sevenperforce

Well, yes and no. Monopropellants, strictly, describe a propellant made up of a group of molecules that break down into smaller molecules exothermically. For example, two H2O2 molecules decompose into two H2O molecules and one O2 molecule, releasing energy in the process. This decomposition reaction happens slowly at room temperature but happens explosively at high heat or in the presence of certain catalysts like silver or titanium. Similarly, one N2H4 molecule decomposes into one N2 molecule and two H2 molecules, releasing a tremendous amount of energy; this reaction does not happen at all under ordinary conditions but happens explosively in the presence of a catalyst like iridium. Monopropellant reactions are still a chemical reaction that is no different, essentially, than bipropellant reactions. The only distinction is that instead of the reactant molecules being stored separately, they are stored within a single, large, unstable molecule. It takes energy to separate a single H2O2 (hydrogen peroxide) molecule into a water molecule and an oxygen atom, but that oxygen atom will subsequently react with another oxygen atom to form diatomic oxygen, releasing much more energy than it took to separate the two from their host molecules. It takes energy to break the four hydrogen atoms clean of their N2H4 host (that's hydrazine), but the energy obtained from those two H+H reactions and the free nitrogen bonds forming into triple-bonded diatomic nitrogen is far greater than the losses. Some chemical explosives are monopropellants. Famously, TNT is made up of a single molecule, C7H5N3O6, which decomposes with extreme prejudice into diatomic nitrogen, elemental carbon, carbon monoxide, and diatomic hydrogen. Nitroglycerin is made up of C3H5N3O9, with similar results. However, neither TNT nor nitroglycerin are as brisant or as energy-efficient as anatol and ANFO, both of which contain a mixture of an oxidizer and a reagent. TNT and other high explosives have MUCH better performance than gunpowder. By gunpowder I assume you mean black powder; modern gunpowder is made from nitrocellulose. Energy content and ease of detonation aren't inversely related or anything; there are plenty of high explosives that are extremely shock-sensitive, like ammonium chlorate, nitroglycerin, silver fulminate, diazomethane, and ammonium permanganate. The distinction, typically, is that chemists formulate high explosives as either primary or secondary so that they can create controlled explosions. If you collect a dime-sized of, say, silver fulminate, it's likely to detonate under its own weight. However, you can stack up a whole industrial-sized pallet of RDX or TNT without ill effects. So what you do is pack a bunch of RDX around the thing you want to destroy, then run your electrical detonator line to the tiny pellet of silver fulminate you've stuck to the side. Both RDX and silver fulminate are of course vastly more brisant and more energy-efficient than gunpowder. Technically, you can "stably" combine kerosene and liquid oxygen. It forms a dirty grey gel. This gel is an incredibly powerful explosive, but it's also cryogenic so it's not exactly useful in ordinary applications. Solid rocket fuel is usually made of ammonium perchlorate composite. It is a mixture of an explosive oxidizer (ammonium perchlorate), a weakly reducing binder (hydroxyl-terminate polybutadiene or polybutadiene acrylic acid acrylonitrile prepolymer), and a violently reducing fuel (usually powdered aluminum). It's basically the ingredients of a bomb, but put together in such a way that it burns in a controlled fashion under a set temperature and pressure.

Rocket engines essentially are a continuous non-nuclear explosion. Technically they're deflagration rather than detonation (although, as @Shpaget points out, there is very promising research into continuous rotating detonation engines; in fact they have even flown test articles). If your question, then, is whether explosive pulses can be a viable propulsion mechanism (through a pusher-plate or otherwise), then you need look no further than the V-1 Buzzbomb used in WWII by the pedants. It was the world's first operational cruise missile and was powered by an extremely simple pulsejet engine, using repeated fuel-air explosions where the fuel was gasoline). A pulsed detonation engine like this is quite simple and lightweight for the amount of thrust it produces, but it is much less efficient than a proper jet turbine engine. There are also a number of designs that use explosive pulses because we are unable to sustain a continuous explosion, like Z-pinch fusion. For non-nuclear explosions to be a viable propulsion mechanism with a pusher-plate arrangement, they would need to be (a) at least 165 times more energetic per unit mass than a hydrogen-oxygen reaction and (b) not be amenable to a traditional propulsion arrangement. Example: suppose that scientists discover an exotic metastable state of superfluid liquid helium in which the atoms exhibit persistent quantum superposition such that a single atom can be host to several billion different degrees of freedom rather than the 3 degrees of freedom in a typical monatomic gas. This would allow you to pump almost infinite amounts of energy into the helium without increasing its temperature or pressure, making it an explosive with potency rivaled only by antimatter itself. Ordinarily, if you wanted to use this for propulsion, you would simply have a big tank of this stuff connected to a small valve that would allow the metastable helium to enter a combustion chamber little by little, where it would lose its metastable nature and explode in the chamber. This would work well. However, maybe the metastable helium has a minimum volume requirement in order to stay stable, comprising such a quantity that the resulting explosion on destabilization would be vastly greater than what any chamber could withstand. In that case, you might need a pusher-plate or other pulsed propulsion setup.

