Paul Allen's new rocket-launching plane
arstechnica.com
arstechnica.com
1. +230m/s orbital velocity [1]
2. +9km launch altitude [1]
3. can launch in more weather conditions
4. 1st stage can abort mid-launch without loss of payload.
Of these, the most important is the increase in launch altitude, because that means stage 2 launches in air of greatly decreased density, leading to the following efficiencies:
A. a vacuum-optimized nozzle can be used, increasing engine performance
B. less aerodynamic/drag energy losses
C. max Q should be lower, allowing full throttle usage, reducing mass needed for structural elements, and reducing vibration stresses on payload
P.S. Thank you KSP for teaching me this stuff! :D
Bonus: video of pegasus air launched rocket: https://www.youtube.com/watch?v=jz_5hnIw2jc
For example Cape Kennedy is 28 degrees north. If you fly to the equator, you get another 200 m/s.
Depends a bit to which orbit you're launching though.
Another good thing is flexibility if you are more constrained by launch windows in a stationary launch location. Hard to explain.
cos(28)=0.88
0.12*(2*pi*7000 km) = 5300 km
5300 km / 24 h = 220 km/h
220 km/h / 3.6 m/s/km/h = 61 m/sfor example the a340-600 has a MTOW of 380 tons but a Max landing weight of 265 tons. Just a bit of a difference.
[1] Figures given in pounds on wikipedia, the kg version looks like a "dumb" conversion: https://en.wikipedia.org/wiki/Boeing_787_Dreamliner#Specific...
This is very useful if say you want to very quickly and quietly launch something towards another satellite.
People are making comparisons with the Spruce Goose, but the Glomar Explorer may be a better analogy.
https://en.wikipedia.org/wiki/Pioneering_Spirit_(ship)
There are videos of this thing lifting an entire oil platform at once.
Surely you could already do that with Orbital ATK's Pegasus? 1000 lbs is plenty of explosives. It seems to me this is just a bigger version of that.
edit: CIA and kursk submarine, was it the kursk doesn't sound like the right time
edit: nope Kursk is 141, K-129 is the one they wanted (Glomar Explorer)
I like the jet-rotor tip helicopter too
edit: haha yeah beast (photo of the helicopter with massive rotors spinning I think at 88rpms)
https://s-media-cache-ak0.pinimg.com/736x/04/f9/9b/04f99b793...
Launching a Pegasus XL has to be just a demo. That's normally launched by dropping it from an old Lockheed L-1011; it doesn't need this monster. They need something bigger to launch. There was a contract with Pegasus for a "Pegasus II", but that didn't work out. The "Dream Chaser" mini-shuttle is a possibility, if that ever gets built.
The 747 engines are probably used. There are lots of retired 747s around, many at the Mojave boneyard. Great aircraft, but a fuel hog by current standards.
Correct. They just flew in two used 747s and cannibalized them for parts. http://www.seattletimes.com/business/boeing-aerospace/paul-a...
>The crew’s flight deck is literally that of a 747.
>Allen bought two used jumbo jets formerly flown by United Airlines and cannibalized them for parts that account for about half the empty weight of the Roc.
>So although the shell of the cockpit and all the rest of the plane’s body is new — hand-built by Scaled Composites from carbon fiber composites — various key pieces and systems, including avionics, hydraulics and fuel subsystems, are salvaged 747 parts. BAE Systems was subcontracted to disassemble the 747 and install its systems on the Roc.
>The cockpit seats look old and used because they are. The seats as well as the controls the crew will manipulate and the windows they’ll look out of all came from the 747s.
>So did the plane’s six Pratt & Whitney engines, which are already refurbished, cleaned, wrapped and set aside in a corner, ready to hang on the airframe when it’s finished.
