Startup SpinLaunch Inc. has received $40 million in funding. The company intends to use a centrifuge to catapult small payloads to the edge of space:
The company remains tight-lipped about exactly how this contraption will work, although its name gives away the basic idea. Rather than using propellants like kerosene and liquid oxygen to ignite a fire under a rocket, SpinLaunch plans to get a rocket spinning in a circle at up to 5,000 miles per hour and then let it go—more or less throwing the rocket to the edge of space, at which point it can light up and deliver objects like satellites into orbit.
[...] Over the past few years, the rocket industry has become quite crowded. Following in the footsteps of Elon Musk's Space Exploration Technologies Corp., dozens of companies have appeared, trying to make small, cheap rockets that can be launched every week or perhaps even every day. These smaller rockets have been built to carry a new breed of shoebox-sized satellites—dubbed smallsats—that are packed full of imaging, telecommunications and scientific equipment. The small rockets, though, are really just miniaturized versions of the large, traditional rockets that have flown for decades. SpinLaunch is an entirely new take on the rocket-launch concept itself.
[...] SpinLaunch has a working prototype of its launcher, although the company has declined to provide details on exactly how the machine operates or will compare to its final system. The startup plans to begin launching by 2022. It will charge less than $500,000 per launch and be able to send up multiple rockets per day. The world's top rocket companies usually launch about once a month, and most of SpinLaunch's rivals have been aiming for $2 million to $10 million per launch for small rockets. If the startup were able to reach its goals, it would easily be the cheapest and most prolific small launcher on the market.
NextBigFuture puts the velocity at up to 4,800 km/h (3,000 mph) instead.
(Score: 5, Informative) by Hartree on Sunday June 17 2018, @08:11AM
This sounds a lot like an idea called the "Slingatron" from the late 1990s. Here's a paper on it.
https://pdfs.semanticscholar.org/5b2f/34d124d1e4982a56b6513232c9fa9406a943.pdf [semanticscholar.org]
(Score: 4, Interesting) by anubi on Sunday June 17 2018, @09:09AM (9 children)
This is interesting study of atmospheric drag.
I know good and well how things burn up coming in from orbital space to the surface. And that's with a constantly decreasing total energy due to atmospheric drag.
We will have even more drag going back up, as we hit the densest part first, before slowing down due to atmospheric drag.
Our escape velocity is about 25,000 miles per hour? How long will any object hurled at even 3,000 mph stay at that speed, given the drag of our pretty dense atmosphere?
Tempted to pull out my old FORTRAN golf-ball trajectory program from university, and change its variables to a 3000 mph drive, and see how high it goes. Well, theoretically, I did demonstrate a 100 yard putt with it. Goes to show a foible of computer simulations, eh... oughta bring a chuckle to any golfer. How hard this theoretical ball had to be hit and what it would have to be made of to stand the stresses of being hit can only be stated in scientific notation. And would only work if the elevation of the putt was higher than the elevation of the hole. I never said it would go into the hole... just make it over the hole.
If anything, a rail gun built into a mountain? Colorado? Mt. Everest?
3000 miles per hour is roughly 4400 feet per second.... [calculateme.com] which is around the speed of small arms fire [wikipedia.org]. So its kinda like firing your hunting rifle into the air during fourth of July celebrations next month...
"Prove all things; hold fast that which is good." [KJV: I Thessalonians 5:21]
(Score: 2, Interesting) by Anonymous Coward on Sunday June 17 2018, @11:55AM (7 children)
Further considerations, what kind of release mechanism has the required strength, accuracy and repeatablity to make this work? 5000 mph is ~2200 m/sec which is the release/launch velocity. Make a WAG at a 50 meter radius centrifuge.
Mechanism strength: Say the rocket mass is 10 kg. Centripetal acceleration is v^2/R or 2200^2/50 = 96800 m/sec^2 or about 10000 g. The release mechanism has to hold on to a mass that appears to be 100 000 kg.
Say that the release has to be within .1 degree (0.00175 radians) of the desired point (1/3600 of the circular path). This means that the aiming error at 100km (100 000 meters or the "edge of space") is about 175 meters which seems acceptable(?)
Circumference is 2PiR or 314 meters so the release is within 314/3600 = 0.087 meters or 8.7cm
1/2200 m/sec = .00045 seconds/meter. To release in 0.087 meters x 0.00045 means the release has to be within 0.000039 seconds. 39 microseconds isn't any problem for computer clock speeds, but it is damn difficult to imagine a mechanism that is strong enough to restrain the rocket and release that precisely. One thought would be a strong brittle material that fractures almost instantly when the surface is scratched -- perhaps there are high strength steels that work this way? Tempered glass might meet the timing requirement, but it's not very strong compared to metals.
Did I make reasonable assumptions? Did I run the numbers correctly?
(Score: 2, Interesting) by anubi on Sunday June 17 2018, @12:11PM (2 children)
Got a good point there... its tricky enough making a good release mechanism for a trebuchet.. which this technique reminds me of... slinging and releasing a mass.
Superconducting electromagnetic?
"Prove all things; hold fast that which is good." [KJV: I Thessalonians 5:21]
(Score: 0) by Anonymous Coward on Sunday June 17 2018, @12:46PM
What is a typical release speed for a trebuchet? How accurate can they be made (or, what is an achievable window for the release time)?
(Score: 2) by PiMuNu on Monday June 18 2018, @09:47AM
Superconductor is incorrect material. You need a fast ramping magnet; superconductors are typically not great for this as they have lots of stored energy which it is hard to dump. Typically, the superconductor is on a copper substrate which takes the stored energy during a quench event (sudden loss of cryogens), but this process takes as long as a few seconds. Most superconducting magnets have heaters to speed up the quench process in the event of an unplanned quench.
