Che fine ha fatto SpinLaunch, lo sparasatelliti elettrico?

Pubblicazione iniziale: 2023/09/08 3:40. Ultimo aggiornamento: 2023/09/11 18:10.

Quasi due anni fa ho scritto un breve articolo a proposito di SpinLaunch, una società privata che propone di lanciare piccoli satelliti usando un sistema a frombola, con un braccio rotante che viene accelerato ad altissima velocità e rilascia il satellite all’istante opportuno; il satellite parte quindi come un proiettile balistico, attraversando gli strati più densi dell’atmosfera senza usare propellente, riducendo (in teoria) i costi e l’impatto ambientale di un lancio tradizionale basato su propellenti chimici.

Da quell‘articolo è passato un bel po’ di tempo senza novità significative. SpinLaunch ha pubblicato alcuni video dei lanci suborbitali di prova che ha effettuato, come questo, che si riferisce a un test del 27 settembre 2022, in cui ha lanciato dei carichi sperimentali di NASA, Airbus U.S., Cornell University e Outpost. Non è stata indicata la quota raggiunta, ma le notizie (Space.com) parlano di accelerazioni di circa 10.000 g e di una dimostrazione della realizzabilità di carichi utili in grado di sopportare queste sollecitazioni usando componenti standard. I carichi sono stati recuperati con successo. Questo è stato il decimo lancio di prova effettuato e il primo con carichi di terze parti.

La notizia più recente pubblicata nella rassegna stampa sul sito di SpinLaunch risale al 9 ottobre 2022, quasi un anno fa, e parla di questo decimo lancio di prova, notando che la velocità massima raggiunta dal veicolo è intorno ai 7.500 km/h. Notevole, ma per un’orbita bassa servono 28.000 km/h (che non è necessario raggiungere con il braccio, dato che il progetto di SpinLaunch prevede che il veicolo abbia un motore a propellente chimico che si accenderebbe dopo il superamento degli strati atmosferici più densi, accelerando fino alla velocità orbitale) e secondo i calcoli di Bigthink un lanciatore orbitale sottoporrebbe il veicolo e il suo carico a 50-100.000 g.

Un articolo del 2 settembre 2023 su The Space Bucket (con relativo video) riferisce che SpinLaunch sembra aver trascorso quest’ultimo anno andando in cerca di finanziamenti e di un sito dove costruire la versione orbitale del proprio lanciatore, senza effettuare altri lanci. I siti presi in considerazione includono l‘Alaska, l’Australia e le Hawaii (il lanciatore dimostrativo attuale è nello stato del New Mexico, negli Stati Uniti). SpinLaunch attualmente dichiara nelle sue FAQ che “il primo Sito di Lancio Orbitale è in via di selezione finale in un luogo che verrà rivelata prossimamente situato in una regione costiera degli Stati Uniti” (“The first Orbital Launch Site is in final selection in a soon-to-be-disclosed location in a coastal region of the United States”), ma questa FAQ risale a oltre un anno fa.

Un servizio video della CBS News datato 3 settembre 2023 (cinque giorni fa) mostra immagini del lanciatore, dei lanci e dei veicoli/proiettili di prova e intervista il fondatore, ma non fornisce novità.

Il 7 settembre 2023 Chris Bergin di NasaSpaceflight.com ha riportato su X (l’ex Twitter) un estratto di un comunicato stampa di SpinLaunch che annuncia una partnership di investimento da parte della giapponese Sumitomo Corporation. Non vengono indicate cifre.

I brevetti US10059472B2 e WO2019164472A1 di SpinLaunch descrivono la sua tecnologia. Altri suoi brevetti riguardano pannelli solari irrobustiti e volani giroscopici per la regolazione dell’assetto.

Il metodo di lancio di satelliti prevede che quando il veicolo viene rilasciato venga rilasciato contemporaneamente anche un contrappeso, per impedire sbilanciamenti del braccio rotante. Il veicolo procede per inerzia per circa un minuto, raggiungendo circa 60 km di quota e quindi superando gli strati più significativi dell’atmosfera, e poi accenda il proprio motore a razzo per circa un minuto per accelerare fino a circa 28.000 km/h; un’ultima accensione di circa 10 secondi consente l’inserimento in orbita intorno alla Terra.

E questo, per il momento, è tutto: per il futuro, il sito di SpinLaunch descrive il progetto del lanciatore orbitale, denominato L100, che avrebbe una camera a vuoto circolare obliqua con un diametro di circa 90 metri, collegata a un tubo di lancio dotato di portelli ad apertura e chiusura ultrarapida (e, presumo, molto ben sincronizzata) che impedirebbero la ripressurizzazione della camera dopo il lancio e in teoria consentirebbero una cadenza di lancio molto rapida (un paio d’ore fra un lancio e l’altro).

