Boeing’s space plane returns after nearly a year

This isn’t our usual gig, but we’re fascinated by this story. Kudos to Boeing on this one.

17 Comments on “Boeing’s space plane returns after nearly a year

  1. Then there is the dream chaser from Sierra Nevada. Not like the shuttle but still wings and runway landings. SpaceX seems to lead the competition by far now. The private space race is exciting IMO. Hope for even more compatition form Europe and Russia.

    What is your favorite of the current 4? I like SpaceX, impressive work!

    • If SpaceX can make the Dragon’s integrated Launch Abort System (LAS) work properly with the further potential to use the LAS engines to land Dragon back on land propulsively, then SpaceX is IMO the best system in the near term.

      However, If Reaction Engines can make their novel precooler and heat exchangers work efficiently and reliably, then the likelihood of Skylon becoming a disruptive technology increases considerably. This would mean, of course, the end of all current launch vehicle providers, unless these companies would make the transition from actual launch vehicle providers to Skylon launch vehicle operators; or Skylon spaceliner operators. 😉

      • Ahh that Skylon would be a very cool vehicle, the first single stage to orbit? All previous atempts have failed sadly, X33 being the last attempt. I remeber there was a lot of buzz in the 1990´s about SSTO.

      • Due to the high propellant consumption of rocket engines, a high fraction of the launch mass of single stage to orbit launch vehicles, has to be propellant in order to achieve orbital velocity. If you use kerosene fuel (and liquid oxygen) as propellants in a SSTO vehicle, about 94 percent of the mass at lift-off would have to be propellant. This means that only 6 percent is allowed for everything else (i.e. inert mass and payload where the former includes tanks structure engines and equipment). Switching from kerosene to liquid hydrogen on a SSTO vehicle, and using very high performance staged-combustion cycle, high chamber pressure rocket engines, usually having labour-intensive processing requirements (i.e like the SSMEs on the Space Shuttle), would lead to a propellant mass fraction of about 88 percent; slightly better than using kerosene, but still not good enough for “everything else”.

        Hence, if you want to reduce the propellant fraction in SSTO vehicles, you’ve got to use some sort of air-breathing engines part of the way up into orbit. Skylon will require an overall propellant mass fraction of about 79 percent, which of course is high compared to current long ranges airliners, but by noe means unfeasible.

  2. Most rocket designers would kill for a mass fraction of 6%. In reality the mass fraction achieved is about half of that. The Falcon 9 1.0 (the version that just launched the dragon) weighs about 333,000 kg at launch and delivers about 10,450 kg to low earth orbit for a mass fraction of about 3%. One thing though about Skylon, the real killer app is re-usability and not Skylon’s ability to obtain most of it’s oxidizer from the atmosphere. Skylon’s approach is just one of several approaches though.

    In terms of Fuel cost it’s cheap and Musk says Falcon 9 burns around $200K in fuel per mission so most of the $55 million price tag for a basic Falcon 9 mission is in throwing away the hardware every trip and in the manpower and effort to launch assemble and launch the Falcon. Elon estimates that a fully reusable Falcon 9 would save about 50% in overall costs if flown infrequently (due to the fixed labor and site costs) and reduce costs a 100 fold if he could reduce the turnaround to less than a day for the first stage and several days for a reusable upper stage. Musk believes the Falcon 9 first stage can be made reusable for 40% hit in payload on the first stage and an unspecified additional amount for the upper stage. So SpaceX is taking the opposite approach from Skylon, sacrifice more payload and probably getting no more than 1.5% for re-usability. But hey the space business is the opposite from the Airline business in that Fuel is about the lowest cost item.

    One more thing Skylon better hurry up if they want to compete. SpaceX is already working a prototype re-usable first stage called Grasshopper and could be flying it as early as this year, although 2013 is certainly more likely. A fully re-usable Falcon 9 would be every bit of the killer app in space that a fully re-usable SSTO would be.

    • John, I was talking about total propellant mass as a fraction of Gross Lift-Off Mass (GLOM), not payload mass as a fraction of GLOM!

