> A terrestrial rocket has to push through a plug of air equivalent to a 30-foot column of water, and physics dictates that the smallest vehicle capable of moving all that atmospheric mass without paying a penalty in momentum is about 30 feet long.
This doesn't make sense - they're talking about momentum, but then conflating that with a length rather than a mass
The momentum to length thing comes from Newton's approximation for impact depth, which basically states that the distance a high speed projectile can penetrate something is a function of its density and length. Something like an asteroid or needs a certain minimum length to make it through the atmosphere without slowing down below the speed of sound.
It has nothing to do with rockets though, as rockets are not high velocity projectiles - they are powered vehicles and most of the time they are in the atmosphere they're not even going that fast. For space launch this condition would only be relevant if you're being shot out of a massive cannon or something.
Excluding the atmosphere, delta vee is exhaust velocity times the logarithm of the mass ratio. The mass of a pressure vessel is proportional to its volume. So if you double the propellant, also tank volume doubles and also tank mass doubles. No clear scale related terms there. Gravity also is not affected by scale, gravity losses are strictly related to mass.
But air resistance is different. It is a surface effect and thus related to scale. It's roughly proportional to the frontal area of the rocket. Hence, if you grow a rocket and keep its shape intact, the air resistance grows only as square, but the other factors like mass grow as cube. So bigger rockets have a smaller proportion of their delta vee spent on air resistance losses.
That's also why sounding rockets are so skinny. Air resistance dominates there. (They are also accelerating faster since they are rail launched and don't have active gimbal guidance, which makes it even worse)
You can think of a medium sized orbital rocket, and then if you shrink it, the only way to keep its payload to total mass rate the same is to keep the air resistance losses same related to the total mass - ie keep its original length and only shrink it sideways.
There isn't really a theoretical minimum size, but with the best chemical propulsion possible you're going to have an initial to final mass ratio of about 5. Now that final mass needs to include the structure of the vehicle, so you can cheat a little bit with staging but you can't make it go to zero.
In practice, the smallest orbital launch vehicle ever launched required 2600 kg to launch a 4 kg payload. Turns out that the smaller a rocket is, the less mass efficient it gets.
I'm not sure there is an "oversupply" of small launch, most of these companies haven't yet had a successful launch and probably will run out of money before they do!
It tickles me that even the smallest operator on this list could launch me (in a suitable person-shaped container) into space, and some operators could also even recover me before I hit the ground.
Almost all the energy/velocity needed for orbiting is "sideways" to the earth. Imagine needed to throw a baseball fast enough that it goes fast enough that the barely visible curve of earth falling away the same as the dropping of the baseball from gravity.
The ISS moves around the Earth at approximatle 28,800 kilometers per hour (17,900 mph).
For comparison, a bullet fired in a vacuum at only 2,000 kilometers per hour will reach about 15km up, or a reasonable ballon launch height.
Just to be clear, space vehicles are not "hypersonic". They travel these speeds outside of substantial atmosphere, where the speed of sound is undefined. Mach number is relative to local air conditions, ergo their Mach number isn't 25. It's undefined. Mach Number is a meaningless quantity. It's not accurate to say that a space vehicle travels Mach 25, only that it's speed is 25 times higher than the speed of sound at sea-level conditions. Saying this can be a useful reference to give context to the insane speed of orbital vehicles (as was done by the commenter above you).
I say all this to give context about the surge in hypersonic research in recent years as being something distinctly different than achievements made in space-vehicles. It's not really accurate to say it's been given the "buzzword" treatment simply because things in space already move so fast. There's a distinct advantage (particularly for military objectives) to flying above Mach 5 in the atmosphere.
Using a simple back of the envelope calculation on energy requirements can tell you why. A 500km orbital altitude requires about 7.66 km/s of velocity. That's 29,337 kJ of kinetic energy per kilogram, and only 4,905 kJ per kilogram of potential energy. You don't perform an air-launch to for the potential energy savings: you do it for the reduced air-loads and energy loss due to drag from beginning your trajectory in a thinner atmosphere. This trade-off is almost never worth the added engineering complexity, as seen by vertical launch completely dominating the launch space compared to air-launch.