I don't think there is any plan for any portion of Orion's heat shielding to be reusable. I admit to some surprise in how toasty it got, particularly with respect to the backshell. I was under the impression that because the backshell only needs to reflect the heat incident on it from the plasma sheath, it would have a silvery metallic heat shield, not the same ablative material as the main shield. Then again, maybe this WAS a silvery metallic heat shield and it just gets precisely this fried every time.

Yep, looks like that'll do it. That's what we're going for, right there.

Perfect. Except that I'm in Indiana this weekend. Crap.

Scott, Kirsten, and I did. I don't know about the others.

And Mark Rober has what, 23 million? Yusaka speaks and understands English, but not super well; I doubt he spends much time on the English-speaking side of YouTube. I'm not sure that "regular people" were ever discussed or contemplated; it was about art and outreach and imagination. I certainly get the appeal of going beyond the usual NASA channels and selecting someone who isn't already practically an astronaut, but I think there's a lot of missed utility there. Someone with a hard science background, particularly in physics or chemistry or engineering even at the bachelor level, has a huge advantage in being able to understand what's actually going on without having to learn it all along the way. And while learning it along the way can help you to be a better communicator, that sometimes is just a poor substitute.

For @JoeSchmuckatelli: as @kerbiloid's equation implies above, the odds of winning is the probabilistic inverse of that. You have a 100% chance of either winning or losing. And as @K^2 correctly notes, your odds of losing 80 times when you have a 99% chance of losing each time is going to be 0.99^80 or 45%. So if you have a 45% chance of losing every time, and you have a 100% chance of either winning or losing, then your chance of NOT losing every time (i.e., winning at least once) is 100% - 45% = 55%. Yes. He goes over other options too (realistic alternatives exist to putting engines on the Moon). EDIT- This only for moving the Earth away from the Sun though. It would not be possible to move it to another solar system without killing everything on it and making it uninhabitable, at which point it might have made more sense to build ships and travel to a star system with a rocky planet (still living in domes, but less effort than moving the planet). If we wanted to move Earth to another solar system, it might be advisable to simply make it a moon of Jupiter and then fly Jupiter itself into another solar system, using a fusion candle. What is a fusion candle, you ask? I'm glad you asked. (Even if you didn't ask.) A fusion candle is thusly named because it is burnt at both ends, not unlike how a candle can be. You grab a large iron asteroid, reinforce it quite thoroughly, and hollow it out. Build city-sized fusion engines at both ends and giant intakes in the center, and then carefully sink it into a gas giant. Uranus would be best for this (let's leave jokes about sinking something into Uranus aside) because it has the lowest rotation speed of any giant in our system and the lowest mass and thus requires the least effort to move around. As the asteroid sinks into Uranus, fire up the intakes to suck in the Uranian atmosphere and ignite the lower fusion engine, holding the asteroid aloft on a column of its own flame. Then fire up the other engine, balancing appropriately, and start flying. It's a slow and steady process, but it's doable. You can then use its gravity to come in and grab Earth and haul it off to a new home solar system. And you're carrying your own mini-sun with you, which isn't the worst of all possible arrangements.

