The R-56: “Yangel Works for Us”

Thre possible arrangements of the R-56 rocket

Three possible arrangements of the R-56 rocket. The one on the right is the “4-4-1” module arrangement initially favoured, while the one at centre is close to the monoblock version finally settled on (it is missing the flared skirt necessary to house all of its engines). Original source unknown.

What it was: A four-stage rocket proposed by OKB-586 in the early 60s. It was aimed at the Moon, despite having a payload of 40 to 50 tonnes, making it much lighter than any of the Saturn V, N1, or Energia. It still would have lifted more than any rocket being flown in 2016.

Details: In February 1962 Nikita Khrushchev organized a meeting of the USSR’s Defense Council with the main missile designers in the Soviet Union at his dacha in Pitsunda (a resort town in the Georgian SSR) for the purpose of rationalizing their missile and space programs. The main players were Sergei Korolev with OKB-1 and Vladimir Chelomei with OKB-52, but a third invitee was Mikhail Yangel, the head of OKB-586.

While Korolev had surged to the head of the Soviet space program post-1957 and initially stood first in ICBM development, all based on variations of the R-7 rocket, by 1962 he had lost leadership in the latter to Yangel. The previous November his R-16 had become operational, and its use of storable propellants made it more militarily desirable than the liquid oxygen-using R-9 that was OKB-1’s response. Though the R-9 could be fuelled and fired in roughly the same amount of time, the feeling among almost everyone but Korolev was that storable fuels were the way forward when it came to developing a nuclear strike capability that could be used with little notice.

Meanwhile a fourth man and his bureau was working behind the scenes. Valentin Glushko had been trying to make large engines that used LOX for oxidizer. The tremendous vibration in his prototypes led to combustion instabilities that caused, as they say, “rapid disassembly”. Convinced that the problem could not be cracked, he had come around to storable propellants, and this had become a problem between him and Korolev. OKB-1 was pushing ahead with the N1 and, while storables were considered for that project, the writing was on the wall: Korolev wanted LOX and kerosene, or LOX and liquid hydrogen. A few years previous Glushko could have pushed back effectively, but Khrushchev had been downsizing the USSR’s military aviation efforts, and underemployed bomber-designing bureaus had been growing new departments devoted to rockets—the N1 would end up flying, for sadly abbreviated distances, using engines developed by Nikolai Kuznetsov’s OKB-276.

Glushko hedged his bets by teaming up with Chelomei on the UR-700 and the UR-500, which were aimed at the 70+ and 20-tonne payload targets set by Khrushchev. The former was to be a super-heavy interplanetary space launcher and the latter was a combination heavy LEO space launcher and ICBM. The smaller of the two figures was apparently selected due to the test of the RDS-220 hydrogen bomb (better known by the name given to it in the West, the “Tsar Bomba”) a few months earlier. This 100-megaton demonstrator had come in at just under 27 tonnes, and it was thought that refined versions with about half the yield would come in several tonnes less than that.

These two rockets were OKB-52’s proposal to the Defense Council meeting. OKB-1 countered with the already-underway N1 and, for the smaller launcher, the N2, which was essentially the N1 with its tetchy first stage removed. Seemingly out of worry that OKB-1 would still prevail, Glushko had arranged for another card in his hand—Yangel.

A relative newcomer to the space side of missile work, Yangel had earned a reputation as someone who listened to the military with the R-12 and R-16 missiles, in contrast with Chelomei and Korolev, who were viewed to varying extents as prima donnas, or at least less than entirely focused on military applications of their rockets. Yangel parleyed this approval into an unmanned satellite launch that was to go ahead the next month: Kosmos-1, the very first mission of the soon-to-be-ubiquitous Kosmos program that represented the large majority of Soviet launches from 1961 until the fall of the USSR. Yangel was interested in extending his nascent space work into manned programs, at least to the extent of designing the rockets for them, and he and Glushko had initially worked on creating a rocket, the RK-100, using the same storable propellant engines that OKB-52 was designing for Chelomei. If Glushko failed to unseat Korolev through Chelomei, then teaming with Yangel would give him another bite at the apple.

The RK-100 was a clustered rocket and Yangel was reportedly displeased with the particular design that his OKB-586 came up with. In any case the first comprehensive space policy statement by the Soviet government, made in 1960, ruled out any possibility of it going forward. At this point the focus shifted to another Yangel-Glushko collaboration. Once again a clustered approach was used. Working on the base of a booster “module” resembling the smaller rockets with which OKB-586 had had success, this new rocket consisted of four modules on the first stage, four on the second, and then a core booster being the third and final stage. This proposal was dubbed the R-56, and Yangel brought it and another design, the R-36, to the conference.

