Every time there’s a test of a scramjet, there’s associated overhype about how great it will be for space access. The upcoming X-51 flight is no exception:
Ms. Waldman said in her report that as scramjet technology is developed testers believe that in the near future it could be used to aid warfighters as a weapons delivery system. She said officials believe that in the future the scramjet technology will make space access easier.
“The application really is all about space lift,” Mr. Brink agreed, and said, “This is the one, I think, in the Air Force Research Lab we’re most excited about.”
Mr. Brink pointed out that they currently transport payload into space with the shuttle, which has to carry all of its oxidizers for the propulsion concept. He said the shuttle is a pure rocket system and said if they can incorporate scramjet engine technology into the space lift systems, they wouldn’t have to carry the oxidizers and could carry more payload instead.
Yeah, if there’s any payload left after you count the weight of the engines, which have terrible T/W compared to rockets, and all of the extra drag you incur staying in the atmosphere to collect the oxygen. I’ve discussed this more than once in the past. I’ve never seen a hypersonic airbreathing conceptual vehicle design that was an improvement in performance over a rocket for a space transport, at least if there was any analysis more serious than the above performed on it. Scramjets have plenty of utility for military applications. I wish that people selling the program didn’t always feel the need to oversell it. And if we had a smarter space media, they’d get called on it.
I have this vague memory of an article by an apparently sane individual that claimed that *if* you were trying to match orbits with a specific object already on orbit, and wanted to avoid the ‘one launch window per day’ problem, you could use an air-breathing, cruise mode ‘first stage’ to get into position, then launch your rocket.
The booster in the text was strictly of a “White Knight” style subsonic airframe, not a scramjet.
Is there any civilian usage for such a beast? Would the old “Orient Express” space planes have benefited from such an engine?
We could just try to bring back the SLAM (pluto) project
It seems to me that if the first few seconds of thrust came from a ground launch system of some type it would significantly add to the payload of any rocket.
Have you actually done any math to support your supposition? Taking into account all of the costs, and the value of “significant”?
Rand, I agree with you but not completely.
I’ve long been of the opinion that getting out of the atmosphere as soon as possible is the first order of a launch vehicle.. so much so that I believe people waste way too much time on supersonic aeroshells when they could just suffer the gravity penalty and build a bigger rocket.
That said, KISS means you never get much rocket development being done. If it works, it works and there’s easier ways to reduce costs than screwing around with something that aint broke.
Scramjets are one of the few places where rocket development is still going on… my government, Australia, has no launch capability but dumps funding into scramjet research cause its still considered cutting edge and makes them look good.
Maybe that kind of R&D will continue until scramjets are actually practical for space launch.. maybe some of those R&D dollars will go into improving t/w.
For the moment, scramjet research serves the same purpose that traditional rocket research served in the 50s and 60s, it’s a hook you can hang your we’re-a-smart-nation hat on.
Scramjets are the technology of the future…and always will be.
Scamjets… Here is a new tech air launch option that does kinda make sense:
http://vimeo.com/6194911
It is a fast, cheap and highly controllable six propeller model electric helicopter (only need watch the first 45 seconds). By my figuring one can get a substantial rocket (it is sufficiently scalable) to 10km plus altitude for very little in the way of development or operating costs. Cost is currently about $100/kg of rocket vehicle lifted – $1m lifts a ten ton rocket (one person) at ~$200 to charge between flights. And the batteries are getting better, 20-30km will likely be possible in the next 5-10 years and lithium batteries with ~10,000 cycle life are already available.
With such a launch assist, I suspect a SSTO delta clipper type vehicle becomes practical. And you could develop it at small scale due to much lower aero losses on the rocket vehicle. No altitude compensating engines, likely much lower range costs, better aborts, etc.
I would also likely favor such a system for reentry vehicle landing. Perhaps slightly heavier but much safer and more capable.
Pete –
The guy has 4 propeller and 8 propeller versions as well.
Pretty cool.
Getting any helicopter to 30,000 feet plus is very difficult – the only way to keep the rotor tip Mach numbers low enough while still generating enough lift is to drastically reduce the disk loading, and if you’re doing that by adding rotors, the parasitic weight of all of the extra drivetrain kills you. Looks like a non-starter to me.
And I agree with Rand – scramjets get you nothing for access to space; their only use is for long-range hypersonic cruise at high stratospheric altitudes. And I question their military usefulness even in that role – you can’t make them stealthy (can you imagine the IR signature? and there aren’t any radar absorbing materials that can take the heat), and the big military need is for persistent stare, not super-quick-reaction standoff strike (that’s what those things called ICBMs are good for anyway, and there are studies at putting conventional munitions on them).
Have you actually done any math to support your supposition? Taking into account all of the costs, and the value of “significant”?
I haven’t done any of this type of math since college over thirty years ago, but you certainly have a point. I don’t know what percent increase of payload to LEO might be considered significant; perhaps anything over ten to twenty percent?
I checked the numbers on the F1e (round figures):
500 kN of thrust; 3500 kG of liftoff mass; 1000 kG to LEO.
I really have no way of determining what kind of launch assist platform or what it would cost and agree that’s some pretty important info. Let’s call it a big spring that provides 500kN. So how many seconds would it have to work to give us 1100kG to LEO? All I remember is F=MA and Vf=V0+1/2At^2.
Any rocket scientist out there that can solve for how many seconds?
