Are radically reduced costs to space imminent with hypersonic flight?

RGClark · 2024-06-12 10:33:03

Several companies are investigating hypersonic air-breathing vehicles. These are vehicles that draw the oxidizer from the air like jet engines. Reaching hypersonic speeds would be to Mach 5+. The easiest approach to these speeds is to use ramjets. Scramjets can reach even higher speeds but are a much more difficult technological task.

I found the most promising approach that taken by Hermeus. They were able to keep costs down by using already existing turbojet engines and modifying them to operate in a ramjet mode:

Building the World's 1st Hypersonic Airplane | Hermeus.
https://youtu.be/jdKUN2V0PMM

But a key fact about the hypersonic air-breathing vehicles is how much they can subtract off from the delta-v needed to reach orbit.

Mach 5.5 is about 5.5*340 m/s = 1,870 m/s. But ramjets can also operate at quite high altitudes, well above standard turbojets, to about 30+ km, 100,000+ feet. The equivalent delta-v for that altitude is about 800 m/s. Then ramjets can supply 1,870 + 800 = 2,670 m/s delta-v to orbit. It turns out the trajectories that use aerodynamic lift can subtract some amount from the needed delta-v to orbit, so call it 9,000 m/s needed for orbit using lifting trajectory.

Then the delta-v that needs to be supplied by the upper stage would be 9,000 - 2,670 = 6,330 m/s. For expendable rockets the first stage commonly supplies about 4,000 m/s total delta-v speed+altitude with about 5,000+ m/s being supplied by the upper stage. But for rockets for which the first stage will be reusable like the Falcon 9 and the Superheavy booster, they supply a smaller total delta-v speed+altitude to the flight of about 3,000 m/s. This is so they can boost back to the launch site more easily.

Then this means the air-breathing hypersonic vehicle can supply about the same total delta-v, speed+altitude, as the reusable booster. But the majorly important advantage is the air-breathing vehicle can be reused thousands of times, compared to only a few ten's of times for the rocket booster. Since the first stage by virtue of its large size commonly takes up 75% of the cost of a launcher, this means the cost of launch will be greatly reduced when that first stage can be reused thousands of times.

The implications of this will be immense. It means for example despite SpaceX spending billions developing the SH/SS it may already be obsolete just by 2025 when hypersonic vehicles become operational and are used as the first stage of an orbital vehicle.

I think the hypersonic advance will be successful, and will thereby be used to launch payloads to orbit as a first stage, greatly reducing costs.

However, I think it possible it will also make possible another key advance: combined air-breathing/rocket SSTO's. The promoters of Skylon have argued this will be possible with their Sabre engine able to reach Mach 5.5. The problem with Skylon is it's projected $12 billion development cost. But by adapting already existing turbojets to ramjets, and utilizing existing airframes plus an existing hydrolox stage for the rocket portion of the flight this cost can be radically reduced, perhaps to only 1/100th of that, a few hundreds of millions of dollars.

See discussion here:

Low cost approach to winged, air-breathing and rocket SSTO's, Page 1.
https://exoscientist.blogspot.com/2024/ … d-air.html

Bob Clark

Last edited by RGClark (2024-06-12 16:53:04)

GW Johnson · 2024-06-12 12:32:48

Point 1 -- hypersonics is being oversold in the press releases. It has been possible to push payloads hypersonic for many decades now, using rocket propulsion; it's just that the powered range is short. The now-retired AIM-54 Phoenix missile used by the Navy peaked at Mach 5 speeds way back about 1980. It had almost 100 mile range with a peak altitude of near 100,000 feet, on a long arcing trajectory that came back down on its target. The other decades-old hypersonics technology is the "hypersonic boost-glider"; in the 1970's we called them "MaRV" for maneuvering re-entry vehicle. These are simply aerodynamically-maneuverable warheads on ICBM's. There is not the time available for true maneuvering, this technology really only makes the targeting more accurate.

Most of the "hypersonic" things being touted today are either rocket-propelled tactical or strategic cruise missiles, or else maneuverable warheads on ICBMs. My, how things stay the same when they change!

The airbreathing hypersonic things are lagging behind, precisely because scramjet, while sort-of operational, is still technologically difficult and still very "iffy" in terms of reliability. Plus, they are still trying to combine it with turbine, when the inlet duct and thrust nozzle geometries are, quite frankly, utterly incompatible. Fool's errand, that!

Point 2 -- It takes big heavy wings to lift big heavy weights in any sort of airplane, no matter how it is powered. Why carry those big heavy wings to orbit and back, when you do not need them above the stratosphere? That kills any rocket mass ratio you could have. I'll give you one guess as to why horizontal takeoff vehicles to orbit have never materialized. And why it is unlikely they ever will.

Point 3 -- All airbreathing propulsion has an altitude limit called "service ceiling", no matter what kind of airbreather. That includes piston, gas turbine, ramjet, and scramjet, or anything else you want to dream up. This is because the thrust it produces is proportional to the combustion pressure, in turn a fairly-constant ratio to the local atmospheric pressure, at ANY altitude. So, thrust decreases as air pressure decreases. Period! End of issue!

Weight does not decrease with decreasing air pressure. So there is an altitude at which you can no longer climb, nor accelerate. Period! End of issue! And with airbreathing jet propulsion of just about any kind imaginable, it is nearer 80-100,000 feet than anything much higher. Riding only airbreathing propulsion to the edge of space at near-orbital speeds is technically-ignorant marketing bullshit trying to land a pointless technology development program from the government! Always has been, always will be!

Point 4 -- it has been established over the last few decades that what things are the most important for getting things to space from an airplane are: (1) speed at staging off the airplane, (2) flight path angle at staging high enough to go ballistic without lift, and (3) altitude at staging, in THAT ORDER of effectiveness. Not reversed, that order!