What he didn’t do was go head-to-head with Korolev and Chelomei. As initially conceived the R-56 would slot into the space between the 20 and 70 tonne launchers, lifting 30-40 tonnes or so, while the R-36 was much smaller than any of the other rockets mentioned, aiming for a sweet spot in automated satellite launches around 1-2 tonnes to LEO.

The meeting did not go well for Yangel’s crewed space ambitions as by April a turgidly named decree called “On the most important projects of intercontinental ballistic and global missiles and carriers of space objects” was issued. It instructed the bureaus involved to go for the N1 as a space vehicle, the UR500 (which would eventually become the Proton) as both a space vehicle and ICBM, and the R-36 solely as a missile—though it too would become a satellite launcher one day, the Tsyklon. However, in the few weeks of space between the original meeting and the decision, Glushko began lobbying the Strategic Rocket Forces and Dmitri Ustinov about not only the “4-4-1” module version but one with a “7-6-1” configuration that he said would lift 70 tonnes—obviously the direct challenge to the N1 and UR-700 that Yangel did not make himself. His efforts paid off. While not authorizing the R-56, OKB-586 were given permission to at least study the “4-4-1” configuration.

A year later, in 1963, the order for the R-56 was revised to specify that it should lift 40 tonnes to LEO. While Yangel’s bureau studied modular rockets that could handle this new requirement, for all intents and purposes they went back to the drawing board and settled on a completely different approach: a four-stage “monoblock” arrangement, to use the Russian term. This is the familiar, boosterless approach where each stage is singular and is merely put on top of another singular stage—the Saturn V being the most famous example of this. The first two stages of this R-56 did the heavy work of getting a payload into orbit, while the third was used to get it to geosynchronous orbit, if that was the intended destination. The optional fourth stage would be for the extra push needed on lunar and planetary missions.

The first stage would be outfitted with sixteen RD-253 engines, the same one to be used on the UR-500 (which had six) and which was ready to fly in July, 1965. This cluster of engines was actually wider than the intended 6.5-meter diameter of the first stage, so it was installed with a short skirt which enclosed 8.2 meters at the base. The second stage had one of the same engine, equipped with a modified bell tuned for operations in vacuum, as well as a small steering engine that produced 15% of that stage’s total thrust. The third stage tapered from 6.5 meters down to 4 meters in diameter, which was the gauge of the rocket up to the top of its 67.8 meter tall stack. Loaded up with Glushko and Yangel’s preferred N2O4 and UDMH, it would weigh in at 1421 tonnes. Compare this with the Saturn V’s 110.6 meters and 2970 tonnes, or the Energia’s 2270 tonnes (not counting Buran) and 58.765 meters. While not in their class, this new R-56 was heading in their direction. If it had been built to spec, it would have been able to lift a little over 46 tonnes to a 200-kilometer orbit when launched from Baikonur, or 12.6 tonnes to the Moon.

What happened to make it fail: All the meetings and decrees regarding the Soviet space program failed to straighten out the USSR’s lunar program. At the end of 1963, multiple boosters and spacecraft were still in play, and the Soviet leadership had still not even formally authorized an attempt by their country at the Moon landing. In an effort to finally settle things, in March 1964 Yangel proposed to the Military-Industrial Commission that Soviet space efforts be split three ways: OKB-1 would work on the lunar spacecraft, Chelomei’s group would get the automated probes to the Moon and the planets, and he would build the rockets.

The Commission turned him down, reasoning that too much work had been put into the N1 already for it to be replaced now. There was reportedly also some discomfort with the fact that the R-56 would need two launches (at minimum) for a Moon mission, which implied a docking in orbit at a time when the first Soviet docking was more than three years in the future.

Yangel then petitioned in succession both Dmitri Ustinov and Leonid Brezhnev (seven months from becoming leader of the USSR, but then in charge of the space program and a native of Dnepropetrovsk where OKB-586 was based). Neither would back him, and the R-56 was formally cancelled by another decree, “On speeding up work on the N1 complex”, that was made on June 19, 1964.