Fast ramping magnets are typically normal conducting. Try google search for "kicker" magnet. Pulse lengths of microseconds and field strengths up to a Tesla are possible. The total volume of the magnet is important,
Having said all that a release mechanism using a normal conducting pulsed magnet would seem quite practical. 10s of tonnes of force are readily accessible. I guess on rockets they use explosive charges (and that probably works, it has been used since the 50s).
(Score: 4, Interesting) by Immerman on Sunday June 17 2018, @02:06PM (2 children)
I would imagine they'd go for a much larger radius - a centrifuge arm could be as simple as a carbon fiber cable, and every doubling in radius halves the acceleration and thus the required strength and cross-sectional area (and price of materials per foot) - so basically your centrifuge arm could cost very roughly the same amount regardless of length. And the longer the sling, the lower the acceleration for the payload - quite important if you want to avoid liquefying it in the process. I doubt there's many rocket control electronics that could survive 10,000g. 100g maybe, so call it a 5km radius centrifuge? That might be viable. Something a little more modest like 1km probably wouldn't be too difficult to make happen. I mean really, what do you need? A flat plain, a long rope, an elevated central spinning motor, and a round dirt track and cart for the first little bit of acceleration before centripetal force pulls it off the ground. After that you just need very limited control surfaces on your rocket could keep things stabilized.
Of course a rope can't directly apply shear forces, so you'd still need at least a bit of an arm to pull it off true - unless it's only there for the centripetal acceleration - rotation might actually be provided by jet engines or similar near the rocket.
I don't know that accuracy is all that important - sounds like the primary purpose is to get projectiles out of the atmosphere - the part of the flight that consumes most of the fuel in a rocket launch, but imparts only a small fraction of the total orbital energy (as I recall, low orbit energy is about 10% altitude, 90% speed). So you could fling it in only very roughly the right direction (say +/- 30 degrees), and let the rocket worry about the course corrections on the fly.
As for a release mechanism - with much larger tolerances you don't need the precision. Perhaps something as simple as a carbon-fiber cable attached to the rocket with explosive bolts? Then for launch angle - at the easiest, you could leave that to the rocket's control surfaces. More complicated you could incline the entire centrifuge, or have a ramp outside the circular track, and extend the arm slightly just before launch (or alternately, use a circular track with a fold-down section that releases the rocket onto the launching ramp when ready. You'd probably need maglev track at those speeds though, and that'd be some insane levitation force, and no doubt a price to match.
Of course, on release you've got the counterweight issue as well - suddenly your centrifuge becomes VERY imbalanced, at very high speed. Could get ugly. One possibility is to make it symmetric - your counterweight is a second rocket with the same mass, and both get released simultaneously in opposite directions. If your goal is just to get out of the atmosphere, and you assume minimal remaining velocity after that, that might not even be much of a problem.
(Score: 3, Interesting) by PiMuNu on Sunday June 17 2018, @02:55PM (1 child)
It is not too hard to make an electromagnet with these sorts of holding forces. One can make a pulsed magnet with pulse length of few 100 ns and ~ Tesla field strength with a modicum of expertise.
(Score: 2) by Immerman on Sunday June 17 2018, @04:41PM
Really? Wow.
Wait "holding force" is the force required to pry a magnet off a flat steel plate with complete contact - how much of that force can really be translated to a maglev support force?
(Score: 2) by choose another one on Sunday June 17 2018, @03:44PM
> Did I make reasonable assumptions? Did I run the numbers correctly?
Suspect you are in the right ballpark - in particular the g force is in the range I've seen quoted elsewhere for this type of system.
I reckon your aiming window is probably too small though - even at hypersonic speeds variable atmospheric conditions will probably give you more error than that over 100km, and if you assume a second stage rocket you can make adjustments with that burn.
Holding, and releasing, 100 tonnes is not a massive problem - a big jet engine is half that in thrust, or take a look at the Saturn 5 hold down systems or the pyrotechnic nuts that held the shuttle on the pad (against 1million pounds or so of SSME thrust - they wouldn't hold once the SRBs lit, but weren't designed to).
Releasing with microsecond timing might be, but I suspect (see above on aiming window) it might only need millisecond, and pyros (explosive bolts etc.) can do that, I think. You might even be able to get away with last century mostly mechanical stuff, I seem to remember (although this is memory from decades ago) that with ejector seats once the initial squib went off everything was timed and triggered, sometimes mechanically, by how far up the rail the seat was, and those sequences were timed to the millisecond [you'd better hope they are accurate or you'll be punching a hole in the canopy with your head (or similar) which doesn't usually end well].
The bigger problem is the g force, which I am pretty sure is in the right ballpark. Designing payload for 10g (which is some margin over eg. Falcon 9 max) is going to be an awful lot different to designing for 10,000g. There is a reason centrifuges are used to separate payloads into component parts. While I am sure it _could_ be done, I am not at all sure it could be done without increasing payload mass (and/or cost), a lot. Payload mass is of course what you are paying to launch. Not real sure of the benefits of designing a cheap (per/kg) launch system that requires payloads to be much heavier / more expensive.
(Score: 4, Interesting) by richtopia on Sunday June 17 2018, @04:44PM
While I haven't heard of SpinLaunch before, space cannons have had received a good deal of research. The most feasible idea currently is to suspend the launch cannon in water to assist with supporting the structure. Escape velocity is very difficult on earth, so the cannon is supposed to be a first stage followed by a chemical rocket to enter orbit.
Surprisingly enough, this is not designed for people. Or anything with structure really. But the estimated costs are much cheaper than chemical rockets, so liquid supplies (fuel, O2) could be put in orbit and dock with a traditionally launched rocket.
https://en.wikipedia.org/wiki/Space_gun [wikipedia.org]
(Score: 1, Funny) by Anonymous Coward on Sunday June 17 2018, @10:55AM
"Catapult, you say? 5,000 miles per hour, you say? Sir, I would like to launch this modest, yet hardy, boomerang ..."
