thus the spaceplane would be lighter and more efficient.

    And I have to wonder, would this spaceplane design really be able to contain over 500 tonnes of hydrogen-oxygen propellant mix in internal tanks anyway? The Shuttle ET contained a bit over 725 tonnes, so this Soviet spaceplane has to have dimensions almost as big as the ET, which was of course much bigger than the Orbiter. And yet bring tanks, engines, crew and cargo downmass (if any) safely back to a landing. It is one thing to make a tank, another to enclose it in a recoverable envelope–and putting hydrogen tanks within a structure poses another problem, which is that hydrogen leaks–having a separate disposable tank means the leaked hydrogen dissipates in the atmosphere, but enclosing it in thermal protection systems for reentry means it is trapped and can build up to pose serious hazards. All of this means the mass per tonne of enclosed propellant is higher than with a simple disposable tank. Can it really close to just 100 tonnes on orbit, or even landing?

    The more generally known MAKS shuttle was going to be broadly similar in concept, being subsonically launched from a giant airplane, but would have a disposable tank with an Orbiter style separate recoverable spaceplane on the tail end of the tank. That I suppose would close with a superior mass fraction to orbit, and greater reusability than the American STS, because the airplane substitutes for the SRBs. The question boils down to, is making a heavier spaceplane that includes its own tanks for reuse but has inferior payload fraction and is overall heavier than a disposable tank system going to cost more money to develop and operate (the plane must be bigger after all) than the cost of making and disposing of a fuel tank every mission? At realistic flight rates I would bet on the MAKS between them, and on a simple booster rocket based system in preference to either.

    • All very good points, but I think I can see the logic behind Tupolev making this proposal.

      I suppose the key thing to recognize is that for the Soviets (and the Americans too) the manned space program through the mid-70s was not about efficiency or practicality, but rather it was a way of signaling their superiority to the world. In the particular case of the USSR this was coupled with a lessened emphasis on economic utility. As I mentioned in my reply to Matthew above, it’s perfectly characteristic of the Russians of the time to go for something awe-inspiring.

      The proposal also came at a time when the air ministry was trying to pry the space program away from the artillery arm of the army, and they did so by essentially proposing *everything* they could think of. This is one corner of that.

      I think it’s also diagnostic of their thinking that another aviation bureau, Sukhoi, considered a hydrofoil launched orbiter, the Albatros I discussed a few months back, at about the same time. It would have had even fewer advantages than the AKS/OOS combination. But they considered it nevertheless.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s