Non è difficile notare che quello che Spinlaunch ha costruito e propone di costruire è non solo un lanciatore di veicoli spaziali, ma anche in sostanza un cannone elettrico, che spara proiettili ipersonici. Se li lanciasse angolati, anziché verticalmente come sta facendo, avrebbero una gittata non trascurabile, e sarebbero dei proiettili assai difficili da intercettare. La soluzione di usare un braccio rotante, invece di un acceleratore lineare come nei cannoni elettrici militari (railgun) attuali, riduce il picco di energia elettrica richiesto, dato che il braccio può essere portato alla velocità di lancio gradualmente. Le applicazioni e implicazioni militari, insomma, non sono trascurabili, anche se le velocità di lancio raggiunte finora da SpinLaunch sono inferiori ai 12.000 km/h dei railgun lineari. Non a caso Wired.com nota che nel 2019 il Dipartimento della Difesa statunitense ha siglato un contratto con SpinLaunch per lo sviluppo della sua centrifuga.

Inoltre un sistema del genere sarebbe molto interessante se installato sulla Luna o su altri corpi celesti che non hanno un’atmosfera densa o non ne hanno affatto: costituirebbe un metodo efficace per lanciare carichi senza dover portare o fabbricare propellente, senza sollevare polvere superficiale al decollo e senza contaminare la zona di lancio con il proprio scarico di gas combusti, evitando la necessità di collocare il sito di lancio lontano dalle strutture abitate (sarebbe sufficiente metterlo dietro una collinetta che conterrebbe eventuali malfunzionamenti catastrofici). L’energia per alimentarlo sarebbe elettrica, e quindi generabile in loco usando dei pannelli solari e un sistema di accumulo.

Questo video è lungo ma contiene moltissime immagini e informazioni tecniche molto utili:

Questi sono i sottotitoli inglesi del video, che ho ripulito da alcuni errori:

0:00 In the middle of a global pandemic, a ragtag group of welders, heavy machine operators and builders were brought together. They were given plans for a mysterious structure. With little information on what exactly they were building, their only clue was scrawled across the top of their plans. SpinLaunch. A space catapult.

In this exclusive, behind the scenes documentary, I talked to the people behind this new innovative company.

SpinLaunch is attempting to subvert a problem that plagues the space industry: the rocket equation. The rocket equation has been the tyrant engineers have feuded with from the dawn of the space age. A simple equation that describes how much fuel a rocket needs to carry a payload to its destination.

The tyranny of this equation is that the fuel needed to deliver the payload is a payload itself. A compounding problem that makes rockets more fuel than rocket. Typical rockets are more than 90% fuel.

SpinLaunch is trying to change that paradigm, by imparting as much velocity as possible to the payload on the ground, eliminating as much fuel as possible from the rocket's weight while greatly reducing the size and complexity of the non-reusable components.

Their plan? To spin a small rocket in a centrifugal mass accelerator under a vacuum up to an astonishing speed before releasing it, punching through the thickest layers of our atmosphere at hypersonic speeds, gaining 72 kilometers of altitude off nothing but pure kinetic energy, before splitting its fairings and unveiling a substantially miniaturized two-stage rocket to continue its journey into orbit.

You may question the numbers here, but this isn’t the first time this has been done. Project HARP, standing for High Altitude Research Project, managed to get a projectile to 180 kilometers of altitude with a high-powered gun. They achieved a muzzle velocity nearly identical to Spinlaunch’s planned launch velocity. However, scaling a kinetic energy launch system up, to launch a 10-tonne projectile, needs SpinLaunch’s technology.

Of course, all of this is easier said than done. The challenge facing SpinLaunch’s engineers is immense. This endeavor demands several new key enabling technologies.

"So in terms of key enabling technologies, carbon fiber certainly takes center stage." 

That’s David Wrenn, SpinLaunch’s VP of Technology. I spoke with him on the SpinLaunch factory floor about the carbon fiber reinforced plastic that they are using for their tether.

"You know, in terms of its strength-to-weight ratio, it's it's essentially unmatched by any other material on Earth. And the amazing thing is that it's actually available in industrial quantities now and engineering tools and simulation methods exist to really quickly iterate, understand what a composite structure will do. And then we have structures and test rigs like this to validate in the real world that the components actually fail at the expected loads. So just to give you a sense of the strength of carbon fiber, this is a pultrusion. So this is made by taking carbon fiber tow, essentially spools of carbon fiber, and pulling it through a heated die with a resin bath. Basically the resin gets impregnated into the carbon fiber tow as it gets pulled through the heated die. And you get these really nice, highly unidirectional structures that can be used for, you know, Spin Launch. Right? And so as small as this cross-section is, right, I think this is about an eighth of an inch thick and only a few inches wide, this can do just under a quarter million pounds of total load capacity, which is... which is really, really impressive for what this is.”

"I’m trying to convert to metric in my head. It's like, oh, yeah, that’s a lot."