(Score: 2) by choose another one on Sunday June 17 2018, @12:00PM (33 children)
Not sure how they are going to handle range safety. Seems a little obvious that in a centrifuge system "down range" is basically all directions.
Almost all large rocketry stuff is done with "down range" as large areas of water (ocean), for safety reasons, the only obvious way for doing that would be to launch in the middle of the ocean. That is feasible for conventional rockets, not sure if you can get an anchor that would take the strain for a massive (presumably unbalanced) centrifuge. Beyond that, where are you going to get the energy from in the middle of the ocean - you'd need a lot, it won't be on tap out there so it needs to be transportable and have a means of converting it into motion fast. Sort of like rocket fuel and a rocket?
(Score: 1) by anubi on Sunday June 17 2018, @12:13PM (30 children)
Weebils wobble but they don't fall down?
Its gonna be fun trying to counterbalance this thing... do you balance it before or after its shot its load?
"Prove all things; hold fast that which is good." [KJV: I Thessalonians 5:21]
(Score: 2) by Immerman on Sunday June 17 2018, @02:11PM (29 children)
Launch symmetrically - your counterweight could be a second rocket, released simultaneously in the opposite direction. After all, the goal is just to get out of the atmosphere - 5,000 mph = 2.2km/s. Not even 30% of orbital speed, and you will have lost most of that to air resistance before exiting the atmosphere.
Easiest of course if the rockets are launched horizontally, and are responsible for guiding their own path upwards with control surfaces.
(Score: 2) by HiThere on Sunday June 17 2018, @06:35PM (28 children)
If you're going to launch horizontally, then you'll need to punch through a lot more atmosphere than if you launch vertically, but construction should be a lot easier. And it does allow symmetric launches which, as you point out, is a major advantage. If you're depending on the payload to make a right angle turn using atmospheric planing, they you'll lose a lot of the energy in that turn. It might be better to just bore straight ahead, and let the curve of the Earth move the atmosphere away.
Do note that this is single stage to space, not single stage to orbit. When you get out there you won't have orbital velocity, so that would need to be handled separately, if desired. If you launch horizontally, either you are launching into a polar orbit, or one of the payloads will be moving against the rotation of the Earth. So they'll have a lot different velocity in an orbital direction when they reach space (even though neither will have orbital velocity).
I wonder how hard it would be to build it with a horizontal axis and launch vertically, and then hold it together long enough to launch the second payload? That would allow launching through a range of vertical angles. Ideally you'd be able to rotate the platform it sits on an launch at any angle above the horizon. (But I'll sure bet the first model doesn't have THAT kind of refinement.)
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by Immerman on Sunday June 17 2018, @11:32PM (27 children)
>If you're going to launch horizontally, then you'll need to punch through a lot more atmosphere
You missed the part where the rockets
>are responsible for guiding their own path upwards with control surfaces.
They launch horizontally, but immediately start turning upwards. Of course you do waste quite a bit of the initial speed in the turn that way.
I'm not sure there's anything to be gained with an adjustable tilt - your final target is orbit after all, and even if you somehow kept the full ~2.2km/s from launch you still need to more than triple that speed to establish a low orbit - plenty of wiggle room to adjust for the target orbit. I would think you'd want to just optimize the tilt for the most efficient suborbital launch possible.
Now, if you were using a (presumably maglev) track rather than a centrifuge arm you wouldn't have the same imbalance issue, as the track would already have to be designed to withstand the extreme stresses on its own, and counterweights wouldn't be relevant. Of course you'd probably need either a really large slope or a large underground ring track for that.
(Score: 2) by HiThere on Monday June 18 2018, @05:37PM (26 children)
"Adjustable tilt"?
I think you've got the wrong image of what I was talking about. Think of a vertical loop on a rotor...and remember that not all orbits are parallel to the Earths rotation. So there could be advantages to launching at an angle of, say, 47 degrees up, 23 degrees South. (That wasn't a serious proposal. First they need to prove the launcher works at all.)
Also remember that any adjustment that the cargo has to make will cause either increased package weight (for fuel, engine, etc.) or entail loss of velocity because of air friction, airfoil machinery, etc. So the less of that is needed the better.
Also remember that this thing is not going to achieve orbital velocity. It may send things 50 miles up, or even more, but their horizontal velocity will be based around the the velocity of the rotation of the surface of Earth at the latitude of launch. Plus any non-vertical component contributed during the launch. At the speed required for launch you're going to want as much streamlining as possible. And a heat shield. Wings are extremely dubious. IF you assume that there is a rocket first stage, then you can use exhaust flaps to steer the package somewhat, but if you do that you're significantly increasing the mass of the launch package.
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by Immerman on Tuesday June 19 2018, @02:18PM (25 children)
You're right - I must have fixated on the "range of angles" and missed the horizontal axis. You're thinking of something gimballed rather like a giant high-speed Ferris Wheel I presume?
Of course in addition to the difficulty, a horizontal axis means you can only launch in one of two directions, though at any angle. I suspect that's a bigger issue than only one launch inclination. After all, you have to impart enough additional momentum to reach a nice circular low orbit before you can go beyond it, otherwise you'll end up crashing back into the Earth on the next pass. And as you say, not all desirable orbits are around the equator.
Regardless though - the hard part is getting to low orbit in the first place - once you're there then there are many far more efficient ways to adjust your orbit (ion drive, geo-magnetic drive, momentum wheel "elevators", etc), it's getting (and staying) outside the atmosphere that's horribly inefficient. If we're in a position where shipping bulk cargo to orbit makes sense, then I think it's reasonable to assume the existence of orbital "tugboats" to get it where it needs to go from there. And with an acceleration of 1000s of g's, bulk supplies are going to be the rule - even building a second-stage rocket that can survive the first stage launch is going to be an engineering nightmare, you're not going to be launching many satellites, and definitely not people.
