7 ms·
The Tyranny of the Rocket Equation (2012)
- dang 7y agoA thread from 2018: https://news.ycombinator.com/item?id=17425156 https://news.ycombinator.com/item?id=17425156 From 2014: https://news.ycombinator.com/item?id=8537390 https://news.ycombinator.com/item?id=8537390
- adrianN 7y agoIt's time we start manufacturing rockets a bit higher up in the gravity well.
- Nodraak 7y agoIt's not only about altitude (going to space is easy), it's also about horizontal velocity (staying in space is hard: 8 km/sec is a lot). I dont recall the exact number, but I think that altitude is 50% of the energy, the other 50% is horizontal velocity. So building a rocket on top of the Everest would save us maybe 5% (10 km vs 100 km altitude), that's not much
- adrianN 7y agoI was thinking at least LEO, not Everest.
- NeutronStar 7y agoThat's still only 10% less gravity, the thing is the ISS goes really fast sideways. Being high isn't enough.
- marcosdumay 7y agoLEO is as much a speed as it is a height.
- adrianN 7y agoObviously? When you want to manufacture things you probably don't want to fall back down right away.
- marcosdumay 7y agoWell, I misunderstood your gravity wheel comment for a static position too.
- Nodraak 7y agoIndeed, sorry. I focused on "higher up" while the important part was "gravity well"
- loeg 7y agoOP wasn’t talking about just altitude.
- petschge 7y agoThe split is actually much more uneven, and heavily on the side of kinetic energy. An orbit at 200 km altitude has an orbital velocity of 7.79 km/s. Potential energy is given by m * g * h (to good approximation, as h is much smaller than the Earth radius of 6300 km) and kinetic energy goes as 0.5 m v^2. Per unit mass (that appears identically in both energy forms) we have potential energy of g * h = 9.81 m / s^2 * 2e5 m = 1.96e6 m^2/s^2. For kinetic energy we find 0.5 * (7.79e3 m/s)^2 = 3e7 m^2/s^2. So only 6% of energy are potential energy, 94% are kinetic energy. For a relatively high orbit with 1500 km and 7.12 km/s the ratio becomes a more even 37% to 63%. If we include the extra 1.5km/s of delta-v that we loose to drag it becomes 28% to 72%. At the typical parameters of stage separation from the first stage the split is 3% to 97%. This is also why replacing the first stage by an airplane (that only gets you altitude, not that much speed) does buy you a lot less than you might think at first. We still need a rocket to go to space, even if you start 12km up.
- ta1234567890 7y agoThank you for the insightful and illustrative calculations :) What would happen if we were able to make it all the way up to space, vertically, but not gain any horizontal speed?
- Tuna-Fish 7y agoIt would only negate 6% of the cost of going to orbit. And you'd have to pay that 94% pretty quickly to not fall down. Except that it would be better than that, because rocket nozzles work best at a specific external pressure, and getting to launch your rocket in vacuum means you can design your engines for strictly that. There are occasionally some ideas of launching rockets from tops of tall mountains, like Kilimanjaro. The advantage directly gained from being 6km closer to space is negligible, but the advantage gained from being able to use more expanded nozzles would be substantial -- albeit likely not worth having to haul your rocket up a mountain to launch it.
- oh_sigh 7y agoYou would just fall back to earth, gaining a lot of speed and then getting destroyed by the atmosphere. Even the force of gravity where the ISS is is pretty similar to at sea level.
- drjesusphd 7y agoUsing ISS parameters, it's more like 8:1, with it taking much more energy to get to speed than altitude. "Low earth orbit is halfway to anywhere in the solar system."
- pjc50 7y agoCurrently all the stuff we want to launch is on the ground, though. Until we have a space elevator, or we're aiming to launch something huge to the outer planets, it doesn't gain us much.
- twic 7y agoJust build taller factories.
- gmiller123456 7y agoActually, it has benefits. Even if you have to launch something from Earth's surface, once in space, the fuel to move it elsewhere can be transported to meet it there much cheaper. The obvious uses are interplanetary travel, circularizing a parabolic orbit, refueling existing satellites.
- polynomial 7y agoNaive question, but don't we have to get all the parts and supplies up there in the first place?
- adrianN 7y agoa) it's easier to lift many smaller parts than to lift one big rocket b) we might start mining raw materials on asteroids or Luna.
- loeg 7y agoAsteroid mining is maybe viable.
- tomxor 7y agoYes, and how much depends how self sustaining you can make the solutions. To illustrate it with an extreme: Some initial mass of machinery must be pushed out of the significant gravity well called Earth at great cost. After which replication of more machinery, manufacturing of rockets and mining of raw materials could be done in much smaller gravity wells (e.g the moon) at significantly less cost to reach orbit again. If this is achievable, your only useful remaining mass to transport from Earth are humans, _less_ the usual long term life support equipment, the payload is relatively tiny... and if you are thinking far ahead enough even those are self replicating :)
- dredmorbius 7y agoThe general alternative falls under the acronym ISRU: in-situ resource utilisation. That's the goal of utilising materials sourced from space environments themselves (Moon, asteroids, etc.) for both structure and fuel for a spaceship. Given that space is mostly, er, space, you've got a bit of a challenge. You have to find non-space stuff that's made of the things you're interested in, and have the capacity to convert it into the forms you need. Which still represents a considerable challenge.
- rkangel 7y agoWhat that article demonstrates is that the important thing is to get fuel from a source a bit higher up the gravity well - that's the dominant mass. We 'just' need a way of making fuel from asteroids or on the moon.
- tim333 7y agoOr Mars. I think Musk has an oxygen and methane plant penciled in for there.
- loeg 7y agoGotta start asteroid mining to make that viable. Not to mention solving the difficulties of keeping humans in space healthy for your construction force (or fully automate it, I suppose).
