Soyuz rocket flaw found?

By Phil Plait | September 13, 2011 3:00 pm

Various media (like the New York Times) are reporting that the flaw that sent a Soyuz rocket crashing into Siberia last month has been identified. Judging from the telemetry, Russian space agency officials are claiming it was due to a clogged pipe that kept the third-stage motor running. When the pipe stopped feeding kerosene to a turbine, power failed to a fuel pump, and the third stage burn stopped.

The rocket had supplies destined for the International Space Station. The loss of the supplies is not really the issue, however. A similar rocket is used to bring up astronauts, and until and unless the problem is fixed, it’s too risky to send anyone up. This may mean the ISS will be de-crewed in the next couple of months.

That’s the immediate issue. However, this rocket failure is a bit more problematic. The flaw appears to have happened in the manufacturing process, and that’s the sticky point. If it were something procedural setting up for launch (tanking up the rocket, for example), that’s a relatively easy fix. If it were some malfunction in the machinery used to make the rocket (again, possible to find and fix in general), the flaw would’ve turned up more often. That leaves something that happened by accident, and that’s not terribly reassuring. How do you know when a problem like that is mitigated? What was the exact circumstance that led to the pipe being faulty?

Until the Russians can figure that out, implement a fix, and make sure that problem is eliminated — and that other similar mistakes are prevented — it’s too risky to allow astronauts to fly on board the rocket. And that, currently, is the only way to get humans up to the space station. The only other rocket currently capable of getting up there is the SpaceX Falcon 9, which has not yet proven its reliability, and in any case has not been cleared by NASA to carry humans (the SpaceX Dragon capsule could theoretically do that, but needs to pass a set of strict regulations to be (pardon the expression) street legal).

I’ll note the Wall Street Journal is reporting that with the flaw being found, Soyuz flights could resume as early as October. Interestingly, the WSJ article reports that NASA is optimistic, while the NYT article says the Russian commission that investigated the crash is more cautious. The WSJ also says NASA may have a statement out this week, so we’ll see. I don’t see any mention of this on the NASA site (an ISS telecon will be held on Tuesday, September 20), but I’ll keep my eyes open.

Also note that three of the six astronauts on the ISS will be coming home in a Soyuz capsule on Thursday, September 15th (note the capsules are unrelated to the rocket that failed). They plan to undock and make the de-orbit burn at 03:06 UT Thursday (23:06 Eastern US time Wednesday night) and land in Kazakhstan 04:01 a.m. UT. That will leave three astronauts on the ISS; they will either have to come home to Earth by mid-November due to the limited fuel lifetime of their return capsule, or have a new capsule put in place. And that won’t be possible until flights resume.

This whole thing is still unfolding, so stay tuned — and you should follow Universe Today as well, which tends to have information on these topics up pretty quickly.

Image credit: NASA


Comments (40)

  1. t-storm

    I think everyone saw something like this happening but not before the shuttle’s brakes had a chance to cool off.

  2. Orlando

    Off-topic: What do you think of this?

    IL&M? SETI? Kepler? WTF!

  3. andy

    Despite these problems, Soyuz is still amazingly reliable (ESA describes it as “the most reliable means of space travel”). It is quite an old design: the Russian philosophy with their launch vehicles appears to be “if it ain’t broke, don’t fix it!”. This seems to have worked somewhat better for them than the American programme which ended up with the amazingly expensive Space Shuttle that never really delivered on what it was intended to do when it was first proposed.

  4. JC

    If they’re that worried about the rocket, couldn’t they send up an empty capsule ? If it fails launch, well, it’s fall back to plan B.

    Otherwise, they have a brand new capsule for the astronauts already up there. Sure, it’d be an extended mission for those folks already up there, but at least they’d have a ‘fresh’ capsule.

    ( I’m assuming that the primary reason to evacuate is because the capsules already up there are reaching the end of their operational life. If there are other reasons, I guess just sending up a new capsule wouldn’t be enough to avoid evacuating, eh? )

  5. Chief

    re 2, andy

    The proposed shuttle back in the very early 70’s was nothing like the current shuttle. Everyone putting their fingers in the design so it can do multiple functions increased the payload bay to the point that a requirement for strap-on boosters was required. I would have preferred the launch to mirror what has been built for spaceship one design, ie assisted lift to low earth orbit by returnable booster plane and the shuttle to continue from there.

    I find it interesting in the wording by NASA on the problems in the failed launch on stage three as feel good while the russian attitude is play it safer. Kinda a reversal of thinking from the early days.

