sevenperforce

You'd have to give an example; I don't think I've seen it. That might be difficult since the recent FH didn't have any views of the upper stage after staging.

The payload was just very heavy. At 6.6 tonnes, first-stage recovery would have necessitated an orbit with an apogee lower than a proper GTO (compare the Galaxy 33/34 mission back in October, where the 7.4-tonne payload could only be lofted to around 20,000 km; the same was true for SXM-7 and SXM-8). Usually, GTO launches have an apogee even higher than the 35,800 km of GEO. A higher apogee allows for a bi-elliptic transfer, which saves propellant on the payload and thus increases the lifetime of the satellite. The customer paid extra to expend the booster. Example? I don't believe they ever would push them beyond design limits. They don't want to risk damage to the engine before the payload has separated.

How? All-moving control surfaces on aircraft are lift surfaces. By moving in their rotational axis, their angle of attack changes, which changes the amount of lift they generate, producing forces about the center of mass to induce pitch, yaw, or roll. The flaps on Starship are not lift surfaces. They are drag surfaces; essentially nothing more than airbrakes. They do not produce lift (other than body lift, which is irrelevant here), and so changing their angle of attack doesn't do anything. The use of airbrakes to control aircraft is not new; flying wings use it for yaw control. The B-2 bomber has split ailerons; if it needs to yaw to the left, it will deploy the ailerons on its left wing in opposite directions, which increases drag on the left wing, which pulls the aircraft to the left. But in that situation, adding another axis of rotation wouldn't do anything, because the ailerons are not producing lift in the yaw plane, only in the pitch/roll planes. What you are suggesting is that changing the angle of attack on the Starship flaps would alter their control authority. It will not, because changing the angle of attack of a control surface does not produce new control forces unless that control surface was already providing aerodynamic lift. Here is a CFD simulation showing that the flaps are not producing any lift: If you do the calculation the amount of propellant used for landing is far less than 30 tons. The 30 tonnes of propellant is used not only for landing, but for the deorbit burn as well. A deorbit burn requires around 100 m/s of Δv. Starship will need roughly 7 tonnes of propellant residuals. So only 23 tonnes can actually be used. An empty Starship has a mass of just over 100 tonnes, approximately, plus the 7 tonnes of residuals. For the tanker variant, 27 tonnes of propellant gives around 630 m/s of Δv, which is certainly more than needed. But Starship needs to be able to return downmass as well. With a notional 50-tonne return payload, that's less than 450m/s. So after the deorbit burn, that's less than 350 m/s to execute the flip and come down to a soft landing while also fighting gravity drag. As you yourself pointed out, the lowest you could expect for that whole process is 250 m/s. Does that settle the controversy?

Really intriguing! I like this kind of conjecture. One correction/caveat. The concept of a 3D expanding balloon is often used to illustrate the idea of metric expansion of 2-space, which can then be used to conceptually understand the metric expansion of 3-space. However, the expansion of 3-space itself is not occurring "in" 4-space. The universe is (as far as we can tell) topologically flat; it should not be supposed that the 3D universe is expanding in a 4-dimensional hypersphere. So with that understanding: yeah, I can't think of any specific reason why the stress-energy momentum tensor that defines spacetime curvature in general relativity could not be the result of 4-dimensional acceleration. Whether true or not, it's certainly a great way to conceptualize it. However, if there is actul physical acceleration in 4 dimensions which causes gravity,, such 4-dimensional acceleration would be independent of the observable accelerating metric expansion of space. And we should be able to know this anyway because the rate of expansion of the universe has changed through its history, but there is no evidence that the degree to which mass curves space has ever changed through the universe's history.

Apart from the extraordinary weight penalty of having flaps and hinges which move in multiple axes rather than just one, adding another axis of control authority to the aft flaps won't help at all if the moment arm from the center of mass isn't long enough. That's just basis physics. Locating the landing propellant in the nose rather than keeping it in the LOX tank pulls the center of mass just far forward enough that the aft flaps have control authority. Without this, the aft flaps would have zero control authority, because the engines are very heavy and so the center of mass would be between the flaps themselves. Giving the aft flaps a new control axis wouldn't help because zero times anything is still zero.