As for a balloon launch, well, balloons have absolutely dismal maximum payloads, so there's no way you're going to be able to hoist a vehicle with enough propellant to reach upwards of 7 km/s underneath a balloon.
That savings has to be worth the building of expensive infrastructure at altitude, which is unlikely to happen. Air density at 10k feet yields is only 25% less than sea-level, so you're not going to see substantial performance gains. It's much better to focus efforts on choosing a launch location as close to the equator as possible where you can shoot due East (to take advantage of Earth's rotation) without flying over-top of people. This will yield much greater energy savings. There's a good reason why the world's launch locations are where they are.
you get more savings being near the equator than being high. if you can manage both you get a very small win that might not be worth it if you are far from infrastructure. better to add a little more fuel than make the satellite get shipped to your remote equatorial mountain
Excuse me but I really appreciate your replies, so my reading of what your saying is the launch is less about reaching a height but instead reaching a velocity that can escape earth’s gravity and launching from a higher altitude just decreases the distance the rocket has time to reach this speed?
Things in orbit like satellites and the moon have not escaped earth's gravity at all. There's lots of gravity in orbit. Being in orbit is when the centripetal force from an object's tangential velocity matches the force of gravity. In other words, get thrown so fast that the arc of your trajectory continually misses the Earth as the surface curves away from you.
The only reason why you need to go up at all is the pesky atmosphere works to slow you down, making it difficult to maintain such high speeds. Satellites have to be up high enough that drag becomes less of an issue. If it weren't for the atmosphere, you could achieve orbit 10 feet off the ground if you moved fast enough.
This is a concept known as a rockoon[0]. Unfortunately high altitude balloons become really big really quickly, and even if you were starting at orbital altitude you would still need a decent sized rocket to reach orbital velocity.
There's actually a small Spanish company, Zero 2 Infinity, pursuing that seriously for small satellite launch and I was sort of surprised not to see them mentioned in the article. As other replies mention, you don't save much in terms of overall energy by launching form higher up. However, there are three big pluses to launching from high up in the atmosphere.
First, rocket engines are more efficient when they run in a vacuum than when they run at sea level. With SpaceX's Merlin rockets one running at sea level has an ISP of 282 but a vacuum optimized Merlin running in space has an ISP of 348 giving it almost a quarter more impulse for the same fuel expenditure.
Second, you don't waste fuel fighting drag as you work your way up through the atmosphere. That can easily be 100s of m/s of delta-v wasted.
Finally, launching higher up means saving weight on the rocket structures. You can make your rocket more nearly spherical and save weight on tank area. You can use a lighter tank structure since air resistance forces aren't an issue. And you can light your stages in parallel rather than series because they don't have to be stacked on top of each other or be optimized for different pressures.
It's been interesting to see the number of small-launch companies increase over the past few years. I'm not sure if the market will support the number of companies currently vying for a slice . I'm curious to see which ones are able to make it and which ones won't.
Here's an idea...maybe companies could play "nice" together...and have like a rocket "meeting place" where rockets basically go after they deliver payload. Then spacex's BFR goes up, some robot collects them all inside the ship and loads them...brings them home.... saving lots of money for everyone and spacex maybe pockets a retrieval fee per rocket based on weight or something....
This is assuming small rockets are also expensive and could save money if they could reuse them.
Most of the rockets listed are 2-stage rockets. The first stage never makes it to orbit - it just falls back down to earth after dropping off the second stage.
The second stage often does make it to orbit, where it deploys the payload and then usually becomes orbital debris. A few upper stages (falcon 9, for example) do a small engine burn to reenter the atmosphere where they burn up.
Collecting upper stages would be... hard. Every rocket is on a different trajectory. Making some vessel that could 'collect' them would have to visit every launch trajectory that they'd each taken, which is a lot of orbital maneuvering and coordination. Then the problem of docking and collecting the stages and packaging them for reentry would be a whole series of ridiculously tough engineering challenges.
The upper stage is much smaller than the first stage, so it's usually significantly cheaper than the first stage.