After the Moon program was finally approved in August of 1964, Yangel’s bureau was assigned to work on the terminal descent/ascent engine for the LK-1, the program’s lunar lander. It thus had the distinction of being one of the few pieces of the Soviet Moon landing craft to make it into space, as it was tested successfully in orbit three times in 1970-71.

What was necessary for it to succeed: The main problem with the R-56 program seems to have been Yangel’s willingness to let go, as opposed to the on-rushing bulls that were Korolev and Chelomei. If he’d been willing to push harder or been a little luckier during the 1962 meeting he might have won the day—Sergei Khrushchev specifically says that he thinks his father would have picked the R-56 at that time if Yangel had presented first rather than last.

On the other hand, even down to 2016 no-one has ever built a rocket with a payload capacity in the 40-50 tonne range (SpaceX’s under-development Falcon Heavy is closest, at 54.4 tonnes). Smaller is fine for almost all launches, and crewed missions absolutely require more if going to the Moon or beyond (barring the construction of a larger craft using multiple launches, which has also never been done). There’s good reason to believe that even if it had flown, the R-56 might have ended up not being good for much of anything.


“Heavy Launch Vehicles of the Yangel Design Bureau, Part 1”, Bart Hendrickx. Journal of the British Interplanetary Society, vol. 63, Supplement 2. 2010

“Heavy Launch Vehicles of the Yangel Design Bureau, Part 2”, Bart Hendrickx. Journal of the British Interplanetary Society, vol. 64 Supplement 1. 2011.

Nikita Khrushchev and the Creation of a Superpower, Sergei Khrushchev. Penn State University Press. 2001.


9 thoughts on “The R-56: “Yangel Works for Us”

  1. Interesting!

    One minor point though – I always heard the cut down N-1 referred to as the N-11, rather than the N-2. And russian friends who follow this stuff did not correct it?

    I could be wrong though…

    Did a CGI version a few years ago.

    • I’ll have to check out your CGI — do you have a gallery online?

      The N-2 name comes from Bart Hendrickx’s paper and he’s generally a meticulous researcher. I’ll do some investigating, but I suspect one of two things is going on. Either the N-2 name was used at that early point, and then the N-11 used later (the N-2 not having been approved back in ’62 suggests maybe two different proposals for the same thing) or else it’s a confusion between the Roman numeral II and the Arabic number 2, i.e., it could be N-II. If you suddenly see it change in this post, you’ll know I found something definitive!

  2. I entered the figures for stage masses, propellant, and thrust from Encyclopedia Astronautica for a launch to 56 degrees and 200 km altitude in the Silverbird Launch Performance Calculator, and it estimated a mere 34 tons to that orbit from Baikonur. (I did not try for Kasputin Yar since I don’t know the azimuth restrictions there; Baikonur launches go at 56 degrees and higher to avoid trajectories over the PRC). I should note at least one obvious typo at EA–the thrust figures for the first stage don’t jibe with 16 RD-253 engines at all! The above disappointing estimate is after correcting that. To be sure, the total mass of the rocket did come very close to the 1421 ton figure you give for the total rocket.

    I found that by adding about 60 tons distributed between the two standard upper stages and doubling the thrust of each, the 40 ton target could be reached, though not for the polar orbit EA cited as the goal. These kludges would be reasonable to achieve the target.

    It also seems to me that with such a rocket, and the 40 ton payload including a hypergolic 4th stage with an engine similar to the (doubled) third stage engine, assuming that this stage requires about 10 percent of its mass to be dry structure, and a 4 km/sec delta-V budget to achieve both trans-Lunar injection and injection into a low Lunar orbit, such a stage ought to be able to put between 8 and 10 tons there. This is enough for a Soyuz type craft that could use a few tons of hypergolic propellant to send itself back to Earth, or for a lunar landing craft similar to the LK eventually designed by Mishin along the lines indicated by Korolev. With some 9 tons or so to play with, a crasher stage would allow a lander slightly heavier than LK as ultimately designed. Therefore, with R-56 rockets of this augmented design that does really achieve 40 tons in LEO, it might have been possible to do a lunar orbital mission analogous to Apollo 8 with one launch, or an LK single cosmonaut short landing with two launches–either the Soyuz and the lander launch separately to LEO with their separate supplies of TLI/lunar injection propellant and then dock to consolidate the TLI fuel and boost to the Moon and low Lunar orbit together as with Apollo, or more straightforwardly each craft boosts from LEO to lunar orbit separately. Presumably the lander would be sent on first, and if it is successful in reaching lunar orbit, the Soyuz could then be launched after it, for a lunar orbit rendezvous.