What are the possibilities here? (A) Eric Berger, who we know is quite intelligent, has had a complete mental breakdown and thinks something is possible when it absolutely isn't (B) The Defense Department has a new, secret version of nuclear thermal propulsion with performance rivaling that of Project Orion (C) "around cislunar space" simply means "around cislunar space" and does not mean "from LEO to cislunar space" It absolutely does not.

This might be better placed in the regular Chinese space program spot, but I'm trying to figure out the minimum amount of work China would need to do to make this not happen again. If they simply added propulsive vents to the tanks and some basic star trackers to be able to maintain pointing, surely it would be able to use propulsive vents to point retrograde. Then firing additional propulsive vents should be enough dV to lower perigee, right? If the tanks are pressed to something on the order of 2 bar and the stage has 2% propellant residuals, how much dV would that provide if you just vent it all?

One thing that would be super nice would be if all of the entry points so far -- Cote d'Ivoire, the Maldives, Borneo, and now the Pacific Coast of Central America -- were along the same great arc. That would suggest that China designs the re-entry such that the vehicle is at least expected to come down within a single orbit. Unfortunately, those four points are not at all along the same great arc.

A more-rapidly rotating planet will tend to become progressively more oblate. The increased distance to the center of the Earth and the increased centrifugal force at the equator combine to have progressively lower and lower effective gravity at the equator. That's about the only change. My dude, the atmosphere rotates with the Earth.

Definitely coming down much more rapidly this time, likely thanks to the low perigee. Looks like it will come down over DC (where I am) if it is just 27 minutes later than the current estimated splashdown point. In a 6-hour window. Lovely.

The Epstein Drive in that series is a torch drive that employs Brachistochrone trajectories. An airbreathing gas core MHD engine could get very high specific impulse and reasonably high thrust, yes, but thrust drops off once you're out of the atmosphere. Of course, that's perfectly fine. You don't need high constant acceleration for some specified period of time. I don't know why you keep insisting otherwise. Your total dV is your total dV, whether you dump it in 70 minutes or 20 minutes or fifteen hours. Saying "speed and velocity" is like saying "temperature and hotness" -- one includes the other. If your velocity vector is the same, then it needs to be your velocity vector relative to something. Is it your velocity vector relative to the Sun?