The Seattle Times article says that the Vulcan rocket will have a mass of 375 tons. (Probably US tons?) Assume it's got the same payload fraction as a Soyuz, about 2.5%, which means it can put 9.4 tons into orbit. (A Falcon 9 can put 22 metric tons into orbit)
According to some comment on stack exchange, the world collectively orbited 255 tons of stuff in 2007. Assuming every payload can be launched on a Vulcan rocket, (Which they can't) and assuming a generous turnaround time of one week, (Allen says it should be faster than that) then it would take only a single plane 27 weeks to satisfy the launch needs of the entire world. You could double the amount of stuff launched, and still only need one plane.
There's probably not going to be a lot of airframes to spread development costs across.
The biggest thing going against it is it's not clear this scheme can drive the costs down enough to double or triple the launched mass per year. There is a limited number of things we can do in space. A significant part of the cost of any single probe is the launch.
If we can reduce launch costs enough (maybe making the booster reusable) and through building a standard probe with pluggable sensors, drive down the building costs, we could have dozens of probes going places we currently can't afford to go. We spend enormous effort in shaving off every single gram from anything that goes to space. If going to space gets cheaper, we can spend a lot less time shaving probes and more time probing.
Rocket exhaust?
This vehicle is going to be incredibly expensive, compared to both other air launch options and dedicated smallsat options. Although this can change, the current plan is for Stratolaunch to use OATK's Pegasus-XL payload, which provides an indicator for cost. Pegasus-XL launches are incredibly expensive ($337.3K per kg). Now, this is due to a variety of reasons (the L-1011's incredibly high maintenance costs, OATK's expensive labor structure, low launch cadence). Vulcan will bear similar costs--the inefficient OATK overhead tied to Pegasus-XL, Stratolauncher is a one-of-a-kind aircraft and increasing maintenance costs (despite using 747 engines). Due to significant development delays, the company has yet to develop or execute a customer strategy. Depending on how much more PA pumps into it, the company will not be price competitive in the market.
For comparison, other small launchers charge ~$25K/kg-$41K/kg (e.g., Rocket Lab, Virgin, Arianespace), and most of these will be able to launch US payloads. Virgin's comparable airlauncher delivers slightly less mass (15 kg less), but is priced at ~$40K/kg. Virgin is already a leg up as it has engaged commercial, civil govt., and mil-govt. customers.
Lastly, it's worth noting that this vehicle still has significant work to be done--look at the wings in the picture.
you seem to know a lot, any chance we could talk about this topic, would love to hear more of your thoughts on the topic? I don't see an address to contact you.
Thanks!
scaled composites, which built this also built the GlobalFlyer with similar configuration.
Imagine that thing, fully loaded taking off into turbulent air. Scary.
http://www.stratolaunch.com/gallery.html
the black is cfrp under the white paint.
Scaled Composites has some secret CFRP processes of construction which non Americans are not allowed to know about under ITAR regulations, so i cannot tell you haha.
Everybody knows that. We just don't want you to know we know ;-)
Again, I know that much smarter people did the math, but from the eye check this does not look like a tenable design.
Imagine differing oscillations of pitch or yaw from each fuselage happening at the same time... and at just the right/wrong frequency for the connecting structure.
I'm sure I'm just overly paranoid, but often we find out long after something is built that some "little possibility" was overlooked.
Exactly. I expect they have real-time computer control of all control services that quickly adapt the relative attitude of both fuselages to prevent over-stressing the wing, but what about "rogue waves" of turbulence hitting each fuselage differently. What's the safety factor on both the adaptive controls and, as a backup, the structural integrity of the wing?
No. This is breakthrough science. Scaled proofed a specific implementation of one of its materials.
twin tail booms have a long history.
e.g p38 with twin engines.
Burt Rutan's key realization was that the connected skinny horizontal tail doesn't really add much torsional stiffness and it's easy to obtain the required stiffness with the wing center box.
1.3 million pounds
65 feet
2,000 nautical mile
I'm sorry, I know it has been mentioned a billion times already, but can someone make a browser plugin that converts everything on a website to normal SI units? As a non-american I have no idea how big this aircaft is or how much it can lift - sure I can look up conversions(I have) but it would be awesome if there was an automatic website conversion option.