There are physical limitations to rockets that we’ve basically already reached. Scramjets don’t have these same limitations. The limitations they do have are not well characterized. Breakthroughs for example that greatly reduce the dry weight of a scramjet would not, AFAIK, violate any law of physics, whereas physics says we can’t greatly reduce the propellant per payload from today’s chemical rockets. Thus, it is well worth studying scramjets based on potential future benefits they might have for getting stuff and people off this planet as well as for the nearer-term benefits you cite.
There are physical limitations to rockets that we’ve basically already reached. Scramjets don’t have these same limitations. The limitations they do have are not well characterized. Breakthroughs for example that greatly reduce the dry weight of a scramjet would not, AFAIK, violate any law of physics, whereas physics says we can’t greatly reduce the propellant per payload from today’s chemical rockets.
In other words, you are completely unfamiliar with the literature in this regard, and don’t understand anything about the basis of current launch costs…
haha, googaw, why don’t you just call it a week?
Getting any helicopter to 30,000 feet plus is very difficult – the only way to keep the rotor tip Mach numbers low enough while still generating enough lift is to drastically reduce the disk loading, and if you’re doing that by adding rotors, the parasitic weight of all of the extra drivetrain kills you. Looks like a non-starter to me.
Designing this type of electric helicopter for 30,000 feet is actually relatively straight forward. Air density at that height is roughly a third that at sea level – tripling the total disc area is one possibility (this costs little, there is no extra drive train), though one would probably instead use more power and maybe top out with a slightly horizontal trajectory (want a fast vertical climb anyway). Note that propeller thrust is in this case somewhat proportional to vertical velocity (F=mdot.dv).
Helicopter gas turbine power drops off roughly in proportion to air density and this is perhaps the main reason why helicopters can not be easily designed to fly high – battery power does not similarly decrease with altitude.
The point of an air launch is to start high off the ground where the air is thin. You’re doing what the first stage does, but with a reusable mothership instead of an expensive booster.
This is where I think scramjets might come in, eventually. Not for the payload, but to squeak a little more altitude and speed from your carrier aircraft.
I think that there is something of a scramjet mafia/cult in the aerospace community, much as the vaunted fighter mafia seems to run the USAF.
A co-worker some years back compared scramjets to Jason Voorhees of the Friday the 13th movies. No matter how many times one kills him, he always comes back.
Of course, I’m still a fan of the original Orion Project, but I’m a hopeless romantic…
‘ “…the vehicle will be accelerated by a solid rocket booster up to about Mach 4.5. Once it reaches that speed the booster will drop away and the vehicle and the engine will ignite and accelerate the vehicle up to Mach 6,…”
…
that conventional turbine engines are physically limited to about 2.5 Mach or 2 ½ times the speed of sound.
“The scramjet in the X-51 will be able to take in the air flight speeds over Mach 4 and up to Mach 6,” ‘
With a conventional rocket how much fuel is used taking the rocket from Mach 4 to Mach 6?
I would guess that the rocket fuel it takes to get a rocket from Zero to 500 mph is more than is used to get a rocket from Mach 4 to Mach 6.
Before I call it a week, somebody has to actually point to something I said that is wrong. And yes, I do understand the literature and the bases for launch costs (note the plural) much better than Trent or Rand.
There are physical limitations to rockets that we’ve basically already reached.
Not completely true. There are lots of things that could be done with rockets to improve performance. Staged combustion improves performance, yet a lot of engines do not use it. Full flow staged combustion engines promise even more reusability and high performance.
Altitude compensating nozzles are another thing that could be further researched.
Thrust augmentation nozzles (TAN) and LACEs to scoop oxygen from the atmosphere are another possibility if you want to do in atmosphere travelling. Reaction Engines is one company working on LACE like technology. Unfortunately development has usually been as painful (or more) than that of scramjets.
There is a lot of secondary evidence that it is possible to build an SSTO given modern construction techniques and materials, plus high performance engines. Doing it TSTO minimizes risk at increased cost.
You would be better off developing reusable, low maintenance, low weight, thermal shielding however. That is one of the major maintenance costs of the Shuttle orbiter. That and the hypergolic RCS/OMS. SSME costs are pretty low, especially after the upgrades. Liquid fuel is cheap. The drop tank and the solids are expensive as well.
IMO one of the main issues with doing Shuttle is that it was perceived as a development program for a useable system. STS is more of a prototype than a useable system. They should have made it a spiral program where the system was progressively optimized until an optimum was reached. However they made it too big. Too expensive for an R&D vehicle.
When folks look at the cost of going into space, they get all hypothetical about exotic ways to reduce fuel costs for getting there, forgetting that for most launch vehicles that the price of the fuel itself is usually one of the least important factors for getting something into orbit… at least from a cost perspective. Heck, for most launch vehicles, the cost of the catering services to feed the ground launch crew usually runs a larger price tag.
From this viewpoint, a big dumb booster that is cheap to make and simple in its design is, at least for now, the better route for going into space. On this issue, I’d have to completely agree with Rand here that the scramjet or any other fancy way of getting into space is inherently better. The White Knight air-launch works in the sense that it avoids having to deal with the bottom layers of the troposphere (aka “weather”) and it is possible to “launch” in less than ideal conditions and in more places than one fixed spaceport. That is about all it really does in terms of providing better launch conditions, as even for sub-orbital flight it doesn’t really add all that much more thrust compared to launching something from the ground. Except for some very small mircosats, I don’t see the White Knight being used for orbital spaceflight.
Launching from a lighter-than-air vehicle like the JP Aerospace guys are doing can scale to the sizes necessary for a heavy launch vehicle for manned spaceflight. Even then, they are simply getting out of the atmosphere and providing a high-altitude launch platform like the White Knight. If there is going to be a significant improvement in launch costs from something other than a missile launch-type system, I would expect it to come from something like that rather than something similar to a scramjet.
