Scaled Composites' Dirty Little Secret

The public is incredibly easy to fool. Way back in 2004, Mike Melvill made history by flying Burt Rutan's beautiful creation SpaceShipOne across the unofficial border to space, twice, and later that year Brian Binnie did it again, winning the Ansari X-Prize and raising the hopes of all that private access to space had finally arrived. But how did they do it?

There is no question that Burt Rutan is a natural genius at aircraft design. His true innovation on SpaceShipOne was the shuttlecock styled effortless reentry system, and in particular, the ease of replacing these two large booms after a few flights to mitigate wear. SpaceShipOne/Two is a glider, and just like the Space Shuttle the wings are "only" used on the way down. There are wings used on the way up, of course, they are on WhiteKnightOne/Two, but once separated from the carrier aircraft the only lift is generated from the rocket.

As smart as Burt Rutan is, he's not a rocket guy. For SpaceShipOne he turned to Tim Pickens, a gifted rocket experimenter who became Scaled Composites' Propulsion Lead Engineer in 2002. It didn't last long, he was only there for a year, but in that time he gave them a motor. Think about that for a minute - it's truly remarkable. Because of that incredible pace the motor design was as simple as possible but no simpler, as they say. Being a hybrid, it shakes like a solid, along with half the reusability of a liquid.. and for some reason it can't restart and had half the performance of both. But it did the job.

Even before winning the prize, Scaled Composites signed a deal with Richard Branson to supply Virgin Galactic with a design for a new set of vehicles and form The Spaceship Company to put them into production. I'm not sure anyone thought it would take as long as it has, but Rutan set to work and delivered the scaled up vehicle designs with the creative names of WhiteKnightTwo and SpaceShipTwo.

Unfortunately, those designs also included details for RocketMotorTwo, which Scaled Composites decided to do in-house this time. Three years passed. In 2007 there was an accident which killed three and injured three more. Although the investigation cleared the company of any wrong doing, it was apparent that they were in a hole that they weren't climbing out of quickly enough.

In mid-2008 WhiteKnightTwo was unveiled. The media ate it up, but some of the statements made by Virgin Galactic's Will Whitehorn to promote the utility of the vehicle were a bit concerning. Everyone wanted to know: where's SpaceShipTwo? And the response seemed to be "we don't need it."

Late last year SpaceShipTwo was unveiled and Virgin Galactic had a big party. Since then, there has been captive carry tests and we're told there may be drop tests by the end of the year. It's all getting very exciting!

Oh I'm sorry, I was talking about the rocket wasn't I? I guess I got distracted by all the shiny white aircraft that I forgot all about it. Well, it seems Virgin Galactic have too.

Sometime between the 2007 accident and today Scaled Composites figured out that they're an aircraft company, not a rocket company, and decided to call in the Sierra Nevada Corporation to get back on track. Since then they've done hot fire tests which, of course, have been complete successes. Oh, they're using ablative nozzles too? Eww.

I've asked a lot of rocket professionals about these scraps of information and, in private, they've all told me the same thing. In public, "how would Jeff Greason say this?" is the what-would-Jesus-do of the space community, and this is how he said it:

In a sane world, a company that has a vehicle without an engine would purchase one from a company like XCOR, which specializes in engines.

So why can't Scaled Composites just buy the best rocket engines in the world (and yes, XCOR's engines really are that good). It would seem to be the rational decision. Apparently, the answer is simple: Rutan sold Branson a hybrid, so he has to deliver a hybrid. It doesn't matter how much hard won experience tells him that hybrids are not safer than liquids (or solids!). It doesn't matter that there's now acceptable engines that you can buy off-the-shelf that just weren't available 10 years ago when Rutan made his decision to go it alone. To change the deal now would loose face and more than likely have contractual implications.

And that's Scaled Composites' dirty little secret.


  1. Anonymous4:27 PM

    Disturbing. I hope they're more on the ball than they appear.

  2. Peter Lykke5:02 PM

    A couple of points:
    The hybrid motor for SS1 was developed by Spacedev after they had won a competition with another supplier (forgot the name of that one). Later, of cause, Spacedev was bought by Sierra Nevada Corporation.

    Hybrids are really not a bad choice for a plane like SS2 - it is the most simple, throtleable motor you can get.