Speed is obvious: higher is better. But you must pay for it!

Path angle is NOT obvious, until you think about what it takes to pull up steeply at high speed in very thin air. It takes big heavy wings to generate enough lift to pull up at all, in thin air, and it takes a long time, since the path radius of curvature and resulting path length are large. All that time you are inherently taking on very large amounts of drag-due-to-lift, which can far exceed the zero-lift drag of any vehicle configuration! And THAT adds greatly to your drag loss you have to cover! The faster supersonic you are, the worse this gets, exponentially! This is EXACTLY why the air-launched Pegasus was no more successful than it was, even with subsonic staging from the carrier plane, under 40,000 feet.

Point 5 -- If you want to launch a rocket stage from an airplane, the configuration should indeed resemble the B-58, where that pod underneath fills the same role, but is your rocket-to-orbit instead of a dropped bomb. You can stage supersonic, but you need to do it at relatively low altitudes down nearer 20-30,000 feet where the plane can reach climb path angles on the order of 45 degrees. The B-58 climbing that steeply might only reach about Mach 1.5, which is only about 0.45 km/s velocity. Point is, the airplane has to make the pull-up, so that the dropped rocket needs no wings.

Point 6 -- The other ~9 km/s has to come from the rocket stage. You won't do that single-stage with anything less than LOX-LH2 at any sort of believable inert mass fraction. If you use LOX-RP1, or likely even LOX-LCH4, your dropped stage needs to be a 2-stage rocket. Reusability will be in serious question, in either case. (Pegasus was multi-stage. It was a solid.)

GW

Last edited by GW Johnson (2024-06-12 12:59:21)

Mars_B4_Moon · 2024-06-12 13:03:32

The Rockwell X-30 was a the 1980s part of a United States project to create a single-stage-to-orbit spacecraft, scalled down NASA X-43 was an experimental unmanned hypersonic aircraft the Soviets or Russians for a while worked on the Tupolev Tu-2000 as a repose to X-planes.

https://web.archive.org/web/20161228024 … -2000.html

Galactic 07 crewed sub-orbital spaceflight of the SpaceShipTwo-class VSS Unity
https://x.com/planet4589/status/1799468227447476573

Stratolaunch launches 1st rocket-powered flight of hypersonic prototype from world's largest airplane
https://www.space.com/stratolaunch-firs … est-flight

Lockheed X-7s an American unmanned test bed of 1950s ramjet engines, L-171 was initially designated the PTV-A-1 by the USAF, "Flying Stove Pipe" was designed to test Ram Jet engines.
https://web.archive.org/web/20201027171 … e-exhibits
Speed 2800 mph or 4506 kph

There is also a lot of gossip about X-Planes out there, the USA, maybe China, the United States SR-91, Mach 6 a hypersonic plane that doesn’t officially exist some refer to it in media reports as the “Aurora” and VentureStar a proposed replacement of the Space Shuttle carrying 20,000 kg (44,000 lb) to LEO.

https://interestingengineering.com/mili … -the-sr-91

Last edited by Mars_B4_Moon (2024-06-12 13:30:23)

GW Johnson · 2024-06-13 07:34:41

X-30, aka NASP, was to use scramjet to extreme altitudes and very high speeds. Scramjet was absolutely unready to apply in the 1980's. Plus, see what I said about ANY airbreather at extreme altitudes in post #2 above, my point 1 in that posting. To that I would add that the max potential Isp of even a hydrogen-fueled scramjet is pretty much equal to the Isp of a LOX-LH2 rocket ay only about Mach 10.

That very low airbreather Isp at extreme speeds is not just low inlet pressure recovery (which is also inherent), it is also caused by the inability to actually burn fuel with ionized plasma from the inlet, instead of real oxygen-bearing air. Oxidation has to do with the outer electron shell. Once ionized, that shell is damaged or absent. The chemistry "quits". Ionization/recombination energies cannot be converted to thrust by a nozzle, only enthalpy can. Once ionized, nearly all the KE of the oncoming stream goes into ionization/recombination, not enthalpy. That has been well-established for decades, now.

In short: the X-30 as-conceived could never have worked. That was a giant corporate-welfare program.

There has been talk for at least 4 decades now about this-or-that replacement for the SR-71, often now termed "SR-72", long ago often termed "Aurora". Yes, there have been experimental planes aimed at that goal. No, none of them yielded usable results, until possibly recently. This thing Lockheed has been hyping of late they term "SR-72" was to have a combined-cycle gas turbine/ramjet/scramjet engine.
They might have succeeded after a couple of decades trying, if they just tried combined cycle gas turbine/ramjet and separated the scramjet portion almost entirely. If they did succeed, they'll try to keep it secretly flying for a while, like they did the SR-71. It'll be a few years before we know.

I know about the old X-7 and the stovepipe ramjets that pushed it. Those have a flight speed limitation of about Mach 3-ish, due to melting of the flameholders and perforated air-cooled combustor liners and nozzle hardware. Above about Mach 3-3.5, there is simply no such thing as cooling air. That has been known since the 1950's. We've known how to get around those limits since the early 1970's, using sudden dump flameholders, and ablatives for our combustor liners and nozzle hardware. Nothing else actually works.

EXACTLY those technologies were in ASALM-PTV, a ramjet cruise missile prototype, flight tested successfully in 1980. This thing was boosted to ramjet takeover (NOT scramjet!!!) at Mach 2.5, then flew in ramjet power to cruise at Mach 4 and 80,000 feet. It would then dive onto its target at an average Mach 5. In one flight test, we had a throttle runaway failure, and it reached an unintended Mach 6 at only about 20,000 feet! It would have melted and broken up, had this been more than a several-seconds-long transient event.

