Jump to content

Bunsen

Members
  • Posts

    152
  • Joined

  • Last visited

Everything posted by Bunsen

  1. The shift in center of thrust usually falls within the engines' gimbal range -- the remaining engines just readjust so the net thrust still points through the center of mass. It's more complicated if it happens early in the flight, because that shift also changes the angle of attack and then you have to account for aerodynamic effects too. On an outboard booster, propellant leveling might require shutting down a counterpart on the opposite booster, but maybe there's enough control in the crossfeed system to handle that. The other problem is the risk of an engine failing violently enough to damage its neighbors or their plumbing. I know the Falcon 9 has some compartmentalization to address this. I don't know how violent an event it can handle, nor how far down the probability scale such events are.
  2. So the best we can hope for, even if SpaceX manages to cut the cost of a Falcon 9 launch by 50% (which I doubt) is a 10% savings for the average space project. But there are feedback effects within that cost structure. Make putting mass in orbit cheap, and the rest of the engineering balance shifts away from custom hyper-optimized designs toward simpler manufacturing, wider design margins, easier analysis, simpler testing, more parts commonality, more mass production, et cetera. Busses get cheaper. Payloads get slightly cheaper. Integration gets cheaper. Ground systems probably don't change much. It will certainly take time for the industry to adapt to those ideas, and some areas will have less to gain than others, but those other "fifths" won't just stay fixed.
  3. Yeah, it only makes sense if the angular rate sensors aren't the only source of attitude information. It seemed to be exerting correct proportional control, but the differential part was backwards. That would require having an absolute attitude reference (horizon sensor, maybe?) that it trusted for knowing its orientation, but using the rate gyros as the dominant source of angular velocity information. I can see how that would be useful; gyros could give more precise instantaneous readings of angular rates, and the absolute reference would let you null out integration errors. I would tend to think that a huge conflict between the model and a sensor would ring some alarm bells and tell the filter to ignore that sensor, but I don't write those programs so I don't really know. Maybe they depend on redundancy to spot errors, and only notice if there's disagreement between the same type of sensor. But if they all get installed by the same crew...
  4. Mine was launched on a Sunday Punch-based rocket, with some other add-on leg parts. It was mostly a success.
  5. Anatoly Zak is corroborating that, but doesn't give an original source: http://www.russianspaceweb.com/proton_glonass49.html#culprit It certainly fits the otherwise-unexplained aspects of the rocket's behavior, though. Specifically,
  6. Can't speak for the rest, but I'm just jealous of anybody with a computer that can handle those things. I've also heard that multiple docking ports like that are of limited use -- only one of them will actually dock, the others will just be magnet-forced toward each other but fairly free to slip around. I haven't tried this myself, as my old-ass computer gets sluggish long before I find the need to multiply ports like that.
  7. When you buy your parts from Jebediah Kerman's Junkyard and Spaceship Parts Co., they usually don't work until you spend a while fixing them up.
  8. I don't think that's the reason. Within a planet's sphere of influence, the physics engine treats that planet as the center of the universe and moves the rest of the solar system around it.
  9. It can be done, but data throughput becomes a major constraint and requires huge amounts of hardware at the Earth end. The plans I've heard of call for buying time on radio telescopes and using very low data rates. The alternative is to take advantage of large, long-lived spacecraft near the same destinations and use them as repeaters, which is already happening at Mars (not for cubesats, but you get the idea). While that does you no good for exploring totally new places like asteroids or Neptune, it could let you send instruments to places like Saturn or Mars without each one going through the multi-decade development cycle of a major probe.
  10. There are two reasons for that. One, they don't use propellant crossfeed -- the outer tanks are only feeding the outer engines, and the center core lasts longer because it runs at reduced throttle for much of the time. Two, real rockets tend to have higher propellant fractions than Kerbal hardware (about 0.88 for the Delta IV CBC), so they can generally burn longer per stage. The fact that it's burning LH2/LOX, which gives really high Isp, also helps with that.
  11. I'm not so sure that the giant black rectangles are still necessary for roll monitoring. Look at modern carrier rockets -- the Atlas V and Delta IV, in their single-core SRB-less variants, as well as the Falcon 9 and a bunch of smaller vehicles, don't appear to have roll-distinguishable features beyond the logos. Either they trust the on-board systems more now (probably not) or the ground tracking is good enough to work with rather small markings. I agree with Nibb31 about the patterns being strictly aesthetic.