    But the rate of the hot fire tests are disturbing - try to look at the dates in the link to scaled you provide. There should be a lot more, as a hybrid is a trial-and-error thing

  3. Yeah, Spacedev was the subcontractor Tim Pickens chose to build his design.. I didn't mean to imply that Tim carved the motor from cheese or something.

    As for those reports, they're incredibly sparse in both number and detail. Hopefully SNC are providing more detail to SC, but even the little that is leaking out looks bad. How many times do people have to try ablative nozzles before they realize it's unreliable?

  4. AshleyZ5:59 PM

    Wow, they've only done 4 firings in a year and a half, despite a concept of operations that's supposed to involve multiple flights per day? I guess that gives XCOR and Armadillo some breathing room.

  5. Dave Salt6:41 PM

    Whitehorn gave a talk at SpaceOps 2010 in April and during questions I asked him about the engine's status. After a few minutes of bluster in which mentioned how they'd looked at a range of designs, he said that everything was well but I got the distinct feeling he wasn't "comfortable" with his answer.

    Beyond questions of engine status, my biggest gripe with their design is that it doesn't seem capable of rapid turn-around and re-flight (i.e. several times a day), which will either constrain their operations or demand a much larger fleet. Either way, it seems to me that their system is going to be far more expensive to operate that an equivalent liquid, which will make them much less capable of competing with ventures operating XCOR or Armadillo vehicles.

  6. Dave, heh, I put my hand up to ask a question at ISDC and a bunch of people around me started shuffling away as they've heard me rant on about the rocket before.. much less politically correct than this I should add.. but I promised I wasn't going to ask about it. :)

    And yes, the number of seats in SpaceShipTwo vs Lynx is directly related to turnaround time. If you can fly more frequently you don't need as many seats, and that has a knock-on effect on the experience, etc. As far as I can tell VG (or whoever is going to do operations..) will pull the spent propellant grain out, complete with the ablative nozzles and replace it after every flight, then refuel the liquid nitrous.

    The shuttlecock booms will presumably be good for tens of flights.. I'd love to get up close to SpaceShipOne and see how much of the skin is left under the paint. That's not to say it isn't a brilliant design, it really is, it's the lightest heat shield that's ever flown, right? It just comes with a trade like everything.

  7. Anonymous10:06 PM

    Isn't it a bit odd to say
    "the best rocket engines in the world (and yes, XCOR's engines really are that good)"
    Doesn't sound very impartial.

  8. Anonymous1:57 AM

    A few points:

    >The hybrid motor for SS1 was developed by Spacedev after they had won a competition with another supplier (forgot the name of that one)<

    Environmental Aeroscience Corporation (eAc) was the second team using a cylindrical port fuel grain design.

    >It doesn't matter how much hard won experience tells him that hybrids are not safer than liquids (or solids!).<

    Because of the apparent simplicity of the hybrid design, some people believe they can jump right in and build giant hybrids without truly understanding what’s going on from the experience of many subscale tests. Bad designs are still bad designs! Watching that Heat1 firing video was painful! Without generating debate, my experiences with Liquids, Solids & Hybrids tells me Hybrids are safer….now define “safer”!

    It’s certainly not hard to argue that hybrids can be developed faster at higher thrust and burn time levels than liquids. There certainly were time issues with the SS1 Xprize flights. Why change what works?

    >So why can't Scaled Composites just buy the best rocket engines in the world (and yes, XCOR's engines really are that good).<

    I believe part of the issue is the use of the integral composite oxidizer tank as part of the airframe. Rutan/Pickens had concerns about cryogenic issues with composites (that is what they do after all!) This pointed to N2O as the oxidizer of choice rather than LOX. Other that the fact they have a pressurized 800 PSI tank (which they thought they could handle) N2O was considered the lowest risk.

    The SS1 motor had a peak thrust of 15K and a burn time of +60 seconds with a total impulse requirement of 850K Lb-secs. I would estimate the peak thrust for SS2 at 50K for 60 seconds and a minimum total impulse of 2,500K Lb-secs.

    With all due respect to XCOR & Armadillo, It’s easy enough to have high reliability with a few thousand pound thrust engine you test it a hundred times. I don’t believe they have fired anywhere near the thrust levels or total impulse required for SS2.

    I can imagine that if SS2 is successful (TBD!) and there is sufficient demand for multiple flights per day, that Rutan might eventually try a liquid for “Fill & Fly” operations. Until that day I believe Hybrids offer a quicker path to space, particularly if Scaled is building the composite CTN to sell to VG as a continued revenue stream.