  6. Bob Portnell

    I’m with JC: Set up a Soyuz crew capsule with the Progress automated docking software, and send it up empty. Valid fix for on-orbit lifetime issue, and, hey! useful flight test of reworked Soyuz booster stage issue.

  7. Jamey

    “Until the Russians can figure that out, implement a fix, and make sure that problem is eliminated — and that other similar mistakes are prevented — it’s too risky to allow astronauts to fly on board the rocket.”

    So what you’re saying here is that any chance of a failure greater than zero is too risky. That we should just give up on manned space flight forever. Too bad.

  8. vince charles

    “The only other rocket currently capable of getting up there is the Space X Falcon 9”

    No, the only other rocket _claimed_to_be_ capable of getting up there. Elon Musk is a shameless self-promoter, and when necessary, liar. The fact that he succeeds more often than not put his individual statements in the “Valley of Death.” You can’t dismiss him outright, but you can’t take him on his word, either.

    In this particular case, the SpaceX Falcon 9 has launched twice, with no major anomalies. However, one aspect of its design is tolerance to certain minor anomalies, and questions have been raised about that. Even if the two flights had been flawless, two flights is not a statistically-significant record. Not even by the standards of the Shuttle, which was considered rushed after only four test flights.

    In this particular _particular_ case, the SpaceX Dragon on the second flight was far from human-rated. It was more than a pressure vessel and heat shield, but that’s not saying much. In particular particular PARTICULAR, SpaceX is nowhere even close to a launch abort system. There is no agency in the world anymore that will place its crew on a capsule with no abort system. And no, you cannot simply add some existing abort tower to some capsule, any more than you can swap capsules between boosters arbitrarily.

  9. don gisselbeck

    When was the last time no one was in orbit?

  10. Robin

    The WSJ statement that NASA is optimistic does not mean that NASA believes a resumption of flights is imminent. NASA’s statement to the WSJ could just mean that NASA is optimistic because the problem was identified and that the first step–and possibly the most important step–toward resumption of flights is complete. Now the Russians can move forward with inspections to see if the problem exists in other launchers, with the fix, and with testing.

    Skipping inspections and testing after the fix in favor of a flight test is not something any respectable engineers and test managers are likely to do, especially when it comes to manned flight. All people involved in space flight understand that risk is never zero. It’s stupid to suggest as much. They do however understand that there is a lot to be lost when something goes wrong and fatal with a manned flight. Just how much do you expect the public to pay? It’s not very realistic to expect public support not to decline if we start doing things the reckless way.

    The attack on Musk isn’t even worth discussing and is entirely self-serving.

    It’s going to be a few years before SpaceX is ready to loft men in space. They’ll need to have a series of successful launches and successful dockings at the ISS. The Dragon’s first cargo configuration docking is due to happen in November. As mentioned, SpaceX still needs to carry out the testing for its novel escape system. While SpaceX is ahead of the game so far, they still have some significant stuff to do before manned missions can start.

    It would seem prudent to man rate Atlas V’s and Delta IV Heavies as well. I think it’s better for everyone and with more people working to the same end, the chances of success should increase. Alas, bureaucracies, Congress, and NASA haven’t always displayed prudence.

  11. dcsohl

    Don@8: On October 31, 2000, Soyuz TM-31 lifted off carrying the crewmembers of ISS Expedition 1 to their new home. They docked with the ISS on November 2, and the station has been continuously occupied since that time. So the last time no one was in orbit was launch-time of Soyuz TM-31 — 3969 days, 18 hours, 53 minutes and counting. It would be a shame if that streak came to an end over this.

  12. Nobody

    Clearly, the answer is that we (humanity) need more human-launch providers. But I suspect that to widen that foundation, we need a need (you read that correctly) to put more people in space. We don’t have that need :(

  13. Messier Tidy Upper

    This is somewhat tangential to the topic at hand but still of interest hopefully for those who haven’t already seen this.

    I’ve been feeling very pessimistic about the US manned space program (heck all manned spaceprograms really) lately – with the Space Shuttles now retired, ‘Constellation’ scrapped and now the ISS facing being decrewed thanks to the Soyuz failure – but this news :

    does provide some cause for optimism. :-)

    Good to see a human spacecraft actually under construction for once – but will we ever see it fly and, if so, when?

  14. andy

    IL&M? SETI? Kepler? WTF!

    My guess is they’ve found a circumbinary planet of some kind, and are going to try to tie it into some lame 1970s sci-fi movie that seems to still have an inexplicably large fanbase over 30 years later.