Basically, recovery of the upper stage is much harder than recovery of the first stage and has less upside.
Still no backers for my "Big Arse Slingshot to Orbit" idea, alas.
We might just be getting to the point where useful functions can be packaged small and tough enough to survive a cannon type launch. Artillery to orbit might be made to work today.
I'm not convinced that Spin-Launch is anything but vapour-ware. At risk of sounding like an arm-chair engineer, the G-forces experienced by the payload during it's rotational-phase must be astronomical. Not to mention the aero and thermal loads of slamming into the thick lower atmosphere at a substantial fraction of orbital velocity. The aerospace startup industry is no stranger to dumb ideas getting inexplicably large amounts of start-up capital, so capital itself isn't a reliable measure of their engineering expertise.
We shall see though.
EDIT: g-force calculation in a centrifuge is velocity^2 / radius. A 1st stage probably needs at least 4 km/s. A 10 km radius centrifuge yields a gigantic 1600 gs at that speed. So they're either building a centrifuge the size of a medium sized city, or their stage 1 speed is so slow that why bother with the added engineering complexity of this portion? This is without even considering the thermal and aero problem that emerges with high-speed flight at sea-level upon exiting the centrifuge. I think I'd keep my wallet closed on this one if I was an investor. Is there something I'm missing somehow here?
I agree, it's hard to imagine a payload that could withstand the g-forces required to reach space through a centrifuge launch. I think what interests investors is that in the low likelihood they succeed, they could create an order-of-magnitude reduction in launch costs. Some investors seem to have decided that was worth the bet
This is like saying that in the low likelihood that someone succeeds in jumping across the Atlantic ocean in a single leap, they will bring about a revolution in distance jumping. But while I don't think we've seen the limits of human performance in athletics yet, jumping oceans is not happening soon. Neither is centrifuging space payloads into orbit, because the engineering problems are simply too many magnitudes beyond what modern materials can handle.
If I were generous, investors in this company are taking a bet on the team and hoping that an eventual pivot pays off. If I wanted to not be generous there is probably some joke about Theranos, uBeam and FOMO in there.
The reason these guys are in deep trouble is SpaceX. Current pricing is 1 megabuck per 200 kg for small customers (5 kUSD/kg). That's today's price on a rocket with a substantial launch history of success.
Third-party observers estimate that SpaceX's own cost is more like 1 kUSD/kg. So, they've got some room to move their prices if the competition gets close.
But the competition isn't even remotely close, at 2-3x the price. Astra's aspirational goal is 5 kUSD/kg, and 12.5 kUSD/kg initially. Alpha: 15 kUSD/kg. Vega: 15 kUSD/kg. They are default-dead.
I think there's room for both. Going on a $ per kg metric is drastically oversimplifying the other factors that go into launch vehicle selection.
SpaceX rideshare is great if you have a mission that you want to launch to a popular orbit or an orbit near enough to a popular one that you can close the gap with on-orbit propulsion. However, the cost savings come with disadvantages. You don't get to pick your launch date, and if you miss your launch, SpaceX isn't waiting for you. It could be a long time before the next rideshare that's going to the orbit you want is available.
If you want an oddball orbit, or you want the flexibility to set your own launch date and maybe delay it if you have program delays, the smallsat launchers could make a lot of sense even if they are 5-10x the cost on a per kilogram basis.
Space missions are really expensive, and it's not the launch that's the driver of that anymore - it's the engineering, customization, and other overhead involved with developing and operating a spacecraft. You can (and should) use off-the-shelf hardware such as a finished satellite bus from somebody like SSTL or Blue Canyon, but you still have to develop the concept of operations, requirements, payload, operations plan, operations infrastructure, and staff the operating team (which is often a 24/7 endeavour, or at least an 8/7 endeavour with on-call night staff).
To put it another way: if your total lifetime mission cost is going to be $50M-75M (not an unrealistic figure for a novel SmallSat mission) and you're looking at $1M for SpaceX versus $7M for RocketLab (to use the numbers from the article), the answer may not be immediately obvious. There may be serious advantages to a dedicated launch at the cost of growing the program budget by about 10%. It's totally situational.
