Alternative rocket builder SpinLaunch completes first test flight

[Ziggurat]

Even for a single launch, SpaceX's existing rideshare program would probably be cheaper

Ride sharing will probably always be cheaper for small sats, for the same reason that heavy lift is cheaper for bulk cargo. The advantage is never going to be cost per kg. But what an individual launch provides that ride sharing can't is flexible scheduling. Same as a shuttle vs. taxi to the airport. You can't get a ride share whenever you want one, you can only get one when they are ready to send one. But if you're the only cargo, you can go pretty much whenever you want.

That flexibility of scheduling has value, and if this works (still a big if at this point), that's their selling point: customers who want to launch something small, but need to launch on a tight schedule of their choosing. SpaceX isn't targeting that market (why would they?), so other players have a chance in that space.

 

[theprestige]

Military value: rapidly replacing spysats and comlinks in wartime. Rapidly deploying anti-satellite payloads in wartime.

"Ecological" value: rapidly deploying cleanup sats to help recover from Kessler syndrome during and after a war.

 

[cjameshuff]

Cost or price? Note that some of us (at least me) are engaging future scenarios where this works and has a future, so long term cost (one that assume the initial investment has been recovered) is interesting too.

For Falcon Heavy, actual base launch price. For Spinlaunch, $0.5M is the targeted per-launch price, giving $2500/kg.

Starship would have the highest fraction of propellant costs of any operating system, with per-launch propellant costs of about $0.5M...still not enough to make a difference. A minimally-successful Starship that just replaces Falcon 9 would have similar launch costs, in the area of $20M, with per-launch pricing similar to the Falcon 9 that would be $500-700/kg.

Starship's lower cost estimates are more like $2M, so they could potentially knock another zero off that price. This is likely overly optimistic...but so is the $0.5M quoted for Spinlaunch. And again, most payloads will need to be heavier to survive launch on Spinlaunch. Even simple containers of water will need to be heavier.

 

[cjameshuff]

That flexibility of scheduling has value, and if this works (still a big if at this point), that's their selling point: customers who want to launch something small, but need to launch on a tight schedule of their choosing. SpaceX isn't targeting that market (why would they?), so other players have a chance in that space.

Except Spinlaunch isn't targeting that market either...they aren't going to keep that big centrifuge operational with just a $500k launch every month or so, Spinlaunch needs high launch volumes. They've been talking about launching every couple hours, they need customers who want to maintain their own megaconstellations. Those same customers don't need to launch individual satellites on demand and would get better prices by buying regular flights of larger launchers.

 

[cjameshuff]

Military value: rapidly replacing spysats and comlinks in wartime. Rapidly deploying anti-satellite payloads in wartime.

"Ecological" value: rapidly deploying cleanup sats to help recover from Kessler syndrome during and after a war.

Again, it's not actually good for sending large quantities of stuff into orbit. A single rocket launch can deploy weeks of Spinlaunch payloads. Aside from that, the Spinlaunch sling is a huge target, if anything even bigger and more vulnerable than a launch complex for conventional rockets. Something like Electron or Astra is better suited for the former problem, and a heavy lifter like Falcon or Starship for the latter.

 

[dasmiller]

The initial spin launch is suborbital. If the payload isn't caught & boosted on the first pass, it'll hit the ground.

When I wrote this, I was thinking of the 10,000 kg spinlaunch thing, not the 200kg to LEO. Of course, the 200kg payloads would not be suborbital; my apologies for any confusion this may have caused.

 

[RecoveringYuppy]

It was in the context of the orbital tugs and, in particular, a tug carrying up many Spinlaunch payloads in one trip, rather than raising them one at a time. But maybe I misunderstood the tug idea.

We now use rockets to get things not just out of the atmosphere but to their final destination. This leads to an upper stage that isn't taking advantage of potential fuel savings and isn't recoverable/reusable. If our rockets simply delivered things to a minimal orbit and then let something else take over both fuel saving and reusability could be addressed.

That something else could be powered by fuel efficient ion thrusters like our deep space missions use and similar to some of the thrusters GEO satellites use for long term station keeping.

For example a satellite intended for GEO would be delivered to LEO and then met by the tugboat which docks with it and delivers it to orbit and then returns to LEO for it's next job.