  12. If your velocity measurement is in "surface" mode and your prograde velocity indicator is dead at the top of the navball, then you're co-rotating with the planet. Pointing straight up during launch keeps you pretty close to that. The atmosphere is also rotating with the planet, so you don't lose any of that free delta-v by minimizing drag. And on a vertical launch, I think the aerodynamics keep you nearly co-rotating with the atmosphere, but the Coriolis-induced lag at 10 km altitude would only be something like 3 m/s so it's almost impossible to notice. The problem with immediately pitching over is that you spend a lot of the early flight just fighting gravity and drag, so you can accumulate horizontal velocity pretty quickly while your vertical velocity doesn't change much. Sometimes that ends with struggling to keep the rocket's nose pointed high enough against the horizontal relative wind. That's less of an issue if you maintain a good TWR and don't overdo the aerodynamic stability. A perfect launch trajectory would include some immediate pitch, but the ideal angle is so tiny it's very hard to get it right by hand. Going straight up a few kilometers until the correct pitch angle is big enough to see costs almost nothing in delta-v, so most of us do it the slightly dumber and much easier way (though personally I find 10 km to be a bit much, 5 or 6 km is usually workable). Almost all guided rockets have roll control; even if there are no roll maneuvers it's necessary to prevent small aerodynamic or thrust asymmetries from inducing spin (just look at what happens in KSP with side mounted boosters and not enough struts). Atlas had a pair of little (heh, "little" means only a thousand pounds of thrust each) vernier engines with wide-angle gimbals, plus gimbaling of the outboard engine nozzles while they were still attached, and Titan had 2-axis gimbal control of each of its nozzles.
  13. Just make sure you have one of these watching
  14. Looks like that came from an Interfax wire piece (one example of the whole article). But, um, does that smell like a complete load to anybody else? The rocket had enough control authority to change direction twice, and enough thrust that it was not climbing perceptably slower than a normal launch (as best I can measure by watching youtube videos anyway -- any difference in the travel at 5 seconds is less than 1 second's worth). I know Roscosmos's approach to range safety is, how to put this politely, casual by the standards I'm used to, but "Oops, there was a transient thrust error at liftoff, but it's gone now, better zigzag a couple times and crash anyway" seems like an odd way to program a contingency procedure. If the rocket has enough thrust to fly, and enough control authority to decide to fly away from the pad, wouldn't the safe thing be to burn off as much of that magnificently toxic propellant as possible before saying hello to the ground? Or at least stick as close as possible to the correct flight path for that magic 45 seconds, then give up? I really don't buy that the thing was following any sort of programmed course of action. The article continues: It looks like the blame-vectoring and ass-covering have reached full throttle already. I fear that further technical information is going to get filtered and mangled according to the politicians' choice of whose fault it was.
  15. As soon as the tail end of the plume clears the buildings, around 0:03, you can see the exhaust is already deflected significantly to the right. By 0:04, the rocket is visibly leaning to the right. It then swings left, but further, and back to the right for the last time, probably overwhelmed by aerodynamics at that point. An amplifying oscillation like that makes me think somebody dropped a negative sign or plugged a gyro in backwards. edit: Ooh, got a link for that? We've been pretty starved for actual info here.
  16. I see two exhaust plumes on opposite sides of the rocket that don't seem to be vectoring. That might be normal behavior if you're right about this: A yaw error, from the camera's point of view, would require thrust vectoring along a plane perpendicular to the gimbalability (yes that's a word now, dammit) on the leftmost and rightmost engines, so they wouldn't participate. If there's also a pitch error, then those engines' vectoring would be visually inconspicuous from our point of view.
  17. Another reason for the KSP's unusually vertical launch profiles is that Kerbal rockets are aerodynamic crap compared to Earth rockets (which sorta makes sense when you look at the sizes), so they suffer more severe drag losses low in the atmosphere. That makes it both hard and wasteful to pick up much horizontal velocity early in the flight.
  18. "What? No, this mission was always called the Mars Climate Impactor. Your press kit must have a typo."