  15. DrFlimmer reports that a new crew could go up as soon as November 12, pending a good launch of a Progress vehicle on October 30.

    If all goes well, the ISS would not be abandoned, and be restored to 6 people on board with another launch around December 20.

  16. “Until the Russians can … implement a fix, and make sure that problem is eliminated … it’s too risky to allow astronauts to fly on board the rocket.”
    Well, maybe not desirable to fly astronauts but also not particularly risky, as the Soyuz has quite, um, survivable abort modes, other than the retired shuttle. See also

  17. It appears that this was a one-of manufacturing flaw (something went wrong in this particular manufacturing process for this particular rocket engine used in this particular launch), not a structural manufacturing flaw that affects all similar rocket engines. It wouldn’t surprise me if it was simply a human error.

    Those are things you can *never* completely prevent (all you can do is check and double-check, and train your workers, but that still is not 100% effective). With a long line of launches, it is bound to pop up once at some point.

    It means chances of a similar malfunction are small. It means there is nothing structural in the rocket engine design that makes it dangerous. It means they can resume launch, and that this is quite safe.

    Also, it is *not* that the Russians have a history of this kind of one-of flaws with their rocket engines.

    So contrary to Phil, I’d say this is actually reassuring. These rockets *are* in itself quite safe. Of course *any* rocket basically remains a bomb you strap yourself to.

  18. flip

    @Jamey, #7

    I think you just invented a strawman.

    …I agree with dcsohl though: I really hope the record for orbital occupation doesn’t end (this soon). Although obviously, I’d prefer it if everyone were safe. Space flight records come second.

  19. Ferris Valyn

    Dr. Plait

    You are wrong in your article. You said

    “The only other rocket currently capable of getting up there is the Space X Falcon 9, which has not yet proven its reliability, and in any case has not been cleared by NASA to carry humans”

    Thats not true. The Ariane V & H-2 from Europe and Japan both have sent stuff to ISS, (and Proton, from Russia, has launched whole modules, although I don’t know if the upper stage of Proton is similar to the Soyuz or not).

    Additionally, the Atlas V & Delta IV could put payloads at ISS (and both have flown multiple times – Atlas V is 27, and Delta iv is 16, I believe).

    The real issue is a dearth of human-capable spacecraft (which you do point to)

  20. MattF

    Jamey: So what you’re saying here is that any chance of a failure greater than zero is too risky. That we should just give up on manned space flight forever. Too bad.

    No, Jamey, that’s not what he’s saying there. That’s your over-reaction to it.

    The problem is that the chance of failure is currently unknown. Until we know how to mitigate the risk — not eliminate it, mind, just mitigate it — it’s irresponsible to charge ahead with the assumption that we know it’s a one-off.

    In my opinion, it’s likely a one-off. But I’d rather have evidence of that before strapping people to the machine again.

  21. Ferris Valyn

    #8 Vincent Charles – You are ignoring the Atlas V and Delta IV. Both can reach ISS just fine

  22. Ben H.

    “The only other rocket currently capable of getting up there is the Space X Falcon 9”

    If you’re going to say the Falcon 9 is capable of visiting ISS then the Delta and Atlas rockets are as well. They just need to be man-rated just like Space X! No big deal…

    Tongue-in-cheek aside, your statement is false. The Soyuz is the only rocke that can currently take people to ISS anytime in the near future (ie, 4-8 years).

    If you mean that it’s the only other rocket that can go there at all (manned or not) then you are wrong as well – the Japanese have their H-II rocket for HTV cargo and the European’s have their ATV vehicle launched on the Ariane 5.

    – Ben H.
    NASA JSC, Houston, TX

  23. Ferris Valyn

    Ben H – Sorry, but manrating an Atlas V or Delta IV is NOT that big of a deal (at least as far as anything can be not that big of a deal in spaceflight).

    Atlas V has 27 successful flights, and Delta IV has 16. Those are not nothing. And we trust not just hugely expensive satellites on it, but satellites that are vital to warfighters on the ground.

    They can be ready to put humans in space in 3-5 years. Not some time after 8 years

  24. Ben H.

    Your statements are 100% in agreement with my point and I agree with you. I admit my time estimate may have been a bit of a shoot-from-the-hip.

    My point was that if your basis of a rocket being able to go to the ISS is simply if its large and powerful enough, listing just Space X is completely disingenuous. But being “currently capable” of going to the ISS involves a contract with ISS program, a successful flight test program, etc etc. Which is a multi-year endeavour as you point out. There are currently NO rockets capable of going to ISS with crew other than the Russian Soyuz family of vehicles.

