sevenperforce

The most probable chain is going to be the use of small, unmanned, one-way probes to visit and analyze candidate asteroids, followed by a larger roboprobe (perhaps using a solar sail) to shepherd the candidate asteroid to a central processing site in orbit. Without a non-rocket-based launch system, it doesn't make sense to do processing on the surface. If you can use a maglev on the moon to fire your cracked propellants into orbit, great; otherwise you're better off doing the refining in orbit to begin with.

Awww geez I have to make the wings reliable too??

Ah, makes sense. see what I did there Also note the phrasing and response to Scott: So it sounds like he's saying that if they decide to give early (read: unmanned?) Starships pad abort capability, they would make the vacuum engines dual bell. They are of course already planned to be fixed. Taking the dual-bell route shaves off efficiency slightly for lunar ops but it's not too bad. They might be able to also use this for single-stage P2P operation, igniting all six-seven engines on the pad for liftoff thrust and then shutting off the core engines as the VacRaptors expand into the bell extension. His phrasing would suggest that for early manned Starship flights (e.g., #DearMoon), they might launch and fuel Starship first and then use Falcon 9 with Dragon 2 to send up the people, if there is no pad abort

Well, he often uses imprecise terms, but he doesn't usually use precise terms (like dual-bell) imprecisely. A dual-bell design would not necessarily be visible in any of the renders. It would show as a very slight inflection point inside the bell and would probably not even be visible in photos unless you had a high-contrast view of the naked engine before it was installed. I think the dual-bell is intended solely for inducing controlled flow separation if the VacRaptors were ignited in a pad abort. In a pad abort, you want all the engines to basically act as if they were SL engines. That's true for any abort system, though. Exactly. Dragon 2's LES doesn't help if the pressure vessel goes boom, as it did on the test stand. The escape towers on Orion and Soyuz won't help if the escape tower buckles, folds, and detonates when it impacts the fairing. The pusher escape motors on Starliner won't help if the propellant tanks in the service module explode. The big distinction is between Starship and the Shuttle. The Shuttle had no abort mechanism whatsoever while the boosters were firing. Starship does. Even airliner-level reliability won't help an airliner if the wings break apart. While not untrue, it should be noted that the risk of catastrophic detonation is rather higher for cryo liquids than for hypergolic liquids. For hypergolic liquids, the propellants begin burning at contact, which tends to disperse them all rather rapidly. For cryos, the liquids may have an opportunity to mix before ignition, which makes the fireball much larger. In either case, though, you have plenty of time as long as you trigger LES at first anomaly. They don't pay that much.

Unless Elon is using wildly imprecise terminology, dual bell means there will be a convex inflection point in the bell which produces controlled flow separation. Pioneered here: https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19940018584.pdf

There are only forty remaining clumps of needles remaining. All the lone needles have long since been deorbited by sunlight pressure, and the forty clumps that remain are tracked as they also deorbit. The first shell that will be deployed goes at 550 km. Those are the ones that will decay within 5 years if left uncontrolled, and which are unlikely to cross the path of any uncontrolled debris on their way down since any uncontrolled debris below them will also decay rapidly. Subsequent shells at 1150 km and 340 km will not be placed until later in the program; the bulk (7500 of them) will go to 340 km, where the decay time is even shorter and the risk of collision with uncontrolled debris is dramatically less. The 2800 sats that will go to 1100 km are the tricky ones, yes. They are the potential Kesslers.

What if neither sat has control? That's going to be an exponential decrease in likelihood. A very low probability times another very low probability is a much, much lower probability. Also, remember that satellites don't decay "up" in space. Starlink sats will decay in five years without control input. If an uncontrolled, decaying Starlink sat is descending, anything in its path is going to have active propulsion, or it wouldn't have stayed in that orbit. The 2009 collision happened at nearly 800 km, in a region where atmospheric drag is virtually nonexistent.

If I could build whatever I wanted, I would use a pair of air-augmented pebble-bed nuclear turboramrockets, mounted on rotating nacelles, for liftoff and touchdown. Then I would use a magnetic aerospike antimatter-catalyzed pulse fusion drive for vacuum operations.