That hypersonic airbreathing speed record stood until 2004, when NASA broke it with its hydrogen scramjet X-43A, in 3 second burn at Mach 7, boosted all the way to test speed by a rocket, and conducted at just over 100,000 feet (where the scramjet had insufficient thrust to accelerate) in order to limit aeroheating to survivable values. ASALM would cruise for several minutes at Mach 4, 80,000 feet, using a synthetic kerosene as its fuel, and demonstrating steady-state heat protection. We tested the combustor ablative liner as good for 15 minutes, steady state, by retaining the charred-through char as insulation. I'm not going to tell you exactly how we did that, but it WAS a breakthrough!

I have yet to see the limiting concerns properly addressed for airplane launch of an orbital rocket. I detailed those in point 4 of post #2 above. The easiest solution is a low-supersonic airplane with mixed turbine and rocket propulsion, carrying the orbital rocket on its belly. It must pull-up sharply to a path angle near 45 degrees, while maintaining speed (only the mixed propulsion can do that), and drop its rocket directly onto the gravity-turn ballistic path that lets the rocket function without wings for pull-up. If done at Mach 2, that's about 0.6 km/s speed at rocket ignition on a gravity-turn trajectory, leaving roughly 8.5-8.7 km/s delta-vee to be had from the rocket. You could do that single stage with LOX-LH2, or maybe even LOX-LCH4. Might have to be 2-stage if using LOX-RP1. Probably little or no reusability potential there.

Mixed jet/rocket propulsion was a highly-successful technology used for the high-speed research planes from the late 1940's into the early 1970's. The most striking example was the NF-104.

GW

Last edited by GW Johnson (2024-06-13 07:55:59)

tahanson43206 · 2024-06-13 11:37:40

For GW Johnson re #4

Thank you for mentioning ionization as a restraining factor in attempts to run air-breathing engines at high velocity.

This is the first time I have seen that particular factor mentioned as an explanation for why flight above a certain velocity with a traditional air breathing system is less and less efficient.

In the case of the British Skylon design, my understanding is that the intent is/was to cool the incoming air so that it can be efficiently combined with fuel to create thrust. I deduce that electrons would return to their proper place in the oxygen shells due to cooling, but would appreciate your evaluation of that scenario.

(th)

GW Johnson · 2024-06-13 23:52:00

My understanding of Skylon and its SABRE (liquid air cycle) engines is airbreathing flight to only about Mach 5 at only about 100,000 feet, during which atmospheric oxygen gets stored in oxygen tanks that are empty at takeoff. At the M5/100k point, those engines shift to LOX-LH2 rocket propulsion, with which the vehicle pulls up steeply to get onto a non-lifting thrusted gravity turn trajectory. It can pull up, and it can accelerate upward along the steep path, PRECISELY because the propulsion is rocket. Rocket takes it to orbit. Single stage is possible because it tanks the oxygen during the early ascent, not before takeoff.

The SABRE engines are liquid air cycle engines. The air is specifically cooled to liquify it, and separate the oxygen from the nitrogen by way of the different boiling points. (This only works with super-cold liquid hydrogen as fuel.) In rocket mode, you dump the nitrogen, and burn the oxygen in the engines with hydrogen fuel from the tanks. In airbreather mode, you dilute the oxygen with at least some of the nitrogen. The combustor and nozzle are more akin to a liquid rocket engine than any gas turbine engine ever built.

This was an old concept dating to the 1950's, but has never been made to work until now. SABRE's makers seemingly have solved the problem of getting really fast heat transfer rates, and they have seemingly solved the problem of how to deal with the moisture in the air which condenses first and freezes all over everything. But, this thing has yet to fly, which is the real proof of the solutions.

I have no problems with any of that, but I do think they have not thought through the whole airframe layout problem, or else its designers had zero understanding of the realities of entry heating. They do NOT have a survivable heat protection scheme, because of shock impingement heating that WILL happen during entry! I have talked about this issue with that design before, here on these forums. It is a FATAL design flaw!

Those tip mounted engine pods have inlet compression spikes whose conical shock waves WILL impinge upon the wing leading edges (there is NO way around that with the parallel-nacelle shape). Those impinging shocks will cut the wings off in a matter of seconds during entry, no matter how thick a piece of carbon ablator they put on those leading edges. This was shown by the well-documented damage to the X-15A-2 when it reached (only !!!) Mach 6.7 at (only !!!)100 kft back in 1968, with an experimental nacelle mounted in parallel to the fuselage, on the ventral fin stub under its tail. Skylon will hit entry heating at Mach 25. Crudely, the expected intensity of stagnation heating varies as velocity-cubed during entry. That will be ~50+ times the heating rate the X-15A-2 dealt with.

Shock impingement heating does not increase gas (or plasma) temperatures, but it does increase the heat flow rate per unit area, right in the zone at the impingement, by a factor crudely near 10. Because of the effect of 4th-power re-radiation rates, the equilibrium temperature of the affected surface will be above all known material melting or ablative destruction points. We've already seen it!

GW

Last edited by GW Johnson (2024-06-14 00:02:56)

tahanson43206 · 2024-06-14 06:26:29

For GW Johnson re #6

Thank you for the overview of the Skylon concept and for a reminder of the power of the physics of sound waves the vehicle will encounter if it ever tries to return from orbit.

However, my original question was about the effect of ionization you had identified as a problem in Post #4.

I read #6 carefully and did not find a reply to the ionization question. It is possible this question has never been asked before, and there may not be any data one way or the other.

However, there is a new "problem" or issue that #6 reveals...

You've suggested that it might be possible to separate nitrogen from oxygen while cooling gases at supersonic speeds.

That seems (to me at least) on the challenging side.

By any chance, to you have a reference with details about that part of the operation?

