SpinLaunch completes first test flight with rocket-flinging launch system
space.com
space.com
Professor James Longuski and his students at Purdue University have done quite a bit of research on this idea over the years. They call it a tether sling. To keep the tip acceleration (v^2)/r low, you want a large r. Some papers:
https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=long...
Note 1: Professor Longuski was my PhD advisor, but I never did any research on tether slings myself.
Note 2: Others have also researched tether slings. The papers linked above give many citations to related research.
To picture a tether sling, imagine a lighthouse on the Moon with a giant arm sticking out of its side, and the arm can revolve around the lighthouse like a merry-go-round. The thing to be launched goes at the end of the arm. Just spin up the arm and let go!
When something is rotating as you said, their mass is away from the rotation point, you can increase the rotation speed by moving the mass closer to the rotation point.
When the skater spins with their hands open they gain rotation speed by just gather their hand around the body.
When the kids spin on the round game, they get closer to the center to gain speed and away to slow down.
Is this a thing? Does this actually translates to increased centripetal force? Why they didnt get more speed by using this method?
One of the major limitatinos of the SpinLaunch approach is the g-forces. As it is, objects launched by SpinLaunch experience a g-force on the order of 10,000G. For a constant velocity, this force goes up as the inverse of the radius. As a result, you want the mass to be as far away from the center as possible so that the g-force experienced is minimal.
The other problem is that this approach is not actually an efficient way of gaining speed. Conservation of energy is still a thing. The rotating mass experiences an apparent centrifugal force pushing it away from the center of rotation (the 10,000Gs that I mentioned above). In order to move move the mass closer to the center, you must counteract this apparent force. At that point, you are likely better off taking the energy you would spend pulling the mass towards the center and simply apply it directly towards increasing the rotational speed.
Actually it wouldn't speed up. The speed of the mass traveling on it's circular path around the center remains the same. The orbit just becomes smaller and thus the path becomes shorter. The mass now makes more rounds in the same time. It is thus spinning faster around the center but traveling at the same speed on its path around it.
If the object goes in a circle, the radial force always acts perpendicular to the direction of motion, and cannot change the linear velocity of the object. The centripetal force does not do any work, the energy of the rotating body remains fixed.
But there is a subtle difference when one does pull the object closer to the axis. First, it is easy to notice that one does expend work. This energy must go somewhere, and there is nowhere else for it to go except into the kinetic energy of the moving object.
But how exactly is this energy transferred? If you think about it, when the object is pulled in, it no longer goes in a circle, but follows a spiral. Its velocity is no longer strictly perpendicular to the direction of the force. If you integrate this seemingly small effect, this is precisely what makes pulling on the sting to increase the velocity of the object.
Ignoring the mechanism, a formal calculation, from conservation of angular momentum or conservation of energy would immediately tell how much the linear velocity of the object would increase when it is pulled closer to the axis.
I was quite surprised to discover how cheap it is to fling chunks of the moon into space. I mean... I'm not sure why you'd bother, but still, it interested me.
(My working: 2380 m/s escape velocity, 0.5 * 2380^2 energy in Joules required, convert to kWh, googled battery and solar panel costs, totally didn't consider efficiency anywhere)
It is easier to send something to low Earth orbit from the Moon than from Earth.
Raw material for space station construction.
You wouldn’t bash in a window at work if the office is too hot. It is super easy to see that that is a hard to undo intervention you might regret later. The climate of a whole planet is much more complex than that, much less well understood and we only have one. Be suspicious of any plan which doesn’t have “knobs” we can adjust as we learn more.
Also we now maybe have the tech to make moon trebuches to fling dirt to shade our planet. Do we have the tech to clean it up too if it blows back in our face? Will we always have the tech? Civilization is not a straight linear progression. Even if we could do the adjustment now who knows maybe thousands of years from now humanity won’t be able to do the same. So it is better if we design our interventions such that they are making the climate stable even without continous tweeking by a technologically advanced civilization. I don’t know how this would be possible without properly designed feedback loops. And I don’t know what feedback system a dust cloud could have.