But operational-orbit failures will be rare, and failures which involve the loss of comms will be only a fraction of those, and failures which involve propulsion system failure will be an even smaller fraction. The tiny, tiny number of Starlink sats with prop failure will be quickly identified and their decay rate can be simulated well in advance, based on existing measurements of decay as provided by the tests they're currently running. There will be plenty of time to look at their descent path and figure out if other satellites need to perform avoidance maneuvers.

*If not controlled*. They have fuel to boost for their assumed useful lifespan. Exactly. They carry onboard fuel to adjust trajectory and stay in space, and enough to perform a rapid deorbit if there was a problem. The only time that they will deorbit passively is if they somehow lose electrical power and become junk. They are probably programmed to do a rapid deorbit if they lose comms as well.

This sat was probably already on its way to being deorbited. After all, it was selected to be deorbited because there was something wrong with it. So altering its course would have made the experiment worthless. Anyway, as the updated presser revealed, they would have coordinated with ESA if necessary.

I agree, though the better view suggests strongly that it is a movable piece within a larger flaperon, and the landing legs themselves are somehow fixed. I never liked the idea of a single failure point that takes out both maneuverability and landing capability. SpaceX chose three satellites -- either ones that weren't operating perfectly well, or some other approach -- to test passive decay. Avoidance of Kessler syndrome requires that the satellites will passively decay if left without control. These three sats are being used for this purpose, to show that even if SpaceX completely lost control, they will all fall out of the sky on their own in a controlled fashion. In this case, SpaceX likely rejected any suggestion that a collision was likely and accordingly declined to move their sat, as doing so would ruin its data. Even an inclination change would alter numerous factors (like insolation, exospheric temperature, etc.) and make the whole thing harder to work out. ESA was free to move its satellite if it was concerned...which, evidently, it was. Sats are low enough that they decay on their own without control input within a few years. So, not a problem. More likely, Falcon 9 or FH drops off a stack of replacement sats with larger propellant reserves directly in an eccentric MEO (perhaps with a slightly higher apogee). The cluster of sats would hang around up there until they were needed, at which point they could use their onboard propellant to adjust inclination at apogee to match plane with the derelict sat's former orbit. This is dV-cheap, since it is done at a high apogee. Then, they lower their perigee just enough to use atmospheric braking to circularize. The replacement sats wouldn't have quite the same lifetime as a brand new one dropped off directly in the desired orbit, but that's fine.

Incompletely-cured melt glass from a meteor impact?

You can also shortcut by simply putting the patient in a high-oxygenated hyperbaric chamber. No radiation needed, and the microbes can't reproduce when exposed to high levels of oxygen. Not as good a prognosis as amputation, but better than the alternative when dealing with infections in the torso.

Hmm. I'm still not sure what you're intimating. There is no "mixing" with "hot preburner exhaust" because hot preburner exhaust is the only thing that enters the combustion chamber. The methane preburner exhaust comes out at 774 K; the oxygen preburner exhaust comes out at 748 K. Autoignition temperature of methane is well over 800 K.

Trebuchets are too complicated. If we have metamaterials with this kind of strength, let's just go for a good old siege onager.

There's no extra fuel/oxidizer; that's why it's a Full Flow Staged Combustion cycle. Something like 90% of the LOX and 10% of the CH4 runs into the oxidizer preburner; something like 10% of the LOX and 90% of the CH4 runs into the fuel preburner. Then 100% of both exhaust flows is routed into the combustion chamber, so you have a purely gas-gas reaction in the chamber, leading to very high efficiency. All of the propellant goes through either one preburner or the other. Now, they may do film cooling injection with a very small amount of still-liquid fuel, but as far as I know the cooling is purely regenerative, by looping the propellant through the engine nozzle and chamber walls. Despite the high preburner temperatures, you don't reliably reach autoignition temperatures for the mixed exhaust flows, and so you use a separate igniter. The dual-redundant igniters are not purely spark igniters; rather, they are spark plugs used to ignite a gas-gas torch fed from the autogenous pressurization tapoff. The mixed preburner exhaust hits the torch flame and, well, kablooey.