My understanding of the design is that the goal is to reduce the relative velocity of the atmospheric gases to subsonic, so that they can mix with fuel and achieve combustion. That is not the same thing as liquefying the incoming gases.

(th)

tahanson43206 · 2024-06-14 06:43:09

For GW Johnson re #6

This is a separate question from the one about ionization...

In #6 you have reminded us of the impracticality of heading straight on into the atmosphere at the velocities where sound physics will cut off any material known to humans.

However, the return flight might employ the same technique as is (apparently) being used by SpaceX for Starship.

In the case of Starship, my understanding is that the vehicle will present the broadest possible area to the atmosphere, to maximize the opportunity to reduce velocity by creating the maximum possible drag.

My question for you in ** this ** post is why the Skylon flight planners would do anything different from SpaceX?

The vehicle could be given a deorbit burn, and then oriented to present the entire fuselage to the atmosphere for as long as necessary to arrive at a velocity where operating the engine might make sense.

It seems to me that there might not be much fuel left at that point, so the vehicle might operate as a glider.

(th)

GW Johnson · 2024-06-14 09:06:06

The ionization thing happens any time you try to combust fuel with air hot enough to be ionized. Down in the stratosphere, that starts happening about Mach 7 or 8 with inlets that try to slow the air subsonic. Scramjets have a different post-capture inlet shape that leaves the air supersonic, just slower supersonic than the oncoming speed. That delays the ionization problem to the vicinity of Mach 10. The static (thermodynamic) temperature of a subsonic stream is very near its total (stagnation) temperature. The static temperature of a supersonic stream is very much lower than the total temperature. In both cases, the temperature ratio total/static = 1 + constant*Mach number-squared, where the constant is computed from the gas specific heat ratio "g" as const = 0.5*(g - 1). This math fails once ionization becomes significant, because the equation of state PV = nRT is no longer valid. The chemical identity of the gas is no longer what it was because of the ionization. Chemistry IS the electron shell, and that is different after ionization begins.

Skylon is intended to leave the sensible atmosphere at speeds not much above Mach 5, where the magnified heating from the impinging spike shocks upon the leading edges, is still tolerable. The faster you go, the more intolerable this becomes, exponentially, because the basic heating is so much larger at higher speeds. You cannot take advantage of "leaving the sensible atmosphere" during reentry descent, BY DEFINITION. You WILL see about orbital speeds about halfway down in altitude from the entry interface, which for Earth from LEO is about Mach 12-13 at about 50-60-some km. The heating there is truly intense, and shock impingement magnifies that problem by a factor on the order of 10.

My point: there is a very good reason why no entry-capable spacecraft has ever flown with any sort of parallel-mounted nacelles or any other structures that could shed shock waves toward the main body. It is the shock impingement overheating risk during reentry. This puts heating into the affected material at a rate that is an order of magnitude (or more) faster than it can be removed by any conceivable means. That affected material then almost-instantaneously soaks out in the local affected zone to the effective driving temperature of the plasma sheath, which at 8 km/s is about 8000 K, at 7 km/s is about 7000 K, etc.

You tell me what materials might still be solid at such temperatures, not to mention structurally strong enough to resist wind loads. See why I talk about the impinging shocks cutting the wings off in mere seconds?

Re-entry vehicles that are lifting bodies or have wings enter the atmosphere at a high angle of attack, but not dead broadside! The space shuttle came back at a very precise 20-40 degrees AOA (40 at peak heating), higher or lower would cause hot plasma impact directly on the windscreen, causing immediate loss of the vehicle and crew. Even "Starship" comes back at 60 degrees, not dead broadside. I do not know the entry AOA for Skylon, but it would be in that same class.

Higher AOA will rip the wings or flaps off. Remember, peak entry gees are usually in the 3-7 gee class (which is why entry angles are always shallow, steep entries are hundreds of gees). Like in airplanes, you can only pull so many gees without ripping the wings off. Making it strong enough to resist more gees usually makes it too heavy to fly. 6+ gees is a lot of gees for any wings to resist. You don't do that at spacecraft-type vehicle inert mass fractions. Skylon CANNOT come back dead broadside, where the shocks don't impinge on the wings. The wind loads would rip the wings right off.

The main known way around these heating and structural integrity conundrums during reentry would be to slow way down before hitting the atmosphere, from about 8+ km/s to something nearer only 1 or 2 km/s (Mach 3 to 6, instead of 25). But that's a big engine dV requirement, which defeats the whole purpose of wanting to aerobrake instead of burning propellant.

There is a second: fold the wings into the wake, and come back dead broadside. But peak gees will be much higher that way. It might still be too much, if there are masses hanging on those wing tips.

GW

Last edited by GW Johnson (2024-06-14 09:28:28)

tahanson43206 · 2024-06-14 09:44:32

For all... here is a link that Google found to a site that shows the Skylon shape, including the parallel engine pods that GW has pointed out are the Achilles's Heel for Skylon safe return.

https://www.cnet.com/pictures/skylon-sp … -images/3/

The site offers multiple images. Number 8 is the one that GW has pointed out as of concern.

If I understand GW's presentation correctly, the shock waves from the tips of the parallel engine pods at the extremity of the winds will sheer off the wing.

The artist who created the rendering would have had no way of knowing that, and the Skylon engineers would (most likely) have missed that detail from the reams and reams of engineering knowledge that they would have studied.

For GW:

If the engine pods were dropped (ejected) from the wings before re-entry, would the problem you've described go away?

I understand this is a costly solution, but I'm trying to put brackets around the problem.

***
Also for GW ...

Is the problem you've identified addressable by changing the shape of the engine pod prior to re-entry? I'm thinking of putting a cap on the tips of the engines. Could a cap be designed to cause the shock wave energy to move away from the wings and out away from the aircraft?