    It did seem strange that this could be done with hypergolic propellants, but the Soviets did push for high ISP in the propellants they chose to use, and so the third stage engine in Yangel’s proposal would have had ISP of 327 seconds, which is considerably better than American made kerosene-oxygen engines such as the F-1 achieved. Also, hypergolic propellants store densely so the dry mass of the stage could be lighter. Thus they hypergolic option is more competitive with even hydrogen-oxygen than it looks at first glance.

    This said, I personally think hypergolic rockets on such a large scale–close to 3 times the mass of Proton after all!–should be avoided, because of the severe hazard such huge masses of severe and active poisons pose. I can say that, but in the 1960s neither American nor Soviet authorities would listen to these qualms. A catastrophic pad failure would be bad, at least twice as bad as a Proton failure, but if these are no more common than Proton failures I think the Soviets would soldier on with them.

    You raise the objection that the global market has not actually found a use for a 40 ton to LEO launcher, but I wonder if even at this late date, a market as such effectively exists. It certainly does for certain niches, such as launching geosynchronous orbit com-sats. I estimate, with Silverbird calculator’s help, that the three stages of the standard rocket would send some 15 tons on a transfer orbit to an equatorial geo-synch orbit, and even bearing in mind the rather severe inclination change from 56 degrees to zero, at altitude this is much moderated and a circularizing 4th stage ought to put at least 6 tons useful payload there. I do think the market could find that attractive, especially if it were for sale in the mid-70s. And of course the Soviet program in particular was decoupled from markets as such even more than Western programs. In the case of the Soviet’s own programs I suspect it would be a case of “we built it, and they come” with heavier payloads than historical. Certainly the Salyut space station program could have benefited from 40 ton LEO stations, and if they took to assembling larger stations out of separately launched blocks as was done with Mir, the resulting station could outmass Skylab after just three launches. The Soviet military would also find uses for it. I figure basically the Soviet program would simply be larger, in tonnage, in proportion to how much the R-56 outperforms Proton.

    Depending on how fast Yangel’s team could shake this rocket down (Proton’s career is not encouraging, it did not become the rocket we know today until the mid-70s) the Russians might have been able to scoop the Americans at every stage of the Lunar program, doing a flyby before we could (without adopting the panic expedient of developing Gemini to do so quick and dirty) followed by a Lunar orbital mission (one R-56 to full 40 ton to LEO spec could do that, as opposed to two Proton launches and EOR required to marginally enable it) before we could do Apollo 8, and possibly even a landing with two launches and LOR. Now even if Apollo 11 were beaten to the Moon, we would be able to argue the Apollo landing missions with two astronauts staying several days were scientifically superior. But with enough R-56 launches even this could be countered and topped. Lunar Gemini shortcuts might have worked for the less ambitious missions, but a Gemini based landing mission would be very marginal and not ready much earlier than Apollo was either. Even if delays mean the Soviets cannot beat Neil Armstrong to the Lunar surface, going with three launch missions that pre-position substantial resources for a minimal two-cosmonaut lander can be competitive with Apollo missions, and a 4 or 5 launch mission could pull ahead.

    Interestingly I considered what difference a hydrogen fueled upper stage might make and only got 2 more tons of payload. A hydrogen/oxygen second stage on the other hand, if we could assume the engines only get a modest 400 sec ISP in vacuum and somehow or other we maintain the full thrust equivalent to 2 RD-254 engines can add more than 10 tons to the payload,

    Now a 25 percent increase is something to fight for, but by and large the hypergolic systems would have offsetting advantages.

    As I mentioned, the data published on R-56 are rather hit and miss and probably do not indicate any sort of properly developed design. But there is reason to think Yangel’s people could have shaken it down had the regime chosen it, and had they been serious about being drawn into Kennedy’s challenge, or even about a methodical progress toward a Moon landing and space station development for purely internal reasons, this might have been the right path for them. I personally would prefer to see something kerosene-oxygen based instead of hypergolics, but plenty people like hypergolic rocket systems just fine.

    • Wow, excellent analysis Shevek! Thanks very much.

      All I have to add to that is that the Chinese have had success with manned launchers using hypergolic fuels. So while it may not be clear that it’s equivalent to cryogenics or liquid oxygen and kerosene, it can certainly be done.

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