BTW, here is the mathematical proof of why scramjets are not useful for getting to orbit – though someone already pointed it out non-numerically in noting that acceleration is provided only between mach 4 and 6.
All jet/rocket engines create thrust by throwing stuff out the back faster than they are moving. For a non-rocket to beat a rocket, mass*velocity of the jet exhaust has to match mass*velocity of the rocket exhaust at the same propellant feed rates, even assuming a magical jet with the thrust to weight ratio of a rocket.
At low speeds, there is no contest. The intake air can be considered essentially at rest, so the jet thrust is:
(mass of fuel/s+intake air mass/s)*(exhaust velocity)
while a rocket’s thrust is:
(mass of propellant/s)*(exhaust velocity)
in a perfect world, the exhaust velocity is related to the square root of the energy in the fuel in both cases. So the equations simply to:
jet: (mass of fuel/s+intake air mass/s)*(K * sqrt[mass of fuel] )
rocket: (mass of propellant/s)*(K * sqrt[mass of fuel] )
It is readily apparent that not having to carry the air with you is a huge boost in this case – there is really no downside. This leads to jet engines that use only a tiny amount of fuel to accelerate a large mass of air.
The problems start at high speed – when you can no longer neglect the relative velocity of the air you are trying to use. Now the equation for the jet shifts a bit:
jet: (mass of fuel/s+intake air mass/s)*(K * sqrt[mass of fuel] ) – (intake air mass/s * intake velocity)
rocket: (mass of propellant/s)*(K * sqrt[mass of fuel] )
This is essentially saying that the jet only gets the benefit of the additional velocity added to the air, while the rocket still receives the full benefit. The math clearly shows that once (intake velocity)>(K * sqrt[mass of fuel] ), the jet has lower performance than a rocket. K varies between fuels, but all fuels run out of power well below orbital velocity. The easiest way to state it, is this:
Before a scramjet reaches the velocity of a rocket’s exhaust, the rocket is more efficient.
Hopefully that is not too much math for everyone. Scramjets can still be used in a first stage, of course – but there they lose out to simple jet engine types. Please let me know if you spot any errors in this.
“This is where I think scramjets might come in, eventually. Not for the payload, but to squeak a little more altitude and speed from your carrier aircraft.”
The ‘more altitude’ part is good (the expansion of a fixed nozzle on your second stage/orbiter can be a more efficient compromise between launch altitude and vacuum, rather than sea level and vacuum, if you don’t want altitude-compensating engines).
But ‘more speed?’ That sounds nice at first glance too, but in addition to the issues already mentioned, hypersonic separation of the stages can be a very tricky thing, that you may not want to do on a regular basis…
These sorts of debates use to interest me until about 20 years ago, when I finally realise that the problem of space access is really one of economics rather than technology. The simple fact is that there’s a myriad ways to orbit (TSTO, SSTO, pure rocket, rocket-airbreather, etc.) but what really counts is the cost of development and operation. On this count alone, scramjets look awful because their likely development costs are huge – just look at what was ‘invested’ in X-30 before it was cancelled.
However, the biggest problem with scramjets is that they work best as ‘cruise’ propulsion but incur terrible mass penalties when used as ‘accelerators’ because of the need for variable geometry on both intakes and exhaust nozzles. Rockets are quite the reverse and, more importantly, are essentially off-the-shelf technology. Moreover, as any propulsion system will need to be both durable and maintainable, the relative simplicity of rockets gives them a major advantage over the relative complexity of the variable geometry mechanisms that would be required for a space launcher using scramjets.
So scramjets are a dead end. What about a vertical half mile of electric catapult? Holes and rails are relatively cheap to produce with no technical hurdles. Would a balloon launch be better? I’m not up on the math, so I have to rely on the expertise of others?
Sramjets are for cruisers point to point, missiles not orbital.
Air breathring rockets (modified RL-10A-5, RS-68, etc..) dating back to the 1950’s (John Ahern) do have an excellent application to orbital which reduces oxidizer and mass of the entire vehicle at launch. Skylon sabre is one example of this combined cycle approach. There are two possible appilications LACE (Liquid Air Cycle Engine) and deeply cooled (Rudakov, Balepin). Best application I have seen is an air-launch concept using deeply cooled rocket to M5.5 then pure rocket to orbit. LACE reduced the OWE by 150 metric tons allowing for a transport carrier aircraft (AN-225 orbital down to a 737-400 point to point) to act as the first stage (piggyback similar to early shuttle glide test). Upper lifting body (AFFDL shaped) capable of placing 8 crew into LEO. The entire system is reusable, runway landing, quick turn-around, low noise take-off, capable of flying from several locations etc….true CATS true RLV simple sexy and scrubbed concept waiting to fabricate and fly.
There are over 200 737 classic airliners in storage at Evergreen facility in Marana Arz. able to accomidate a stage-n-half (drop tank) sized version.
Oh yeah the old heavy LACE argument. The Japan, NAL, Mitsubishi and ISAS LACE unit was light weight lifted into place by a single technician. Compare that to the redustion of 150 metrc tons. India’s Hyperplane utilized most recent LACE technologies resulting in a very robust system design. The AFFDL flat bottom pointed trapezoid shape is utilized to reduce heating, and wetted area over a cylindercial type Skylon shape.
