It has been a while since I've commented, and the reason for that is that the last couple canon posts have posed some conundrums that I have trouble processing. I'm confused by the Soviet attempt at recovery for instance. In general it isn't problematic, as we don't know yet how well it works. It is all very well to proclaim reforms, but making them actually happen, and then having them actually work, are very different things. But what throws me is the East German brainwave of selling their petroleum allocation. One might suppose the Kremlin planners were too hidebound and conservative to think of simply trying to sell their petroleum production on the world market directly, i guess. They designed their input-output material flow models in an era where oil was cheap for one thing. This might also explain why they allocate more petroleum to their satellites then the latter would strictly need--if they were able to adopt Western conservation methods adopted after the oil crunch of the Yom Kippur War. (The second oil crunch of the late 70s was due to the Iranian revolution, which is not happening here--yet--due to heavy US intervention, which the Soviets tolerate quietly in return for a free hand in Afghanistan). But how much oil can the DDR sell, without sacrificing production goals the stuff was allocated to permit? If it is possible for them to then use the money from selling the oil to simply purchase on the Western market the items that oil was supposed to support producing in East German factories and fields--well that's a really glaring commentary on how extremely inefficient those productive processes were in the DDR! A combined play would be to sell some of the oil, and use the money not only to purchase the shortfall in planned items but also to renovate the processes to be more efficient. If they just shut down the factories and farms and bought the items, not only would it be doubtful they could come out ahead but they'd be violating the premise of a Socialist People's Republic, in which everyone is guaranteed jobs; even if the workforces idled were still paid the same wages and able to buy the same goods, it leads directly to social unrest even so. Anyway the oil is not drilled in the DDR and the only reason I can see for the Soviet hegemons to allow this to happen is if they realize that the satellites can perhaps manage to close deals with Western bloc nations that the Soviet Union itself would not be allowed to make. So it boils down to a money laundering scheme, but it is hard to see how the satellite regimes are allowed to pocket more than a tiny token of the proceeds. The oil comes from Russia after all. That was the least of the puzzles that bother me though. Going on in order of ascending puzzlement, I don't see how Columbia can manage its modest Lunar orbital mission on the propellant margins the Wiki page gives it. 1900 m/sec total delta-V strikes me as quite inadequate; it allows practically nothing for mid-course corrections and even with allowing exactly zero for those, it doesn't seem quite enough to go into low Lunar orbit and then out again and back to Earth. We could free up a little margin by supposing that the Columbia capsule is sent, not on the fast TLIs of OTL's Apollo, but on a minimum energy Hohmann orbit. Such a TLI would save very little propellant in the launch itself but it would have the ship approaching the Moon at at a significantly lowered speed, knocking a few hundred m/sec off the maneuvers into and out of LLO. However there are two prices; for one thing the transit to and from the Moon is slower by a couple days. For another, there is no possibility of a free return orbit; these require "ballasting" the approach to the Moon with the extra velocity the faster orbits provide, and a minimum energy approach would absolutely require a successful rocket burn to achieve a survivable return path to Earth, even if the orbital mission were aborted. If their rocket fails they are simply dead. It better not fail! Another way to get more margin for Columbia, and allow for faster transit and possible free return orbits, would be if the "low" Lunar orbits were a lot higher than those used by Apollo OTL. Apollo of course was aimed single-mindedly at landing men on the Moon and had to go for low orbits just a hundred miles or less above the Lunar surface. Columbia can opt for higher orbits since there is no immediate prospect of landing; if the Soviets were in a better position to compete there might be a "race" to achieve the closest orbit but the Russians can't even put anything manned into orbit at all. So I guess Columbia might be marginally able to claim to make low Lunar orbits after all--if we push the mission toward the margins in several ways. But then the talk of extending Columbia into a Lunar landing program is, as the author has later clarified, still just talk. It is perfectly clear to me that the tonnage of Columbia, which clearly should be greater to perform the mission of going down to low altitudes to retrieve a lunar landing crew, is far too low to serve as a suitable LM. That is, presumably if Columbia is the most the Americans can yet place in lunar orbit, a custom designed LM of the same mass is simply too small, about half the size of the OTL Apollo LM. Thus a mission to the Moon designed around the existing Minerva rockets would have to rely not on two but three launches, or even four, to provide for an adequate set of craft to get the job done. The more launches, the more problematic linking everything together suitably, before all the hydrogen boils away, becomes. I think it might still be done, but it is going to be a major project, not a simple extension of Columbia. So onward to the fourth and most troubling problem: The Air Force proposal for a successor to DynaSoar launched with a single fuel tank and reusable engine from the back of a subsonic airplane strikes me as preposterously impossible! The problem, in a nutshell, is gravity loss. I can believe in a disposable tank holding a couple hundred tons of hydrogen-oxygen