      The propellant mass fraction of, for example, the Space Shuttle and the three stage Soviet/Russian Proton-M launch vehicle (not including the propellant mass of the Briz-M upper stage), with GLOMs of 2,040,000 kg and 690000 kg respectively, is (/was) a little more than 85 percent for the Shuttle and about 90 percent for the Proton.

      As for SpaceX, Elon Musk gets my kudos for having tenacity and huevos, as well as ingenuity. However, without having gotten NASA as a prime customer in 2006, SpaceX would in all likelhood have ceased to exist by now, and without the loss of the Shuttle Orbiter Columbia in 2003 and the subsequent decision to end the Shuttle program by 2010, the Commercial Orbital Transportation Services (COTS) wouldn’t have been started in the first place. Also, the COTS Demo flight 2 was only the third overall Falcon 9 launch (in two years). Even if the Falcon 9 should better the 95-97 percent average current success rate of expendable launch vehicles, there’s still a long way to go reliabilty wise. Furthermore, reusability of the Falcon-9 is IMO in an “unreal” exploratory CAD engineering phase. If you look at the second stage, it’s conceptually shown re-entering ballistically from orbital velocities, without even a ballute, for thereafter to land propulsively near the launch site. Now, look at how the Dragon capsule was scorched on the windward side during reentry. Then imagine how the sides of the 2nd stage and the Merlin rocket engine would fare. And then you’re supposed to re-use this thing within hours. Yeah, right!

      The Space shuttle was declared “operational” after 4 missions, and the Falcon 9 is now apparently “operational” as well. Compare that to the thousands of hours of testing that is required to certify a new airliner.

      It looks as if Reaction Engines is planning to undertake a more airliner-type testing regime for the Skylon. The planned test flights are alone going to be more than twice the number of flown Space Shuttle flights (over 30 years).

      Quote from: baldusi on 05/03/2012 04:35 PMQuote from: Hempsell on 05/03/2012 12:22 PM
      How many prototype crafts are you expecting to use? What’s the expected life of the airframe and engines? And the time between rebuilds of the engine? I’m trying to grasp the level of reusability

      The qualification flight test programme has two production prototypes (there are also two earlier full scale development vehicles which are probably not orbital). One is a pathfinder that undertakes the scoping test flights the other is a workhorse and it puts in a solid 204 flights to prove the airfame specified life.

      Around 30 of the pathfinder flights are abort tests and do not reach orbit and so are not counted in the flight statistics. So we have a total of around 380 orbital flights available from the two airframes.

      Once the workhorse has done 204 flights and the overall programme has around 300 orbital flights we will have proven the specified mission success reliability (99/%) to a better than 80% confidence level, assuming a perfect flight record, and at that point SKYLON can be made operational. Although we have assumed we would still do the remaining 80 or so test flights.

      In the more likely event of some test flight aborts we have the additional orbital flights (up to the total of around 380) to prove the reliability to the required confidence level.

      If an operator wants to start operations without the full proof of the airframe life or with a low proven mission success rate they can pick an earlier point in the test flight programme to begin their operations.

      The 1% mission failure rate is a failure to deliver the payload and not a vehicle loss statistic. Unlike expendable vehicles this is not the same because as you point out we have a full abort capability from every point in the mission.

      The specified vehicle loss rate is better than 1/10,000 flights but it is not possible to directly prove than with any practical test flight programme, so it is inferred from the test flight anomaly record.

      Both these are specified values for entry into service proven by the test flight programme. It is expected that the real inherent reliability will be much better than that and that, as the operational record is generated, the Weibull function will improve it further in mature operation.

      I cannot comment on details of the Sabre 4 or its relation with other pre–cooled engines – sorry.

      The heat exchangers are eyewateringly expensive so we have not yet identified any non-space applications were it makes economic sense – not even jet engines.

      • OK, got on propellant mass fraction, normally you see payload mass fraction as the standard reference so I was a little confused about the 6% mass fraction. In terms of the re usability issue, though, its difficult to understand why you would pick out an underfunded company like Reaction Engines as more likely to pull of a game changer than SpaceX.