Deeply cooled, LACE, rocket is the choice for orbital. You can take it further applying augmented nozzle technology etc.. RL-10A-5, RS-68
Robert Horning,
[[[From this viewpoint, a big dumb booster that is cheap to make and simple in its design is, at least for now, the better route for going into space.]]]
Exactly why I like ocean launched designs like the Sea Dragon. Simple and effective.
Regarding scramjets vs. rockets in the atmosphere: why not use a first stage (or zeroth-stage) with an aerospike engine instead of a scramjet?
Quoting Wikipedia here: The disadvantages of aerospikes seem to be extra weight for the spike, and increased cooling requirements due to the extra heated area. Further, the larger cooled area can reduce performance below theoretical levels by reducing the pressure against the nozzle. Also, aerospikes work relatively poorly between Mach 1-3, where the airflow around the vehicle has reduced pressure, and this reduces the thrust.
But all these disadvantages also seem to apply to scramjets, in spades. And an operational aerospike engine will require a hell of a lot less development effort than an operational scramjet. (Apparently the aerospike engine was the one novel feature of the X-33 that wasn’t a problem.) I honestly don’t understand the love affair with scramjets.
So … Would using aerospike engines on the first stage of a vertical-ascent design or in a supersonic White Knight make any sense?
I also should note that I agree with Rand regarding scramjets. They are useful for traveling in the atmosphere at high speeds (like a cruise missile, reconnaissance platforms, or bomber would do) but there is no point in using them for space launch. The velocity range where they work is too narrow. Even pulse detonation engines would do better at space launch. At least they can work from lower initial velocities.
Breakthroughs for example that greatly reduce the dry weight of a scramjet would not, AFAIK, violate any law of physics, whereas physics says we can’t greatly reduce the propellant per payload from today’s chemical rockets.
Googaw, the dry weight is driven largely by the inlets. There are only two ways to reduce that weight: 1) develop some sort of “unobtanium” that makes a large inlets much lighter, or 2) reduce the physical size of the inlet.
1) turns out to be very difficult within the laws of physics. 2) is much easier. Making an inlet smaller not only reduces the dry weight but also reduces drag. Unfortunately, it also means you collect less air, but that’s okay because you can carry a bit of oxidizer to compensate for that. The smaller the inlets get, the better the structural performance gets. The optimal scramjet has no inlets at all. That’s called a “rocket.”
David,
First, let me say that I agree with both you and Rand that scramjets are of no use for earth to orbit propulsion. Indeed, I would go further and say that no airbreathing scheme is of any use in earth to orbit propulsion.
However, the following is incorrect:
in a perfect world, the exhaust velocity is related to the square root of the energy in the fuel in both cases. So the equations simply to:
jet: (mass of fuel/s+intake air mass/s)*(K * sqrt[mass of fuel] )
For the case of the scramjet or any airbreather for that matter there is an additional source of energy – the kinetic energy of the incoming air. At scramjet velocities this is extremely important and is why inlet design is so critical for scramjets. If the inlet is inefficient the kinetic energy of the incoming air is largely wasted.
Therefore your conclusion:
Before a scramjet reaches the velocity of a rocket’s exhaust, the rocket is more efficient.
is not in general correct.
Here is the standard derivation for scramjet potential:
Overall efficiency is defined by rate at which useful work is being done divided by rate at which available energy is being consumed so we have:
K = T * v / mdot * (q + v^2 / 2)
K – overall efficiency
T – scramjet thrust (N)
v – scramjet velocity (m/s)
mdot – fuel consumption (kg/s)
q – lower heating value of fuel (J/kg)
Note that the coordinate system is relative to the earth.
T/mdot is the definition of specific impulse so we have
K = Isp * v / (q + v^2 / 2)
where Isp is in units of (Ns/kg or m/s)
Rearranging gives
Isp = K * (q / v + v / 2)
So with H2 fuel (q = 120,000,000 J/kg) and at a velocity equal to rocket exhaust velocity (v = 4500 m/s) a 25% efficient scramjet is still providing
Isp = 0.25 * (120,000,000 / 4500 + 4500 / 2) = 7200 Ns/kg or 740 s
Note that a H2 fueled scramjet, if its efficiency is high enough, can theoretically provide high Isp at orbital speeds and beyond. Emphasis on “theoretically”.
Of course, a 25% efficient scramjet at M~15 is a pipe dream and there are all the other issues (drag, heat flux, low density propellant, etc) to be addressed.
But the above explains why the scramjet continues to fascinate and keeps coming back into fashion at 20 to 25 year intervals as a new generation forgets the downside in the interim.
When folks look at the cost of going into space, they get all hypothetical about exotic ways to reduce fuel costs for getting there, forgetting that for most launch vehicles that the price of the fuel itself is usually one of the least important factors for getting something into orbit… at least from a cost perspective. From this viewpoint, a big dumb booster that is cheap to make and simple in its design is, at least for now, the better route for going into space.
No, Robert, we don’t forget that at all. Big dumb boosters are not cheap to make or simple to design. 50 years of history shows that. Ask the old timers if the Saturn V was cheap to make or simple to design.
General Dynamics once did a study comparing the development cost of the X-15 to an expendable rocket of similar size and performance (the Atlas A). They concluded that the X-15, while more complex, was chaper to develop. An independent Air Force study, using different methodology, reached the same conclusion.
The late Dr. Max Hunter said that people who think expendable rockets are cheaper to develop forget that in order to design a rocket, you have to test a rocket. Expendable rockets are much more expensive to test, because you have to build a new copy for each test flight — and you lose much of the data because you don’t get the engine or airframe back to study. Look at the number of test flights White Knight and SpaceShip One conducted, for $25 million — 66 flights of the first-stage carrier aircraft and 17 flights of the upper-stage SpaceShip One. To conduct a similar program with an ELV, you would have to build 66 first stages and 17 upper stages. If you could build each “dumb” stage for only a million dollars, you would still spend $83 million.