Mass drivers are powerful weapons. Haven't you read the classic "The Moon is a Harsh Mistress"?
Hmm. The Moon Is a Harsh Mistress builds a compelling case to do so.
For sending civilian payloads into space, it has a very low chance to reach a price point that is competitive with reusable rockets. You either need an absolutely gargantuan centrifuge -> high capital costs - or you need to subject your payload to thousands of Gs. "The payload" is in this scenario must include 1 ton liquid fuel rocket for every 200Kg you want to put into space, or 1 ton solid for every 100Kg.
You have to ensure this "second stage" will never explode inside your centrifuge despite being subjected to orders of magnitude higher loads than the typical rocket. This is possible but very expensive, you will need huge structural mass to hold what would appear to be thousands of tons of propelant at 1G. So the mass fraction goes down, the costs of the rocket and centrifuge "spin" out of control, and you provide a paltry orbital capacity of a few tens of Kg and only for specifically engineered satellites that no existing manufacturer knows how to build. A commercial dead end.
For military applications on the other hand, it's close to perfect. You don't need orbital insertion, you can accelerate your payload silently without allowing for detection in the early phase of the attack, your projectile starts at essentially cruise speed and can only be detected by infrared emissions due to atmospheric heating for a few seconds until the launch ablative heatshields are ejected. It's a fantastic first strike weapon.
Starship can deliver a payload of 150 tons to Orbit, fully reusable.
SpinLaunch needs a non-reusable second state and can deliver at best a few 100kg.
So what do you think is cheaper, launch 1 Starship or 500+ SpinLaunch non-reusable rockets?
The HARP program in the 1960s experimented with gun launched Martlet series of rockets. They tested several designs using solid fuelled rockets, but the program was cancelled before the liquid fuelled version was ready for testing.
I will change my mind when they offer payload to orbit at a price no more than half of what SpaceX can do. Currently, they launch 51 Starlink satellites of 260Kg at a price point of $50 milion, or about $4000/Kg.
Any competitor looking to deploy a high-G constellation will need much better prices, less than $500K per 260Kg satellite, and Starship is on course to slash another zero from that.
They might well be able to compete based on differentiating factors other than cost, such as availability. Small sats riding on an F9 are generally on ride-share programs and can have to wait years for a slot. Unless you're commissioning a whole F9, you take what you can get. With this thing they can throw up a single small satellite whenever you like, and pretty much as many or as few as you like.
Price may well be comparable to reusable F9, after all this thing puts the "upper stage" to about the same altitude as an F9 second stage, and as with this the F9 second stage isn't reusable. Also the "first stage" of this thing is a catapult that's functionally infinitely reusable, whereas F9 first stages have only racked up a maximum of 10 launches so far. This catapult should theoretically manage thousands of launches.
Starship is a real game changer for sure, but we'll have to see how that pans out. It's still a long way from being a functioning service doing daily launches.
With this it's huge, cannot be hidden or moved and the orientation would be obvious from a spy satellite. Any serious enemy would just have systems like this under 24/7 observation and would launch retaliation the moment it was fired, or even looked like it was going to be fired at them.
Other applications perhaps.
I'm no expert, but I thought ICBM launches were detected using thermal IR sensors on satellites. There wouldn't be any for this launcher. There would still be a heat signature from the second stage, but it wouldn't need to be anywhere near the launcher itself.
As for launching, you are guaranteed to produce a thermal signature in the first few seconds of flight, since you hit the high density atmosphere at a 3-5 Km/s speed. You will leave a distinct, large and highly detectable infrared signature and the enemy will be alerted: https://youtu.be/JAczd3mt3X0?t=274
This is good spit balling take, but an bad top comment. I assume on HN if someone says something is for the military it has to be up-voted without thinking?