Oh dear, this again. Acceleration is NOT a feature of a "rocket drive" at all. A rocket engine produces thrust. Acceleration is the result of dividing thrust by mass. Let me say it again. Rocket engines DO NOT produce acceleration; they produce thrust. Period. And because the mass of a vehicle changes over time, the acceleration of a vehicle changes over time, assuming constant thrust. Specifically, it goes up. Since you said "3g max acceleration" I'm going to assume (probably wrongly) that you're thinking of something sane. For a vehicle to have a 70-minute burn time with 90 tonnes of propellant and achieve 3 gees at burnout, then (clearly) it must be burning 21.4 kg per second and it must be yeeting those 21.4 kg/s out the back end at a sufficient exhaust velocity that the impulse to the dry mass of the vehicle is 29.43 m/s2. This gives a range of values depending on the dry mass of the vehicle and thus the corresponding mass ratio. Here are some comparisons: Ratio of Fuel to Dry Mass Thrust (kN) Isp (sec) Initial Acceleration (gees) Total Δv (km/s) 24:1 (typical for some chemical upper stages) 110 465 0.12 14.7 6:1 (typical for a laden first stage rocket) 441 2,100 0.43 40.1 3:1 (enough thrust to get off the ground) 1,177 5,606 1.00 76.2 1:1 (typical for aircraft like a 747 or a B-2) 2,649 12,618 1.50 85.8 Having a specific impulse in the range of 2,000-5,000 seconds isn't completely out of the question; that's the theoretical specific impulse range of your typical open-cycle gas-core nuclear rocket. But of course a gas-core nuclear rocket sprays out rather unpleasantly radioactive exhaust, and it's doubtful that you'll be able to get the amount of thrust you need. The NASA studies of gas-core designs found they wouldn't produce enough thrust to even lift half the weight of their pressure vessel alone, and that's before factoring in the weight of a moderator/reflector and radiators. If you want to get up above 12,000 seconds of specific impulse, you're going to need a better energy source. Antimatter could do the trick. If you had perfect conversion of potential energy to kinetic energy, then you'd be accelerating your 21.4 kg/s of propellant to 124 km/s, requiring 164.5 GW of power. That's going to require you to burn about one milligram of antimatter per second, or a total of 4.2 kilograms over your entire journey. Of course, you won't, because the maximum efficiency of any heat engine (the Carnot cycle) is around 70%. That means 30% of your energy is lost to heat, so you're actually going to need 6 kilograms of antimatter. But that means we're going to have to have a way to reject 70.5 GJ of thermal energy per second. There are two ways you can do this. The first is by using an open-cycle cooling loop where you're just dumping part of your propellant overboard. That propellant will of course achieve very high temperature (that's the point) so it will basically be an entirely separate rocket engine similar to a regular NTR, except the heat it's receiving is waste heat from the antimatter rocket rather than from a nuclear reactor. Assuming liquid hydrogen propellant and a coolant exhaust specific impulse on the order of 1000 seconds (but now benefiting from the 70% limit of the Carnot cycle), each kilogram of coolant is carrying away 69 MJ of heat and producing an impulse of 9.8 kN. But at this rate, you'd need to be dumping over 1000 kilograms per second just to deal with the waste heat of the engine, and our propellant budget is only 21.4 kg/s. So we turn to the second approach: a closed-loop cooling cycle where the coolant runs through a series of radiators and is then injected into the reaction chamber as the main source of propellant. That will work, but it will approximately double the weight of your engine. But let's suppose we ignore all of the waste heat issues entirely. Some antimatter engines have T/W ratios on the order of 4:1, and so if we're just gonna handwave and say this is achievable here, we'll end up with an engine that weighs around 67 tonnes. This leaves 23 tonnes for payload, structure, and propellant tanks. If you double the amount of propellant then you cut your acceleration in half. If you double your thrust to accommodate this, you use up your propellant twice as fast, and you're right back where you started. You can't exactly "lighten the ships mass" when your engine and propellant alone are 87% of the ship's laden mass. And if you're trying to accelerate at 3 gees the entire time...that makes no sense. Engines don't produce acceleration; they produce thrust. As your mass goes down, your acceleration will go up. So space is accelerating past you, but you're not gaining kinetic energy? Ok, let's imagine that. If you have this, you don't need 70 minutes of whatever wacky acceleration you're imagining. I'm assuming you can't activate this impulse warp in atmosphere? So you just need about 2 km/s to get out of the atmosphere and point in the desired direction, and then engage impulse warp. At three gees of acceleration, that's going to be a burn time of just about 68 seconds, and you're going to only need a fuel to dry mass ratio of about 1:1. The problem is this: when you drop out of warp, you said you have the "actual velocity and momentum vecotor [sic]" from the beginning. But actual velocity relative to what? Relative to your home planet? Relative to your destination? Earth is moving at 30 km/s around the sun. If I point my vehicle toward Jupiter and warp toward it at 3 gees, then drop out of warp near Jupiter, I'm going to be outpacing Jupiter by a whopping 7 km/s. And that's if Jupiter is perfectly lined up with Earth so that our velocity vectors are pointing in the same direction. If our velocity vectors are pointing in different directions, it gets even worse. Why? Why would multiple nozzles be any different from a single nozzle? In terms of thrust and heat, this makes no sense at all. Increasing the number of engines doesn't make it less efficient. It increases your dry mass but the specific impulse isn't going to change. Please just learn the rocket equation.