  19. Why the hell would... I guess the old Soviet "NEEDS MOAR RADIOISOTOPES" design philosophy is still going strong? Edit: Just read Kryten's edited post. Heh.
  20. Those use cesium-133, which is stable. Atomic clocks are based on microwave-frequency transitions of nuclear spin states, not on nuclear reactions or decay. I can't imagine any reason there would be nuclear material on board that rocket, so I'm chalking that up to RT's usual standard of reporting.
  21. DAMN YOU ASAS! Um, I mean, hope nobody was hurt. That visible stream from one side made me think engine failure at first, but it almost looks like it's coming from between two of the side-mounted UDMH tanks and the color is consistent with N2O4, which is stored in the central body. I'm starting to think that's a secondary consequence of a failure that started right at launch -- it looks to me like the rocket is beginning to yaw even as it leaves the first camera's field of view. I don't see how a pure guidance failure would damage a tank that quickly, so my wild-ass guess is something went sideways with an engine's gimbal or support structure and the guidance system went crazy trying to correct for it. edit: I'm now reading references to a normal nitrogen tetroxide purge that happens soon after launch, so that plume may be entirely unrelated to the failure. In that case, I'll just blame ASAS.
  22. The thought of a Kerbal flight profile like that happening at the modern, familiar Space Coast stimulated an idea that still has me grinning dangerously. Let me paint the scene: [visual: Distant, faintly hazy view of a large, black-and-white painted rocket on Pad 39B. Most of the umbilical booms have retracted. A gentle breeze stirs the tree branches and carries wisps of fog away from the cryogenic tanks. The Big Countdown Clock in the foreground passes T-00:20:00.] [audio] "Hullo! Scott Manley heeir!"
  23. One should be careful to say exactly what they're doing experiments on. What White is testing is the spacetime curvature induced by ordinary, positive mass, and everyone already knows how this will turn out because it's already been measured with more sensitive instrumentation in many other contexts. There are no plans to measure the unknown part, because the necessary material is hiding somewhere between Jimmy Hoffa's body and a bunch of honest politicians. This embodies the same error as above -- the consequences of the Alcubierre metric are not in dispute, nor is the way that spacetime curves around normal matter. Nobody is arguing with that part. The existence of matter with negative mass would be brand new physics. That's the speculative part, and it's indispensable for the whole idea. The fact that it also depends on well-understood science does not make the scheme any more solid. That's a strong statement, and you'll find quite a number of physicists who disagree1. It's not a question with a mathematical resolution, true. But you'll also find nothing in the equations that directly prohibits the existence of magnetic monopoles (div(B)=0 is an empirical statement, not one derived from other more fundamental laws), yet those seem to be in short supply for some reason. The absence of a known mathematical prohibition does not guarantee, nor even suggest, existence. Alcubierre's work can be seen as reducing the question of "Can global causality be violated?" to "Can negative mass be created?" What evidence we have thus far points toward the negative on both counts. The understanding behind it is weak enough, particularly in light of cosmological-scale negative curvature, to sustain doubt, but it's a gross mischaracterization to claim that the existence of suitable negative mass is more than speculation. 1Myself among them, obviously. My reasons for it could be approximately summarized as "Nature likes single-valued solutions." This line of discussion heads rapidly toward philosophical wankery, though, so I'll freely admit that it's unresolved. But I've got twenty bucks on causality until I go back in time and tell myself not to place that bet.
  24. [citation needed] I say that because while there are lots of hypotheses about microscopic effects of general relativity (i.e. anything having to do with curved spacetime at the quantum level, including microscopic wormholes), none of them have been observed. Such observations would be huge news for the field of quantum gravity -- like Nobel Prize huge. Many of those hypotheses conflict with each other, too, so we can't even make reasonably precise predictions about whether spacetime even forms closed loops, or their size and duration if it did, or much of anything else in the absence of some observational evidence. As for the causality question: If you can use it like everybody thinks of using faster-than-light travel (i.e. you can start in one place, turn on the warp drive, go somewhere, turn it off, and be somewhere else before a beam of light would have gotten there), then it violates causality. This is a trivial consequence of any information or energy going from point A to point B faster than light, no matter what the mechanism.
  25. Oh man, I should not have read this thread. Now I'm gonna have to bust out the C64 emulator again.
×
×
  • Create New...