***
Finally for GW ...

If using propulsion to reduce velocity prior to entering the atmosphere is the solution, then all we are talking about is the cost of propellant.

Can the Skylon carry enough rocket propellant to drop it's orbital velocity low enough to survive re-entry?

***
Perhaps a combination of reducing the orbital velocity using the engines, and altering the shape of the vehicle at re-entry time, might insure the survival of the vehicle?

(th)

Last edited by tahanson43206 (2024-06-14 09:58:18)

GW Johnson · 2024-06-14 11:33:15

For GW:

If the engine pods were dropped (ejected) from the wings before re-entry, would the problem you've described go away?

Ans.: yes, but it is hard to imagine how to recover them, or how to get them to survive entry tumbling. They are the most valuable portions of the vehicle, even more so than standard rocket or jet engines. That also precludes post-entry propulsion for Skylon, reducing it to a dead-stick glide landing without much divert capability at all.

Also for GW ...

Is the problem you've identified addressable by changing the shape of the engine pod prior to re-entry? I'm thinking of putting a cap on the tips of the engines. Could a cap be designed to cause the shock wave energy to move away from the wings and out away from the aircraft?

Ans.: No. Even a blunt object sheds a detached bow shock, which only a bit more remote from the object (about a cross section dimension away) is just about the same angle and shape as the shock shed by a sharp object.

Finally for GW ...

If using propulsion to reduce velocity prior to entering the atmosphere is the solution, then all we are talking about is the cost of propellant.

Can the Skylon carry enough rocket propellant to drop it's orbital velocity low enough to survive re-entry?

Ans.: No, not in anything resembling its current form. Wig = 275 tons, Wp = 220 tons, and Wpay = 12 tons, per the article. The "allowance" for inert + payload is ignition minus propellant = 55 tons. Subtracting 12 tons of payload means their design concept's inert mass is 33 tons (edit correction 43 tons), an inert fraction some 12% (edit correction 15.6%) of ignition. The thing as it is is 80% propellant. Bear in mind the dV to deorbit from LEO is only about 0.1 km/s. The dV to "survive" entry without serious aerobrake heating is 6-7 km/s.

They think (per the article) that they can use carbon composite for exposed airframe structure and meet that inert percentage with little in the way of heat shield, based on a BS argument about "lower ballistic coefficient" (which in turn tells me no one on that team knew anything substantive about entry heating). But that carbon composite material is limited to failure of the epoxy matrix at material soak-out of only 300 F. It requires protection, even on the lee-side, just as SpaceX found out with their "Starship", which is one of two good reasons they went to stainless steel.

I see little evidence they had anybody on the team that came up with this concept who actually understands reentry. The only way I see to "save" the design is to fold the wings into the wake zone, which cannot be done with a round body. The cross section will have to be almost triangular to do that, with a rather flat belly in that view.

That required cross-section shape will play merry hell with pressurizable round tank shapes that must fit within an airframe instead of actually being the airframe. Between that and adding both a heat shield PLUS INSULATION BETWEEN IT AND THE COMPOSITE AIRFRAME, I would expect a vehicle inert mass fraction nearer 25-35% than 12% (edit correction 15.6%), leaving only 65-75% propellant even at zero payload. There went your dV capability!

One other thing: there are 3 sharp stagnation points on this design: the nose tip, and the two spike cones for the engine inlets. The spikes must be sharp to function properly as supersonic inlets. The nose tip does not have to be sharp, but rounding it does increase drag coefficient. The problem with "sharp" in entry is high heating rate. The old first-order heating correlation equation that still works today says stagnation Q/A = constant * (atm density/nose radius)^0.5 * velocity ^ 3. I use that, and so did Justus and Braun in their seminal paper about entry, descent, and landing on multiple planets. It is from the 1953-vintage warhead entry work of H. Julian Allen. It worked pretty good then, still works pretty good today.

Note the dependence of Q/A upon 1/square-root-of-nose-radius. When that radius goes to zero, that factor in the heating correlation goes infinite. For entry from LEO, when the nose radius is about the same as the cross section dimension of the object, peak Q/A at about 7-8 km/s and 50 km altitude is in the hundreds of Watts/sq.cm (cm not m!!! so these heat flux rates are 4 orders of magnitude larger than most Earthly heat transfer applications). At 10% of that dimension, it's in the low thousands. At 1%, it's in the high thousands. And THAT is why nothing sharply pointed is still sharply pointed after an Earth reentry! (And that assumes no shock impingement phenomena multiplying those heating rates further.)

Do you see a problem with wing leading edges when that "nose radius" ratio is nearer 1% than 10%? Guess why space shuttle leading edges were carbon-carbon slow ablators and not the ceramic tiles. The two-piece Tufroc tiles on the X-37B can survive that kind of stagnation heating, but only just barely (!!!), and even then only on a rather blunt nose tip.

GW

edit corrections done at 3:26 PM CDT 6-14-2024

Last edited by GW Johnson (2024-06-14 14:27:08)

tahanson43206 · 2024-06-14 15:30:44

It's past time for enthusiastic SSTO true-believers to weigh in! I'm trying to carry the load but am obviously way out of my league.

***
So! For GW ...

If the answer is to carry enough propellant to perform a 6-7 km/s deorbit burn, then let us assume a way exists to permit that.

Can you envision a successful flight under those circumstances?

It seems to me that the mass of the vehicle will be significantly reduced at de-orbit time.

All the fuel and oxidizer needed to reach LEO is gone.

The amount of fuel and oxidizer needed to remove forward momentum by 6-7 km/s is much less than was needed to reach orbit in the first place.