True, but there is terribly little of the flight envelope where you need active ejection/escape from a failing Starship. Starship's failure modes are: Catastrophic structural failure on the pad Failed engine startup during abort from Superheavy failure Catastrophic structural failure during boost Failed engine startup after separation Catastrophic structural failure during the burn to orbit Premature engine shutdown during the burn to orbit Catastrophic structure failure on orbit Failed de-orbit burn Heat shield burn-through on re-entry Loss of multiple flaperon control during re-entry Loss of multiple flaperon control during EDL Failed engine startup during landing Premature engine shutdown during landing Several of these (7, 9, 8, and 10) are not helped by any ejection/abort/escape system (although 8 could potentially be solved with a rescue mission). Most (1, 2, 3, 12, and 13) can be solved by ejection seats. Granted, there is a brief portion of boost where the vehicle is too high and too fast for ejection seats, but still low enough in the atmosphere that dynamic forces will rip the vehicle apart. In which case. . . . The failure modes from above that remain (4, 5, 6, 11, and unejectable portions of 2 and 3) are survivable with ejection seats if the cabin area has enough structural integrity to survive vehicle breakup. Even if Starship goes to pieces, an intact cabin area will follow a generally ballistic trajectory until it reaches a low enough altitude that the ejection seats can safely fire. For 4 and 6, the vehicle may even be able to dump its propellant and use a flaperon-assisted glide to perform a controlled descent to ejection altitude. We know from Challenger that it's perfectly possible to survive a mid-air vehicle breakup shortly after MaxQ. The astronauts on that Shuttle were alive until the cabin hit the water. That was because the Shuttle cabin was horribly designed. They could have built the cabin in such a way that there were ejection seats in the mid-deck, but they chose not to, because they were fools. Also because ejecting into the exhaust/debris field of the SRBs would have been suicide, but that's beside the point. It's standard in crewed spaceflight, no matter how unlikely an engine failure may be. Think about it. If your car engine fails, you coast to a stop. If your plane engine fails, you glide to a landing. If your helicopter engine fails, you autorotate to a landing. If a rocket engine fails, however, you need a way to get to the ground safely. Starship is the first crewed vehicle which plans propulsive landings on Earth, so it needs a contingency for the highly unlikely event that an engine fails. As @SunlitZelkova points out, I am quite familiar with using social media for science education. I have about seven times as many TikTok followers as Tim. Granted, he has a lot more YouTube followers than me, but still. Not begrudging it at all, mind you. He's a nice guy and he does a great job, for the job he does. Tim dropped out of college to pursue photography, and that's totally fine . . . but I think it's a real shame that out of eight slots and two backup slots, DearMoon won't have a single person with an actual science background. And it wouldn't have to be me, either. Kirsten Banks, Emily Calandrelli, Lena Vincent, Bill Nye, Scott Manley, Mark Rober, and Hank Green all have real hard science degrees and extensive experience in the intersection of science education, public outreach, and social media. Any of them would do a better job. Sending Tim along for DearMoon instead of any of the above is kind of like hiring Colonel Sanders to take you on a culinary tour of the world when Gordon Ramsay, Anthony Bourdain (RIP), and Alton Brown are all ready and standing by. LOL, I mentioned him (above) before I saw this post. Seems obvious that Tim got the spot because he knows Elon.

SpaceX is already designing pressure-fed landing engines for Lunar Starship HLS, so these could work. This would, of course, require adding dedicated pressurized CH4 tanks to the crew cabin section and dedicated helium COPVs to maintain press. The main problem is figuring a way to separate the entire nose of the vehicle, including a portion of the heat shield, in a safe and reliable way. What do you do with the forward flaperons? Do they come with, or are they supposed to separate too? It's a huge amount of complexity and new failure modes in an area where you want the maximum structural integrity in the first place. For something this massive, the best you could hope for would probably be drogue-assisted propulsive landing. Is DearMoon back to using falcon heavy? No, it never was. Yusaku Maezawa originally contracted to fly around the moon on a modified Crew Dragon (uprated ECLSS, heat shield, etc) on an expended Falcon Heavy, but ever since DearMoon was announced (simultaneous to the announcement of the slimmed-down 9-meter Starship) it has been a Starship project. The plan to expend the Superheavy (announced at some point within the last 18 months, I can't remember exactly when) means there is no loiter time or need for propellant refill in LEO.