"Yeah. I think that's like a million newtons or something like that. So and then if you look at, you know, this is where it gets interesting is, is can you build really thick cross-sections of carbon fiber? You can see there's there's really great..."

"Is this pultruded as well?"

"So this is not pultruded, this is laid up and then it's... it's cured in an autoclave. So this is essentially a subsection of the laminate that you would see at the root of the tether on the suborbital system."

"Okay. So this is, this is..."

"So this isn't this is not unidirectional or pultruded, but if it were, if you were to take this the same cross-section here and basically make this a pultruded fiber, if you just stacked up multiple pultruded sections, you would get about 9 million pounds of total load capacity through this cross-section. So it's, it's impressively strong for what it is."

5:12 "I mean, it's, it's heavier than you’d expect."

"Yeah."

Da qui in poi non ho avuto tempo di impaginare per bene, ma ho sistemato gran parte degli errori:

Like holding a carbon fiber brick.

I don't think there's... I've never seen an application where that much carbon fiber has been laid up.

It's, it's rare to see carbon fiber this thick.

Yeah.

It really is rare. The final fully scaled tether for SpinLaunch’s orbital system is likely going to be the single strongest tensile structure on earth. Let’s do the math on that.

SpinLaunch aims to [release?] its aeroshell, containing the miniaturized rocket system, at about Mach 6, that’s roughly 2 kilometers per second. With a radius of 45 meters, the tether will need to spin 450 times per minute to attain that velocity. At that rate the g loading on the tether will be 10,000 gs. Meaning this aeroshell is going to exert a force 10,000 times greater than its weight due to gravity.

The aeroshell with the payload and rocket is going to weigh approximately 10 metric tonnes. So that means the tether, at the tip, is going to need to be able support 100,000 metric tonnes, or 100 million kilograms. To put that into context, a fully loaded Falcon 9 weighs about 0.55 million kilograms, so this tether is going to need to support the equivalent weight of 182 Falcon 9s.

This is going to require a hefty piece of carbon composite with cross-sectional area of at least 0.23 meters squared. That explains the brick of carbon fiber we saw. That brick could support about 4.1 million kilograms. So the full scale tether will need to be 24.4 times this size at its tip, but that’s just the tip.

This equation tells us why carbon fiber is so vital to this endeavor.

Because each section of the tether has to support the section above it, its strength to weight ratio needs to be exceptional. If we calculate the tether area near the hub for the same carbon composite, the tether only needs to increase in area by 2.5 times, at about 0.56 meters squared. We would of course need to add a safety factor of at least 1.5 to this, increasing these dimensions by 50%.

I have skimmed over this equation here, but if you want to learn more about the engineering of this system, and energy of getting to space in general, I have created an entire course on Brilliant to partner this video, and you can sign up for it with the link in the description.

That design is perfectly feasible and is reflected in SpinLaunch’s renders. We even have the manufacturing skills necessary to build even larger composite structures, thanks to the wind industry.

So, this is all well and good, but spinning a carbon fiber composite up to Mach 6 isn’t possible in air. The aerodynamic heating would destroy it. So, to solve this issue, SpinLaunch created a massive vacuum chamber around its tether.

"You know, there's a bunch of things at the beginning of SpinLaunch that were nonstarters for a lot of people, like even just building a large-diameter vacuum chamber. You know, people were telling us, you know, the one behind me here would cost tens of millions of dollars to build. And we ended up doing it.

You know, we had this really, really kind of scrap-heap mindset. And we ended up doing it for less than a couple million dollars with ten people. Right. Which is unheard of.

There's some large industrial vacuum chambers out there. But there's, you know, quite a few of the really large chambers around the world are for aerospace applications. And so they're achieving extremely high levels of not only vacuum, but cleanliness. And so the cost is proportional to that. And it's kind of exponential. You know, they're achieving vacuums that are on the order of ten to the negative 8 millibar torr. And, you know, typically we're operating at about a million times worse than that."

SpinLaunch is breaking new ground with this kind of vacuum chamber. Typical large-volume vacuum chambers, like the world’s largest one at the Space Power Facility in Sandusky, Ohio, are designed to simulate the vacuum of space.

Those require an extremely low pressure vacuum, with tight tolerances and control of contamination. They even need specialized tools like lamps to simulate the radiation and heat emanating from the Sun and cryogenic cooling to simulate the heat of space. The people that built these facilities are the industrial experts SpinLaunch had to draw from, and most thought they would never be able to build a vacuum chamber this large on their budget.

But SpinLaunch had some things on their side. They didn’t need that extreme of a vacuum, as their goal is not to simulate the vacuum of space. Their goal is to minimize drag and the power required to overcome it, minimize the aerodynamic heating that would destroy the tether, and eliminate all those pesky aerodynamics effects like flutter. That means SpinLaunch could use cheaper materials like mild steel, where ultra high-vacuum chambers need more expensive specialized processed materials to avoid outgassing, where gases within the metal in the form of oxides, or simply dissolved within the metal, are released into the vacuum.