Note that for some repetitious applications the tugboat doesn't have to be with the payload for it's entire journey. If you were routinely delivering stuff to some high orbit like ISS or Mars you could do it with just two tugboats. One meets the stuff at LEO and moves it into a transfer orbit. Then at the destination orbit another one meets the payload and moves in to the circular orbit of it's destination.

Conceivably such a tugboat would lead to some new options for our deep space missions by allowing final assembly or deployment to happen at the ISS
or someplace where actual humans could examine them.

I don't know of anyone pursuing this idea now since there are bigger fish to fry on the reusability/cost saving front but if Spinlaunch is successful it could drive this.

When I wrote this, I was thinking of the 10,000 kg spinlaunch thing, not the 200kg to LEO. Of course, the 200kg payloads would not be suborbital; my apologies for any confusion this may have caused.

Might be off topic for this thread but some people have considered such a thing. Something like Spinlaunch simply throws something nearly vertically out of the atmosphere where it is then intercepted by a vehicle that gets it up to orbital speed. I'll see if I can find a name for this idea.

Oh, and it's another problem for the "bulk materials" application, because it can't launch multiple payloads per day to the same orbit. You'll be limited to, say, 100 kg/day to the ISS orbit.

This seems wrong on multiple fronts.

In fact it's completely wrong that this won't be able to hit the same orbit several times per day. In fact it's actual limitation is likely the opposite: that it can't hit a variety of orbits.

I think what you are trying to say is that it can't hit certain orbits by itself without an (additional?) upper stage.

Do they claim that their full scale operational model will be able to reach ISS any time soon? I would think that orbit will be too high for them to reach directly.

And this isn't "another problem" for the bulk materials applications I'm thinking of. It's an advantage. You would very much want to hit the same orbit over and over again in such an application. And pretty much any orbit would do.

SpaceX has averaged nearly 2300 kg/day for all of 2022 so far, all of which could in principle have gone to the same destination, and is still scaling launch rates up.

That's impressive but if Spinlaunch works as they claim they could easily scale up to that and beyond in short order. Again, assuming they actually deliver.

 

[theprestige]

We now use rockets to get things not just out of the atmosphere but to their final destination. This leads to an upper stage that isn't taking advantage of potential fuel savings and isn't recoverable/reusable. If our rockets simply delivered things to a minimal orbit and then let something else take over both fuel saving and reusability could be addressed.

That something else could be powered by fuel efficient ion thrusters like our deep space missions use and similar to some of the thrusters GEO satellites use for long term station keeping.

For example a satellite intended for GEO would be delivered to LEO and then met by the tugboat which docks with it and delivers it to orbit and then returns to LEO for it's next job.

Since you'd have to keep sending up fuel for the tugboat, you'd be sorely tempted just to send it up with the payload, and deliver it all the way to the orbit you desire all at once. No waiting for the tug. No messy orbital rendezvous and docking added to your mission profile.

 

[RecoveringYuppy]

Since you'd have to keep sending up fuel for the tugboat, you'd be sorely tempted just to send it up with the payload, and deliver it all the way to the orbit you desire all at once. No waiting for the tug. No messy orbital rendezvous and docking added to your mission profile.

And no fuel saving or reusability. Way to miss the point. Obviously it would make sense to send the fuel up in bulk which is another point of efficiency.

BTW standardized docking rings already exist. People might recall the questioning of the decision to include one on JWST since there was doubt we'd ever send anything out to repair/restock it.

ETA: And BTW it's not like the current mechanism never fails. The docking might be the lesser risk compared to trusting an upper stage that has to work the first and only time it's used.

 

[dasmiller]

We now use rockets to get things not just out of the atmosphere but to their final destination. This leads to an upper stage that isn't taking advantage of potential fuel savings and isn't recoverable/reusable. If our rockets simply delivered things to a minimal orbit and then let something else take over both fuel saving and reusability could be addressed. That something else could be powered by fuel efficient ion thrusters like our deep space missions use and similar to some of the thrusters GEO satellites use for long term station keeping.

For example a satellite intended for GEO would be delivered to LEO and then met by the tugboat which docks with it and delivers it to orbit and then returns to LEO for it's next job.