    People like Phil who post statements like what I quoted above are giving the impression that Space X is the only company out there with a rocket with the technical specs and preparedness to go to ISS, and it makes it sound like they can send people up there tomorrow. This is a distortion of the technical environment – as you so helpfully elaborated on. Just because Mr. Musk sat down last December after their first flight and said “we can fly crew tomorrow!” doesn’t mean its really that easy. I love what Space X is doing and hope them the best (hell, I even think it would be cool to go work for them some day!) but the impression everyone has that they are going to save America’s space industry before the end of 2012 is grossly out of touch with how these complex expensive programs evolve.

    – Ben H.
    NASA JSC, Houston, TX

  25. Ben H.

    Note also that Space X DOES NOT HAVE A CONTRACT to fly humans to ISS.

    – Ben H.

  26. Ferris Valyn

    I fully agree that, while SpaceX is impressive, the idea that they alone will save NASA is false (certainly not by the 2012 date).

    However, I would submit that it is Commercial spaceflight in general that will save NASA. And my ire was specifically raised by your “not capable for 4-8 years”

    I have no illusion that any of the commercial providers can not have a vehicle flying by the end of the year that can get astronauts to ISS (because, as we’ve both said, you need a spacecraft, not just a rocket). But I have no doubt that they can within 3-5 years.

  27. Ben H.

    3-5 years is possible if properly contracted and funded by NASA and congress. Right now its looking like CCdev funding will be much too low to meet the TECHNICALLY feasible goals you identify. Unfortunately, more goes into it than technical feasibility, which is partly my point.

    And the difference between 3-5 and 4-8, in the context of my original point, isn’t that important. A launch 3 years away most certainly will not help the current issue with Soyuz rockets. I would imagine with one unmanned flight and no crew contract, it would be at least 3 years for Space X as well. However, Phil seems to think that it’s simply a matter of NASA signing a paper they are sitting on and Space X can come save the day.

    – Ben H.

  28. @#3 Andy – so the ESA says that the ONLY means of space travel is the most reliable means?

  29. Robin

    @VinceRN (#25): That is a misreading of the ESA statement by you. Soyuz’s safety record extends back to 1971, when the last fatalities on a Soyuz flight occurred. Soyuz fatality rate is lower than that of the Space Shuttle (which is of course retired) and also lower than Apollo’s. No one outside of China has access to their launcher, and its 3 manned Shenzhou launches may not be considered statistically significant (purely a mathematical construct for reliability studies, not a measure of success with Shenzhou).

  30. andy

    @VinceRN: No, where did you get that impression? I certainly never said that, nor did ESA.

  31. Jeremy

    I third JC (#4) and have been thinking the same thing. Any indication that this idea has been considered? or am I missing a flaw in this?

  32. vince charles

    10. Robin Said:
    September 13th, 2011 at 7:52 pm

    “The attack on Musk isn’t even worth discussing and is entirely self-serving.”

    Sounds like you have no idea what sort of whoppers come out of Elon Musk’s mouth (/keyboard). And yes, that’s whopperS, because one would be excusable, but their pattern isn’t.

  33. Buzz Parsec

    Human-rating a launcher (and a spacecraft) isn’t just a matter of launching a bunch of them and saying “See, it works!” In fact the standards for human-rating are actually a little vague because NASA never had to publish them before for other people to implement. So it’s kind of a moving target right now.

    However, the Falcon 9 and Dragon were both designed to meet all the applicable standards (which have to do with things like structural margins, redundancy and systems monitoring) available at the time, except, obviously, the current version of the Dragon doesn’t have an escape system. Neither the Atlas 5 nor the Delta 4 were designed to be human rated, so upgrading them will be expensive. Atlas 5 has some human-rated heritage (its engines were originally built for the ill-fated Soviet N-1 moon rocket) and already has a lot of the needed redundancy and monitoring built into it, so ULA says it will be much cheaper to human-rate than Delta 4, which basically has to start from scratch. (ULA is a partnership of Boeing and LM which assembles and launches both rockets.) One problem is the Atlas isn’t powerful enough to launch many of the commercial space ships, including the Orion, without solid strap-on boosters, which many people would prefer to avoid. The Delta 4 Heavy can launch any of them, but has a much longer and more expensive path to being human rated. (The current plan is to launch the 1st test flight (no people on board) of the Orion in a couple of years on a D4-H, but it doesn’t need to be human-rated to do that.) Falcon 9 has a huge head start on this, since it can and already has launched a Dragon without assistance. I don’t know if the Falcon 9 is officially human-rated yet, but it is pretty close.