Even if you look at subsonic platforms — cruise missiles vs. piloted aircraft — you will find that expendable vehicles, while they may be simpler, are not cheaper and easier to develop. Aircraft have redundant systems for a reason; it’s not just unnecessary complexity. That’s why some UAVs are initially developed and tested in a manned configuration with a temporary cockpit. It saves time and money during the development program.
As for the “big” part, here is where you go seriously astray. Making a rocket bigger can improve the structural efficiency — i.e., the propellant ratio. But as you noted, propellant is a minor part of ELV launch costs. So, you’re building a lot more expensive hardware to save on cheap propellant. Even if the marginal costs are slightly reduced, the development and capital costs will skyrocket. Again, look at the Saturn V.
The White Knight air-launch works in the sense that it avoids having to deal with the bottom layers of the troposphere (aka “weather”) and it is possible to “launch” in less than ideal conditions and in more places than one fixed spaceport. That is about all it really does in terms of providing better launch conditions,
It’s not “all” it does. Air launch also allows igniting the rocket engine at lower atmospheric pressure, where rockets perform better. It may also eliminate the requirement for altitude-compensating rocket nozzles or multiple engines. Depending on the design, it may also sizing the wings and landing gear for landing weight, rather than takeoff weight.
The sad thing about these sorts of discussions is that they highlight just how little of the design-space has been explored over the last half century. Worse still, with the exception of a few entrepreneurial companies, there has been an almost complete absence of any serious effort to develop reusable launcher technologies during the past decade.
This is one of the main reasons why I think the new Obama space policy is a genuinely positive step.
Exactly why I like ocean launched designs like the Sea Dragon. Simple and effective.
Tom, I suggest you talk to anyone who’s served in the Coast Guard or Navy. They’ll tell you that search and recovery operations at sea are not always effective and they are never simple.
Note that SpaceX has not yet successfully recovered any of their stages, as they originally planned. It also appears that Robert Truax may have encountered some problems when he was dropping his Sealar models from helicopters for the US Navy, although there’s not enough public information to say. Mercury lost a capsule (and nearly lost an astronaut with it).
The sad thing about these sorts of discussions is that they highlight just how little of the design-space has been explored over the last half century.
I think that it has been well explored, but over too longer time for everyone to remember and analyse the lessons learned. Experienced rocket engineers up until about Apollo had been through many rocket vehicle developments, nowadays, an experienced rocket engineer is lucky to have been through one rocket vehicle development program. The vehicle turnover is just not there, and with it the experience. Nowadays, NASA engineers can not even redo Apollo, let alone take a giant leap from something like Apollo to the shuttle and beyond.
I am not expecting any serious advancements until the prototyping rate that existed pre Apollo is again obtained. Hence my hopes for New Space, which is trying to get back up to prototyping speed.
Payload to dry mass ratio is kind of important for a launch vehicle, and it is really bad for a Scramjet. Scramjets are also extremely expensive pieces of dry mass, not like cheap propellant tanks. They have very narrow speed ranges, they must fly up through the atmosphere at reentry velocities, they are seriously aerodynamically constrained, they must carry rocket engines anyway, and the list goes on. Scramjets kind of have to turn the entire vehicle into an inside out rocket engine which also has to fly like an airplane.
Yes with technological advances there is significant room to reduce the dry mass of a Scramjet, but the same is actually also true for rocket vehicles, and much easier. There is yet a huge amount of room to reduce tank and engine mass for rocket vehicles.
The LACE engine vehicle designs are interesting:
http://en.wikipedia.org/wiki/Reaction_Engines_Skylon
But even here one is talking about tens of billions of dollars and decades to develop. In comparison, with a good approach the development of a CATS RLV rocket vehicle is probably now in the $100m and 5 year range.
Even in the very long term I see no advantage in a Scramjet, extra LOX is much cheaper than extra LH2.
Pete, please note that Skylon’s engine cycle is not LACE: the ingested air is not liquified and is not collected. Sabre (a.k.a. RB-545) essentially adds an intake/cooling/compression kit onto a conventional rocket engine in order to allow it to use atmospheric oxygen (injected at 90K and 150bar) during the initial part of the launch trajectory, up to around Mach 5 at around 30km altitude.
It was conceived by Alan Bond as a way of reducing gross launch mass in order to enable a practical SSTO but, given the novelty and complexity of the air-breathing kit, has always been somewhat controversial. This is why Reaction Engines has focused the majority of their efforts to date upon demonstrating the pre-cooler technology by building flight-weight heat-exchanger elements.
Ed: Saturn V was not a big dumb booster. It used five engines with turbopumps in the first stage alone. The idea behind a big dumb booster is that systems with more parts are more expensive and have a higher probability of failure. That it is better to have a less efficient and larger design, rather than having a more efficient albeit smaller design.
Another thing that people often forget is that you can do static testing of rocket engines and rocket stages. So the advantage of being reusable for testing purposes is not as big after all. Propellant leaks could still lead to a loss of vehicle in a reusable, as could engine failure, structural failure, or staging errors where there is a collision between stages. Even guidance errors could lead to reusable vehicle loss (e.g. Airbus crashes due to fly by wire bugs).
There will be some reduction in costs for a reusable, like you have no need to build a test stand for the vehicle, you only need the launch platform or runway, plus as you said you can do post-flight reinspection on every flight (where the vehicle does not explode).