Note that for some repetitious applications the tugboat doesn't have to be with the payload for it's entire journey. If you were routinely delivering stuff to some high orbit like ISS or Mars you could do it with just two tugboats. One meets the stuff at LEO and moves it into a transfer orbit.

Then at the destination orbit another one meets the payload and moves in to the circular orbit of it's destination.

I think you did an unintentional bait-and-switch there. Ion thrusters, for example, are not well-suited for the kind of transfer orbit you're talking about. Frex a typical GEO satellite needs to fire its ion thrusters for months to do orbit raising, and it does it by spiraling out. Transfer orbits are more efficient, yes, but a transfer orbit would take much, much longer. I haven't run the numbers (if I feel inspired, I may) but I'd guess that if you had two reasonably-sized tugs, you'd be able to lift maybe 4 spacecraft/year.

Conceivably such a tugboat would lead to some new options for our deep space missions by allowing final assembly or deployment to happen at the ISS
or someplace where actual humans could examine them.

Generally true, but not unique to Spinlaunch.

I don't know of anyone pursuing this idea now since there are bigger fish to fry on the reusability/cost saving front but if Spinlaunch is successful it could drive this.

It's certainly been studied, but AFAIK the numbers never looked promising. As technology evolves, the numbers will change and it may get better.

Might be off topic for this thread but some people have considered such a thing. Something like Spinlaunch simply throws something nearly vertically out of the atmosphere where it is then intercepted by a vehicle that gets it up to orbital speed. I'll see if I can find a name for this idea.

Bear in mind that in this scenario, the interceptor will be moving >7 km/s with respect to the Spinlaunch payload, and that it'll have to get that payload up to 7.5 km/s before it falls back down, which will likely mean several g's of acceleration.

In fact it's completely wrong that this won't be able to hit the same orbit several times per day.

Only if the Spinlauncher is on the equator.

 

[RecoveringYuppy]

I think you did an unintentional bait-and-switch there. Ion thrusters, for example, are not well-suited for the kind of transfer orbit you're talking about. Frex a typical GEO satellite needs to fire its ion thrusters for months to do orbit raising, and it does it by spiraling out. Transfer orbits are more efficient, yes, but a transfer orbit would take much, much longer. I haven't run the numbers (if I feel inspired, I may) but I'd guess that if you had two reasonably-sized tugs, you'd be able to lift maybe 4 spacecraft/year.

That's a design decision. Frex works that way because it can. If you want a faster transfer orbit you'd equip the tug with bigger or more thrusters. There are still some advantages to this concept even if you equip the tug with chemical rockets. You could design these to do fast transfer orbits or slow spiral outs, whatever works.

Generally true, but not unique to Spinlaunch.

Yes, of course.

It's certainly been studied, but AFAIK the numbers never looked promising. As technology evolves, the numbers will change and it may get better.

I think it's more the market that needs to evolve. There just isn't enough market to justify this now.

Bear in mind that in this scenario, the interceptor will be moving >7 km/s with respect to the Spinlaunch payload, and that it'll have to get that payload up to 7.5 km/s before it falls back down, which will likely mean several g's of acceleration.

In this context that means it would be a piece of cake. We're talking a scenario that involves a launch from Earth's surface of 10K gs.

Only if the Spinlauncher is on the equator.

No matter where it is. You and cjameshuff seem to be making some unstated assumptions about this.

 

[dasmiller]

That's a design decision. Frex works that way because it can. If you want a faster transfer orbit you'd equip the tug with bigger or more thrusters. There are still some advantages to this concept even if you equip the tug with chemical rockets. You could design these to do fast transfer orbits or slow spiral outs, whatever works.

As long as you're not assuming that you get the mass efficiency of ion with the high thrust of chemical rockets.

In this context that means it would be a piece of cake. We're talking a scenario that involves a launch from Earth's surface of 10K gs.

Catching an object that's moving 7.5 km/s relative is a "piece of cake?"

No matter where it is. You and cjameshuff seem to be making some unstated assumptions about this.

If it's at, say, 35 deg north latitude, then it'll be launching into a 35 degree inclined orbit. If it does two launches an hour apart, those two payloads will both be in 35 degree inclined orbits with a 15 degree difference in RAAN*. That means they're in significantly different orbits.