It also makes the process of drawing a vacuum much easier. Drawing a vacuum isn’t as simple as just turning on a pump and leaving it on long enough. The more air you draw out, the harder it becomes, as you are not only working against a continually growing pressure gradient, but statistical probability.

The first stage of drawing a vacuum is to remove the bulk gas. At this stage the gas is a viscous fluid, and the molecules within the chamber interact with each other often. Here we can use traditional fluid flow pumps, like a positive-displacement pump, that mechanically moves molecules out of the chamber, and higher-pressure air at the back of the chamber forces more air to fill the space created, allowing more air to be pumped out.

But as gas is removed from the chamber, the distance between the molecules increases. This is called the mean free path: the distance a molecule can travel without colliding with another molecule.

Now, pressure is really just molecules colliding, and as collisions become more infrequent, the pressure gradients that are needed to achieve equilibrium begin to vanish, meaning it takes longer and longer for equilibrium to be established, and the rate the pump can remove molecules lowers, as there are simply fewer and fewer molecules near the pump to remove. At some point viscous flow stops entirely and we enter a flow regime called molecular flow, where the distance between collisions is actually larger than the internal dimensions of the vacuum chamber. Meaning, the molecules are statistically more likely to just bounce inside the chamber with nothing forcing them towards the exit. At this stage it is impossible to actively pump the molecules out.

The molecular pumps needed for this flow regime instead act like some kind of Venus flytrap, waiting for a molecule to enter it, and then its job is to prevent the molecule from returning to the original chamber. Turbomolecular pumps are basically multiple levels of turbines that knock molecules in one direction and prevent them from traveling backwards. These pumps require insane rotation speeds, anywhere from 36,000 rpm to 72,000 rpm, and need incredibly tight tolerances too; they are trying to pump individual molecules after all. So, it goes without saying, these kinds of pumps are expensive.

All the while, outgassing and other leaks are actively working against the game of pure chance. Creating a high vacuum requires extreme precision in manufacturing and design. SpinLaunch didn’t need any of this.

Mark Sipperley, the director of Engineering at SpinLaunch, walked me through the vacuum pump station at the New Mexico site.

"Here in the vacuum plant the, the most familiar thing would be the tube that came out of the chamber and then runs underground in this manifold. So this is the very end of it. So off of this vacuum manifold, we then have a series of three different types of pumps. Up on top we first have roughing pumps, which pull the atmosphere of like one atmosphere down to about 30 millibar. Those are dry screw pump that are essentially like overlapping lobes. It's it's one form like a turbocharger.

Then the next stage is we have this Roots pump, which is, which is like a yeah, another shape of a turbocharger. It's like this the rotating twin screw pumps okay.

So that kicks on at about 30 millibars.

So it's mostly only in it's 30 millibars below.

But you'll notice that each pump, it exhausts into the pump like a pump, that's a slightly higher pressure.

So this Roots pump only works in 30 millibars below, but it can't exhaust all the way up to one atmosphere.

So that's backed by another piston pump

So this this would also be like a great roughing pump.

So when we turn on the first system, we have nine Edwards GSX pumps up there, and then this piston pump.

Sorry, both of these piston pops when we get down to 30 millibars, which we turn on a series of Roots pumps, which is this guy right here. And we have another smaller one on this one. These pistons are also running as well. And then once we get down below one millibar they're going to turn on these vapor diffusion pumps, which only are really effective down at the very low pressure. Those work like oil jets. So you vaporize oil, you shoot it down a series of channels and it grabs onto the air molecules, runs them down a series of tubes and then you have these cooling loops that will then condense out the oil and then the water, or sorry, the air progressively makes its way through like a long path, and it eventually goes to a Roots pump, which can grab on to it, and then it goes to a piston pump, then all the way out.

So we talked before, you know, vacuum may not be the best description. Most people think of vacuum like a hard vacuum like in millibar, in torr is what most people are used to like. E to the minus -7 torr, like seven zeros are six zeros. And the number that's like true vacuum. That's hard vacuum, that's where you test like, you know, like electric propulsion systems and like high-end space features. That is nowhere near the atmosphere that we need. And that's that vacuum is also really expensive to get to. We have to follow a lot of stringent rules like you can't use steels, you have to use aluminums and use coatings. You know, even like putting your fingerprint in an atmosphere in a vacuum that low, will take weeks to boil off. We only require the equivalent to the minus 3 torr or as a .01 millibar or .1 millibar. Today we're going to be running at like one millibar. We only pull the vacuum that we need, because a vacuum is expensive. So it’s is closer to describing it like a high atmospheric chamber than a vacuum chamber specifically. And then it will be the same.

And again, that's all driven by the aerothermal. That's all that's the only vacuum that we needed to accomplish. So on the orbital system, we'll probably pull a very similar vacuum, we don’t have to go much deeper, there’s no benefit to going lower."