(th)

RGClark · 2024-06-14 23:01:32

GW Johnson wrote:

My understanding of Skylon and its SABRE (liquid air cycle) engines is airbreathing flight to only about Mach 5 at only about 100,000 feet, during which atmospheric oxygen gets stored in oxygen tanks that are empty at takeoff. At the M5/100k point, those engines shift to LOX-LH2 rocket propulsion, with which the vehicle pulls up steeply to get onto a non-lifting thrusted gravity turn trajectory. It can pull up, and it can accelerate upward along the steep path, PRECISELY because the propulsion is rocket. Rocket takes it to orbit. Single stage is possible because it tanks the oxygen during the early ascent, not before takeoff.
The SABRE engines are liquid air cycle engines. The air is specifically cooled to liquify it, and separate the oxygen from the nitrogen by way of the different boiling points. (This only works with super-cold liquid hydrogen as fuel.) In rocket mode, you dump the nitrogen, and burn the oxygen in the engines with hydrogen fuel from the tanks. In airbreather mode, you dilute the oxygen with at least some of the nitrogen. The combustor and nozzle are more akin to a liquid rocket engine than any gas turbine engine ever built.
…

Skylon doesn’t use air-liquification. It does use a precooler to cool the air that entered at ramjet speeds and got slowed down by the intake ramps to subsonic speeds. Using this approach the air can then be sent to the usual compressors used on turbojets. When it switches to rocket mode it uses both hydrogen oxygen stored on board at launch.

Bob Clark

tahanson43206 · 2024-06-15 09:09:25

For RGClark re #13

Thank you for pitching in to help with the progress of this topic.

It is helpful to be reminded of how Skylon is intended to work, and how it is designed.

GW Johnson has pointed out what appear to me to be fatal design flaws that would cause destruction of the vehicle upon re-entry.

GW Johnson has offered a solution that would allow Skylon to return safely from LEO, by reducing it's velocity to a level that it's existing airframe can handle.

The dv change is (apparently) on the order of 6 - 7 km/s. LEO (of course) is around 9 km/s.

The needed dv change for safe return would be possible if enough oxygen and LH2 could be carried to LEO to accommodate the burn.

The existing vehicle appears to be just barely able to reach LEO with propellant loaded at launch. I haven't seen any data to support this guess, but for this discussion, I will assume the prediction is accurate.

In order to perform a de-orbit burn of 6-8 km/s the vehicle would need a supply of propellant.

In other topics in this forum, members have proposed a variety of methods of giving a vehicle momentum at launch to help to improve performance of the system.

Skylon has the distinct advantage of a design for horizontal take off and landing. Thus, it is a candidate for any of the proposed horizontal launch assist concepts that have been explored in the forum.

I'd like to toss out this one, that I have not seen previously here in the forum, or anywhere for that matter: One of the modern high speed trains might be enlisted to provide an initial horizontal velocity on the order of: 300 km/h.

While that velocity is only (about) .08 km/s, it is never-the-less potentially enough to give the Skylon design a running start that would (potentially) allow it to save on propellant at launch, and thus have enough to return safely from orbit.

The mass of the vehicle in LEO will be much less that it was at launch, so less fuel will be needed to perform the de-orbit burn that would have been true at launch.

(th)

Last edited by tahanson43206 (2024-06-15 09:10:10)

GW Johnson · 2024-06-15 09:45:08

I checked, Bob is right, SABRE is not a liquid air cycle engine. I honestly thought it was. It does have some sort of translating nozzle design, connecting to two different combustion chambers, one for the airbreather at lower expansion ratio, and one for the rocket at higher expansion ratio. The precooling is how they can do axial-flow compression to very high pressures, and why they can do this past about Mach 3-ish, where the blading in conventional axial-flow compressors on conventional gas turbine engines gets too hot and self destructs under the loads.

That being said, I don't think there is a practical solution for the Skylon airframe (as it is usually depicted) to survive re-entry, not with those engines on its wingtips. I think better thought went into the engine than did the vehicle design. The Wikipedia article about Skylon has the BS about reduced ballistic coefficient reducing entry heating. Use my entry spreadsheet and run the trades yourselves. Yes, lower ballistic coefficient raises deceleration altitude. But only by a little bit here at Earth. So the peak heating is only lowered a little bit. The effect is dramatic at Mars. Not here.

Not with those pointy spikes and nosetip, and the necessarily-low radius wing leading edges! The engine inlet spikes must be sharp, or else they cannot properly provide the external shock wave compression function that they must serve. Blunt them, and the shock detaches as a bow wave out front, which greatly increases the drag of the engine nacelle. But as I wrote in the earlier post, sharp things cannot stay sharp after being through entry. Not and survive the heating.

It also claims a magic skin material which will survive the (unrealistically-low) heating: some sort of ceramic matric carbon fiber composite. To the best of my knowledge, there is as yet no such material, other than as a lab curiosity. Even if it did exist, the shock impingement heating raises the equilibrium leading edge material temperature very close to soaking-out at the effective plasma sheath driving temperature: something in the 7000-8000 K class. Not even a ceramic matric carbon fiber composite (should one actually exist) will be able to withstand that!

Multiple layers of foil between the skin and the load-bearing internal structures and aluminum tanks are supposed to be "insulation" preventing heat from conducting inward from the hot skin. There's better insulations than that, but the whole concept ignores conduction between the skin and the load bearing structures, passing through whatever things physically hold the skin on the airframe at all.

For those who don’t believe in shock impingement heating, the damage it did to the X-15 is documented in NASA Technical Memorandum TM-X-1669, “Flight Experience With Shock Impingement and Interference Heating on the X-15A-2 Research Airplane”, written by Joe D. Watts, dated October 1968. It is available publicly on the internet as a pdf file.