I'd be surprised if a unit-converter plugin didn't exist already, but it should definitely preserve the original measures since the units do have contextual significance. e.g. nautical miles in navigation
edit: here's one https://chrome.google.com/webstore/detail/autoconvert-auto-c...
The Falcon 9, by comparison, uses the first stage to get many times higher, but spends a huge amount of it's fuel during that first part of takeoff. For smaller launches, the plane-drop seems a much better idea.
... Do they know something we don't?
Clearly there's something I'm missing here (or they wouldn't have built it), can anyone fill me in?
[0] https://en.wikipedia.org/wiki/Air_launch_to_orbit
[1] https://en.wikipedia.org/wiki/Pegasus_(rocket)#Launch_histor...
https://en.m.wikipedia.org/wiki/Scaled_Composites_White_Knig...
As others noted, this has been done before, but it's not too common. I think that's mainly because rockets need to be really, really big to get a decent-sized payload to orbit. Launching from an aircraft would let them be a bit smaller, but they still need to be huge. Pegasus (the one currently in use) is a small rocket and can only orbit a small payload. This plane is designed to carry about 500,000lbs of rocket. For comparison, a fully loaded Falcon 9 is about 1.2 million pounds, and that's only a medium-sized rocket when it comes to orbital launchers.
Amusing aside: in the 1970s, the US Air Force tested the feasibility of air-launching ICBMs. The test was carried out by loading a Minuteman ICBM into a C-5 Galaxy transport, flying it over the ocean, and then essentially shoving the missile out the back door and lighting it up. In this case, the advantage wasn't efficiency or payload, but rather the ability to move ICBMs around to make it more difficult to destroy them in a first strike. They decided it wasn't worth all the trouble in the end.
>Atmospheric and gravity drag associated with launch typically adds 1.3–1.8 km/s to the launch vehicle delta-v required to reach normal LEO orbital velocity of around 7.8 km/s (28,080 km/h).
So it'll help. Also significant is the ability to base out of an airport and launch closer to the equator.
[1] https://forum.nasaspaceflight.com/index.php?topic=9959.msg18...
The majority of the horizontal delta-V is produced in the second stage. The job of the first stage is to a large extent just to lift the second stage out of the densest part of the atmosphere, to minimize drag for the second stage. A plane can do the same thing, and possibly more cheaply in certain scenarios (well, that has to be Allen's plan anyway).
The big advantage of airlaunch is the Isp improvement from launching at altitude. Since the rocket equation is exponential with respect to the ratio of delta V and Isp improving Isp even a little has big effects, and reducing delta V even a little does too.
But the big advantage of ordinary vertical launch rockets is that it's fairly straightforward to just make them bigger. The Saturn V was able to launch Skylab with a 6.6 meter diameter. And rockets could easily be scaled up to launch even larger diameter payloads. That's pretty difficult with something like Stratolaunch. Even with basically the largest aircraft ever made it still has a pretty small payload mass and fairing size.
One other cool thing about stratolaunch is that it has the ability to launch out of a lot more locations since it doesn't depend on a launch pad and tower. However, I suspect that's not going to be of huge importance.
I'd like to see what they can do with this technology but I suspect it'll be underwhelming.
Also, I wonder how this compares in capability to the XS-1 currently under development, which will be a vertically launched first stage rocket that can fly back for a runway landing.
Edit: just found the link below from last year on Space.com.
http://www.space.com/32680-stratolaunch-rocket-vulcan-aerosp...
With a jet engine, air is your friend, because it is the matter you spit out the back of the engine. And at the much smaller speeds of an airplane, mechanical and thermal stresses are much smaller.
It's prohibitively expensive and logistically challenging to build on a mountain, and mountains on Earth don't go nearly as high as planes can.