I almost bet that they did not consider the cost of the B-52 launch platform when they did that X-15 cost estimate however. Nor pilot training costs (which the Atlas did not require).
So while reusables could end up cheaper for space launch in the end, this is not proven at all. Especially not when maintenance of a reusable costs more than building a new vehicle.
SpaceShipOne did not solve the orbital launch and reentry problem. It can reach less altitude than a lot of sounding rockets.
This does not make it less awesome, but it is chalk and cheese.
I was hoping someone would weigh in with comments on the X-30 National Aerospace Plane (NASP). Also known as Copper Canyon.
When that was first proposed, I thought it was a wonderful, very promising concept, and the best way forward. (In my defense, I’ll mention that I was in Jr. high at the time).
The Mach 8 thermal barrier issue was what threw me for a loop, when it came up. (By this time, I was in high school, and had an interest in science). The claim was that the drag issue was solvable, by a clever trick; reclaim all the energy lost to drag by using hydrogen to cool the heated aircraft surfaces, and then using the hot hydrogen for combustion, thus “recapturing” and using the energy, and thus “solving” the mach 8 barrier issue.
Err… Well, to me at that time, that just made no sense whatsoever. One operating assumption seems to be that ALL the energy loss via drag, air displacement, etc, it converted to airframe heat. The second seems to be that all that energy is 100% transferable to thrust, and the third seems to be that this energy-collection apparatus would have zero mass. Let’s just say I had issues with those three assumptions.
The other issue that puzzled the heck out of me; the NASP design concepts used hydrogen. They were also very slender and sleek. That made no sense at all, given that I knew the relative sizes of the hydrogen and 02 tanks in the shuttle ET. The Oxidizer massed more, by about a factor of six, but the hydrogen took up one hell of a lot more volume (about two and a half times as large). So, you’d still need the bulkiest fuel (hydrogen), which did not seem to fit at all with the sleek designs.
I was, as time wore on, more and more receptive to the theory that the NASP was never intended for SSTO, but was instead a development program for a successor to the SR-71 or some other military application. (Goals I would be fully supportive of, so my speculation was not politically motivated)
Later still, I have my doubts on that, based on the fact that by “hiding it in plane sight” who the heck are you fooling if it obviously cannot, even theoretically, do what you claim? It remains an enigma to me.
Question: was the NASP concept as preposterous as it sounded to me? And if so, why on earth would it be pursued?
My admittedly uninformed current opinion on Scramjets is this; I think it might, with massive technical advances in several areas, be possible to make use of it, after extraordinary development cost, for SSTO, provided one omitted reusablitly or payload of any sort, giving you a capability of putting just an empty fuel tank in LEO, at a cost not much more than an order of magnitude higher than today’s least-affordable methods.
Pete, please note that Skylon’s engine cycle is not LACE: the ingested air is not liquified and is not collected.
Yes sorry, brainfart…
Ed: Saturn V was not a big dumb booster. It used five engines with turbopumps in the first stage alone. The idea behind a big dumb booster is that systems with more parts are more expensive and have a higher probability of failure.
Five F-1 engines on the first stage alone makes the Saturn V a big rocket, IMO. I’m well aware that engineers who built the Saturn V (and other big rockets) made different design choices than those advocated by “Big Dumb booster” advocates. I also suspect they had reasons for those design choices. Von Braun may have been a lot of things, but he wasn’t stupid.
So, Southwest Airlines must be very expensive and have a high probability of failure, because they fly airplanes with a lot of redundant systems?
Reality does not seem to conform to that view.
Another thing that people often forget is that you can do static testing of rocket engines and rocket stages. So the advantage of being reusable for testing purposes is not as big after all.
No, people don’t “forget” that. No professional engineer would think that static testing is a substitute for flight testing / road testing / sea trials. That belief is unique to the rocket field, which grew out of the German amateur rocket clubs. It’s one reason why rocket engines and stages fail at a much higher rate than aircraft, cars, or ships.
Even guidance errors could lead to reusable vehicle loss (e.g. Airbus crashes due to fly by wire bugs).
The Scarebus has serious design flaws. no doubt. Even with those flaws, it is orders of magnitude more reliable than any expendable. If Scarebuses failed at the same rate as ELVs, there would be hundreds of crashes every week.
Is that the best argument you can come up with?
I almost bet that they did not consider the cost of the B-52 launch platform when they did that X-15 cost estimate however. Nor pilot training costs (which the Atlas did not require).
And I almost made an easy buck off a sucker. 🙂
So while reusables could end up cheaper for space launch in the end, this is not proven at all. Especially not when maintenance of a reusable costs more than building a new vehicle.
Maintenance does not cost more than building a new vehicle. Don’t confuse religious belief with facts.
SpaceShipOne did not solve the orbital launch and reentry problem. It can reach less altitude than a lot of sounding rockets.
No one said SpaceShip One solved orbital launch and reentry problems. It wasn’t intended to. Sounding rockets don’t solve those problems, either. Can you tell me which “Big Dumb” sounding rockets are cheaper than SpaceShip One?
Can you point to *any* expendable vehicle that is cheaper than reusable vehicles in *any* regime?
On the one hand, you dismiss SpaceShip One because it hasn’t reached orbit. On the other hand, you want us to accept cost claims about Big Dumb boosters that have never left the drawing board, let alone reached orbit. (When expendable rockets are actually built, they don’t turn out to be phantasmagorically cheap, and we’re told that they aren’t really Big Dumb boosters.)
googaw, wow, you’re dumber than I thought.