This is another one of those technical issues that the internet made a big deal out of, without fully understanding what SpinLaunch actually needed out of the vacuum chamber.

One of the other primary concerns expressed on the internet was the tricky and unique problem of a vehicle traveling at hypersonic speeds from a vacuum into a thick sea level atmosphere.

To begin with, we need to prevent air from rushing into the vacuum chamber once the vehicle is released. SpinLaunch is aiming to be a high-frequency launch system, capable of launching multiple satellites per day, holding the vacuum between launches to decrease energy and time costs. However, the primary concern is the disastrous effects that air would have as it meets the tether spinning at hypersonic speeds. This would be an incredibly expensive single shot system if this was allowed to happen.

To solve this problem, SpinLaunch needed a way of sealing the chamber extremely quickly after launch, so inside this long tube attached to the vacuum chamber is a double-door airlock, with doors on either end of the tube. This tube is also under vacuum during spin-up. As the vehicle is released, using a release mechanism that SpinLaunch kept hidden from our cameras throughout the shoot, it passes into the exit tunnel, where the first door rapidly closes behind it. As this first door is closing, the second door will begin to open. The atmosphere will begin rushing into the tube and give the aeroshell its first taste of the hypersonic flight regime it will be flying in. The first and second door need to close quickly enough to prevent air from entering the vacuum chamber.

This is not an easy problem. Millisecond delays that may seem trivial in most cases start to mount up when the vehicle travels this quickly. The time it takes for an electrical signal to propagate, the time it takes to overcome the inertia of the door, the time it takes for a proper seal to form. All these problems become matters of survival at these speeds.

Once again, SpinLaunch are keeping their cards close to the chest on this one, but they did give me a demonstration of the door closing in their factory and engineering hub in Long Beach, California.

"Well, it’s it's moving really fast. And so when the... when the you know, without specifying, is he going to do a countdown?

You're ready for a countdown just let me know.

Yeah.

So basically what's going to happen is, you know, this is going to be filled with, you know, for lack of a better word, like a black door which basically you'll see that like you can pass through this with a vehicle and then in an instant it's going to be close.

Okay.

So and again, it's like fast in the blink of an eye.

So you'll see a little bit of settling as a, as a after it closes.

But it's, you know, basically 95% close within, you know, 30 milliseconds."

Oh wow, okay.

Closing the airlock.

[Static]

Speaker Closing airlock in five, four, three, two, one.

[LOUD BANG]

It's pretty fast.

Yeah that is not what I was expecting.

[Laughing]

Yeah, it's fast.

So that actually closes.

Like, it's actually hinged.

There's a pivot.

Yeah, there's a pivot involved.

Yeah I wasn’t sure if it was going to be a sliding thing.

But the hinged one makes sense as well.

Yeah, that wasn't that what I was expecting.

Yeah.

So it’s 100% reusable, so you can set that back up again and do it again and again and again.

So that's a key aspect of it is that you don't have any major consumables in the process.

So, so that's fast. Visceral.

All right.

[Laughter]

It’s a door closing.

I don't know what to ask.

It's really important not to let everybody back in.

So that's you know, that's why we have it.

Oh, everybody jumps.

You can’t not.

Yeah, yeah.

"So, you know, the airlock is a really critical subsystem of the overall, you know, of the overall architecture as you travel from vacuum into the atmosphere because the tether is still rotating at high velocities, you want to maintain the vacuum inside of the vacuum chamber. And so the airlock is your first line of defense for that. And so we have multiple redundant airlocks just like what you see here that the vehicle passes through and it subsequently closes behind the vehicle, you know, preventing the air from in-rushing and reentering into the vacuum chamber. And so that the exit tunnel is really the only portion of the chamber that experiences a rise in pressure."

I imagine that allows you to reset and like increase frequency of launches as well. If you're not having to re…like..

"Yeah, totally. So you can, you can do, you know, you can essentially provide a, you know, an airlocked space for the end of the tether as well. And so you can basically just re-pressurize that space as you load in new vehicles. It's possible you could do vehicle integration in vacuum. But currently we're... we're anticipating actually repressurizing a small portion of, you know, interfacing around the tether. Repressurising a small portion and integrating the vehicle without it being in vacuum or."

What do you actually see the like how many launches a day do you think you can manage? I think that's like one of the advantages of this that you can yeah.

"I think on the very high end, it's upwards of ten. I think on the low end, it's, it's, it's about five is a pretty good nominal target for us. We see viability there."

In SpinLaunch’s public videos, the secondary air lock has simply been sheets of mylar. This is one of the few problems that becomes easier as the launcher scales. As the exit tunnel grows in length, it will take air longer to reach the door at the base of the exit tunnel. SpinLaunch have only just begun these one-third scale tests, with their fastest launch to date at 1.6 Mach, slowing ramping up the speed of launch as they test their systems. This prototype launcher features some other simplifications compared to their final planned configuration.