I wish him all the best, even if he annoys me somewhat from time to time. (Yes, it's a little sour grapes. I would have preferred to be on that flight. Tim does a lot of research but he lacks some of the fundamental physics and engineering knowledge that you really need to understand what's going on and explain it properly.) While that's true, that word be a catastrophic failure - you need a total failure in 2 of 3 sea level engines, which have already been fired at the beginning of the mission (thus if there was issues that could risk that, you would have had whole days or weeks to address it and find countermeasures), and much more if the vacuum engines can be used instead both in the 6-engine and 9-engine variant. I'm not saying it's impossible, but it's a very, very low chance in a later starship flight when crew is already launching Assessment of risk is a tricky thing. Calculating the Loss of Crew ("LOC") risk for a launch requires you to sum the probabilities of every possible failure mode, with all of their interactions and dependencies. Merely summing the probabilities is difficult enough; adding in the interactions and dependencies makes it incredibly hard. The equation looks something like this: LOC = Σ1..n(PFn * Σ1..m(1-PCn_m)) Here, PF1 is the probability of the first failure mode happening, PC1_1 is the probability of the first contingency for the first failure mode working, n is the total number of failure modes, and m is the total number of continencies for each given failure mode. You can add new abort/contingency/escape systems which increase the m for certain failure modes, but each new system also increases the global n, so you have to make sure that the result after the system is added is actually lower than the result before the system is added. In the spacecraft world, we have had two instances where the abort/escape system itself directly caused a failure, which either resulted in fatalities or would have resulted in fatalities (the Soyuz Dec. 1966 incident and the Dragon April 2019 incident). And of course there are numerous examples from the world of aircraft. There are certain failure modes which we accept as having no meaningful contingencies. In commercial and general aviation flights, catastrophic losses of structural integrity in the wing (like this or this) or in the tail section (like this or this) are understood to have no contingencies, so you just try to reduce the absolute odds of those failure modes. The Space Shuttle was particularly bad in terms of LOC because there were MANY significant failure modes, almost none of which had any contingencies. The only significant failure mode that could be survived was a single-engine loss several minutes into flight. An earlier single-engine loss, or the loss of a second engine at any point, and the vehicle was doomed. Any booster failures were also instant LOC. With Starship, there are at least contingencies for most failure modes, assuming a nine-engine upper stage. With 33 independent engines, Superheavy can survive several engine failures without even having mission loss, and given a reasonably beefy interstage, even a catastrophic structural failure on Superheavy will allow Starship to act as its own abort system and execute a RTLS and nominal landing. That alone makes it much safer than the Shuttle. If the failure is on Starship, however, that's another matter. Starship can survive losing up to two of its six total engines on ascent and it can survive losing up to two of its three central engines on landing. All three central engines are ignited at landing, too, so if one or two fail to ignite or shut down immediately after ignition, it's no big deal. The engines have the same frag shields that have been used and tested on Falcon 9, so even a catastrophic engine failure (as in the CRS-1 and the March 2020 Starlink launch) is fine. All four flaperons have dual-redundant driver mechanisms, and Starship has acceptable control authority even if one of those four flaperons completely locks up. So Starship can lose between 3 and 5 of its 8 flaperon drivers without losing the vehicle. Yes, any catastrophic structural failure in the tanks or downcomers will result in LOC. But tanks and downcomers are literally just steel cylinders, really the simplest possible structure in aerospace. "The best part is no part" is helpful here; the failed strut in CRS-7 only existed in the first place because the F9US needs helium COPVs inside the tanks, which Starship doesn't need thanks to autogen press. I'm not saying it's impossible for the main structural components to fail, but the failure of any primary load-bearing metal structure in a commercial aircraft (like an engine pylon, wing spar, or tail) is unsurvivable, too. As long as the crew escape system didn't further increase possible failure modes, sure. All the pyros and contingency engines and COPVs and other systems necessary to achieve an independent capsule abort system like F-111 or @tater's Ithacus add to failure modes, though. If we suppose that Starship's risk of a catastrophic structural failure is 1% and the independent abort system's chance of correcting a catastrophic structural failure is 95%, but the independent abort system has 5 separate systems which each have an 0.1% chance of creating their own failure mode for which there is no contingency, then you basically break even. Ejection seats a la Gemini would be a better solution, even though they introduce their own failure modes too. You need to pull 135 m/s2 -- about 14 gees -- to slow down by 90 m/s over a 30-meter distance. So if you could reliably land tail-first and perfectly spread that deceleration out, such that the entire vehicle acted as a crumple zone for a titanium-bathtub crew module, then sure, it could be survivable. The problems are (a) engineering a crew module which can survive the ensuing explosion while also falling at an acceptably-consistent rate through the rest of the vehicle and (b) figuring out a way for the vehicle to enter a tail-down configuration without using the engines to initiate the kick-flip. I agree. However, the current plan for DearMoon is to expend a Superheavy to allow a direct launch into TLI . . . so they seem gung ho about it.