> specifically engineered satellites that no existing manufacturer knows how to build.
Your consumer phone could be launched fine as is. Most satellites are not different, this is a known problem with a know solution. There is nothing magical here. Electronics are fine.
> It's a fantastic first strike weapon.
No one wants this. North Korea doesn't want this. It means it will get bombed. They want second strike weapons. Obviously all locations this is at will be well know, to test it has to be fired. Rockets you can just put randomly under 200 tents. https://dailyuknews.com/us-news/china-is-spotted-building-a-...
> This is possible but very expensive, you will need huge structural mass to hold what would appear to be thousands of tons of propelant at 1G.
But you are politically happy to test nukes in this when civilians seem scared of propellant blowing up?
This has clear civilian potential. This has limited military potential outside of the crossover with civilian. The economics is questionable, but if it works it'll be a game changer. It's great people are working on this stuff.
Communication sat buses have critical mechanical components: solar panel assemblies, gyros, reaction wheels and momentum sinks, unfoldable antennas, propellant tanks and valves. Earth observation sats have optics so fragile that even applying 1 G perpendicular to the design launch vector will damage them beyond repair, so they can't be mated with the rocket in the horizontal state.
All these can be probably beefed up and redesigned to withstand thousands of Gs, but it's certainly not a solved problem, and there is very little incentive for satellite designers to attempt it.
> But you are politically happy to test nukes in this when civilians seem scared of propellant blowing up?
A mechanical destruction of the warhead will not trigger a nuclear explosion, nuclear weapons are designed for this, it will just contaminate the launch stand. You also need much less conventional fuel if any since you are aiming at suborbital, and this reduces the loads and the damage after an explosion. Military risk tolerance is high, if there is a 10% chance if internal detonation then each facility will get to launch 10 warheads on average, probably more than they will ever have the chance during a real nuclear exchange - considering spinup time.
An explosion inside the commercial orbital launcher will wipe out a large investment and might put the whole company in jeopardy, the energies are significantly higher while the acceptable probability of failure is much lower.
> The test projectile "goes as fast as the orbital system needs, which is many thousands of miles an hour," Yaney told CNBC
This is an evasive statement. The test projectile is nowhere near as fast as their planned orbital system. The challenges scale poorly. Acceleration (and resulting loads during the spin) scales with v^2. Then once you hit the atmosphere, drag and thermal flux increases with speed even faster.
--
Here's John Carmack (who ran his own rocket company once, Armadillo Aerospace) on Spinlaunch.
https://twitter.com/ID_AA_Carmack/status/1458870561606615046
Why so vague?
It's got a lower lever so the overall momentum might be low enough to make that recoverable. Maybe if it's 10% of the rocket's momentum, going into... yeah, it's hard to imagine that being recoverable but maybe it's just weights and that's good enough.
I'd love to see a real explanation.
EDIT: hmm, or a supplemental weight on the payload side, moving outward the right distance as the payload releases.
It seems like you'd generally want the non-payload rotating bits to vastly outmass the payload, so its release perturbs the whole system to the least amount possible. And then you use regenerative braking to reclaim the energy .
What I came up with was this: the launch arm has a second mass inside it, that will move (via centrifugal force) away from the center, in the same moment the payload is released.
Of course, just having two payloads launch in opposite direction as you propose maybe has much less implementation problems (though I wonder how you stop that second dummy payload in a way that does not cause an earthquake or explosion :)
> Although not shown in the previous figures, the circular mass accelerator structure 150 may comprise a second exit port directly opposite the exit port 115 to capture the counterweight 135 that is released simultaneously with the launch vehicle 105 to minimize an imbalance on the motor at the time of release. The counterweight 135 may be a solid material, or a liquid such as water.
The use of a liquid is a curious idea. Perhaps it could be dispersed in such a way as to spread the force of the counter weight being released across a wide surface area? Like a small explosive forces the liquid out in all directions?