One of the most obvious problems to tackle is the issue of vibration. When a spinning object's weight is not evenly distributed it will vibrate. This is how rumble feedback works in gaming controllers. A simple electric motor with an uneven weight attached. However, with a structure as large as SpinLaunch’s tether, spinning several times per second, any imbalance could shake the entire structure to the ground. This is a major problem, because by design the tether releases a 10-tonne weight right as it hits its maximum velocity. SpinLaunch needs a way to balance the tether after launch.

There is a very simple solution to this problem though. Release a balanced weight from the other side of the arm at the same time. Right now they are simply releasing a counterweight that slams into an armored section of the vacuum chamber. We saw one of these counterweights being manufactured out of fiberglass in the Long Beach factory; however, over the long term having to clean up the mess this creates after each and every launch is far from ideal.

The ideal solution would be to release a counterweight in the form of another launch vehicle after a single half rotation of the tether. The oil-filled journal bearing the massive axle sits upon should be able absorb the force of this imbalance over a period of time this short.

The next issue we need to concern ourselves with is the aeroshell punching into the atmosphere at Mach 6. This, again, is a fairly unique problem. Typically weight is a restraining factor in aerospace, but for SpinLaunch the energy required to spin the aeroshell up to speed is actually rather trivial.

"And I like to use the analogy of like a Tesla, right? So the Tesla Model S Plaid is about 0.7 megawatt. So on the low end, it's about 100 Teslas but it really comes down..."

Is that the full scale?

Yeah, for the full scale.

Yeah, yeah, yeah.

"For the orbital system you're talking about like on the low end. On a very low end. You know, it's probably about 65 to 70 megawatts. And again, that really depends on where you end up with the final orbital tether, you know, whether or not you, you know, what, what safety factor you operate with. What, what, you know, what tether strength you end up with your effective tether, cross sectional strength that all feeds back into itself. And then you have to kind of scale it accordingly. I would say like really conservatively, like, you know, if you wanted to spin up really fast, then you're talking about higher power demand. So whether you want to speed up in an hour or 2 hours, you know, proportionately makes a difference of of how much power that you need. So but, you know, on the high end, you're talking about maybe 150 megawatts of power, which is like... I don't know, maybe in layman's terms, it sounds significant, but, you know, you can you know, there's, you know, there's motor catalogs where you purchase you know, the motor that that has that capacity. Right. And so this is, you know, it's industrial scale hardware and certainly, you know, mostly off the shelf."

Do you need to worry about grid integration at all when you're when you're suddenly drawing that much power?

"For better or for worse, no, because you're typically, you know, particularly for for early, you know, orbital accelerators that we're building, we're expecting them to be in really remote locations, kind of remote coastal locations. Green field sites that don't have substantial existing onsite, you know, resources or power. So you're you're basically, you know, bringing your own power. You you know, and so you have to, you know, decide on, you know, what is your energy source or are you doing energy recapture, you know, etc.."

SpinLaunch claims their total energy demand per spin-up is about 100 MWhrs. The cost per kilowatt hour for industrial facilities is about 6 cent. So that’s a cost of 6000 dollars in electricity cost. That’s insanely cheap. To put that into perspective, 100 megawatt hours is equivalent to about 9600 litres of kerosene, about 8 tonnes of fuel.

For reference, the Electron Rocket from New Zealand's small-satellite launching company Rocket Lab, capable of launching a similar sized satellite, weighs a total of 12.5 tonnes, the vast majority of that weight being its own fuel and oxidiser.

SpinLaunch claims their rockets will need to carry about 30% of the fuel and oxidiser compared to these competitors. They are essentially replacing the first stage of a traditional rocket with an easily reusable kinetic launch system.

SpinLaunch will also be able to recapture a good deal of the electricity stored as kinetic energy in the tether, using regenerative braking, even further reducing their electricity bill.

Because of all this, the limiting factor for SpinLaunch in terms of weight is actually the weight the tether can support, and as a result, it actually makes sense to maximize the density of the aeroshell, because it affects a variable that will drastically improve its ability to punch through the atmosphere:  its ballistic coefficient.

Ballistic coefficient is essentially an object's ability to resist air resistance. Think about how hard it is to throw a feather. No matter how hard you throw it, it’s not going to go very far. It’s got a large surface area for air resistance to act upon relative to its weight. That’s a low ballistic coefficient.

Ballistic coefficient is found by dividing the mass of the projectile by the drag coefficient multiplied by the cross-sectional area. So SpinLaunch effectively wants to maximize the mass relative to the cross sectional area. This is obviously not typical for aerospace vehicles. 