- Gravityloss 7y agoOr at least using the propellants from shallower gravity wells. Indeed, asteroids are the best places. After that come smaller moons. The Moon, Mars, Earth are very deep and steep gravity wells and not practical. The Moon doesn't have proper elements for fuels anyway.
- WalterBright 7y agoAnother possible solution is to use an air-breathing rocket until it gets out of the atmosphere. This saves the weight of the oxidizer. There have been some experiments in this direction, notably Richard Branson's.
- chipsa 7y agoAir breathing saves wet-mass, but increases dry mass. So far, the dry mass increase has been enough to make it not worthwhile. Especially since the change in trajectory required results in much higher drag losses, and more worse thermal performance (rocket gets hotter)
- Retric 7y agoWith the notable exception of anti satellite weapons which have often been fired from military aircraft. It’s cheaper because an individual flight on an existing high altitude high velocity aircraft is cheap. Unfortunately, it would require an enormous number of space flights for this first stage to be worthwhile otherwise.
- rkangel 7y agoThe most advanced project of this nature is Skylon and its Sabre engine (https://en.wikipedia.org/wiki/Skylon_(spacecraft) https://en.wikipedia.org/wiki/Skylon_(spacecraft)).
- toraobo 7y ago> There have been some experiments in this direction, notably Richard Branson's. No? Richard Branson invested in SpaceShipTwo which uses a hybrid rocket motor with liquid oxidizer and solid propellant. It is NOT air-breathing. It's air-launched but the carrier aircraft is powered by traditional jet engines (not rockets).
- marcosdumay 7y ago> but the carrier aircraft is powered by traditional jet engines (not rockets). Hum... That's the name we give to air breathing engines.
- HNLurker2 7y agoDid the slashdot effect just happen or it is just me?
- brohee 7y agoFunny he didn't mention beaming the energy from the ground. At the time of writing (2012) NASA was experimenting with microwave transmission, but it seems they ran out of fund before succeeding... https://en.wikipedia.org/wiki/Beam-powered_propulsion https://en.wikipedia.org/wiki/Beam-powered_propulsion
- sunkenvicar 7y agoSeems there are typos in table 2 rows 4 and 5.
- JanSolo 7y agoDon Pettit is my favourite astronaut. He's easily the nerdiest guy who ever made it up to the International Space Station. The reason he's so cool is that he can present scientific topics in an easy-to-understand way that really captures the imagination. He conducts a whole bunch of personal experiments in orbit and then uploads them to Youtube. Using static electricity to spin water droplets around a knitting needle. Using surface tension to improve coffee usability. Real science, but presented with a wide-eyed innocence that shows how stoked he is to be up there doing all this cool stuff. A great asset for NASA in my opinion. They should fly him all the time!
- LifeLiverTransp 7y agoAnother thing is - to build space cannons - for industrial raw materials, to be prcoessed in orbit. Dont rocket, if you must not. https://en.wikipedia.org/wiki/Space_gun https://en.wikipedia.org/wiki/Space_gun
- Robotbeat 7y agoAn interesting aspect of this that is not mentioned here is that, fundamentally, although the amount of propellants is large compared to the payload, the vast majority of the propellant (and thus the lift-off mass) is oxygen. Liquid oxygen is extremely cheap as it just requires compressing the air (and, of course, airliners also must compress the air to burn with fuel, it's just that they compress it in-situ). Only getting 4% of your lift-off mass to orbit sounds like a terrible deal, but it's really not so bad when you consider the vast majority of your lift-off mass costs just 5 cents per pound, and almost all the rest costs about 10-30 cents per pound. Near the limit of rocket performance would be the 2016 SpaceX ITS (now Starship) proposal which had an expendable capability of about 550 tons of payload for about 8500 tons of propellant (of which about 20% was fuel). That's a fuel to payload ratio of 3:1. (unfortunately, reusable performance is nearly half as bad, so about 5 or 6:1). https://www.spacex.com/sites/spacex/files/making_life_multiplanetary_2016.pdf https://www.spacex.com/sites/spacex/files/making_life_multip... For long-haul aircraft carrying cargo, the situation isn't that much better. Near the edge of their range, cargo airliners have about 2-4:1 fuel to payload capability. And considering that rockets are now starting to use LNG for fuel (often significantly cheaper than jet fuel), the fuel cost for Starship (or similar vehicles) could actually be less than that for a long-haul airfreighter on very long routes. In fact, for single-trip flights half-way across the world (i.e. at the edge of the capability of modern aircraft), the fuel:payload ratio for ITS (Starship) or an airliner would be about the same, might be even better for Starship/ITS. In part that's because launch vehicles stage, which makes them extremely efficient. And Don is a fantastic guy, but I also think he exaggerates slightly the engineering in rocketry vs other fields. The cost per unit dry mass of an airliner and a launch vehicle is approximately the same. The factors of safety are also similar (although usually rockets don't have high cycle requirements...). And in some ways, rockets are simpler structurally as their typical load case is pressurization and axial loading (which is often in the same direction). That means they can use relatively inexpensive thin-gauge stainless steel construction in ways that the more complexly-loaded aircraft cannot. The main issue is we just throw away rockets for the most part. Fix that, and we don't need exotic nuclear propulsion or anything to have low costs to achieve orbit. There is one figure of merit that rockets tend to do much better at: Final dry mass to payload. Because rockets stage off the vast majority of their mass early in flight, only a small part of the rocket dry mass has to go through the majority of the delta-v. A rocket upper stage may be a fourth to a fifth the mass of the payload. With airplanes, the dry mass of the airplane essentially always is greater than the payload mass, and on longer flights this might be 2:1 or greater.