propellant to fuel a single J-2S installed in a trailing reusable manned spaceplane in the 10-20 tonne range all right. Such a rocket setup should be able to push itself to a 7800 m/sec orbital velocity, starting from the 300 m/sec or so that an airplane cruising just below the speed of sound in the stratosphere provides. Indeed we don't want the assembly to actually reach full orbital velocity, not quite--we want the fuel tank to burn up in the atmosphere, otherwise we are cluttering up low orbital space with lots of loose tumbling tanks as so much space junk. So the actual delta-V achieved should be some 50-100 m/sec less, with the spaceplane required to circularize its orbit using onboard auxiliary rockets, presumably hypergolic. But in the interim, it is going to take hundreds of seconds to achieve the necessary near-orbital velocity, and during those seconds it is necessary to climb from stratospheric to a suitably high altitude for sustainable orbit--I guessed about 100 nautical miles, at which the speed is the aforementioned 7800 m/sec, just about. It is not enough then to provide the acceleration needed merely to close that velocity gap--the craft must be lifted well over a hundred kilometers into near-vacuum. And the average speed necessary to do that is dwarfed by gravity loss. Over hundreds of seconds, during most of which the pull of Earth's gravity is countered only a very little bit by centrifugal force, the downward acceleration will mount into thousands of meters per second. Averaged over the 7000+ it needs to make tangentially this is not so bad, but unfortunately we can't average. We can to an extent. If we want a graceful, elegant approach to orbital altitude, we need not thrust at all in the vertical direction as orbital speed is approached--we use the residual net pull of Earth's field to brake the climb to a stop just at the altitude we want. In this last phase of boost, gravity "loss" works in our favor. But this means we have even less time to make up for the downward pull we don't want, earlier in the boost! And where we need the most thrust, at the beginning of the burn, we have the least acceleration available. What we need is to already be moving upward with considerable velocity, a thousand meters a second or more, when we start the spaceplane's installed rockets. There is simply no way to get it with the arrangement offered by DoD a couple posts back. A whole lot of thrust might seem to be the ticket--but thrust costs engine weight, and we can't be planning on installing too much engine weight in the spaceplane. Enough thrust would have to be provided by other engines, which in the DoD sketch have no place to be installed and no prospect of recovery, nor is it clear where their propellants would be stored. Doing it by the brute force of installing another stage messes up the concept in a lot of ways. The booster stage would have to be disposable for one thing. Where to install it? The spaceplane is in a good place vis a vis its own propellant tank; if air launching it would work, then it can easily abort by separating from the tank and maneuvering away. But if a booster stage were installed behind it it would be sandwiched between two tanks and in a terrible position to escape. Should we mount twin boosters on the sides of the tank? That puts the spaceplane right between their two exhaust plumes! No matter how we arrange the booster, obviously it will have very large mass compared to the second-stage tank and spaceplane; we must double or more the take-off weight of an already fantastically large carrier airplane. Is there a way to save the basic concept, of "zeroth" stage airplane carrying the tank and spaceplane to a suitable height and speed to release it to finish the job, with the airplane then flying back to base and only the tank being lost? I say, sure there is--but the airplane would become a piece of really grandiose engineering! It might prove to have other uses, or simply so cheapen orbital operations there is a large demand for many of them operating continually in the launch role--but while the long-run economics might work out to be cheap, the short-run development costs would be colossal and the time frame would be a decade or more I guess. What is needed is not subsonic but supersonic launch--indeed the plane should not launch the rocket until it has not only exceeded the speed of sound by a considerable multiple (Mach 3 or more) but also gained a very large upward velocity, to buy time for the spaceplane's modest rocket to push for near-orbital speeds. Launching from 1000 m/sec instead of the near standing start of 300 that a subsonic launch offers will cut down the necessary mass ratios to achieve orbital speed significantly--which is good because even with an upward toss of a kilometer per second or more, we still need a lot of downward thrust during the burn to to maintain an adequate upward speed. Although a supersonic plane that can raise itself and a hundreds-of-tons payload by aerodynamic lift alone to start climbing at several G's is not impossible (just very very grandiose and expensive to develop!) as it cranks up the lift coefficient to climb it will of course greatly raise the drag; the thrusts needed to climb will dwarf those needed to merely achieve and maintain supersonic speeds flying more or less straight and level. We are not merely climbing; we are pouring on the coal to rapidly achieve upward speed measured in kilometers per second. As it climbs, the angle of flight will be turning vertical even if we have managed as much as 1000 m/sec horizontally and maintain that; we are aiming for launch angles of 45 or even 60 degrees above the line to the horizon. Obviously we can't rely on airbreathing jet engines, even upgrades of the ones used on SR-71 or the Valkyrie B-70 bomber, for this phase--at this point we'd need rocket engines putting out ten or more times the thrust, guzzling down propellant to do it. We'd need them only briefly, a minute or so, then the plane can release its payload which can then boost on to