*Right Ascension of Ascending Node. It's where the orbit crosses the equator going north.

 

[RecoveringYuppy]

As long as you're not assuming that you get the mass efficiency of ion with the high thrust of chemical rockets.

Yeah, definitely.

Catching an object that's moving 7.5 km/s relative is a "piece of cake?"

Compared to what else we were assuming, yes. And note that I was addressing what you said about g forces not what you are questioning now.

If it's at, say, 35 deg north latitude, then it'll be launching into a 35 degree inclined orbit. If it does two launches an hour apart, those two payloads will both be in 35 degree inclined orbits with a 15 degree difference in RAAN*. That means they're in significantly different orbits.

Ah. OK. Significant for some applications. I don't remember the context of your statement but, yes, if that matters then once per day is the limit. However equatorial launches still have a phase angle. So for most applications I have in mind, equatorial isn't different in that regard.

 

[cjameshuff]

In fact it's completely wrong that this won't be able to hit the same orbit several times per day. In fact it's actual limitation is likely the opposite: that it can't hit a variety of orbits.

No. You can go to the same orbital plane, but at a dozen launches per day, Earth will have rotated 30 degrees and the launcher will no longer be in the plane for the next launch, and 12 hours later it'll be in the plane but pointed in the wrong direction. You're not going to the same orbit for 24 hours.

Do they claim that their full scale operational model will be able to reach ISS any time soon? I would think that orbit will be too high for them to reach directly.

The ISS is in LEO, lower than all the currently deployed Starlink shells. They really aren't useful for much if they can't even reach the ISS.

That's impressive but if Spinlaunch works as they claim they could easily scale up to that and beyond in short order. Again, assuming they actually deliver.

...no, they could not. Achieving that launch rate to a single orbit would require a huge array of Spinlaunch centrifuges.

 

[cjameshuff]

And no fuel saving or reusability. Way to miss the point. Obviously it would make sense to send the fuel up in bulk which is another point of efficiency.

Since you have to send up propellant to get the tug to the next payload, and propellant to carry that propellant along with the payload to the destination, you probably are not coming ahead on propellant consumption that way.

 

[rjh01]

<snip>

Catching an object that's moving 7.5 km/s relative is a "piece of cake?"

<snip>

They do that every time they send a spacecraft to the ISS.

The advantage of having a tug boat in space is that there is no need to launch it every time. Plus less junk from discarded rockets in space. The major problem would be that no maintenance would be able to be carried out on the rocket for long periods. Plus being able to refuel it.

 

[dasmiller]

They do that every time they send a spacecraft to the ISS.

That's not the scenario that inspired my remark. The concept was that Spinlaunch would launch something vertically up to LEO altitude and the tug would then bring it up to orbital speed. Assuming the tug was in orbit, the tug would have a lateral speed of 7.5 km/s and the spinlaunch payload would have a lateral speed of around 0. Either the tug would have to slow down to 0 (falling!) and then accelerate back up (an extra 15 km/s acceleration for the tug. It would take less delta-V for the tug to do a round-trip to Mars), or the tug will hit the spinlaunch payload at several km/s.

There may be some technique using a many-miles-long spinning tether, but that's . . . not established technology.

 

[RecoveringYuppy]

The ISS is in LEO, lower than all the currently deployed Starlink shells. They really aren't useful for much if they can't even reach the ISS.

OK. But there is plenty of LEO below the ISS.

Note that I don't see anyone in the thread insisting they actually have a viable market.

...no, they could not. Achieving that launch rate to a single orbit would require a huge array of Spinlaunch centrifuges.

Based on your numbers it would take 23 wouldn't it? I thought you posited 100kg/day. And those could be doing other things the rest of the day.

Since you have to send up propellant to get the tug to the next payload, and propellant to carry that propellant along with the payload to the destination, you probably are not coming ahead on propellant consumption that way.

That depends on what other choices you make. If your imperative were to optimize/reduce fuel cost you can do it.

 

[Blue Mountain]

They do that every time they send a spacecraft to the ISS.

The advantage of having a tug boat in space is that there is no need to launch it every time. Plus less junk from discarded rockets in space. The major problem would be that no maintenance would be able to be carried out on the rocket for long periods. Plus being able to refuel it.