  9. TGV has fired a 30K throttling liquid engine
    and can provide a solution.

  10. I've witnessed a firing of a HTP Rubber Hybrid in this size range. I would not choose a hybrid, it has all the drawbacks of both a a solid and liquid with additional problems like the difficulty of controlling the mixture ratio.

    I would use a HTP hydrocarbon biprop. It could use the SS2 integral metal lined Nitrous tank for the HTP and the mixture ratio is skewed enough that you can build tubular hydrocarbon tanks that bolt in place where the hybrid grain presently goes.
    It would have better density, and better isp.
    It could easily be regenerativly cooled for fill and go ops. Switching to LOX would be a lot harder as the Nitrous tanks is integral with the SS2 airframe and probably not Lox compatible for thermal reasons.

  11. Vito Corleone3:31 AM

    For SSOne, there were time pressures. Everything had to be done by the end of 2004 or the Ansari X Prize would have lapsed. No money to pay back Paul Allen for his investment. So they built an engine that would do the job by the deadline they were given.

    One of the disadvantages of Mr. Diamandis' prize approach: it pushed a great leap in progress while leading SC down a blind technological alley.

    Now they're stuck with a design that might have to be changed significantly to accommodate another type of engine. They've also sold the public and their clients on the supposed safety and environmentally benign nature of hybrids even though I doubt either claim is true.

  12. Anonymous3:44 AM

    One of the advantages of Mr. Diamandis' prize approach: it pushed a great leap in progress that wouldn't be there otherwise.

  13. Sonny Corleone (RIP)6:04 AM

    The guy that VG sent to NewSpace 2010 spent about 10 seconds on the engine. There was a slide with a picture from April 2009. He took no questions - an oddity for a conference of the geeks, by the geeks, and for the geeks.

  14. Sonny, yes, I saw that and cringed. "Oh yes, and we have a rocket engine in development".. it's just such a strange way to go about presenting a space shot.

  15. Anonymous8:21 AM

    If you look at the you tube video of the SS2 test, the motor is mounted VERY low over a field of dirt. When its firing you can not see the quality of the plume as dust obscures it. One of the hard things to get right in a hybrid is the fuel regression rate they tend to go off the nominal mixture rate during the burn. If one had a clean shot of the plume one could tell if this was happening by the atmospheric after burning or the black soot in a severe case. This dust cloud would conveniently hide that.

  16. Oh, btw, regarding that composite oxidizer tank problem that Scaled has, XCOR happens to make a composite LOX tank, one of their other great innovations. Sounds to me like Rutan really really should be talking to Jeff more...

  17. Peter Lykke8:24 PM

    One question pops up from time to time: Did Scaled ever find a root cause for the accident in 2007?
    I have googled and googled, but haven't found an explanation.
    Does any of you know? Or is it just "something about N2O"?

  18. The root cause was complacency with cold N2O. There's an internal report, but it hasn't been released. And if you're the same guy who asked me this on irc, sorry for repeating myself.

    Complacency with N2O cold flow is still pretty common today.. so that report not being released after 3 years is considered by many to be kinda irresponsible. They were fined $25k for violating workplace safety and failing to train workers to handle hazardous materials (so says the LA Times), and there was a report someone obtained from that investigation.

    June last year they produced this document:

    Considering that people continue to seem to have no idea what would be appropriate training for handling N2O, beyond what they teach dental assistants I guess, this is a very good thing.

  19. Peter Lykke9:16 PM

    No, I'm not that guy, and thanks for the answer. So it self- ignites?
    I didn't know that it could.

  20. It's a monoprop and thermal decomposition can cause it to ignite, yes.

  21. Anonymous10:30 AM

    "dirty little secrete": much ado about nada and a feeble excuse for putting iut a blog

  22. First, I mean no offense to the original poster. Everyone is entitled to their opinion. Here is mine:

    I think it's worth noting that, of all the companies claiming to be on the path to commercial spaceflights, Scaled is the only one whose vehicle is not only completely built, but also has flight time. They're also the only private corporation to ever put a man in space. And, they are also one of the only companies working on an actual spaceplane rather than simply improving upon the Apollo design. I think they deserve a lot of respect (they most certainly have mine).