Apart from a LOX-augmented nuclear thermal rocket (LANTR), there really aren't any liquid-propellant engines with enough thrust AND enough specific impulse to get you into orbit. Also, the side-mounted rockets are not inline and so you wouldn't want to use them for acceleration through the atmosphere, anyway. You can put the engines in side-mounted nacelles a la Serenity if you want, but that might be more trouble than it is worth. You need to have a liquid-propellant thruster system somewhere, after all, because you need translation and attitude control. You also need liquid landing engines because you cannot exactly land using pusher-plate retropropulsion. Simulations of exhaust reflection off unprepared surfaces indicates severe damage to engine bells in most cases. Nuclear pusher-plate launch doesn't really pose a major health risk. As long as it's done far enough away from population centers, with "clean" mostly-fusion nukes, the global increase in cancer risk is very low. And you can't use it all the way to the ground, after all -- you're going to have to switch to thrusters at some point because pulse propulsion does not play well with landing pads.

There are a few. Dedicated landing engines can be higher-thrust and lower-efficiency than main engines because their burn time is lower (by analogy, the Puff monopropellant engine had the highest vacuum TWR of any non-SRB back when I was playing regularly). They can also be mounted in a way that protects them from debris on landing, which is more complex to accomplish with a main engine. Dual-thrust-axis landing allows for a more stable landing arrangement with fewer catastrophic failure modes and obviates the need for cranes or complex ingress/egress. If you have a main engine which is known for doing nasty things (like Orion or a NSWR), it allows you to avoid firing that "torchship" engine near the surface. It allows you to better optimize that main engine for lower atmospheric pressures.

With Orion, the distinction is rather unimportant because you are dealing with massively hypersonic particle velocities anyway.

My suspicion is that once shaped-charge losses became involved, a fuel-air approach would have no better specific impulse than chemicals...or, at the very least, not high enough to make up for the increase in dry mass. You also have to deal with the same problem faced by scramjets: you need sufficient mixing at subsonic, supersonic, and increasingly hypersonic speeds. At Mach 2+, the wake shock behind a vehicle would be so rarified that it would be almost impossible to get enough oxygen for a detonation. You would almost need an intake to allow mixing, at which point we are dealing with a hypersonic pulsejet anyway. Of course, if you are building an Orion, you will want to shape the pusher plate in such a way as to take advantage of atmospheric reaction mass. And if you have a pusher-plate Orion optimized to use atmospheric reaction mass, then you could at least get a few pulses out of a fuel-air approach before you were going too fast to allow mixing. With either a conventional Orion or a dual-thrust-axis SSTO, you could conceivably use a fuel-air explosion (or several of them) to get the initial kick up and away from the bulk of the atmosphere, at which point you could light up the nukes. Orions need to launch from the poles anyway to avoid causing EMPs, so once you do the first few pulses with chemical fuel you can light up the Orions in-atmo. If you're using a conventional (VTOL) Orion, you can get almost double efficiency out of the very first pulse because the shockwave reflects off the launch pad, so that would make a chemical "first stage" more viable.

Even 10-20% of usage blows chemical rockets away. To the OP -- I have always been a huge fan of dual-thrust-axis SSTOs. I'm partial to the Firefly. A chemical Orion is a non-starter. Isp is way too low. Generally, blowing something up and using a pusher plate is really inefficient. The only reason it works well with the Orion concept is that the amount of energy is high enough to overcome inefficiencies. Plus, I think you get a boost in atmo thanks to added reaction mass.

The first firing of a dev raptor used TEA-TEB but all the fullscale raptors were spark torch igniters.

A solar-powered ion thruster is analogous....

The reason you don't see stars during the day is because the Rayleigh scattering is brighter than the starlight behind it. Not quite. It's a function of particle size. Particles within the rings are mostly between 1 cm and 10 m with an exponential size-frequency relationship. The odds of a given photon intersecting a solid particle while traveling through a given path can be calculated if you estimate the size-frequency relationship, but it's not at all "X% of the path is blocked". Also, the ring particles are almost exclusively pure water ice...purer than found in most places in the solar system, actually. Though they are not transparent, so you're correct there. The rings are visible because they scatter sunlight, and so there is enough sunlight being scattered toward you to saturate. But you can definitely see through them. Consider this high-contrast Cassini image: From this vantage point, you can see how much light is blocked by the rings, both in their shadow projected on Saturn, and by the visibility of Saturn's disc through them at the top. The B ring obviously blocks more light than the C or A rings, but even it allows a little light through (the apparently-black portions of the stripes across Saturn are actually lighter than the black background.