5 ms·

I agree with this. But Jeff makes a valid point about judging the rocket by its smaller size. The launch of orbital vehicles is a solved problem, but landing th

I agree with this. But Jeff makes a valid point about judging the rocket by its smaller size. The launch of orbital vehicles is a solved problem, but landing the booster is newly under development. The smaller booster is harder to land vertically, so the technological solutions they are testing may prove to be more robust in adverse conditions.

jpgvm 11y ago
I don't think the smaller booster is harder. It's almost by definition easier. It weighs less, doesn't need to reach orbital velocity and can have an engine that throttles all the way down to hover because the MTOW is not that different from it's dry weight. Not reaching orbital velocity is the main thing here, it's what makes all of this easy in comparison. The solutions they are exploring are many times less complicated and difficult than what SpaceX is doing. Saying otherwise is mischaracterising both of their achievements.
- Zikes 11y ago
  > And the vertical landing architecture scales extraordinarily well. When you do a vertical landing, you’re solving the classic inverted pendulum problem, and the inverted pendulum problem gets a bit easier as the pendulum gets a bit bigger. Try balancing a pencil on the tip of your finger. Now try it with a broomstick. The broomstick is simpler because its greater moment of inertia makes it easier to balance. They're specifically doing smaller rockets to tackle the inverted pendulum problem first, then they will scale up. They see that as the bigger challenge, and may greatly affect the design of larger orbital rockets which have a great deal of research available to draw from for other challenges they run into. In the article they say that one of their major solutions is a gimbaling engine, which I imagine is something you would have to design a larger rocket around, rather than add in after the fact.
  - double0jimb0 11y ago
    >> I imagine is something you would have to design a larger rocket around, rather than add in after the fact. Control authority. That is the measure of margin a system has over a feedback mechanism's (in case of a rocket: ability to accurately apply thrust in correct vector at required frequency) ability to dampen uncontrolled perturbations. I'd bet you that SpaceX has just as thoroughly understood and tested their working margins on their landibg system's control authority as has Blue Origin. For this reason I don't believe Blue Origin has any technical advantage over SpaceX in the manner you imply they do. This might be hubris, but there aren't too many things us humans don't understand about rocket design at this point. The last big one was the waterhammer effect in turbo pumps supplying rocket fuel. Understanding 1st, 2nd, and 3rd harmonic vibrations in mechanical systems is standard practice in rocket and any other airframe design (Boeing airframes, missiles/rockets, etc). It's a solved problem. If Bezos is hanging his hat on this argument, he's reached the limit of his omnipotence.
    - Zikes 11y ago
      I'm not implying any advantage, only reiterating that they're starting small and building up. My reply was aimed at "I don't think the smaller booster is harder. It's almost by definition easier." which the article explicitly said the opposite of, which I then quoted. I don't care if SpaceX or Blue Origin wind up on top, I don't own stock in either one, but I'm much more likely to trust the word of an actual certified rocket scientist than an unsourced HN comment.
- zardo 11y ago
  The dynamics problem of balancing an inverted pendulum gets easier as the distance between the COM and the force input is increased. Of course, its also easier to have very fast gimballing on a smaller engine, and a taller rocket bends more which makes it harder to detect exactly where it's pointed in the first place.
  - NamTaf 11y ago
    A rocket isn't an inverted pendulum. This is a common mistake. https://en.wikipedia.org/wiki/Pendulum_rocket_fallacy https://en.wikipedia.org/wiki/Pendulum_rocket_fallacy
    - gozur88 11y ago
      The pendulum fallacy has to do with the tail of the rocket not naturally following the head during launch. Landing really is an inverted pendulum problem.
      - jakub_h 11y ago
        How is even the landing an inverted pendulum problem? There's no instability of an inclined rocket during the landing; there's an instability of a "gimbaled-away" rocket. That's not the same thing.
        gozur88 11y ago
        I'm not sure what you mean by that. How is this different, mathematically, from the classic 8th grade textbook picture of a child balancing a yardstick on his nose?
        jakub_h 11y ago
        Easily. The yardstick is unstable in inclination if it isn't vertical. The rocket is unstable in horizontal position if it isn't vertical. But it doesn't start to rotate as long as the thrust vector goes through the CG. The thrust vector has stable orientation relative to the rocket (if you don't gimbal), whereas the force vector doesn't have stable orientation relative to the yardstick.
avmich 11y ago
Landing a booster was done, e.g., in Delta Clipper program, early 1990's.
rakoo 11y ago
> landing the booster is newly under development. They're already one step further though: actually reusing the rocket