GW

Last edited by GW Johnson (2024-06-15 10:30:33)

tahanson43206 · 2024-06-15 12:12:22

The Skylon vehicle can re-enter backward.

It can use as much fuel as it has on board to reduce dv in the forward direction.

Since it will be pointing aft for the de-orbit burn, it might as well stay in that configuration.

The stern of the vessel and (importantly) the aft ends of the engines, can be designed to blunt or reduce the cutting action of the show waves.
(th)

GW Johnson · 2024-06-15 14:46:59

All objects of any shape shed shock waves at any speed faster than sound. At very fast supersonic and especially hypersonic speeds, that shock wave wraps about the object, but not very far from its surface. For pointy objects, the shock wave is attached to the nosetip, while for blunt objects, it is detached as a bow wave just out front. However the lateral sides of the shock wave "wrap" about the object are roughly the same shape, regardless. It cannot be significantly moved by changing the shape of the object. It simply "is".

GW

RGClark · 2024-06-16 20:09:47

GW Johnson wrote:

…
That being said, I don't think there is a practical solution for the Skylon airframe (as it is usually depicted) to survive re-entry, not with those engines on its wingtips. I think better thought went into the engine than did the vehicle design. The Wikipedia article about Skylon has the BS about reduced ballistic coefficient reducing entry heating. Use my entry spreadsheet and run the trades yourselves. Yes, lower ballistic coefficient raises deceleration altitude. But only by a little bit here at Earth. So the peak heating is only lowered a little bit. The effect is dramatic at Mars. Not here…

Do you have a link to this reentry spreadsheet? I want to examine the feasibility of this claim:

Wings in space.
by James C. McLane III
Monday, July 11, 2011
Wing loading (the vehicle’s weight divided by its wing surface area) is a prime parameter affecting flight. The antique aluminum Douglas DC-3 airliner had a big wing with a low loading of about 25 psf (pounds per square foot of wing surface). At the other end of the spectrum, the Space Shuttle orbiter has a high wing loading of about 120 psf. This loading, combined with an inefficient delta-shaped wing, makes the orbiter glide like a brick. A little Cessna 152 private plane features a wing loading of about 11 psf and modern gliders operate down around 7 psf. A space plane with huge lifting surfaces and a very low wing loading might not require any external thermal insulation at all. Building a space plane with a wing loading of, say, 10 psf should not be an impossible proposition. Perhaps some day it will be done.
http://www.thespacereview.com/article/1880/1

The only aircraft that come close to this low 10 psf wing loading are gliders and small single-engine propeller craft like Cessna’s and Piper’s.

The author mentions the Cessna 152 as close at 11 psf. But if you look at the empty weight it does fall under 10 psf:

You would have to replace the glass windows with metal of course. And you would have to make the landing gear retractable. But the biggest structural issue is those wing struts. As is usually the case with upper mounted wings, it has wing struts to help support the wing weight. These would likely burn off during reentry.

The Piper has lower mounted wing so doesn’t use wing struts. It also has retractable landing gear. The glass windows of course would have to be replaced with metal.

For either of these the wings would have to be made fold-away to fit inside a fairing. The propeller could be made fold-away during reentry to only be used when the craft has slowed to subsonic speed. We might also want to remove the propeller and engine to lower weight and just use a gliding landing.

Bob Clark

kbd512 · 2024-06-16 23:05:40

Dr Clark,

Could you post a link to any supersonic manned aircraft ever built, prototype or otherwise, which has a 10psf wing loading?

I would love to know more about how that was done, if such a vehicle actually exists.

GW Johnson · 2024-06-17 08:59:17

Wing loading works out as a function of size and intended use. Light aircraft are usually in the vicinity of 10 psf. Model aircraft are nearer 1 psf, for the same speed ranges. The older propeller airliners near 20-30-something psf, along with the propeller bombers of that time, all being under 300 mph. Higher-performance supersonic jet fighters today are nearly all in the 100 psf class, give or take.

The delta wing is not easily characterized as "inefficient". It depends upon what you want to do with it. The delta wing has a very severe center-of-pressure travel going from subsonic to supersonic and back. That introduces very severe center-of-gravity issues into any supersonic delta wing fighter or bomber designs, something known since the late 1940's. See XF-92, our delta prototype back then. See also the losses of the B-58 until the in-flight center-of-gravity control was automated to respond to an engine-out deceleration-to-subsonic faster than a human pilot could respond.

Countering that is the advantage of being able to reach angles of attack in the 40+ degree class at landing speeds, without adding a plethora of leading edge devices to keep from stalling. That has also been known since the late 1940's. Swept wings cannot do that. Nor can straight wings. And we know it works: F-102, F-106, B-58.

And adding a sort of strake ahead of the main wing, often termed a "double delta", does help control the center-of-pressure travel better. Such was on the shuttle, and the SR-71/YF-12A, and is on the FA-18, although that one is not strictly a delta. Such is a good add-on for most stub-type wings with swept edges.

The real problem of using a delta wing to achieve a low landing speed without high lift devices is that high angle-of-attack: the pilot literally cannot see the runway directly in front of him. He really needs a window through the forward part of the cockpit floor. Something almost impossible to do, in turn leading to other bizarre design implementations. See Concorde vs Tu-144.

While delta wings can be flown "tail-less", they work with better safety if you also add a horizontal tail somewhere: XB-70, A-4 Skyhawk, Tu-144.

GW

PS: the swept wing also has serious low-speed instability troubles, requiring leading edge flaps, and/or fences, and/or a step in the planform about mid-span, to help control those instabilities. You pay a design price to get transonic drag reduction with sweep. This was extensively explored with the research craft at Muroc Dry Lake in the late 1940's and early 1950's, particularly with the D-558-2 Skyrocket.