        The simple fact is Reaction Engines is in the engineering study phase and even using their baseline 2004 engineering studies they don’t do much better than SpaceX expendables on the cost issue. In 2004 Reaction Engines estimated that Skylon would cost $12 billion to develop (an extremely low estimate for such an advanced launch vehicle design). Even assuming this number and a usage rate of 200 aircraft and no more than 190 million pounds per aircraft, again unlikely given the billion dollar plus cost for something much simpler like a shuttle orbiter the best reaction engines comes up with is about $400 per kg to orbit. Well when Musk is promising prices of $1,000 per kg to orbit on his website using a Falcon 9 Heavy why would anyone put up the tens of billions to develop the Skylon? And the $400 per Kg number is 10 years old at that, Musk’s prices are what he is advertising now.

        I would also certainly agree with you on the difficulty of re-usability and as far as the Falcon 9 second stage goes, Musk did say that SpaceX was hiding the details on it’s design on purpose (he’s a little paranoid about the Chinese). So who can say if their designs are valid or not as they have admitted that the video does not accurately depict the details of the design.

        It just comes down this, SpaceX has real hardware and real re-usability designs and a real funding stream in terms of billions of dollars in existing launch contracts and hundreds of millions in NASA development dollars, with billions more potentially to come from NASA and the Pentagon. Reaction Engines has a contract worth a few million Euros to continue research on the SABRE engine and nothing more. My money is on SpaceX in this case. The fact is that despite Alan Bond’s impressive credentials and impressive work on project Daedalus Reaction Engines hasn’t done anything besides some proof of concept engineering work on the SABRE and some CAD drawings of the proposed and as of yet completely unfunded Skylon.

      • Well, SpaceX maybe a “game changer” for the American aerospace industry where the US government’s traditional “cost-plus” contracting system ensures that manufacturers make a profit even if they exceed their advertised prices. Hence, the contractors incentive is to maximize the cost of a vehicle in order to maximise shareholder value. Furthermore, the typical American aerospace engineering philosophy is to keep tweaking a design trying to squeeze out as much performance as possible. That’s understandable from an engineering standpoint, bit it drives up costs. Higher costs typically leads to a lower launch rate, which of course, is further increasing the cost per launch. In the Soviet Union the production of Soyuz launch vehicles reached a peak of 60 per year in the early 1980s. It has flown more than 1700 times, which is far more than any other launch vehicle. The Soviets/Russians only upgraded the Soyuz LV when absolutely necessary. Due to the lower R&D amortization costs and the high flight rate, the actual cost per launch, in Soviet times, for a Soyuz launch vehicle (excluding payload), was probably under $1000 per kg to LEO. After 1992, launch rates dropped significantly, and even though the Soyuz LV has been upgraded in recent years, the cost per kg is still well under competing American designs. Interestingly enough, SpaceX is quite influenced by the Soviet/Russian launch vehicle design philosophy.

        Currently the Russians charge NASA at an inflated $60 million “market price” per Soyuz seat. In the future, don’t be surprised if market savy Russians will be trying to undercut US cut-price “commerical companies” (i.e. companies wholly reliable on the US government).

        So, I have no doubt that SpaceX can easily beat the LMs and the Boeings of the world on launch vehicle pricing. It remains to be seen how low they can go when, or if, they’re no longer propped up by the US government. Also, previous studies have shown that reusability only makes sense if you have a higher flight rate than some 50 launches per year, which means that unless you create new markets such as space tourism, there is just not going to be the necessary demand to justify reusability. The problem with Dragon vs. Skylon space-tourism-wise, is that the former needs to demonstrate a reliability way beyond current capsule technology, and rocket engine reliability, in order for it not only to be certified by aviation regulatory authorities, but to avoid massive lawsuits after the first loss of vehicle event. However, as already mentioned, Skylon is to have a full abort capability from every point in the flight with a loss rate better than 1/10,000.

        “Even assuming this number and a usage rate of 200 aircraft and no more than 190 million pounds per aircraft, again unlikely given the billion dollar plus cost for something much simpler like a shuttle orbiter the best reaction engines comes up with is about $400 per kg to orbit”.