jpgvm 11y ago

I don't think the smaller booster is harder. It's almost by definition easier. It weighs less, doesn't need to reach orbital velocity and can have an engine that throttles all the way down to hover because the MTOW is not that different from it's dry weight. Not reaching orbital velocity is the main thing here, it's what makes all of this easy in comparison. The solutions they are exploring are many times less complicated and difficult than what SpaceX is doing. Saying otherwise is mischaracterising both of their achievements.

Zikes 11y ago
> And the vertical landing architecture scales extraordinarily well. When you do a vertical landing, you’re solving the classic inverted pendulum problem, and the inverted pendulum problem gets a bit easier as the pendulum gets a bit bigger. Try balancing a pencil on the tip of your finger. Now try it with a broomstick. The broomstick is simpler because its greater moment of inertia makes it easier to balance. They're specifically doing smaller rockets to tackle the inverted pendulum problem first, then they will scale up. They see that as the bigger challenge, and may greatly affect the design of larger orbital rockets which have a great deal of research available to draw from for other challenges they run into. In the article they say that one of their major solutions is a gimbaling engine, which I imagine is something you would have to design a larger rocket around, rather than add in after the fact.
- double0jimb0 11y ago
  >> I imagine is something you would have to design a larger rocket around, rather than add in after the fact. Control authority. That is the measure of margin a system has over a feedback mechanism's (in case of a rocket: ability to accurately apply thrust in correct vector at required frequency) ability to dampen uncontrolled perturbations. I'd bet you that SpaceX has just as thoroughly understood and tested their working margins on their landibg system's control authority as has Blue Origin. For this reason I don't believe Blue Origin has any technical advantage over SpaceX in the manner you imply they do. This might be hubris, but there aren't too many things us humans don't understand about rocket design at this point. The last big one was the waterhammer effect in turbo pumps supplying rocket fuel. Understanding 1st, 2nd, and 3rd harmonic vibrations in mechanical systems is standard practice in rocket and any other airframe design (Boeing airframes, missiles/rockets, etc). It's a solved problem. If Bezos is hanging his hat on this argument, he's reached the limit of his omnipotence.
  - Zikes 11y ago
    I'm not implying any advantage, only reiterating that they're starting small and building up. My reply was aimed at "I don't think the smaller booster is harder. It's almost by definition easier." which the article explicitly said the opposite of, which I then quoted. I don't care if SpaceX or Blue Origin wind up on top, I don't own stock in either one, but I'm much more likely to trust the word of an actual certified rocket scientist than an unsourced HN comment.
zardo 11y ago
The dynamics problem of balancing an inverted pendulum gets easier as the distance between the COM and the force input is increased. Of course, its also easier to have very fast gimballing on a smaller engine, and a taller rocket bends more which makes it harder to detect exactly where it's pointed in the first place.
- NamTaf 11y ago
  A rocket isn't an inverted pendulum. This is a common mistake. https://en.wikipedia.org/wiki/Pendulum_rocket_fallacy https://en.wikipedia.org/wiki/Pendulum_rocket_fallacy
  - gozur88 11y ago
    The pendulum fallacy has to do with the tail of the rocket not naturally following the head during launch. Landing really is an inverted pendulum problem.
    - jakub_h 11y ago
      How is even the landing an inverted pendulum problem? There's no instability of an inclined rocket during the landing; there's an instability of a "gimbaled-away" rocket. That's not the same thing.
      - gozur88 11y ago
        I'm not sure what you mean by that. How is this different, mathematically, from the classic 8th grade textbook picture of a child balancing a yardstick on his nose?
        jakub_h 11y ago
        Easily. The yardstick is unstable in inclination if it isn't vertical. The rocket is unstable in horizontal position if it isn't vertical. But it doesn't start to rotate as long as the thrust vector goes through the CG. The thrust vector has stable orientation relative to the rocket (if you don't gimbal), whereas the force vector doesn't have stable orientation relative to the yardstick.

avmich 11y ago

Landing a booster was done, e.g., in Delta Clipper program, early 1990's.

rakoo 11y ago

> landing the booster is newly under development. They're already one step further though: actually reusing the rocket