(Score: 2) by HiThere on Tuesday June 19 2018, @05:41PM (24 children)
You *can't* get into orbit with this kind of launch. You can get into orbital height, but your velocity components will be all vertical. This helps a lot, but if you want an orbit you're going to need to add that velocity by machinery included in the payload weight.
Back as far as the 1950's there was a souped up weather rocket that could reach space...but all it's velocity was vertical, so it had no hope of achieving orbit. The additional weight requirements are what changed things from a weather rocket on top of a V2 (2 stage) to the three stage mass used to launch the echo balloons. Just reaching high enough is only about half the problem.
This has a lot of implications for the requirements of the Spintronics device. It's going to need to launch something a lot heftier than a cubesat if it wants the thing to go into orbit, even if all that's put into orbit is the cubesat. What it launches is going to need to be a rocket with a cubesat payload that has enough delta-V to achieve about 20,000 mph (depending on just how high it gets).
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by Immerman on Tuesday June 19 2018, @06:16PM (23 children)
Right, you absolutely need a second stage to get from sub-orbital pseudo-parabolic trajectory to orbit - as is mentioned in the summary. Even if you could impart orbital speed, your orbit would still intersect the Earth and make for a bad day all around. You don't really need to get beyond low orbit yourself though - no reason you can't leave that to slower more efficient "tugboat" rockets that live in orbit.
Why are you assuming a vertical launch though? That's going to be far more difficult to deliver, and what would be the point? Especially at 5000mph - you'd launch clear of the atmosphere in about a minute, and then fall straight back down when you ran out of kinetic energy. The one thing you can count on, regardless of trajectory, is that at the top of your arc your vertical velocity will be zero. Your *horizontal* velocity however will be whatever you've given it at launch, minus losses to air resistance during the 60-mile climb out of the atmosphere. Which is presumably why they're launching so fast: 5000 mph, plus the ~1000mph of the Earth's surface is over 1/3rd of the delta-V required to reach low orbit - but only horizontal speed really helps with that, vertical speed just buys you more time above the atmosphere.
(Score: 2) by HiThere on Wednesday June 20 2018, @01:22AM (22 children)
By the time you get to orbital height, whichever direction you start from, your velocity will be vertical. The "space tug" is a reasonable approach, however. But first you need to get the space tug up there, which won't be launched by Spintronics.
Actually, there are lots of momentum transfer devices that could work. The space tug is one, but if they can aim well enough, I'd prefer a pinwheel. The problem is that a pinwheel needs to (over time) lower as much mass as it raises. But it's more flexible in the destination of it's cargo than is the space tug. The space tug is cheaper to set up, but it seems to inherently require exhaust mass, where a pinwheel doesn't. (Think of a multi-armed bolo with a heavy weight at the center spinning rapidly, and a grabber at the end of each arm, that can catch and release. Docking could be a bit tricky, though. But the release could set you on a path to nearly anywhere without loss of momentum.)
For that matter, the arms of a pinwheel could dip into the upper atmosphere. The only problem is avoiding friction, so you need the rotation to decrease the relative velocity between the arm and the air...but it's so thin up there that you wouldn't get much friction anyway.
That said, neither the space tug nor the pinwheel will be available in the near future. If Spintronics is going to depend on a space tug to solve their orbital problem, they're probably a decade too early.
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by Immerman on Wednesday June 20 2018, @04:33AM (21 children)
Why do you think the velocity would be vertical? The *only* launch angle that would result in a vertical trajectory at any time would would be a vertical launch. Just as a tossed a baseball will only ever go vertical if thrown straight upward. At the extreme far end picture a horizontal launch - it'd lose a lot more speed to air resistance, but when it left the atmosphere it'd be hundreds of miles away from the starting point horizontally, but still thousands of miles away from the vertical line paralleling it's path while passing through the center of the Earth.
Certainly any space tug would (probably) have to get up there some other way - this would be suitable only for extremely durable or non-solid materials, and extremely rugged equipment. I'm partial to pinwheels as well - but that's another, potentially even larger project.
And hey, if you have a pinwheel dipping tethers into the atmosphere it could match speeds and effectively roll across the upper atmosphere while the tethers move relatively slowly and almost entirely vertically while within the upper atmosphere. Don't even need a suborbital launch then, just pluck the cargo off a massive high-altitude airship. And if you rode any such a pinwheel to its apex before letting go, you'd be launched completely free of Earth.
(Score: 2) by HiThere on Wednesday June 20 2018, @05:02AM (20 children)
I suppose vertical is oversimplified, but it seems to me that the lowest it could be would be about tangent to the surface of the Earth. True gravity will bend it a bit, but it had better be going fast enough to get through the lower atmosphere quite quickly, so there won't be much bend. There probably *is* a speed that would put it into orbit in certain directions, but that speed would vary critically with very small variations in the weather, so you can't figure on that. You need to be going fast enough to get aloft in mildly adverse conditions.
Additionally, since all the impetus is applied to the payload at the initial launch, it needs to be going extremely fast at ground level, when gravity would be strongest, and have the greatest leverage. So I may be wrong about there being even one exact speed that would put in into orbit. It certainly wouldn't be just a quadratic equation, but something much more complicated, and probably dependent on variables that could not be measured at launch time (and certainly not before).
So, yes, vertical is oversimplified. But it's the best simple approximation. (The lower the angle, the more atmospheric variations will distort the trajectory, also the faster it needs to launch, as there is more atmosphere to get through.)
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by Immerman on Wednesday June 20 2018, @02:10PM (19 children)
You'll certainly want a strong vertical component - that gets you out of the incredible braking forces of the atmosphere faster - you lose speed fast when the compression wave in front of you is turning the air into plasma. It's just that vertical alone doesn't buy you nearly as much - thanks to the exponential nature of the rocket equation every little bit of horizontal speed takes a bite out of the most expensive bit of rocket propulsion needed to reach orbit. And yes, it's trajectory may be all over the place, but you still need your rocket to accelerate itself from 5000mph (minus large atmospheric losses) to 17,500mph (low earth orbit speed) - making any needed course corrections during that acceleration should be trivial.