Florida's air is humid and dense. You get lots of heat gain from that. The density itself matters. Condensation is a huge source of heat. Freezing is another huge source of heat. So you boil off cryogenics and maybe worse. We lost a space shuttle due to a chunk of ice hitting the wing. Falcon 9 rockets need unusually cold filler; if it warms then it expands and won't even fit in the rocket. It probably pours out and gets drained away, the alternative being a rupture.
The USA would do better launching from near Tuscon, Arizona. It's decently high up and dry, which helps keep the cryogenic content cold. We used to do this in fact, before the Mexican embassy complained about defective rockets crashing in Mexico. We chose to move to Florida instead of making a liability payment deal with Mexico.
http://www.zero2infinity.space/bloostar/
The low pressure lets them get away with toroidal tanks, pressure fed booster engines, and a relatively large faring. Or at least that's the plan, they haven't launched anything close to orbital class yet.
But what is the point of a space elevator then?
Every time you get some altitude, you also get a minuscule amount of velocity because the Earth is rotating. A space elevator gets you so much mind blowing altitude that you'll also get all the velocity you need.
Also, the best places to launch from are near the equator where there is a large body of water Eastward. And there are very few mountains in such spots.
Not a dumb question at all. It would actually give you some non-trivial advantage - but altitude matters considerably less than speed when trying to achieve orbit.
Starting a mile higher would allow you to avoid the densest parts of atmosphere, giving you less air resistance and helping with the Max Q, or maximum dynamic pressure (rockets sometimes need to reduce their engine output until they have cleared the denser atmospheric layers, or they would break up due to air resistance stresses).
This does unfortunately not outweigh the cost and complication of moving your launch site on top of a mountain. If you're really after every scrap of advantage, moving your launch site closer to the equator is way better than moving it up (until you already have it at the equator, that is).
That said, remember that the difficulty of elevating yourself above a given height is much worse the lower you are. Using a plane-drop means skipping the very worst parts of the launch. The air is so much thinner.
There's other advantages though. Launch windows get much easier, as the plane can fly in any direction, and stay at a given heading a bit longer.
But the rational part of my brain kicks in and slaps me down for being so arrogant. It feels humbling to be reminded how stupid one can be.
[0] http://www.businessinsider.com/elon-musk-patents-2012-11
[1] https://www.tesla.com/blog/all-our-patent-are-belong-you
The worst thing is, I can't even remember where I have learned that plane-launched rockets make little sense.
Extending my question further, why does this plane have a cockpit at all? Isn't this something you could remotely pilot, especially to mitigate the potential risk of being at 30k feet with thousands of kilos of RP-1 or similar?
Hell, for all we know, this could be remotely piloted.
Yes, I know what you meant. :) Interesting question.
More important than that is that the rocket doesn't have to punch through as dense of an atmosphere when getting up to speed. That means you don't have to reinforce the rocket as much since it's maximum dynamic pressure (maxQ) is lower giving you a better mass ratio and it means that you're not losing as much delta-v to air resistance but that's only another 100 m/s or so for a decently sized rocket like the one being launched here.
But more importantly you can design your engines so that they don't have to work at higher atmospheric pressures.
https://en.wikipedia.org/wiki/Rocket_engine_nozzle
A Merlin engine with a bell designed for taking off from sea level has a trust at sea level of 845 kN but a thrust in Vacuum of 914 kN all while burning the same amount of fuel. And a Merlin with a nozzle sized for vacuum use can go up to 934 kN when firing in vacuum, again for the same fuel flow rate.
(It's possible the article has been updated since you saw it to correct that.)
Have a plane take off with a small load and have rockets connect with it at it's highest altitude and push it into orbit.
expended stages could reconnect with another plane of the same model on the way down.
Allen basically built a museum to (mostly) Hendrix. It's since been re-purposed somewhat, and Wikipedia doesn't have much of the history. But originally the museum was mostly Hendrix. Here are links to a few early articles that make it clearer:
http://www.independent.co.uk/travel/americas/in-search-of-ji...