"If you, if you look at reentry capsules whether it's for something like the Stardust return capsule where it's really, really high velocity or you look at it reentry from orbit for a manned capsule or something like the Space Shuttle, they're typically using thermal protection systems that are extremely low density, like on the order of less than 300 kilograms per cubic meter. It's just basically foam. And so... so typically that means you're making like significant compromises, like. The material often is, You know, brittle or prone to fracture you know, or really expensive or gets worn away. And then you have to replace the tiles, kind of in the infamous case of the Space Shuttle. So what we're dealing with is, you know, you're on the tip of the vehicle you have, you know, materials like copper, which, you know, not only are they, you know, a significantly higher density, right? You're talking about, you know, thousands of kilograms per cubic meter, but they also have really great thermal conductivity. So basically, as you transition through the atmosphere, you have a high heat load, but then you're dumping that basically into heavy, dense materials that have good thermal conductivity."

This is one of those unintuitive consequences of this style of launch. When I first saw the full-scale aeroshell on the SpinLaunch factory floor, I first asked if I could ride it like a cowboy, but then immediately noticed the bi-metallic nose cone. I knew from looking at it that it was made from copper and aluminum, and that struck me as extremely odd. Those metals would melt at the temperatures I associate with hypersonic speeds. But, because SpinLaunch launches at Mach 6, it actually transitions through the lower atmosphere rather quickly, and as a result, the heat generated can simply be absorbed by these large heat sinks. Aluminum and copper's high thermal conductivity means the heat is distributed through the body of the aeroshell before it has a chance to damage the vehicle. 

The hefty carbon fiber shell is also incredibly strong. SpinLaunch has already pulled their smaller scale aeroshells out of the ground, buried several feet deep from the force of impact, and reused them with minimal refurbishment. With a parachute, these aeroshells will be fully reusable with minimal maintenance, especially as they serve no function other than to protect the inner rocket’s stages. This isn’t an intricate mechanical machine.

Launching from the ground at these speeds comes with advantages too. If we plot drag coefficient vs Mach number for a bullet-like projectile, something rather unintuitive occurs. The drag coefficient rises as you would expect up until we hit Mach 1; at this point it starts to fall as Mach number increases. This is the equation for drag. It’s proportional to drag coefficient, air density and velocity squared. With drag coefficient being lower at hypersonic speeds, it actually makes some sense to punch through the thick lower atmosphere, where the high-density air causes drag to rise, as fast as possible.

Deceleration is a function of time after all, meters per second square, meters per second lost per second. Let’s calculate the dynamic pressure this drag would create at launch, and the deceleration it would cause. The dynamic pressure is found by multiplying air density by the velocity squared and dividing by 2.

At sea level, at Mach 6, the dynamic pressure will be 2.6 megapascals. The final aeroshell is 1 meter in diameter and has a drag coefficient of about 0.1, which means the force applied to the aeroshell at launch will be 205 kNs. This sounds like a lot, but here's where the ballistic coefficient comes in. This drag force is being applied to a 10 tonne body moving at mach 6. That’s a lot of inertia.

Force equals mass by acceleration. Acceleration equals force divided by mass. That means high mass equals less deceleration. In this case, deceleration due to drag will be about 19.8 m/s per second at launch, but it will rapidly decrease as we move through to thinner and thinner layers of the atmosphere and lose velocity. In fact, with SpinLaunch’s planned trajectory, we can plot the atmospheric density the aeroshell will encounter over time: halfing in just 5 seconds, and dropping to less that 10% of the original air density in 15 seconds. While gravity will remain more or less constant at 9.8 m/s per second. That means gravity losses form the majority of energy losses in our transition to orbit. In total, SpinLaunch will lose about 150 m/s of velocity to drag and 1000 m/s to gravity.

Satellites like Starlink orbit at 500 kilometers with a velocity of about 7700 m/s, so even if SpinLaunch maintained its 250 m/s velocity from launch up until the aeroshell broke apart, the two-stage rocket hidden within would still have its work cut out for it.

However, now free of the mass of the aeroshell, the substantially miniaturized rocket needs only a fraction of the mass of fuel and oxidiser to rapidly accelerate the 200-kilogram satellite, the largest satellite this system can launch, through the thin atmosphere at this altitude.

We can actually graph the relative velocity of the spacecraft over time. Starting at Mach 6 at launch, and ending up at about 1500 m/s when the aeroshell splits apart. The rocket motors then kick in to rapidly accelerate the satellite to its 7700 m/s orbital velocity.

The physics here absolutely checks out here, but whether the economics and cost of development will be viable is the big question to be answered. SpinLaunch has built a 1/3 scale prototype at a relatively low cost, but the hardest part of this technology is scale. They have reached 1.6 Mach thus far, have tested their satellite components at 10,000 g in their test facility in Long Beach, and are continually upping their test parameters, pushing further and further.