Even if you could launch fast enough to reach orbit without vaporizing your payload - that orbit will still intersect the surface of the Earth (your launcher, if the Earth weren't rotating) so a second stage will always be vital. There's a wide range of such orbits though, the faster you fire, the larger the (increasingly elliptical) orbit, until you manage to leave the atmosphere will full escape velocity (11.2km/s = 25,000mph), at which point you'll leave Earth space entirely and orbit the sun instead. But all of those Earth orbits still include the launch point in their path, which means a crash landing at the end of their first revolution. However, if you were on such an orbit it would only take a relatively small burst of propulsion at the highest point to raise the lowest end of the orbit clear of Earth. So it's well worth considering as a long term goal - but the current proposal is to launch at only a fraction of orbital speed.
For "direct to orbit" launches it's probably better to consider one of the many options that leave the atmosphere (vacuum tunnels included) *before* accelerating to speeds that cause rocks to vaporize in air. But they're considerably more expensive, and any option to launch fuel and other bulk materials into space more cheaply than via rocket is going to be valuable while we're first developing orbital infrastructure. Plus, cheap surface launchers will be far more valuable on the moon and asteroids, where there's no air to worry about and they could potentially launch resources all the way to their destination.
Oh, and you're not going to see much variation of gravity during a launch - the Earth's radius is 6,371km, reasonably stable low Earth orbit is maybe 200km above that - so you're only going 3.1% further away from the center of mass, and still experiencing 94% of surface gravity.
(Score: 2) by HiThere on Wednesday June 20 2018, @05:47PM (18 children)
I agree with your points, but the Spintronics launcher is a kind of catapult, so all the velocity has to be imparted to the payload before the launch.
OTOH, if you could reach orbit (??) I'm not sure where the orbit would be at perigee WRT the surface of the Earth, due to modification of the trajectory by atmospheric friction. It might actually end up a distance above the surface of the Earth.
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by HiThere on Wednesday June 20 2018, @05:49PM
I just noticed I've been getting the name of the company wrong. It's SpinLaunch Inc., not Spintronics.
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by Immerman on Thursday June 21 2018, @02:10AM (16 children)
The Spinlaunch "catapult" is only designed as the first stage launch though - it is designed to launch (incredibly rugged) rockets, which then ignite once outside the atmosphere. Neatly slicing off the ugliest end of the rocket equation. Picture a SpaceX Falcon 9: huge first stage with a much smaller second stage, and a tiny payload. That huge first stage only gets it out of the atmosphere with a bit of a running start - the much smaller second stage actually does the majority of the work of reaching orbital speed and altitude.
As for a catapult-only launch - yes, if you assume some active (or accidental) air control you could indeed end up on an orbit that avoids the Earth's surface - however, your last atmospheric trajectory control happens while still in the atmosphere, and your orbit will re-enter the atmosphere with exactly the opposite angle of attack. Which will slow you down and lower your orbit deeper into the atmosphere. Play things right and you may be able to "skip" off the atmosphere for several orbits, but it generally only takes a very few orbits before you drop below orbital speed and begin final reentry. The maneuver is sometimes used for extended slowing prior to re-entry, for example to reduce a probes highly elliptical interplanetary Martian capture orbit into something slower and rounder before deploying the lander (I think there might be rocket firings involved in that as well though - a little boost each apogee to pull perigee far enough out of the atmosphere that the next dive isn't the last - with apogee dropping substantially on each pass.
To achieve a stable orbit you must accelerate while outside of the atmosphere, which pretty much means some kind of rocket. With a fast enough launch it doesn't necessarily have to be that big, just enough to give a burn at apogee that lifts perigree far enough out of the atmosphere that it can stay in orbit. The faster it leaves the atmosphere the higher the apogee, and the easier it is to raise perigee.
(Score: 2) by HiThere on Thursday June 21 2018, @05:31PM (15 children)
The problem is, if it's only the first stage, then the payload needs to be a LOT heavier. And even for just putting up a b-b the momentum at release is significant. That weight is going to need to be trimmed in every way possible. Which is why it makes a lot more sense if there's some momentum transfer device already in orbit.
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by Immerman on Friday June 22 2018, @02:02AM (14 children)
Certainly - however firing a projectile from ground level with enough speed to reach orbit (i.e. just needs enough rocket to pull perigee up out of the atmosphere) brings it's own massive costs. Even assuming no speed loss to the atmosphere, you'd need to launch 3.5 times faster than the piddling 5000mph they're proposing. That means 12.25 times greater acceleration for the same size centrifuge, or 12.5x larger centrifuge for the same acceleration. And that's going to create a far more intense shockwave in front of the projectile, and a far hotter plasma as a result: think orbital reentry, except you're starting in the densest part of the atmosphere instead of the thinnest.
And of course in reality that fireball dissipates energy (aka speed) VERY quickly, so you're going to have to fire the projectile even faster, which means the fireball will be even hotter and dissipate speed even faster... Which is why all the "direct to orbit" plans involve a either a long vacuum tunnel reaching out of the atmosphere, or even suspending the launching apparatus itself above the atmosphere
As it is though, you've eliminated the far more expensive and less effective "half" of the rocket in the first-stage, in exchange for having to make a far more rugged second stage to survive those 1000's of gs of lateral stress. However, its just occurred to me that the second stage could be an incredibly simple and cheap solid rocket booster with the most rudimentary of control systems. After all, if you're just boosting inert cargo to be collected in orbit, it just has to accelerate in a "straight" line up to orbital velocity once it clears the atmosphere, so that a "tugboat" can come collect it. It doesn't have to be able to carry out a complex mission involving multiple rocket burns, it just has to avoid hitting the ground.
In the future you've also got the possibility of tumbling skyhook "pinwheels" and other far more efficient "second stages" on Earth. Not to mention a catapult is a WONDERFUL solution for the moon (and even Mars) where atmospheric braking is a comparative non-issue.