According to this site [2] that's equivalent to around 200 kg of TNT. Even with the counterweight being mostly water, that's quite a lot of energy to disperse. How does one evenly spread out the water to a surface the size of a football field? Would that even be enough area to prevent a shock wave being reflected back at the launch equipment?
[1] https://calculator.academy/joule-calculator/#f1p1|f2p0
[2] https://www.convert-me.com/en/convert/energy/tntkg.html?u=tn...
* 5 metric tons TNT
* 1/3000 Little Boy atomic bomb
* 4 barrels of crude oil (!)
* Boil 2,300 gallons of water
* 0.25 mg of matter converted to energy
* Enough energy to melt two 11,000 kg iron counterweights, with 2 GJ left over
* A magnitude 3.7 earthquake [1]
[1] https://www.volcanodiscovery.com/earthquakes/energy.html
Doubtful even if pointy. Water is way denser than atmosphere
A school-level physics calculation is enough to debunk it in a minute.
Earth orbital velocity is roughly 8000 m/s. To achieve even a small fraction of it by spinning, the projectile and the device must withstand centrifugal acceleration:
a = V ^ 2 / r
Let's assume we obtain 1/8th of the orbital velocity, which will give a significant fuel reduction, thanks to fuel is exponential to delta V.
r of the full-scale should be 136 m (small-scale diameter is 91 m as in the article, scaled by 3 as said there too, divided by 2)
1000 ^ 2 / 136 = 7352 m/s^2 = 750 g.
I leave to the others to calculate how many RPMs should the device make. There's no material that can withstand such forces for extention. Many projects of energy conservation with flywheel were cancelled because sighnificantly heavy and large flywheels (couple of tons and just about 1 meter in radius) tear themselves apart at 2-3K RPM.
I've not heard of any devices handling 100g over any significantly long periods of time.
Artillery shells, even the fancy ones with guidance hardware, can handle such loads too, in compression.
You, "leave to the others to calculate how many RPMs should the device make"; like the people behind this company that clearly did that math, build a device and then proceeded to launch an object several tens of thousands of feet up. Clearly their math is better than yours.
"I've not heard of any devices handling 100g over any significantly long periods of time."
You just did.
Show me a complex mechanical device, like a rocket, with fuel tanks, fuel pumps and ball bearings, that withstands >100g continuously for >10 seconds. Turbine blade that has 100g at its tip, is not one.
"Don't learn physics, and you'll live in the world of wonders!"
Production of chemical fuel on Moon is going be tricky, there are no fossil fuel deposits there and water is much scarcer than on Earth. Our contemporary chemistry isn't completely ready for such conditions, at least not on industrial scales.
If we could use a mechanical / electrical mechanism to throw things onto the Lunar orbit, it might be much more efficient than, say, making methane in situ using Sabatier reaction. Solar panels are much more productive on the Moon than on Earth.
Actually, I wonder if cold gas thruster can do the same job. (https://en.wikipedia.org/wiki/Cold_gas_thruster) At least some gases are extractable on the Moon surface.
[1] https://digitalcommons.usu.edu/cgi/viewcontent.cgi?filename=...
Did we need to orbit fluidized electronics?
have seen it described as the "canadian iraqi space gun", which causes some people to do a double take
Little joke
However, I think this technology will be extremely valuable for launching material from the moon.
So kudos for investing for this. And I think it will even end up being a profitable investment. Just not for the intended purpose.
G-loading. Rockets are normally rated for force in one direction (down) the same as gravity and launch acceleration. They can only handle a few small percentages of G laterally. This rocket would need that, plus at least a few G of lateral acceleration for spinup and a massive negative G capability for the impact with the lower atmosphere immediately after launch.
Was there a G-meter on this rocket/dart? What did it feel like to go from thousands of mph in a vacuum to suddenly thousands of mph at sea level? 50g? It would be like slamming through concrete. Larger rockets would no doubt feel less of this impact but they would still need structures akin to fighter jets. Those structures would be heavy and likely nullify any fuel savings.
http://www.youtubemultiplier.com/618f3abec89ec-you-spin-me-r...