True, but then you'd need an energy source from somewhere else to accelerate the propellant, instead of using a magnetic nozzle to do it (which I believe was the original plan). More importantly, all your "uber" magnetic fields still won't accelerate your propellant. You'll need a source of energy to do that, just as you would if you didn't have a magnetic nozzle.

For the Merlin 1D, helium is used to spin up the turbopumps first, followed by TEA-TEB ignition of the combustion chamber. It's possible that they simply use a bleed valve from the combustion chamber to the gas generator to ignite it.

Yes, Earshaw's theorem has been fully and completely tested, not only empirically but mathematically. Most things in physics cannot be "proven" per se, but there are exceptions. Certain physical systems can be modeled mathematically and can have mathematical proofs applied to them. Earshaw's theorem is every bit as proven as the fact that pi is an irrational number.

Boy does that backshell look TOASTY.

Directional flashlights?

Even apart from all of the reasons why you can't have 1000 Tesla permanent magnets, what you're trying to do wouldn't even work to start with. Earnshaw's theorem says that you can't have a stable arrangement of fixed magnets, electrostatic charges, or other electromagnetic field interactions which results in a constant force being applied to something without destabilizing torque. So even if you did have super-powerful permanent magnets, they would not allow you to push charged particles out the back of your vehicle without an input of power.

Oh, that's certainly possible. The atmosphere of Mars was much denser in the past, but due to the planet's smaller size and lack of a magnetic field, most of the water vapor that was once in the atmosphere escaped into space. If the polar icecaps on Mars were to melt, they would roughly double the density of the Martian atmosphere, which would cause accelerated greenhouse warming, which would cause sublimation of gases currently adsorbed by the soil, which would really start to crank up the density of the atmosphere. If you're looking at the past, then you could envision a situation where Mars was struck by a particularly large comet a few million years ago. That would probably be enough to do the trick. Assuming magic exists and we can use magic to do magic, I reckon magic is possible.

You do realize that growing ANYTHING requires water, right? And that if you have artificial light, you'd need a source of energy to produce that light, right? And if you have water and a source of energy, you can use the energy to split the water into hydrogen and oxygen, which is literally the most efficient possible chemical propellant? I don't understand how you don't understand this. I really don't. How isn't this just incredibly obvious?

Because he doesn’t know the rocket equation, refuses to learn the equation, and thus fails to recognize that the only things contributing to the total deltaV of a vehicle are (a) the weight of all the crap in it that it intends to shoot out of the back, relate to its dry mass, and (b) how fast the crap gets shot out of the back.

Neutron star collisions are (virtually) always binary pair collisions, not chance encounters. Classically, gravitational drag wouldn’t exist. But under relativity, it does. You know how tidal forces operate, right? And how the Moon’s tidal force impact on Earth’s oceans cause the Moon to drift away from Earth? Well, relativistic gravitational frame-dragging works basically like that, except that instead of the effect being caused by the differences in distance between the near and far sides of an object, the effect is caused by the difference in the curvature of spacetime due to extremely dense objects. And so co-orbiting neutron stars rob each other of energy through this exchange, coming closer and closer until they merge. The same thing happens in our solar system…just at an aggressively lower rate. Earth loses about 3 nanometers-per-second of orbital velocity each year.

What's your yaw control?

How does it handle?

I can't help because I'm not very far from Minneapolis and my telescope isn't very big but this is such an awesome thread.

They sure know their target audience. Let the pork flow!