(Score: 2) by HiThere on Friday June 22 2018, @02:41AM (13 children)
Some form of catapult would certainly be desirable for the Moon, and possibly Mars. I'm not sure about a centrifuge, though. A mag-lev linear accelerator seems a much better model. (You could run it up the side of Mount Olympus.)
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by Immerman on Friday June 22 2018, @04:27AM (12 children)
A centrifuge doesn't need to be an arm-and-capsule model. How would the cost compare if, instead of a long linear mag-lev accelerator, you used a small circular mag-lev track whose circumfrence was a small fraction of the linear accelerator's length? If it takes 100 revolutions to reach full speed, then you've just reduced your required track length by a hundredfold. And mag-lev track isn't exactly cheap.
Of course the mag-lev forces support would be dramatically greater due to the centripetal acceleration - but that's where the arm can come back in, especially in a vacuum. All you need is a strong, passive tether to a central anchor to support the centripetal "weight" of the load, and you're free to use "mag-drive" strictly for stabilization and thrust.
Of course the down side is that the extreme centripetal acceleration rules out passengers and fragile cargo - but even a linear "catapult" is likely to be a rough ride.
So I got curious and ran the numbers (below) and it would seem that regardless of desired acceleration and muzzle velocity, a circular accelerator requires ~12.5x longer track than a linear accelerator. I'm kind of surprised by that result. But, if your cargo can handle 12.5x the acceleration your mag-lev system can deliver, then you break even. Anything more than that reduces track length. And of course on a circular track the linear acceleration can be far more leisurly, so you can also use a far less powerful mag-drive system and reduce costs even further.
Track length required for a given muzzle velocity "v" and maximum acceleration "a"
With a circular track, linear acceleration can be made minimal, and so only centripetal acceleration is relevant
maximum centripetal acceleration a = v^2/r
so r = v^2/a, and track circumference
c = 2pi*v^2/a
With a linear track, the track length required under constant acceleration is d = (1/2)*a*t^2
and t = v/a, so track length
d = (1/2) v^2/a
c/d = 2*pi / (1/2)
(Score: 2) by HiThere on Friday June 22 2018, @05:19PM (11 children)
Thanks for the figuring, and clearly a circular track (I was thinking of a particle accelerator as a model rather than using an arm) has the potential to allow a greater diversity of exit points.
Perhaps the particle accelerator analogy should be taken further. Some of them have a circular section feeding into a linear accelerator, so what you do is run around a circular track many times building up speed until at the final stage you feed it into a linear accelerator. You wouldn't need to worry about "bremsstrahlung", so the acceleration in the circular section could be as slow as you fet desirable. The question here is how many G's can your maglev track take, as centrifugal force is not dependent on powered acceleration. Then you feed it down the linear accelerator to adjust the final speed and direction. But this all needs to happen essentially in a vacuum. Still, doesn't hyperloop depend on a vacuum too? This would be smaller than that.
But I still don't trust firing the final payload through a thick atmosphere. It seems like anything that would reach orbital height would burn up on the way. So you'd need a tough ablative packaging, and you'd need to include an entire rocket.
Still, say you had a circular track 5 miles in diameter, you could build up quite a hefty speed without imposing excessive G forces on the payload. Say you did it just outside of, say, Pikes Peak, you could have the linear segment run up the side giving you structural support for your track, and adjusting the angle of travel. But Mons Olympus would be better with both more height and less atmosphere to traverse. On Earth Everest or K2 would be better, as they give you longer tracks that get you higher above more atmosphere. But even at 5 miles above sea level you've got a lot of friction to traverse.
All of these devices seem to work better in lower gravities, which is no big surprise, as so do rockets.
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by Immerman on Friday June 22 2018, @06:48PM (10 children)
Glad you appreciate the calculations.
That's why you use the cable to provide the many Gs of centripetal acceleration - then your accelerator track can ignore that component entirely. Of course that becomes a much bigger problem on Earth - a single 5 mile diameter vacuum chamber is a much bigger challenge than a vacuum tunnel ring of the same diameter.
I didn't have much luck finding any reference to particle accelerators with linear and ring components, unless you count the (relatively) low-speed "injector" that essentially just gets things moving fast enough for the ring to take over. There's sometimes a linear "exit tunnel", but I think at that point it's just performing beam focusing/aiming, not actual acceleration. Perhaps you're thinking of those? Certainly an aiming "barrel" of some sort would be a valuable thing to have for much more precise orbital entry.
Given the relative inefficiency of ring accelerators at a given acceleration limit, I would think that if acceleration is a concern, you'd go full linear. I mean if it takes 12.5x the amount of track for a ring to reach the same speed, even a ring-style "pre-accelerator" doesn't seem to make much sense.
I'm unconvinced about the "gentleness" of a useful ring accelerator though - assuming a 5 mile diameter (~4,000m radius), and a max acceleration of 4g (I believe that's regarded as roughly the max a typical person can reliably survive for any duration), you'd get a top speed of v = (a*r)^0.5 = (40m/s^2*4000m)^.5 = 400m/s = 895mph. That might get you out of the atmosphere briefly with good aerodynamics (straight up in a vacuum it'd get you to ~250km, using 1/2*mv^2 = mgh) , but it's not going to help much with the 9400m/s of delta-v needed to reach LEO from the surface, you're going to need a much larger rocket to get up to speed. And that square root really sours things - to get 5.6-fold increase to Spinlaunch's proposed 5000mph launch speed, while maintaining a 4g max acceleration, would require a ring 155 miles in diameter. Not impossible, but rather significantly more expensive.
As for atmospheric heating... I tend to agree with you. But we do have some really effective thermal insulation, and even 5000 mph is much slower than the minimum 17,400mph of reentry speed. If they think they can pull it off, I'd love to see them try. I mean low orbit is only 60 miles away, at 5000mph you could be through the atmosphere in under a minute, and unlike during reentry you're trying to *avoid* slowing super-heating the surrounding air to slow down. Oh, and incidentally the heating primarily isn't due to air friction, but rather to the rapid compression of the air in front of the vessel.