You wanna claim more experience than me, that’s not hard.
Rand’s qualifications are well known, what are yours?
“So while reusable could end up cheaper for space launch in the end, this is not proven at all. Especially not when maintenance of a reusable costs more than building a new vehicle”
Cost come down with flight rate, like most systems commercial airline services etc… There is an ideal flight rate and number of vehicles that one must field to maximize ROI. Do the market research and run the numbers to find out if the markets support it.
I believe the Skylon Sabre is deeply cooled not full on LACE. Bond is not new to this having worked on HOTOL etc… He knows the limitations M5.5 range to maximize the effect.
For LACE and deeply cooled technologies there is an extensive data base of NACA, NASA, Marquardt papers, patents etc..from the heady research days of the sixties. Some of the old timers from that era are still around and capable of providing a wealth of experience and advice. Plus new discovers from ISAS Japan, Russia, and India. The R&D foundation has been laid what is needed is to build some flight worthy hardware and go fly it. What is needed is the will to embrace, flight test and apply. “Big and dumb” and super sized hobby propulsion systems style CATS might be simpler, safer and cheaper upfront but does it take us to the next level to true reliable high flight rate RLV CATS?
The SCRAM cycle regime is a daunting challenge I doubt if it has direct applications to LEO. However the materials developed to survive the intense heat and temps certainly may. Air-breathing cycles like deeply cooled, augmented and LACE if applied to existing off the shelf LH2 rocket motors based on current industrial capability could offer profound and sudden quantum leaps forward. The biggest obstacle seems to be awareness.
Maintenance does not cost more than building a new vehicle.
While this may be true 99% of the time, it may not be true in all cases. It depends on what you compare. Maintenance cost for the shuttle is enormous because of the choices made in its design. It is certainly cheaper to put the same mass in orbit choosing some expendable rocket. Alvin Toffler seems to be winning here with regard to the disposable society.
OTOH, if it were true even 1% of the time, you’d think somebody would have built an example by now (forget airline analogies… where’s the orbital example?)
Until somebody actually does it the argument fails. I personally believe in the eventual success of a reusable, but for now expendables are the only examples we have.
OTOH, if it were true even 1% of the time, you’d think somebody would have built an example by now (forget airline analogies… where’s the orbital example?)
They have.
Ken, it doesn’t cost more to maintain an orbiter than it would to replace it. Period.
So, Southwest Airlines must be very expensive and have a high probability of failure, because they fly airplanes with a lot of redundant systems?
Redundancy is a separate issue. I may not be a mechanical engineer, but I am a software engineer. Software easily can have much higher system complexity than a rocket would. We know damned well that the more non-redundant parts a system has, the highest is the possibility for failure. More parts also increase cost and maintenance.
Ever noticed intercontinental airplanes used to have four engines, then three, and now most have two engines? Why do you think they did that? Oh right. It is cheaper.
Or for that matter, try reading about the RS-68 rocket engine design philosophy. Minimizing the number of parts was an important part of it.
It only makes sense to increase the number of parts in the system to implement required features, or when the parts decrease overall costs when considering the universe outside the designed system.
Redundancy is nice sure, and helps increase reliability. But it costs money too.
As for Von Braun, you should take notice he pursued expendable rockets instead of reusable vehicles. There were plenty of other people at the time who favored full reusables. They did not get anyone or anything in the Moon. Or even LEO for that matter. Neither then nor in 50 years. Neither in the US, nor elsewhere. Coincidence?
It was not for lack of trying either. Stalin was so enamored with orbital bombers he even tried to kidnap Sanger from France.
The Scarebus has serious design flaws. no doubt. Even with those flaws, it is orders of magnitude more reliable than any expendable. If Scarebuses failed at the same rate as ELVs, there would be hundreds of crashes every week.
“Fly by wire” problems are hardly limited to Airbus airplanes. As can be seen by the recent Toyota Prius braking software bug. Of course it is always easier to swallow a pilot error, than a computer error. People like control after all.
Most failures in an expendable rocket traditionally happen during the early launches. Ariane 5 is one example. Still the failure rate is higher per flight than an airplane. But airplanes did not begin with the failure rate they have today either. Neither have there been enough flights (hah!), at enough rate, nor enough people, nor enough history at solving the problems.
As for the superiority of reusables versus expendables, I just threw a paper tissue in the waste basket. See, it is easy to make analogies.
“All jet/rocket engines create thrust by throwing stuff out the back faster than they are moving.”
This is not true. One can throw anything out the back at very slow speeds compare to your velocity and still gain thrust.
But the faster you throw anything out the back, the more efficient one is using that mass. Obviously the most effective use of that mass is if it leaves the rocket at the speed of light.
If what you said were true, then we couldn’t reach orbital speed with chemical rockets. One of highest exhaust of chemical rockets is hydrogen and oxygen: 4,440 m/s. Which is slightly more than 1/2 of orbital velocity [around 7,800 m/s].
Wiki:
“An example of a specific impulse measured in time is 453 seconds, or, equivalently, an effective exhaust velocity of 4,440 m/s, for the Space Shuttle Main Engines when operating in vacuum.
An air-breathing jet engine typically has a much larger specific impulse than a rocket: a turbofan jet engine may have a specific impulse of 6,000 seconds or more at sea level whereas a rocket would be around 200-400 seconds. Note that an air-breathing engine is thus much more propellant efficient; this is because the actual exhaust speed is much lower, because air provides oxidiser, and because air is used as reaction mass.”