        First, Skylon is, in fact, much “simpler” than the compromised shuttle orbiter design and the enormous cost per orbiter was partly due to the low production volume (5 space-worthy orbiters built). Skylon will be operated autonomously; no expensive life support system is required unless humans are aboard; the aeroshell is several hundred degrees cooler than the Shuttle during re-entry due to Skylon’s lower ballistic coefficient; the aeroshell is a reinforced glass ceramic composite manufactured in thin sheets and corrugated for stiffness much much more robust than the delicate ceramic tiles on the shuttle orbiters; the pre-burners on the Synergetic Air Breathing Rocket Engines (SABRE) should be much simpler than its SSME equivalent. There are no turbines in the exit flow as the Helium acts as the energy transfer medium for the pump drive. An objective of the SABRE design is to eliminat* any case where you have hot combustion products on one side of a seal with excess O2 or H2 on the other. The SSME has these features and it requires complex seals, purging and monitoring because a failure would be catastrophic.

        As for Reaction Engines and “advertised” cost per kg to LEO, it all depends on how the market would evolve with the availability of Skylon-type vehicles.

        Even if the customer paid all of this, it would still represent a large reduction on current costs and would be a true transport operation. This however is a very naive and pessimistic assumption. The real market would involve benign and aggressive operators with differing flight rates and nationally biased traffic. The total traffic would affect service and facility costs whilst profits and loan repayments would affect operators’ cost. Pricing strategy would create different rates for cargo categories and human transport. We expect mission costs to fall to about $10 million per launch for high product value cargo (e.g. commmunications satellites) $2-5 million for low product value cargo (e.g. science satellites) and for costs per passenger to fall below $100k, for tourists when orbital facilities exist to accommodate them.

      • ” SpaceX is quite influenced by the Soviet/Russian launch vehicle design philosophy.”

        SpaceX actually started with several visits Elon Musk made to Russia in the early 2000s he wanted to buy a Russian Rocket to send a private probe to Mars. So yes SpaceX was heavily influenced by Russian design philosophy and rocket technology. Musk found on his trips, however that the Russians wanted more money than he was willing to pay, and the rest of the story is that he founded his own Rocket company because he couldn’t hire anyone even the Russians to fly a payload at what he considered to be a reasonable price. As you mentioned the key to getting launch costs down in the short term is flight rate. To this end SpaceX’s immediate to goal is to produce 400 Merlin engines a year in 2013 or 14 (more than any other country on earth) and about 40 Falcon 9 cores per year. Once you hit the 40 to 60 core range though you start level off in terms of benefits of producing more cores and you need re-usability to bring costs down.

        I’ll be honest about Reaction Engines though, while I agree they have an interesting approach and it looks like workable design, and I think Alan Bond is one of the great aerospace visionaries of the last 30 years I don’t really think they have a prayer of obtaining the funding necessary to build the Skylon. Skylon may be advertised as simpler than the Shuttle, and in many ways it is, the fact is though that space planes be they the Shuttle or even the X-37 have never been very cheap and have always wound up costing much more than planned. I simply don’t see development ever happening for less than at least twice what Bond estimated 10 years ago, and I don’t see a very large number of vehicles being built to drive vehicle costs down. Especially given that the only single customer in the world large enough and rich enough to fund a project like Skylon, namely the Pentagon would never touch it because of its British heritage and location. So on this who would ever provide the $20 billion plus development price (Bond’s $12 billion is way to low). Along with the money to build the Skylon and operate it at a loss until a market developed to take advantage of its potential for a much higher flight rate and re-usability? Certainly not the ESA who has 1/3 of NASA’s meager budget and is struggling to find a way to fund the essential and must do Ariane 6 project.

        So in the end I’m still betting on Musk’s brute force approach to re-usability vs. the more elegant but more technically challenging efforts to build a reusable winged SSTO vehicle.

      • “Especially given that the only single customer in the world large enough and rich enough to fund a project like Skylon, namely the Pentagon would never touch it because of its British heritage and location”.

        Reaction Engines has made it clear that ITAR restrictions prevent any US involvement in Skylon; except being a customer of course. 😉

        ” Once you hit the 40 to 60 core range though you start level off in terms of benefits of producing more cores and you need re-usability to bring costs down.

        The studies that indicated that reusability only makes sense if the flight rate is higher than 50 flights per year was made for the trade-off between current expendables and a small fleet of Reusable Launch Vehicles (RLVs); A.K.A. Shuttle-2 etc.