(Score: 2) by HiThere on Friday June 22 2018, @07:44PM (9 children)
OK, I just pulled the 5 mile diameter out of the air. 10 or 20 would be fine. And I thought the SLAC had some cyclotron type accelerators feeding it. But of course, the entire reason for it being a linear accelerator was the Bremsstrahlung. In this case it would be to give you an "infinite track" to get up speed on.
Since both acceleration and radius are linear, doubling the radius results in half the acceleration at any particular velocity.
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by Immerman on Friday June 22 2018, @08:31PM (8 children)
The problem is, doubling the radius isn't going to be near enough. Let's say we quadruple it, to 20 miles~=32,000m, and we want to hit 5000mph=2235m/s.
a= v^2/r = (2235m/s)^2/32,000m = 155m/s^2 ~= 15.5g
That'll kill pretty much anyone in quick order - as I recall advanced fighter jet maneuvers are limited to ~10g for a few seconds because even wearing special special pressure suits that's all the young, healthy pilots can safely endure. And "safely" is perhaps a dubious term - my impression is that most any physical stress that causes blackouts tends to cause some permanent damage. Though if the maneuver takes you from being the hunted to the hunter then it's probably a fare tradeoff - being shot out of the sky is generally a far more immediate health hazard.
But hey, another 4-fold increase in radius would do the job - an 80 mile ring doesn't sound so bad, until you consider that it translates to 500 miles of mag-lev track in a vacuum tunnel, capable of supporting 4x the weight of your payload. I'd go with the 40-mile linear accelerator myself, assuming you could manage 4gs of linear acceleration for less than 12.5x the cost per mile.
... just realized that all of my calculations have been ignoring Earth's gravity. 4g of lateral acceleration +1g of gravitational acceleration = 4.123g total, ...so I guess it's not actually that big a deal.
(Score: 2) by HiThere on Friday June 22 2018, @11:32PM (7 children)
Yes, but there's lots of cargo that could be launched at 16G. My two problems are air friction burning things up and the fact that even when you get high enough you aren't in orbit. Plus a huge doubt that they could launch anything very massive, like, say, a second stage rocket attached to the payload.
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by Immerman on Saturday June 23 2018, @03:13AM (6 children)
Absolutely. But how does the amount of 16G cargo compare to the amount of 30G, or even 1000G cargo? Once you're lethal to passengers, then you need to balance the increased cost of a larger ring against the increased amount of cargo it allows you to launch. I wonder just how much sustained acceleration various typical devices could handle - I imagine a hammer could handle considerably more, but how about a laptop? I mean just getting banged around in normal use can easily impart dozens, even hundreds of Gs for a few milliseconds - but that's short-duration surface acceleration - internal flexibility can dramatically reduce how much of that acceleration is applied to more fragile components. No such reprieve for sustained acceleration.
(Score: 2) by HiThere on Saturday June 23 2018, @05:42PM (5 children)
There's also the weight that the system can handle. If the payload is only stressing the system at 16G, then it can handle more massive payloads than if it's stressing the system at 32G or more. And if you need to include an upper stage rocket, that seems to me to be quite important.
Actually, it's so important that your calculations make me a lot more dubious about the practicality of the entire thing (and I was already pretty dubious), but I had guessed the centripetal load at a lot less.
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by Immerman on Saturday June 23 2018, @09:25PM (4 children)
Well, we've got a lot of folks working on quite small orbital rockets, and it seems like a 2-4% payload fraction is pretty standard across a wide range of rocket sizes, so I don't think it actually makes a whole lot of difference just how big it is, except to the amount of payload you can launch at once. If it makes sense to do at at one size, it probably makes sense across a large range of sizes - and the smaller you go, the lower the structural demands there are on the launch system - allowing you to either launch at higher speeds, or shrink your structure and use higher accelerations.
Replace the first stage with a catapult, which judging by the Falcon 9 is about 85% of the total mass, and you've gone from ~3% payload to around 20% payload, greatly shrinking the size of the rocket. The real question is - can you actually build and operate a spinlaunch system so that the amortized cost is less than using a reusable first stage booster?
(Score: 2) by HiThere on Sunday June 24 2018, @12:16AM (3 children)
There's no way you're going to be able to build a spin based catapult able to launch anything even approaching what a Falcon can launch. For the first models it's my guess that their target payload is in the cubesat range. The problem is, if you need to include friction shields and a second stage even a cubesat starts getting pretty heavy. Also lighter payloads are lose more momentum to atmospheric friction, but if you've already got a friction shield and a second stage rocket, I think you'll be beyond the point of inflection in *that* curve. It's still going to be bad enough, though.
One thing that *might* work is to just use the catapult to get the payload up to ramjet speed. That would let you cut down on the maximum stress that your catapult had to deal with, and give you a high speed takeoff assist. In that case, though, it's not exactly a launch system as we were thinking of it, since you still need the, call it the 1.5 stage, of the rocket. I don't know if that would be worth the while, but it might be interesting.
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by Immerman on Sunday June 24 2018, @01:32AM (2 children)
You absolutely won't compete with the F9 on capacity - but I really doubt they're trying to. A startup trying to compete directly against SpaceX is a fool's game - FH is the most powerful rocket we've seen in years, and the BFR promises to dwarf its capacity for a fraction of the launch cost. The high g-forces would likely destroy most satellites anyway. And a satellite launch would need a heavier fully mission capable second stage to get it to the right orbit.