    Furthermore, Scaled has always been very terse when it comes to progress reporting. They've never announced a schedule or given a timeline. I mean, the FAA didn't know SpaceShipOne even existed until they rolled it out of the hanger for flights (which is probably good given all of the back-room sabotaging of private ventures that NASA has done over the years). So the lack of a verbose description of their under-development propulsion system is not unusual. No one at Scaled wants to lose their job over a media quote.

    Finally, if you check all of the comments from experts and the blogosphere prior to SS1's prize winning flights, there were a lot of skeptics jeering at anyone who would dare to claim to actually have a system that could compete with a goverment-run agency, but they all shut up when Brian Binnie stepped out of that cockpit to accept his astronaut wings.

    Now, SS2 is on the brink of powered flights, and the naysayers have returned. Yes, the design is made by a human so it's not perfect. Heck, you can second guess it all you want, but it's pointless to do so. Given how long it's taken for someone to actually provide commercial space access, I'm willing to give Rutan the benefit of the doubt; and what's more, I'm rooting for him the whole way. It's about time, to be honest.

  23. Anonymous9:41 PM

    Some time ago, I wrote to Rutan pointing out that Thiokol, Huntsville (I believe test fired hybrids using a grain of HTPB mixed with 25% solid oxidizer (ammonium perchlorate) to get appreciably higher thrust. Thrust could still be terminated by stopping fluid oxidizer flow. Have not heard a word about this from Rutan or anyone else. Any news/thoughts out there?

  24. Michael1:36 AM

    I couldn't agree more with JD's comment above. Scaled/Rutan is an innovator. Looking back at Rutan's impressive record, nothing he did was really COTS-like. Sure, it may not make sense to use a hybrid. We all have hindsight superpowers. However, he succeeded and he's making it [commercial space tourism] happen. In the process, he'll likely improve upon existing hybrid rocket motor technology, or at least bolster it.

  25. Anonymous5:13 AM

    So, where are tourist's flights? why this shit in pause from 2004? where are god damn flights?

  26. "In a sane world, a company that has a vehicle without an engine would purchase one from a company like XCOR, which specializes in engines."

    I guess it could be equally said that in a sane world, a company that has a engine without a vehicle would purchase one from a company which specializes in airframes (especially a supersonic one). Obviously Jeff hasn't decided to do this with his lynx. This is because for this commercial space industry to succeed we are going to need to learn to do more with less. All those involved are going to have to get comfortable stepping outside of their disciplines for any of this to work. Xcor found people to hire that specialized in airframe design and manufacture just as scaled found people that knew rocket engines. I don't see why any of that is interesting, it seems like a pretty logical and conservative chain of events to me (hardly a dirty secret).

    Xcor talks a pretty big game shooting holes in others' designs when as far as public information goes they have a cockpit mockup and a wind tunnel model. The world waits for Xcor to roll an internally designed built and flown airframe through there doors. Meanwhile Scaled has run rocket engines that they designed and built. As they used to say in little league...check the scoreboard.

  27. Thank you for this blog, QG. I thought it was wonderfully informative, as were the comments. I'd love to be privy to the inner machinations of SC. Lacking that, the scuttlebutt presented here isn't bad.

    It goes without saying that I'm pulling for Rutan, his employees, and his contractors 100%. I wish I could help. Alas, the Feds and their damn rules closed me down years ago.

  28. Your article here is full of inaccuracies and misinformation. Among other items, the 2 most glaring items if misinformation are as follows:

    1. "...just like the Space Shuttle the wings are "only" used on the way down. There are wings used on the way up, of course, they are on WhiteKnightOne/Two, but once separated from the carrier aircraft the only lift is generated from the rocket." - INCORRECT.

    All aerodynamic surfaces of SpaceShips 1 & 2 ARE utilized and are absolutely essential for the flight up prior to reaching the near vacuum at and approaching it's apogee.

    2. "Although the investigation cleared the company of any wrong doing..." - INCORRECT.

    Scaled Composites was, in fact, found to be at fault. The following is but 1 of many sources reporting this:

    Highly recommended that you choose a hobby other than journalism. Writing like this doesn't even meet the standards necessary for publication in a middle school newspaper.

    1. Thanks for the constructive criticism.. in future, can you try not being an asshole about it?

    2. This comment has been removed by the author.

    3. Oh, get over yourself.


Post a Comment

Popular posts from this blog

Disabling OS-X Device Removal Warnings In Yosemite

The Three Amigos

Living Inside An Asteroid