    The European Ariane 5 was originally intended to launch the small Hermes shuttle (long since cancelled), so it was designed to carry humans from the start. I don’t know how much further it has to go to be human-rated.

    The spacecraft are another story. The only one that has flown is Dragon, and SpaceX claims that if humans had been on board, it would have been uncomfortable, but they would have survived. (They might have needed a SCUBA tank and mask for oxygen.) But it is still far from human-rated. There’s no escape system yet (though it was designed with one in mind), there’s no onboard control system (though knowing SpaceX’s heritage, it is quite likely all you need to do is plug a laptop PC into an Ethernet port and run the right software, maybe just a web browser, or maybe it’s an iPad app :-)), there are no acceleration couches for the crew to strap into (which makes launch and landing pretty uncomfortable and potentially dangerous.) The environmental control system (O2 & N2 supply and regulation, CO2 removal, humidity control, heat and cooling, lighting, etc.) is pretty rudimentary, but they’ll need this for cargo flights to the ISS (since otherwise it wouldn’t be safe for the astronauts to go inside to unload it), so that will be ready for the next unmanned test flight. Also, the 1st flight didn’t have solar panels, and so was limited by its batteries to about 4-5 hours in space. The upcoming flight will have the solar panels. I’m sure there are hundreds of other items that need to be finished as well.

    Still, that is much further along than any of the others. The 1st Dragon that will dock with the ISS is supposed to arrive at Cape Canaveral this month (it might already be there.) LM just started constructing the 1st Orion test ship, which is supposed to be launched in 2013 (without people on board.) On the other hand, they have already tested the Orion launch abort system at least once. The 1st Orion test flight will put them at about the same stage as SpaceX was last December, but without a human-rated rocket to launch it.

    The other commercial developers are quite a bit farther behind.

  34. vince charles

    “Neither the Atlas 5 nor the Delta 4 were designed to be human rated, so upgrading them will be expensive.”

    You’d be surprised. The cost has been estimated, and is not noticeably worse than Gemini modifications to the Titan II. You can argue that the Atlas V propellant pairings are more risky to human missions, but in terms of the vehicle modifications, the detection and command systems function the exact same way.

    “Atlas 5 has some human-rated heritage (its engines were originally built for the ill-fated Soviet N-1 moon rocket)”

    You mean Energia/Buran, and also a proposed Soyuz lifter using the Zenit. But that one never flew.

    “One problem is the Atlas isn’t powerful enough to launch many of the commercial space ships, including the Orion, without solid strap-on boosters”

    No, it’s just the Orion/MPCV that’s too heavy. Since the others are short-term, LEO capsules, they are nowhere near as massive as Orion.

    “LM just started constructing the 1st Orion test ship”

    No, LM started constructing the first spaceworthy, livable MPCV. At least one, and more likely two prototypes have already been completed, to various standards, for various tests.

  35. Buzz Parsec


    I heard numbers like $1 billion for the Delta 4, half that or less for Atlas 5. But I’m having a very hard time finding actual numbers. According to;wap2, the Delta numbers are too high and include other things, while the Atlas numbers are too low.

    I thought ULA had issued a press release or testified about it somewhere, but I can’t find it. Do you have a link?

    You’re right about the RD180 engines. I was thinking of the NK-33s, used in the Taurus 2. (The RD180 engine used by the Atlas 5 was derived from the RD170 used on Energia, which was also intended to launch humans.)

    I was unclear about the test Orion. I meant the first one that could actually be launched into space, i.e. the one for the 2013 unmanned test flight. I was trying to distinguish this capsule from later ones that could potentially actually carry people. You are right they’ve built a number of engineering test capsules and prototypes, such as the one used for testing the launch abort system and the one used for water landing tests, and a bunch of others, but none of those was actually intended to travel in space.

  36. Joseph G

    I know I’m not exactly a rocket scientist (heh), and I may be underestimating the difficulties involved, but cripes – the Apollo program went from crazy idea (only one American had been in space when Kennedy made his famous announcement) to moon landing in 7 and a half years. SpaceX HAS fully constructed rockets, and is capable of building a new one every 3 months. I realize that safety is a priority, but why isn’t NASA double-timing their efforts to get the Falcon 9/Falcon Heavy systems human-rated? And fine, if Spacex tech is too unproven, what about the Ariane?

  37. Brad Morton

    Joseph G: very simple, Apollo had $20bn in 1960’s dollars that it spent. I dunno what it would be in today’s money but its got to be hundreds of billions.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


See More

Collapse bottom bar