So don't think satellites - think cargo. Fuel perhaps for starters. Food paste. raw materials of various kinds. Heck, vast quantities of rock or water for shielding. All the sort of stuff we're going to need to get to orbit in large quantities if we're really going to become a spacefaring species. For that stuff, it doesn't really matter how much you ship per load, all that matters is how much it costs per kg to deliver. If a spinlauncher could launch a small rocket every few minutes (or even hours) all day, every day, for only a tiny fraction of the first-stage cost, they could still get a massive amount of cargo into space reasonably quickly, even if the could only manage 100kg per launch. More would obviously be better, but the bigger things get, the harder it is to hold keep them together at high Gs - the old square-cube law at work. In the end all that really matters is how much it costs to get each kg of bulk material into orbit. Heck, the second stage boosters could even be reusable - they don't even need to survive re-entry, just save them up and send them back down en-mass on the next under-booked BFR return flight.
Also remember that what you're familiar with seeing are NOT a air friction shields - if it were the fragile space shuttle tiles would be destroyed on every flight. They're heat shields, to protect from the heat of the plasma fireball created by the bow wave. And reentry vehicles are designed to *maximize* that heating (within survivable limits), since that's what's slowing it down fast enough to not leave a smoking crater wherever it lands. A launch vehicle in contrast will be designed to minimize it. As such I would imagine that ablative shields would be out - as soon as they start vaporizing the surface becomes much rougher, increasing air friction and destabilizing the flight path.
(Score: 2) by HiThere on Sunday June 24 2018, @05:20PM (1 child)
Cargo is worse than satellites. It's heavier, and the amount you can launch is more important.
Your point about heat shields is pretty good, but supersonic airplanes have to do some fancy designing to keep from melting. And not only are their mass constraints are a lot less, they also aren't going as fast.
OTOH, people are starting to talk about hypersonic missiles, so maybe.
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by Immerman on Sunday June 24 2018, @08:51PM
Whats the smallest package of fuel, clean socks, or dehydrated potatoes that it's possible to launch? That's the lower size limit of of your cargo. There'a a wide range of cargo that falls into the realm of "we want to move X amount of stuff in Y amount of time, but don't really care how big the boxes are." That's your target market, especially for an entirely new kind of launch system that should excel at small launches at extremely rapid intervals.
(Score: 0) by Anonymous Coward on Sunday June 17 2018, @01:06PM
I was imagining a relatively vertical axis for the centrifuge, with the tilt of the "disk" equal to the aiming elevation. This means that the elevation angle can be set precisely in advance. As I understand it (without doing any numbers), getting to orbital velocity is the tough part, so most of the velocity will be downrange(?) Then the release timing provides the azimuth.
Most of the centrifuge could be behind big earthworks, with a narrow slot that the launch goes through. Seems safe enough if the launch direction is over the ocean?
(Score: 0) by Anonymous Coward on Sunday June 17 2018, @02:41PM
A nuclear reactor? Lots of those on ships and submarines, and the power ratings for typical marine nuclear reactors are in the range of several hundred megawatts, probably plenty enough for this kind of application.
(Score: 2) by MichaelDavidCrawford on Sunday June 17 2018, @12:36PM (4 children)
I've had the idea for a while now to shoot myself out of a giant slingshot at Burning Man, then parachute safely back to Earth.
It happens that live-human-carrying gliders are sometimes launched this way.
A 50-foot roll of 3/16" ID and 3/8" OD of medical-grade Latex rubber tubing can be had for sixty bucks from a great many scientific supply shops.
If you ask real nice, Kent Elastomers - the company formerly known as "Kent Latex Products" - will be happy to extrude for you any arbitrary length of any arbitrary inside diameter and arbitrary outside diameter industrial-grade Latex tubing.
I rang them up once in the day to ask if there was a cheaper option than buying their tubes at full-retail price. Kent himself answered the phone and said that if I cut them a check for a hundred Samoleons, he'd give me a loosely-coiled cardboard box with one thousand total feet of random lengths of 3/16 X 3/8 tube.
Here's Why I Wanted So Much Rubber [youtu.be].
Rather more realistically, when I finally managed to produce the wherewithal to attend BM again, I'm going to bring some folding tables, lots of folding chairs, lots of Latex and a few 100-packs of 1/4" hardwood dowels.
And a bandsaw. Can't forget Grandpa Crawford's bandsaw!
And a generator. I'll have to buy that one myself.
Yes I Have No Bananas. [gofundme.com]
(Score: 1, Informative) by Anonymous Coward on Sunday June 17 2018, @12:57PM (2 children)
> ... bandsaw!
For 1/4" dowels, a good pruning shear will make a clean cut, no need for the noisy generator.
Try this type,
https://www.bigfrogsupply.com/products/felco-f10-left-handed-pruning-shears [bigfrogsupply.com]
The rotating handle costs more, but *greatly* reduces hand strain when using for an extended period (there is a right-handed version too...) Bought an F10 30 years ago, still works great with a few drops of oil and the occasional sharpening.
(Score: 2) by MichaelDavidCrawford on Sunday June 17 2018, @04:35PM (1 child)
asdfg hjkl;'
Yes I Have No Bananas. [gofundme.com]
(Score: 1, Touché) by Anonymous Coward on Sunday June 17 2018, @06:42PM
Send me the generator (I already have a nice Walker-Turner 16" band saw) and I'll release you from eternal debt.
(Score: 2) by Immerman on Sunday June 24 2018, @01:37AM
Sounds like fun, though you might want to double-check that the chute could open in time to do any good. A glider starts life as an airfoil, a parachute relies on an extended period of high-speed air resistance to unfurl it first. It'd be rather embarrassing to soar up into the sky, and them plummet back down to earth like a rock, with a rope of still-furled parachute falling across the playa around you.
(Score: 2) by requerdanos on Sunday June 17 2018, @08:54PM
This sounds like something that's a very interesting thought experiment that will end up not launching anything.
And, if that's the case, I hereby offer to think very hard and get nothing launched for half that price.
(Score: 0) by Anonymous Coward on Monday June 18 2018, @02:18AM
A Baltimore Trebuchet Club.