orbit on its own. Besides the sheer inadequacy of the jet engines to give the burst of thrust we need, we would also obviously be climbing, in this brief minute of burn, well above the levels of the atmosphere where jet engines could be sustained--indeed, the wings even in their most aggressive angle of attack would be providing little to no lift. In the final boost, the airplane ceases to be an aircraft and becomes, gradually, a ballistic rocket! A drastically suborbital one to be sure, but I suppose it would coast on under its upward inertia to a hundred kilometers or so altitude, and there reach apogee and start curving back down on a steep parabola back into the atmosphere. The thing, to be recoverable, would have to brake hard, presumably by turning to put its flat side flat onto the velocity vector, braking down to more or less sustainable speeds--Mach 2, say, or perhaps even falling below the speed of sound, before restarting its engines and flying back to base. This, I submit, is the sort of airplane it would take to get the job done on the terms the Air Force proposes. Subsonic launch will not cut it! Such an airplane might after all be worth developing. But it is clearly a gigantic project, not a simple kludge of anything off the shelf as the presentation so rosily suggests. It must first of all lift itself and any propellant it contains along with a couple hundred tons of cryogenic propellant tank with a spaceplane stuck on the end off a runway, then climb to the stratosphere with this load on airbreathing engines, then push the whole ungainly mess through the sound barrier and on to Mach 3 or so. We have good reason to believe this speed should be attainable, despite the awkward parking of a tank and spaceplane on top (or below, might be better). But at what mass ratio, of airplane to cargo? It is pretty Utopian already to suppose the ratio might be as low as 1:1. But note that if the orbital part already masses at least 150 tons, and possibly twice that, the TOW is already in the 300-500 ton range, which is to say pushing the limits of anything achieved hitherto OTL. We must bear in mind the carrier plane must not only carry fuel to climb and boost to a straight and level Mach 3, but then carry a lot more rocket propellant to double that on the way to suitable release conditions. I'm guessing that overall, we'd be amazingly good engineers to get the launcher plane down to just twice the mass of what we have to launch, and so we are looking overall at something between 500-1000 tons! And that is not run-of-the-mill subsonic aircraft but something that can match the performance of an SR-71. It is a real monster. If it can be done, it would definitely be something to be proud of, and despite the obviously staggering expense of making and maintaining it, it might after all lower the cost of launches to orbit overall. Well, it had better! But of course this supersonic Goliath is not at all what the Air Force is asking for in the ATL. What they ask for--cannot, as far as I can see, fly at all. There is no way to get enough thrust into the spaceplane to boost to orbit from a puny horizontal Mach 0.9 and a vertical standing start! Something needs to give. Aha, says someone--what about Pegasus? That's real world OTL, it is a rocket launched from a subsonic plane, so there! Well, I should look it up, but I bet right now this is what we find: 1) Pegasus delivers an order of magnitude less mass to LEO than the ATL USAF launcher must, a ton or so. 2) to do so, Pegasus uses a higher ratio of propellant mass to orbited load than the ATL USAF could afford to--the airplane that can raise the necessary propellant to get the spaceplane into orbit would already dwarf the biggest thing in the American inventory and even the most overgrown Soviet design. 3) in some combination, Pegasus either uses engines that put a very high G load on the payload as it reaches orbit, or uses multiple stages, with the major portion of total thrust created being in early stage engines that are disposable. The ATL USAF is claiming they will recover the engines used. My monster supersonic Mothra of a launcher will meet that claim (though an order of magnitude more thrust total will be installed!) Pegasus cannot. Nor can even USAF test pilot astronauts be expected to endure the G loads I believe Pegasus launches put on their payloads. Aha, says someone else--look, the Soviets of OTL had similar plans! Look here at MAKS for instance! Yep--the Russians were thinking about doing something similar, with the spaceplane installed on the tail end of the tank just as the author has shown--I would guess this is really a ripoff of MAKS. But IIRC--I might look it up soon, but right now my browser is overloaded as it is--MAKS was going to rely on an ambitious approach involving tripropellants. The tank would be bigger, but modestly so, because initially the rocket would burn kerosene and oxygen, generating a higher thrust at lower ISP; this high thrust would get it past the crucial crunch phase of low speeds, then it would switch over more or less gradually to burning hydrogen for higher ISP with lower thrust (but now the mass of the assembly would be much reduced). I'm not at all sure it would work, and note how the Russians haven't made it work in thirty years or so, nor has any richer power purchased or stolen the concept from them to realize it. It might be a more sensible investment than others that have been pursued--but there are still some daunting barriers. Such as developing a tripropellant engine for instance. There is no mention of this in the USAF proposal as given thus far. Perhaps the upcoming installment will elaborate to show the idea is indeed a transfer of the OTL Soviet concept. But I would need to see a lot of numbers to be convinced the approach will work! Being flabbergasted that the Air Force could get away with a such a blue-sky, rosy scenario with no numbers as a serious bid for replacing DynaSoar has paralyzed me for weeks now. I hope to see the logjam broken one way or another in upcoming posts.