Edit: looking at the video closer, it looks like there is a slight angle to the launch, tilting away from the building next to it, which makes sense.
For objects above the earth's surface, the radius of the 'orbit' around the center of earth increases, and thus the 'forward' component of the velocity has to increase in order to stay over the same point on the earth. If you shoot something high enough straight up, it will land to the West of the point it was launched from. (This does ignore effects of wind and such, but I believe so did your argument).
There's a point at which you can no longer ignore the rotation of the earth.
Maybe the effect still isn't huge at the scales we're talking about but it's not straight up and straight down.
Edit: that being said, it does look like there is a slight angle.
The idea is, I think, to use this as a first stage. The second stage is a rocket that takes the projectile to orbit.
I suppose it partly depends on how high you can reach (LEO?) with this method and whether your spun rocket can carry a fuel payload itself.
One minor complication: Technically, air resistance applies a retrograde force relative to the direction of the wind. At low speeds this might actually increase your orbital velocity. However, the speeds where this is relevant are so small that they are not worth consideration.
Just the lateral deceleration of the liquid slowing down in the tank would probably provide a couple tons of thrust.
Maglev trains aren't science fiction anymore.
A Starship can launch 150 tons to Orbit, fully reusable.
Now ask yourself, if SpinLaunch even if you assume the largest possible ground station can not launch more then a few 100kg to Orbit and still requires a Upper Stage rocket that has to be thrown away. How could it possibly be cheaper?
So lets assume an absurdly large SpinLaunch system that can get 500kg to Orbit. You still need 300 launches to match what Starship can do in a single afternoon. In a more practical situation its more like 800 launches to match a single Starship.
Now try to think a gigantic stack of 300 SpinLaunch upper stage rockets that would be thrown away including 300 rocket engines, 300 avonics systems and so on. Compared to a Starship that simply lands and is fully reusable.
Not to mention that a chemical rocket is far more flexible in regards to orbit and far more versatile in possible payloads.
At best it can compete for a very small part of the small rocket launch business. But even then, a lot of the time dedicated launches are used for non-standard orbit and SpinLaunch has far less flexibility then a normal rocket.
I really see no way this makes much sense.
great video: https://youtu.be/dqwpQarrDwk
here it can be lifeboats, backup hooks, spacecraft with backup thrusters, separate mechanisms for moving people vs stuff, and potentially many more
Think about it...even a playground wheel can generate enough centrifugal/petal force to make a human feel a little unwell.
Non-trivial fundamental problem that you can't disappear by just inventing a new material/technique or something.
Edit: and not come out as human paste
I do wonder how a Hyperloop kind of long tunnel, vacuumed out would be a less-G substitute.
In Clarke's story, the launch system is located on the Moon rather than Earth; the lower escape velocity and lack of atmosphere makes the fictional engineering a lot simpler (and allows the climax of the story, which I won't spoil).
The idea was crude materials would be cut up from moons and asteroids, and launched using maglev rails towards civilization centrals. Or it can also be used for recovery, so long that the landing spacecraft can successfully mate the train.
Above is portrayed first, in cases with Sci-Fi novels, then of course something terrible would occur, and the story starts to ramp up. A shivering refugee is found inside, one of containers suspiciously veer off course, the crucial docking latch breaks, etc.
This is a hard part (which sounds impossible to me): since all acceleration needs to be done in a spinner, you need to get to 7900 m/s (even assuming no braking from the atmosphere).
And a = v^2 / r will turn living things into a paste for sane r. :(. I would love to be refuted.
The payload would have a rocket stage after reaching a certain altitude to further accelerate.
"Braking" will commence immediately after exiting the spinner apparatus though. And commence rather unceremoniously. I don't think this is going to carry humans in the foreseeable future.