Not really. Astra's (and the entire small-launch market's) niche is different than SpaceX's. The focus of small-launch vehicle companies is not cost-per-kg, it's cost-per-launch. The Ride-Share model for large launch vehicles works great, until launch flexibility, short lead times, and unique destination orbits become key customer objectives. Or to put it another way, SpaceX is like the US Postal Service. Small-launch vehicle companies are like FedEx. Small-launch vehicles are the couriers to space.
It's a great point that these small launch vehicles will never be able to compete with larger vehicles like Falcon 9 on cost/kg. It does seem to me that launch factors besides costs (such as those described by others on this thread) leave room for small rocket players. I think a big question is how many players can exist in that small, dedicated launch market. It's also exciting to see how some of the small launch companies are exploring ways to differentiate themselves to carve out their own protectable segment of the market.
Author here! So exciting to see my newsletter on HN! Thank you for taking the time to read, I really appreciate it. Happy to answer any questions or respond to comments as well!
44 comments
[ 6.0 ms ] story [ 104 ms ] thread[1] https://www.airspacemag.com/space/the-one-pound-problem-7188...
This doesn't make sense - they're talking about momentum, but then conflating that with a length rather than a mass
It has nothing to do with rockets though, as rockets are not high velocity projectiles - they are powered vehicles and most of the time they are in the atmosphere they're not even going that fast. For space launch this condition would only be relevant if you're being shot out of a massive cannon or something.
https://en.wikipedia.org/wiki/Impact_depth
But air resistance is different. It is a surface effect and thus related to scale. It's roughly proportional to the frontal area of the rocket. Hence, if you grow a rocket and keep its shape intact, the air resistance grows only as square, but the other factors like mass grow as cube. So bigger rockets have a smaller proportion of their delta vee spent on air resistance losses.
That's also why sounding rockets are so skinny. Air resistance dominates there. (They are also accelerating faster since they are rail launched and don't have active gimbal guidance, which makes it even worse)
You can think of a medium sized orbital rocket, and then if you shrink it, the only way to keep its payload to total mass rate the same is to keep the air resistance losses same related to the total mass - ie keep its original length and only shrink it sideways.
In practice, the smallest orbital launch vehicle ever launched required 2600 kg to launch a 4 kg payload. Turns out that the smaller a rocket is, the less mass efficient it gets.
Is there a market for this?!
The ISS moves around the Earth at approximatle 28,800 kilometers per hour (17,900 mph).
For comparison, a bullet fired in a vacuum at only 2,000 kilometers per hour will reach about 15km up, or a reasonable ballon launch height.
For some reference, "hypersonic" is defined as "greater than Mach 5", and Google says the fastest rifle muzzle velocity is under Mach 4.
I say all this to give context about the surge in hypersonic research in recent years as being something distinctly different than achievements made in space-vehicles. It's not really accurate to say it's been given the "buzzword" treatment simply because things in space already move so fast. There's a distinct advantage (particularly for military objectives) to flying above Mach 5 in the atmosphere.
As for a balloon launch, well, balloons have absolutely dismal maximum payloads, so there's no way you're going to be able to hoist a vehicle with enough propellant to reach upwards of 7 km/s underneath a balloon.
The only reason why you need to go up at all is the pesky atmosphere works to slow you down, making it difficult to maintain such high speeds. Satellites have to be up high enough that drag becomes less of an issue. If it weren't for the atmosphere, you could achieve orbit 10 feet off the ground if you moved fast enough.
[0] https://en.wikipedia.org/wiki/Rockoon
First, rocket engines are more efficient when they run in a vacuum than when they run at sea level. With SpaceX's Merlin rockets one running at sea level has an ISP of 282 but a vacuum optimized Merlin running in space has an ISP of 348 giving it almost a quarter more impulse for the same fuel expenditure.
Second, you don't waste fuel fighting drag as you work your way up through the atmosphere. That can easily be 100s of m/s of delta-v wasted.