I wonder if one could create a factory in space that would manufacture the fuel. Of course, now the issue becomes transferring the manufactured fuel from the factory to the tug while everything's moving at 7 km/s.

 

[RecoveringYuppy]

They do that every time they send a spacecraft to the ISS.

But that's at a much more relaxed pace.

 

[RecoveringYuppy]

I wonder if one could create a factory in space that would manufacture the fuel. Of course, now the issue becomes transferring the manufactured fuel from the factory to the tug while everything's moving at 7 km/s.

That's not a problem. Rendezvous and docking are well solved procedures.

The 7 km/s issue was me and dasmiller discussing a radical idea for how catapults might work.

And manufacturing the fuel in space is another potential benefit of basing our upper stages in orbit. How well it works depends on where you're getting the raw materials for the fuel.

 

[Blue Mountain]

That's not a problem. Rendezvous and docking are well solved procedures.

The 7 km/s issue was me and dasmiller discussing a radical idea for how catapults might work.

And manufacturing the fuel in space is another potential benefit of basing our upper stages in orbit. How well it works depends on where you're getting the raw materials for the fuel.

Send them up by rocket? Oh, wait ...

 

[rjh01]

That's not the scenario that inspired my remark. The concept was that Spinlaunch would launch something vertically up to LEO altitude and the tug would then bring it up to orbital speed. Assuming the tug was in orbit, the tug would have a lateral speed of 7.5 km/s and the spinlaunch payload would have a lateral speed of around 0. Either the tug would have to slow down to 0 (falling!) and then accelerate back up (an extra 15 km/s acceleration for the tug. It would take less delta-V for the tug to do a round-trip to Mars), or the tug will hit the spinlaunch payload at several km/s.

There may be some technique using a many-miles-long spinning tether, but that's . . . not established technology.

I agree the satellite would need to be in LEO for the idea to work. And you would then want the satellite to be in a higher orbit. There might not be a high demand for this service.

I wonder if one could create a factory in space that would manufacture the fuel. Of course, now the issue becomes transferring the manufactured fuel from the factory to the tug while everything's moving at 7 km/s.

Once we start mining the moon the rules start changing. There will be a demand for moving objects from LEO to the moon. On the moon, they would have the fuel to get the satellite back to LEO. It would not matter how the satellite is launched from Earth. The idea of a tug could then work.

 

[theprestige]

They do that every time they send a spacecraft to the ISS.

They do not, in fact, launch ISS deliveries vertically, and have the ISS slam into them at the orbital speed as it swings by.

Rather, they launch the delivery into orbit. Then the delivery adjusts its orbital parameters until its speed and location match the ISS,

 

[Gord_in_Toronto]

I went looking for an update.

SpinLaunch - Wikipedia

en.wikipedia.org

As of August 2024, no further test flights or other technological accomplishments have been reported by SpinLaunch since its 2022 test flight.

I guess the technical/engineering issues are proving just a little difficult.

The animated graphics on the SpinLaunch website are still as impressive as ever.

 

[Gord_in_Toronto]

Here's an analysis of the current progress (or lack thereof):

Around and around we go but not fast enough.

It's still a wonderful idea though.

 

[arthwollipot]

I discovered recently that I still have a Google Alert set up for Spinlaunch. Someone mentioned it in an article about making a giant space cannon and it made the rounds again for some reason.

 

[Darat]

I heard they were just waiting for delivery of a Brilliant Light Power generator.

 

[Ziggurat]

It's still a wonderful idea though.

It's a neat idea but I have to wonder if it's not out of date already. Spin Launch started back in 2014, when the SpaceX Falcon reusability was still unproven. Back then, beating non-reusable rockets was probably viewed as good enough for financial viability. But that's no longer the relevant benchmark. Reusable rockets have driven costs way down. Can they even compete now? I'm not sure they can.