Part 1: https://yewtu.be/watch?v=9ziGI0i9VbE
Part 2: https://yewtu.be/watch?v=ibSJ_yy96iE
TL;DW:
- Vacuums are hard.
- Rotation is hard.
- Mach 7 is hard.
- Lower-atmospheric thermal ablation is hard.
- Promotional videos long on drama, simulation, confetti, and inspriring soundrack, short on actual data and performance shots.
- Leadership expertise and experience is profoundly lacking.
the article doesn't really go into detail but alludes to the answer to my question being "yes".
I guess the idea is:
you can spin a smaller rocket which requires energy EnergySml. Alternatively you could have a larger rocket which requires EnergyLrg to get out of the atmosphere.
So EnergySml + Fuel in rocket to leave atmosphere << EnergyLrg (which is launch + leave atmosphere)?
---
other questions I had were:
- wouldn't the rocket being spun have to be heavier than the equivalently sized chemical rocket because it has to be able to survive being flung around in a centrifuge?
Using a really big cannon, not entirely unlike in the Jules Verne novel, has been investigated seriously. It has similar efficiency benefits. Some limits on what can survive the acceleration though: https://en.m.wikipedia.org/wiki/Project_HARP
That could be extremely complex, unless (perhaps) the rocket using solid boosters, in which case, you are spinning a highly volatile mega bomb.
https://www.youtube.com/watch?v=Rb6sxy3f7VE
Well train human can survive:
38G for 0.5s
9G for 2min
7.5G for 5min
Not without problems though. He "sustained a fracture of his right wrist during the runs on two separate occasions, also broke ribs, lost fillings from his teeth and bleeding into his retinas that caused temporary vision loss".
This is during short periods (of deceleration), so very different from SpinLaunch's launches.
Thinking about that, would it be possible to "fling" an object trebuchet-style by flying some crazy maneuver with a "hook-plane" instead of a satellite?
The military have probably already tried something like that to send ballistic missiles I guess, and in that case you don't even have to accurately aim as the missile can guide itself to the target. Why isn't this a thing already?
This thing causes 10000 G for a sustained time on the projectile. To put this into perspective above 14 G even fit and trained individuals tend to black out. Blood centrifuges use 1500-3000G for 5 to 10 minutes to separate your blood plasma from the red blood cells. After hours of 10000 G acceleration a human body will be a red goo stratified into density layers at the wall of the vessel.
Punching through a membrane won’t faze the goo much comparatively.
Payload: 200kg vs 300kg
Height: 6m (est) vs 18m
Wet mass: 11t vs 13t
Stage 1 thrust: 75kN vs 230kN
Stage 2 thrust: 5kN vs 26kN
Engines: Pressure fed vs Electric pump fed
The SpinLaunch rocket dumps its shell before turning on its engines, and presumably has much lower gravity losses, so despite shockingly similar wet masses, the rocket and its engines can be greatly downsized.
I don't have any solid opinion about the concept, but it does seem like “smaller tanks” is maybe not the best description of what is a fairly significant design change. You do shed a lot of complexity from the propulsion systems, but I'm far from sure the stuff you add is any cheaper.
[1] https://player.vimeo.com/video/573539093#t=26s [2] https://en.wikipedia.org/wiki/Rocket_Lab_Electron
After acceleration, you have 'jerk' V^3, which is commonly used, and then 'snap, crackle, and pop' which are less commonly used.
https://en.wikipedia.org/wiki/Jerk_(physics)
V^4 is also called jounce, and I know some vehicles have jounce bars, but I don't know if those are truly related.
That first part is correct; gravity is measured in units of acceleration.
But I have no idea what you mean by "velocity squared". If you mean the ordinary sense, of multiplying something by itself, you're wildly wrong. Acceleration is the derivative of velocity with respect to time, dv/dt, and it is measured in units of velocity over time, not units of velocity squared. Its magnitude is unrelated to the magnitude of velocity.