This is a comparison of SpaceX and SpinLaunch’s proposed launch trajectory, but it doesn’t tell the full story of the real driving issue here, economics. A SpaceX launch to low Earth orbit costs about 67 million dollars. The heaviest Falcon 9 payload to date has been 16,250 kg on a densely packed Starlink mission. That equates to a launch price of about 4100 dollars per kilogram.

However, small satellite launch companies, like RocketLab, who offer greater control over orbit and launch schedules, charge about 15,000 to 25,000 dollars per kilogram. Dollars per kilogram is not a perfect metric, but gives us some idea of the competition SpinLaunch is facing.

SpinLaunch’s main competitive advantage is in the decrease of expendable materials like fuel while substantially miniaturizing rocket components. They also have huge potential to launch far more frequently than their competitors, helping the economics of scale to kick in.

SpinLaunch claims to be targeting an ambitious per-launch price in the range of half a million dollars, placing them at 2500 dollars per kilogram.

In my time in SpinLaunch, talking to their engineers, it’s clear they are excited and believe in this company. The basic napkin physics for SpinLaunch absolutely check out, and they are well on their way to solving the engineering challenges, but scaling up this monstrous engineering effort is going to require enormous amounts of investment, and SpinLaunch could not disclose the answer to many of my questions, as they seek patents for their solutions.

I got little info on one of the most difficult parts of the launch system, the release mechanism for the aeroshell; even the 3D models SpinLaunch provided for this video had the release mechanism removed, so we had to model our own along with the internal rocket structure. The design of the satellites is another problem, due to the massive gs the satellites have to survive, but g-hardening isn’t as large an engineering challenge as the internet seems to think.

"The most difficult part. So besides the structure is, is also the reaction wheel. So the reaction wheel is, generates momentum and basically steals the bus. And so it typically is a big mass that's cantilevered up at a certain angle. So which is the one thing that we don't like, you don't want having a big mass sticking up on a can really."

This is what I assumed was going to be like a difficult thing to because it inherently has to be fairly high mass to control the satellite.

Right.

"And so we've done a lot of work to instead of re-engineering the wheel itself and figure out different ways to do that, we basically just took and created it took a clever way of deploying the wheel. So we support the wheel in the flat orientation and we spin. So when it's spinning, it's, it's well supported. The bearings are unloaded and so it can spin and do its thing. And then we deploy the wheel for when it actually used to operate. So it's a it's a simple solution for what could have been a really difficult problem."

And does the the axis of the actual wheel cause any issues when it's like being loaded? I imagine that's a fairly high weight to have on the axle yeah.

"So we, what we do is we unload the bearings and as part of the deployment mechanism we actually move we reload the wheel into the bearings."

Oh, okay. So it's just taken off completely Okay.

"So again, trying to make simple solutions for very difficult problems."

And like those are very simple answers, right?

Like I figured that the like the inertial wheels would be difficult, not like you just think about it as like, yeah, that's actually a fairly easy thing to just not deal with.

You don't have to have it in the exact configuration when you like launch, right? Same with the solar panels.You can have them, like you said, loaded.

"Yeah. So it's the problems aren't necessarily hard to deal with. It's just you have to think differently. We just have to change the way we think about design. So it's a little bit it's not a lot. The nice part is to that over the last 60 years, what people have been trying to do with satellites actually has helped us because they want to reduce mass. They want to make things stronger. So every bit that they're forcing them to deal with shock and vibe already helps us. And it already inherently starts to make them more hardened. Most components, we don't have to do anything to them. Maybe a little epoxy here but like we one of the most surprising events that we have here among the entire team is we took a board that had a password stuck up, you know, maybe a quarter of an inch. And we all looked I was like, okay, that thing's it's going to fly off the wall. And we spun it and we brought it back and it just went over and that was it. And we are like, all right. Our intuition is completely changing. And yeah, and it's because it's, it's so little mass and it's being held on by two pieces of steel. You know, the amount of force that that was really imparting on those two piece of steel was relatively small. So it just bent over and we all kind of like, Oh, yeah, after you think about it, it does make sense. Okay. Yeah, that's right. So our intuition is starting to grow about, yeah, this little connector of the sticking up really isn't that big of a deal. And so that has been in a positive way. Very surprising."

Gs can only create force where there is mass, and it turns out the satellite industry has been finding ways to reduce mass for decades. A simple aluminum can is capable of withstanding 10,000 gs with a basic redesign of its structure. Minimizing weight located on unsupported surfaces lowers the mass available to be multiplied by the gs, and some simple corrugation can help the aluminium absorb some of the loading without buckling. We spun up an off the shelf star tracking camera using Spin Launch's in-house centrifugal accelerator, which can already achieve 10,000 gs, and the camera worked perfectly fine just moments later.

This is a really interesting engineering challenge, that I think the internet is giving a hard time for some bizarre reason, posing questions about basic physics calculations without actually doing the math, and then saying it’s impossible. Even missing the fact that kinetic energy launch systems have already reached beyond the Karman line 6 decades ago. [...]