        Re-using Falcon-9 first stages –assuming that re-usability of the 2nd stages is a pipe dream for the foreseeable future — would certainly help to bring costs down, but by an order of magnitude? Not very likely IMO.

        There are multiple factors influencing how low SpaceX can realistically go on pricing.

        a) How many times can SpaceX reuse and launch the first stage before there’s a failure to recover the stage? NB: New hardware is required if recovery is unsuccessful.

        b) How many times can SpaceX launch a “used” rocket stage without having a failure? NB: In case of the launch being unsuccessful, new hardware is required.

        c) What would the refurbishment costs be for a flown Falcon 9 first stage?

        d) What is the ratio of manufacturing (of new hardware) to infrastructure spending? NB: Infrastructure costs (launch/recovery) don’t go down with reuse, only the manufacturing costs of new hardware will go down.

        e) How is the insurance cost curve going to look like after x number of Falcon 9 first stage flight cycles? Will the insurance be cost-prohibitive after just a couple of re-uses?

        d) Will the satellite telecommunications Industry demand an “all new rocket” for every launch of high value satellites?

        IMO, reducing costs by 50 percent would be an achievement, evenconsidering the continued subsidisation of SpaceX by the US government.

        NB: If fuel depots in LEO would be developed in the mid term, there will be a market demand for launching that “low value” fuel on launch vehicles using recycled stages; which would mean, of course, that launch costs would likely go go down further.

        “I don’t really think they have a prayer of obtaining the funding necessary to build the Skylon. Skylon may be advertised as simpler than the Shuttle, and in many ways it is, the fact is though that space planes be they the Shuttle or even the X-37 have never been very cheap and have always wound up costing much more than planned. I simply don’t see development ever happening for less than at least twice what Bond estimated 10 years ago”,

        Actually, the plan is to to develop Skylon an “commerical terms” with little, or no funding from any governments. That doesn’t preclude an Airbus-type Reimbursable Launch Investment (RLI) financing scheme for say half the amount required. 😉
        In that sense, governments in Europe and elsewhere would help to de-risk the programme and thereby help with getting a multitude of both risk-seeking and risk-neutral investors aboard.

        As for SpaceX presenting realistic and belieavable cost assumptions, apart from the enormous amount of hype; at least Reaction Engines talks about what kind of markets that are required in order to lower costs by up to two orders of magnitude:

        Figure 11 and Figure 12:

  3. Well I think that this antiquated use of Lox/Hydrogen engines to lift off a pad is a poor choice. A space plane off a runway or even better a boosted launch ramp with the space-plane fueled by NERVA type Lox/Hydrogen engines would have a much higher payload and reusability. The NERVA engine has a 10+ impulse to the standard Lox/Hydrogen engine. This could also be supplemented by solid boosters if needed. Dead weight off the pad takes a huge amount of fuel and as a pilot I prefer something with wings. The arena of materials and electronics available today should make such a craft very possible.

    • An aerospaceplane using a Nuclear Engine for Rocket Vehicle Applications (NERVA); or a Nuclear Thermal Rocket engine (NTR), in the Earth’s atmosphere, is a non-starter, not only politically speaking due to the massive environmental implications, but also due to the excruciatingly poor thrust to weight ratio of NTR-type engines. If a NTR engine was to be used in combination with an “afterburning” LOX-Augmented Nuclear Thermal Rocket engine (LANTR), the specific impulse would be only slightly better than state-of-the-art LH2/LOX chemical rocket engines.

    • NERVA could never achieve 10+ the specific impulse of a chemical Lox/Hydrogen engine. Due to material limitations a NERVA design was limited to 850s isp vs 470s isp for the best chemical upper stage rockets (e.g. the latest RL-10 versions and only the extended nozzle version on the Delta IV at that). Anything more than 850s and would melt the 1960s materials used to construct the power plant. Modern materials could get you a little more but not much more. However, outside of the atmosphere this savings was quite significant and NERVA would make a good second stage and it was proposed for that application.

  4. Have you seen the nuclear aircraft reactor project? That would have been cool, a heat engine with gasturbine. The radiation shield was too heavy though 🙂

Leave a Reply

Your email address will not be published. Required fields are marked *