The RS-25 is very close to being a FFSC engine. After all, it has two different preburners and two different turbopumps. The only difference is that both preburners are fuel-rich, because hydrogen's heat capacity and specific energy are both just so much greater than oxygen that running both preburners fuel-rich makes more sense. But sure -- if you replaced the RS-25's fuel-rich preburner attached to the oxidizer turbopump with an oxygen-rich preburner, you could do it. It just wouldn't be quite as efficient because hydrogen just works better than oxygen.

They talk about missions other than Artemis but I'm unsure what other missions they'd need depots for, unless they're talking about Mars. In theory, such a depot could make @tater's dream of a fully reusable stretched Lunar Starship possible by coming up to meet it in MEO.

Yes, but that would cause rotation around the pitch axis because you'd only be venting from one of the vehicles and so you'd have off-axis thrust. OTOH, I wonder if an alternative way to settle the propellants would be to get them rotating together in a penguin roll. Would the centrifugal force then assist in settling the propellant? Hard to know, I think...lots of CoM shifting going on.

Ah, yes, you're absolutely right. A little thrust to provide consistent settling, and then the recipient Starship can vent to lower its tank pressure and thus accept flow from the donor Starship. Depending on the viscosity of liquid methane and liquid oxygen, the constant settling might not even be necessary; once it gets flowing it should stay settled and the tank pressure differential will do the rest.

I think it depends to some degree on the internal plumbing, no? If I recall correctly, this is the propellant fill adapter that the Quick Disconnect arm interfaces with, right? According to the renders I've seen, the fill lines run up the inside of the skirt and go directly into the methane downcomer and the LOX tank, respectively: So if you were to connect two starships back-to-back (which I will call "Thing 1" for no reason whatsoever), open all the valves between these fill lines, and start your settling/prop-transfer burn, the tanker can only fill the depot (or other recipient ship) until the levels are equal. On the other hand, if you connect the ships with one inverted (we'll call this "Thing 2" for absolutely no reason again) then you can transfer as much as you want: The other possibility would be to have some sort of positive displacement pump involved so that the "Thing 1" position could work. This would require an entire new system. However, it might be necessary if the "spray it into the empty portion of the tank" approach in the "Thing 2" position would create too much evaporation.

Oh, that's smart. They'll still be able to utilize a low-level continuous settling burn to effect the transfer. Unclear, however, how a settling burn will operate if they are docking side-to-side now instead of tail-to-tail.

Interesting that this is a significantly lower perigee than in prior launches. With the last one, the initial orbital parameters were a perigee of 182.5 km and an apogee of 299.3 km. By the time the perigee had dropped to ~170 km, the apogee was already down to ~232 km and it was very close to re-entry.

From that article it looks like everything went to GEO together.

I wonder if the entire 3,750 kg payload is being placed into GEO or if parts of it broke off in transit (in LEO or GTO).

Off, usually. Oh, wait, you said MULTIPLY six by nine, not PUT six by nine.

The answer is trivial. The question is the challenge.

It seemed very different to me. The ground tracking cam showed the entry burn on one booster starting at approximately the same time the other started, and the landing burn of one was visible from the other.

People assume that nuclear rocket engines must be wildly, wildly overpowered, but they really aren't. The specific impulse of a rocket engine is primarily a function of three things: chamber temperature, expansion ratio, and propellant molecular weight. Believe it or not, the peak chamber temperature of the RS-25 (3,300°C) is actually 29% higher than the chamber temperature of a NERVA nuclear thermal engine (2,500°C). The reason NERVA gets higher specific impulse is not because it imparts greater energy to its propellant, but because its propellant is pure hydrogen, which has a much lower molecular weight than the mixture of hydrogen and water that comes spewing out the back end of an RS-25. If you're planning a mission and you want as much Δv as possible for a small payload, you're actually going to opt for a high-thrust engine on your upper stage with less efficient propellants, because your mass fraction is going to be the biggest factor. On the other hand, if you have a very large payload and a relatively low Δv requirement, you're going to want to maximize specific impulse, since your mass fraction is limited by the size of your payload. So a nuclear thermal engine is good for sending big payloads, not for sending small payloads fast. An NTR would be great for sending very large monolithic payloads to the moon. It would not get you to the moon in a matter of hours.