On a more technical note, the problem I've always wondered about that I haven't seen any real details on is the release mechanism. Their launch vehicle has angular momentum about its own center when spinning, and if you instantaneously release it (a technical challenge of its own, BTW), it's still going to have that angular momentum. Which means it's going to want to spin, or tumble, as it flies out of the launch tube. We even saw this a bit with the test launch footage, where it breaks through the membrane at a bit of an angle. The fins should quickly stabilize it in air, but at the cost of losing kinetic energy from drag. The ideal solution would be to hold on to the launch vehicle in two places, release the forward part first and then release the back part a tiny bit later. During this interval, you would be applying torque to the launch vehicle to cancel out its angular momentum before it exits the spinner, so that it enters the air going straight and you minimize kinetic energy losses from the fins needing to straighten it out for you. But given how fast this thing spins, the timing on that is INCREDIBLY tight. How do you make a release mechanism that can act that fast and that precisely? Maybe magnetic couplers? I don't know. But it's a really non-trivial problem. Hell, it's non-trivial even if you just want a single attachment point.

 

[Myriad]

On a more technical note, the problem I've always wondered about that I haven't seen any real details on is the release mechanism. Their launch vehicle has angular momentum about its own center when spinning, and if you instantaneously release it (a technical challenge of its own, BTW), it's still going to have that angular momentum. Which means it's going to want to spin, or tumble, as it flies out of the launch tube. We even saw this a bit with the test launch footage, where it breaks through the membrane at a bit of an angle. The fins should quickly stabilize it in air, but at the cost of losing kinetic energy from drag. The ideal solution would be to hold on to the launch vehicle in two places, release the forward part first and then release the back part a tiny bit later. During this interval, you would be applying torque to the launch vehicle to cancel out its angular momentum before it exits the spinner, so that it enters the air going straight and you minimize kinetic energy losses from the fins needing to straighten it out for you. But given how fast this thing spins, the timing on that is INCREDIBLY tight. How do you make a release mechanism that can act that fast and that precisely? Maybe magnetic couplers? I don't know. But it's a really non-trivial problem. Hell, it's non-trivial even if you just want a single attachment point.

We discussed that a bit around post 132, but as you said, no additional details about the release mechanism were available, so not much came of it. The timing is rapid but the exact ideal interval is known in advance (given that they must have accurate readings of the centrifuge's angular velocity or else there's no hope for any of it).

Long ago as a student I did a high school physics lab using an air levitated (think air hockey table) mass on an inclined beam, magnetic release, and electronic timing, to try to get as precise a measure of G as possible. The largest cause of error turned out to be the magnetic release mechanism, because the electromagnet powered down too slowly and too variably. There are probably ways to make an electromagnet behave better, but I'd be wary. I'd suggest that the second (rear) release be purely mechanical with no moving parts (that is, fixed only to the rotor and/or the vehicle). Some kind of shaped cradle that the rear of the projectile slides clear of once the front of the projectile gets clear of the rotor.

 

[dasmiller]

On a more technical note, the problem I've always wondered about that I haven't seen any real details on is the release mechanism. Their launch vehicle has angular momentum about its own center when spinning, and if you instantaneously release it (a technical challenge of its own, BTW), it's still going to have that angular momentum. Which means it's going to want to spin, or tumble, as it flies out of the launch tube. We even saw this a bit with the test launch footage, where it breaks through the membrane at a bit of an angle. The fins should quickly stabilize it in air, but at the cost of losing kinetic energy from drag. The ideal solution would be to hold on to the launch vehicle in two places, release the forward part first and then release the back part a tiny bit later. During this interval, you would be applying torque to the launch vehicle to cancel out its angular momentum before it exits the spinner, so that it enters the air going straight and you minimize kinetic energy losses from the fins needing to straighten it out for you. But given how fast this thing spins, the timing on that is INCREDIBLY tight. How do you make a release mechanism that can act that fast and that precisely? Maybe magnetic couplers? I don't know. But it's a really non-trivial problem. Hell, it's non-trivial even if you just want a single attachment point.

Complicating it further is the fact that rotation has to be very accurately corrected despite the fact that different payloads will have different longitudinal CGs and different moments of inertia (MOIs). These things can be measured, of course, but it means you can't just do a few calibrations shots to get your release system sorted out; it'll have to adapt for different payloads.

We'll add it to the list, along with the fact the hundred-million-kg unbalanced load on the bearing (and the whole building!) after release, the fact that it doesn't actually put things in orbit (it still needs an upper stage) and as someone who did his career in spacecraft design, I don't even know where to start designing a spacecraft that can tolerate 10,000 Gs. I can tell you that it's challenging to make an efficient spacecraft that tolerates 20 Gs.