Redundancy is a separate issue.
No, it is the issue you were talking about. You claimed that “systems with more parts are more expensive and have a higher probability of failure.”
In the real world, more parts (greater redundancy) generally reduces failure.
We know damned well that the more non-redundant parts a system has, the highest is the possibility for failure.
Really? Your write your programs without error checking, to minimize the number of parts? Or do you only recommend that for rockets?
When you design a piece of software, do you ignore other designs that have been successful in the past because you don’t believe in “analogies”? Or do you only recommend that for rockets?
Do you design your program so that it can only be run once, and can’t be tested as a complete system before it goes into production? Or do you only recommend that for rockets?
Ever noticed intercontinental airplanes used to have four engines, then three, and now most have two engines? Why do you think they did that? Oh right. It is cheaper.
It’s cheaper because engine reliability has been improved over many generations and millions of hours of operational experience. And they do acceptance flight tests on every engine before it goes into service. They don’t just do a static test and call that good enough, as you say Big Dumb boosters should do.
There’s also a lot of redundancy within the engine. The engine doesn’t fail the first time one part fails. And there are a lot more systems in the airplane besides the engines. A 777 is a much more complex aircraft than a B-52, even if the B-52 had eight engines and the 777 only has two. It’s more reliable, as well.
Or for that matter, try reading about the RS-68 rocket engine design philosophy.
The RS-68? How is that relevant? Do you consider the Delta IV a Big Dumb booster? If so, Delta IV is no cheap, so this doesn’t prove your notion that Big Dumb boosters will be cheaper than reusable vehicles. If not, why are you bringing it up?
Redundancy is nice sure, and helps increase reliability. But it costs money too.
No, it saves money — that’s why the airlines are willing to pay for it. Having aircraft constantly going out of commission because they have no redundancy costs money.
As for Von Braun, you should take notice he pursued expendable rockets instead of reusable vehicles. There were plenty of other people at the time who favored full reusables. They did not get anyone or anything in the Moon.
Yes, the Moon race was a stupid goal. What’s your point? Because people have done stupid things in the past, you think everyone should continue to do stupid things in the future?
I thought you said Von Braun’s rockets weren’t Big Dumb boosters because he used turbopumps? In any case, his rockets certainly weren’t cheap, so they certainly don’t prove that Big Dumb boosters will be cheaper than reusable vehicles.
Neither then nor in 50 years. Neither in the US, nor elsewhere. Coincidence?
No coincidence at all. When no one with the necessary resources to do X makes an effort to do X, then X usually doesn’t get done. I don’t know why you find that so remarkable — and no, that doesn’t mean it’s impossible to do X if someone who has the resources decides to try.
“Fly by wire” problems are hardly limited to Airbus airplanes.
I didn’t say they were — and again, even with the problems, Scarebus is far more reliable than an expendable aircraft (aka cruise missile) with similar performance. I’m still waiting for you to offer evidence that expendables are cheaper and more reliable.
As for the superiority of reusables versus expendables, I just threw a paper tissue in the waste basket. See, it is easy to make analogies.
Yes, it’s easy to make stupid analogies. You’ve proven you can do that. Saying something intelligent is a bit harder. Please try.
Again, do you have any actual evidence to support your belief that expendables will be cheaper and more reliable?
as a software engineer I giggle that you would ever bring up software in a discussion about reliability. Way to discredit everything else you have to say 🙂
Trent: it is easy to mock software when you do not have a clue about the level of complexity inside it. It is also easy to mock software when you are comparing industries which have existed for centuries to it. Of course in that case your problems have usually already been mostly solved by someone else and then you throw your compendium of pre-existing design rules at it.
“As for Von Braun, you should take notice he pursued expendable rockets instead of reusable vehicles.”
It depends on when you look at Von Braun, or for what purpose. From 1948 through 1956, his designs were all reusable. Even the stages without wings had parachutes and airbags.
The reason reusables played no part in anyone’s moon mission is that we were in a race to get there, and had to pick something that had a high probability of success. The complexities of aircraft development were well known, as were the difficulties of high-speed flight. Developing an aircraft-like vehicle would be daunting, to say the least. What ruled it out completely was size. For even a small payload, a larger airplane than had ever been flown would be required. To put it in perspective, the Saturn V’s dry weight was about equal to the heaviest aircraft gross takeoff weight at the time. With propellant, it’s more than an order of magnitude larger.
On the other hand, it was known that an expendable rocket could be scaled, and there was no physical reason to suspect the existence of a size limit. The advent of LOR allowed us to do the mission with a single rocket of (barely) achievable size. Anything else would have required a Nova (for direct ascent and return), or LEO rendezvous involving multiple launches.
Von Braun’s Mars mission would have required 950 launches, BTW, which is why he was so focused on reusability. With Apollo/Saturn V able to do a lunar landing with one shot, it wasn’t necessary to try to reuse the Saturn. Nevertheless, fairly extensive design studies were conducted to try to get at least the first stage back. Performance impacts were so large, and the reliability of ocean-recovered hardware questionable enough, that these studies concluded that it was not worthwhile.
In “The Promise of Space,” 1968, Arthur Clarke wrote that if geopolitics hadn’t intervened, we would have developed small, rugged, reusable launch vehicles, and built an orbital infrastructure (to paraphrase with modern buzzwords) which would have allowed us to be ready to go to the moon around the end of the century. That was how most people, Von Braun included, thought about the progress of space in the 40s and 50s.
Big expendables (and the need for “heavy lift” in general) were an aberration, born of the urgency of Apollo and cemented into the space psyche.