Finally, launching higher up means saving weight on the rocket structures. You can make your rocket more nearly spherical and save weight on tank area. You can use a lighter tank structure since air resistance forces aren't an issue. And you can light your stages in parallel rather than series because they don't have to be stacked on top of each other or be optimized for different pressures.
https://www.zero2infinity.space/
https://copenhagensuborbitals.com/
https://en.wikipedia.org/wiki/Copenhagen_Suborbitals#History
1st stage test went 9 months ago.
https://www.linkedin.com/company/c6-launch-systems/?original...
It's been interesting to see the number of small-launch companies increase over the past few years. I'm not sure if the market will support the number of companies currently vying for a slice . I'm curious to see which ones are able to make it and which ones won't.
This is assuming small rockets are also expensive and could save money if they could reuse them.
The second stage often does make it to orbit, where it deploys the payload and then usually becomes orbital debris. A few upper stages (falcon 9, for example) do a small engine burn to reenter the atmosphere where they burn up.
Collecting upper stages would be... hard. Every rocket is on a different trajectory. Making some vessel that could 'collect' them would have to visit every launch trajectory that they'd each taken, which is a lot of orbital maneuvering and coordination. Then the problem of docking and collecting the stages and packaging them for reentry would be a whole series of ridiculously tough engineering challenges.
The upper stage is much smaller than the first stage, so it's usually significantly cheaper than the first stage.
Basically, recovery of the upper stage is much harder than recovery of the first stage and has less upside.
We might just be getting to the point where useful functions can be packaged small and tough enough to survive a cannon type launch. Artillery to orbit might be made to work today.
https://en.wikipedia.org/wiki/Space_gun
We shall see though.
EDIT: g-force calculation in a centrifuge is velocity^2 / radius. A 1st stage probably needs at least 4 km/s. A 10 km radius centrifuge yields a gigantic 1600 gs at that speed. So they're either building a centrifuge the size of a medium sized city, or their stage 1 speed is so slow that why bother with the added engineering complexity of this portion? This is without even considering the thermal and aero problem that emerges with high-speed flight at sea-level upon exiting the centrifuge. I think I'd keep my wallet closed on this one if I was an investor. Is there something I'm missing somehow here?
If I were generous, investors in this company are taking a bet on the team and hoping that an eventual pivot pays off. If I wanted to not be generous there is probably some joke about Theranos, uBeam and FOMO in there.
Third-party observers estimate that SpaceX's own cost is more like 1 kUSD/kg. So, they've got some room to move their prices if the competition gets close.
But the competition isn't even remotely close, at 2-3x the price. Astra's aspirational goal is 5 kUSD/kg, and 12.5 kUSD/kg initially. Alpha: 15 kUSD/kg. Vega: 15 kUSD/kg. They are default-dead.
SpaceX rideshare is great if you have a mission that you want to launch to a popular orbit or an orbit near enough to a popular one that you can close the gap with on-orbit propulsion. However, the cost savings come with disadvantages. You don't get to pick your launch date, and if you miss your launch, SpaceX isn't waiting for you. It could be a long time before the next rideshare that's going to the orbit you want is available.
If you want an oddball orbit, or you want the flexibility to set your own launch date and maybe delay it if you have program delays, the smallsat launchers could make a lot of sense even if they are 5-10x the cost on a per kilogram basis.
Space missions are really expensive, and it's not the launch that's the driver of that anymore - it's the engineering, customization, and other overhead involved with developing and operating a spacecraft. You can (and should) use off-the-shelf hardware such as a finished satellite bus from somebody like SSTL or Blue Canyon, but you still have to develop the concept of operations, requirements, payload, operations plan, operations infrastructure, and staff the operating team (which is often a 24/7 endeavour, or at least an 8/7 endeavour with on-call night staff).
To put it another way: if your total lifetime mission cost is going to be $50M-75M (not an unrealistic figure for a novel SmallSat mission) and you're looking at $1M for SpaceX versus $7M for RocketLab (to use the numbers from the article), the answer may not be immediately obvious. There may be serious advantages to a dedicated launch at the cost of growing the program budget by about 10%. It's totally situational.
> the smallsat launchers could make a lot of sense even if they are 5-10x the cost on a per kilogram basis.
In magnitude I think this is way too bullish.