 

[Ziggurat]

Complicating it further is the fact that rotation has to be very accurately corrected despite the fact that different payloads will have different longitudinal CGs and different moments of inertia (MOIs).

I'm not sure that matters too much. There's going to be variations, sure, but within some bounds. And you don't have to get it perfect, the fins are still going to help stabilize the orientation in air. You just want to reduce how much work those fins have to do, you don't need to get it to zero.

We'll add it to the list, along with the fact the hundred-million-kg unbalanced load on the bearing (and the whole building!) after release,

That's definitely a challenge. I've wondered if it might help to simultaneously let the counterweight spool outwards, reducing the force and also slowing down the rotation.

I don't even know where to start designing a spacecraft that can tolerate 10,000 Gs.

I'm sure some payloads can be designed to handle that, but yeah, it both limits what sort of payloads can be launched and adds a design cost to them. Even for suitable missions, that still reduces competitiveness against other launch platforms.

 

[dasmiller]

I'm not sure that matters too much. There's going to be variations, sure, but within some bounds. And you don't have to get it perfect, the fins are still going to help stabilize the orientation in air. You just want to reduce how much work those fins have to do, you don't need to get it to zero.

True, but keep in mind you're starting at about 2700 deg/second and probably want to get down to more like 20 deg/sec. That's a 99% reduction, which means your total error from all the steps (MOI & CG knowledge, clamp release timing, etc) will have to be <1%.

That's definitely a challenge. I've wondered if it might help to simultaneously let the counterweight spool outwards, reducing the force and also slowing down the rotation.

I keep mulling over things like that, but keep in mind that the counterweight is starting at a few km/s. If you simply release it, of course, you'll neutralize the lateral load on the bearing, but then it'll hit something with about 70X the impact of a large battleship shell (seriously. I did the math). As for spooling it out . . . if 'spool' is literal, then bear in mind that the cable holding the counterweight is a minimum of half-a-meter in diameter (assuming really good carbon fiber and no margin for safety) so it probably doesn't roll up gracefully. And also . . . how far can it go?

I'm sure some payloads can be designed to handle that, but yeah, it both limits what sort of payloads can be launched and adds a design cost to them. Even for suitable missions, that still reduces competitiveness against other launch platforms.

Modern 'smart' artillery shells (even going back to the proximity-fused shells in WWII) have electronics that tolerate even higher loads than SpinLaunch, albeit for a very short period of time, so there are probably some things that can be done. But anything like a traditional spacecraft with solar panels and some basic optics and a momentum management system? Very difficult.

Then there's the "relatively small, low-cost upper stage." I can't imagine how you'd get a usable payload fraction from an upper stage that had to survive 10,000 Gs. Any conventional thrusters would simply collapse under their own weight; if they were solid enough to survive, they'd probably melt because they couldn't dissipate the combustion heat.

Finally, on the topic of competitiveness: SpinLaunch advertises a 200kg spacecraft that's fired as part of a 11,200 kg SpinLaunch package. That's a payload fraction of 1.8% to LEO. For comparison, a Falcon 9 has a payload fraction of more like 4%. Pound for pound, do we really think the state-of-the-art composites and release mechanisms of a SpinLaunch vehicle are going to be that much cheaper than the mostly-kerosene-and-LOX Falcon 9?

 

[Ziggurat]

I keep mulling over things like that, but keep in mind that the counterweight is starting at a few km/s. If you simply release it, of course, you'll neutralize the lateral load on the bearing, but then it'll hit something with about 70X the impact of a large battleship shell (seriously. I did the math). As for spooling it out . . . if 'spool' is literal, then bear in mind that the cable holding the counterweight is a minimum of half-a-meter in diameter (assuming really good carbon fiber and no margin for safety) so it probably doesn't roll up gracefully. And also . . . how far can it go?

The computer renders I've seen place the counterweight at about half the distance of the vehicle. If that doubles (which means you don't have to make your containment any bigger than it already is), conservation of angular momentum means tangential velocity will have halved while radius doubles. And with centrifugal acceleration being v²/r, you'll end up decreasing the off-balance load on the bearings by a factor of 8. So you don't have to spool it out very far (nor does it need to be via a cable) for it to make a significant difference. I'm sure it wouldn't be easy to do, but still might be worth doing.

Finally, on the topic of competitiveness: SpinLaunch advertises a 200kg spacecraft that's fired as part of a 11,200 kg SpinLaunch package. That's a payload fraction of 1.8% to LEO. For comparison, a Falcon 9 has a payload fraction of more like 4%. Pound for pound, do we really think the state-of-the-art composites and release mechanisms of a SpinLaunch vehicle are going to be that much cheaper than the mostly-kerosene-and-LOX Falcon 9?

Not pound for pound, no. The question is, are there customers who would pay a premium for on-demand small launches? And will they pay enough of a premium, and at enough volume, to justify Spin Launch? You can launch a 200 kg satellite on Falcon, but unless you go as part of a ride share (which imposes very significant scheduling constraints), you can't do that at anything like Falcon's best per pound price. I suspect that there aren't enough customers like that, but only looking at the best care per pound price won't tell you the whole story.

 

[arthwollipot]

Something new (3 December 2024):

Exclusive: SpinLaunch raises $11M but hoped to raise much more, source says

Space startup SpinLaunch is fundraising again, though a source tells TechCrunch that it was exploring raising a significantly more ambitious sum earlier

techcrunch.com

 

[novaphile]

I can't see any way this could beat a linear accelerator, if you're going for non-traditional acceleration to put your payload into orbit.

A mile of track on an equatorial mountain range should do the trick.

 

[dasmiller]

Not pound for pound, no. The question is, are there customers who would pay a premium for on-demand small launches? And will they pay enough of a premium, and at enough volume, to justify Spin Launch? You can launch a 200 kg satellite on Falcon, but unless you go as part of a ride share (which imposes very significant scheduling constraints), you can't do that at anything like Falcon's best per pound price. I suspect that there aren't enough customers like that, but only looking at the best care per pound price won't tell you the whole story.

Well, how will SpinLaunch's schedule compare to something like the Electron? Electron has already made over 50 flights, lifts about the same payload (225 to 320 kg for Electron, 200 kg for SpinLaunch), weighs about the same (12,500 kg for Electron vs 11,200 for SpinLaunch). Maybe SpinLaunch can do a quick turnaround on centrifugal launch part, but there's still that 2nd stage which is a conventional (if unusually robust) rocket and weighs about 2/3 of the whole Electron rocket. Is the turnaround for an 8 tonne SpinLaunch rocket (not including the 3 tonne sabot) inherently that much quicker than the turnaround for a 12 tonne Electron rocket?

 

[Ziggurat]

Well, how will SpinLaunch's schedule compare to something like the Electron?

My seat of the pants wild ass guess is that they might be able to compete with a non-reusable Electron, but if Rocket Labs gets their first stage to be reliably reusable, they probably can't. I have no hard numbers to back this up, only a gut feeling, so take that for what it's worth (ie, not much).

 

[BillC]

I can't see any way this could beat a linear accelerator, if you're going for non-traditional acceleration to put your payload into orbit.

A mile of track on an equatorial mountain range should do the trick.

No, the power requirements to accelerate a payload to launch velocity in such a short distance are absurdly large.

If we try to reach Spinlaunch's projected 2120m/s launch speed in 1600m, we need an acceleration of 143.2 g, which is manageable, but the kinetic energy of the launch vehicle (11200kg rocket+payload, plus say 1800kg launch sled) is huge: 29.2GJ. To supply this in the 1.51 seconds the sled will take to travel the track length will require 19.3GW, which is more than enough electricity to supply a country the size of the Netherlands.

 

[3point14]

No, the power requirements to accelerate a payload to launch velocity in such a short distance are absurdly large.

If we try to reach Spinlaunch's projected 2120m/s launch speed in 1600m, we need an acceleration of 143.2 g, which is manageable, but the kinetic energy of the launch vehicle (11200kg rocket+payload, plus say 1800kg launch sled) is huge: 29.2GJ. To supply this in the 1.51 seconds the sled will take to travel the track length will require 19.3GW, which is more than enough electricity to supply a country the size of the Netherlands.

I'm crap at physics.

If you increase the length of the track do those numbers come down in a linear fashion?