Last edited by GW Johnson (2024-06-17 09:19:20)

RGClark · 2024-06-17 10:46:06

kbd512 wrote:

Dr Clark,
Could you post a link to any supersonic manned aircraft ever built, prototype or otherwise, which has a 10psf wing loading?
I would love to know more about how that was done, if such a vehicle actually exists.

No supersonic aircraft has had a psf that low. I haven’t even been able to find any subsonic jet aircraft, military or commercial, with a psf that low. Moreover even a single engine propeller craft has to be small to wind up with a psf that low.

You might think you can just scale up the craft. But say you scaled up a 10 psf craft say 2 times in every dimension so its mass is 2^3 = 8 times greater. The problem is the wing area will only 2^2 = 4 times greater, so the psf will be 20 instead of 10.

Bob Clark

kbd512 · 2024-06-17 15:12:52

Dr Clark,

In that case, I'm going to go way out on a limb and say this isn't likely to prove feasible, even if it's theoretically possible. If there are no actual examples where something similar to what's required has been done in the past, then there's probably a good reason. I would opine that the reason it hasn't been done before, is that the wings would get ripped off of a hypersonic airframe with a wing loading that low.

RGClark · 2024-06-17 18:05:11

kbd512 wrote:

Dr Clark,
In that case, I'm going to go way out on a limb and say this isn't likely to prove feasible, even if it's theoretically possible. If there are no actual examples where something similar to what's required has been done in the past, then there's probably a good reason. I would opine that the reason it hasn't been done before, is that the wings would get ripped off of a hypersonic airframe with a wing loading that low.

Still, I’d like to see the numerical simulations or hypersonic wind tunnels tests to see what the max heating would be. If you read the full article I cited, both von Braun and the legendary NASA engineer Max Faget believed it possible:

Wings in space.
by James C. McLane III
Monday, July 11, 2011

http://www.thespacereview.com/article/1880/1

Bob Clark

kbd512 · 2024-06-18 08:29:44

Dr Clark,

I don't know what combination of materials might be suitable for 10psf, but perhaps a CNT composite with aerogel foam insulation could survive this. I still want to know what you intend to do about that "hitting the sensible atmosphere at Mach 10" problem. Assume this could work and you rapidly bleed off about half your speed while still fairly high up. You're then flying through the upper atmosphere, which is getting denser by the second, at significant hypersonic speeds, but your wings are gossamer thin and the kind of lightweight thermal protection required might protection your vehicle up to Mach 5 or so. If you can't reduce your velocity to Mach 3 or so by the time you descend to 120,000ft, you're in for a world of hurt. This is exceptionally risky, even if it can work. What if you're unable to broadside the vehicle due to a control authority problem? You'll be incinerated.

That picture is a reasonably accurate depiction, if this works. You still need significant thermal protection because the surfaces facing the oncoming flow are glowing orange, much like Iron heated to just below its melting point. We have TPS materials that we know will survive if we make reentry at the correct angle, and we fully understand how to maintain control authority in that regime. Using this alternative method is akin to the NASA HIAD "weight shifting" method of reentry control, which is not fully developed. Broadsiding the vehicle is quite similar to that, so a lot more development work is required on the fundamentals of vehicle control. Maybe much of what NASA did to control ADEPT and HIAD is applicable to this very low wing loading design, or maybe not. More testing is required to know for sure.

GW Johnson · 2024-06-19 16:18:49

As I indicated in post 20 above, there is a size scaling effect on the wing loading achievable. This is a square-cube law thing. It is inherent.

Mass (weight) scales as size cubed. Area scales as size squared. Double the size, and weight increases by a factor of 8 (all else equal, although it is usually not). Wing area only increases by a factor of 4.

So wing loading increases by roughly a factor of 8/4 = 2, for a factor-2 increase in size. Wing loading is crudely proportional to absolute size, at the same velocity and altitude.

Does that shed any light on what might be possible?

There is also the design speed range, which is an even stronger effect. W/S = CL q from the lift equation. Dynamic pressure q varies as velocity squared, and as proportional to atmospheric density. CL is pretty much fixed by the range of allowable angles of attack, typically only around 10 degrees wide. The faster you want to fly at a given altitude, the larger is q, and the higher W/S (wing loading) is going to have to be. Period, end of issue.

There is also the structural strength to resist the loads of pulling high gees. That is another square-cube law effect. The gees are a velocity-squared thing at any given altitude. The material stress capability is not! Which is exactly why you do NOT try to fly high angles of attack, when you are flying at high velocities, no matter what altitude you are flying! And yet surviving entry depends inherently upon flying at high angle of attack!

All these effects acting together are what determines your entry design. Focusing upon only one is a serious mistake!

GW

Last edited by GW Johnson (2024-06-19 16:34:39)

New Mars Forums

Announcement

#1 2024-06-12 10:33:03

Are radically reduced costs to space imminent with hypersonic flight?

#2 2024-06-12 12:32:48

Re: Are radically reduced costs to space imminent with hypersonic flight?

#3 2024-06-12 13:03:32

Re: Are radically reduced costs to space imminent with hypersonic flight?

#4 2024-06-13 07:34:41

Re: Are radically reduced costs to space imminent with hypersonic flight?

#5 2024-06-13 11:37:40

Re: Are radically reduced costs to space imminent with hypersonic flight?

#6 2024-06-13 23:52:00

Re: Are radically reduced costs to space imminent with hypersonic flight?

#7 2024-06-14 06:26:29

Re: Are radically reduced costs to space imminent with hypersonic flight?

#8 2024-06-14 06:43:09

Re: Are radically reduced costs to space imminent with hypersonic flight?

#9 2024-06-14 09:06:06

Re: Are radically reduced costs to space imminent with hypersonic flight?

#10 2024-06-14 09:44:32

Re: Are radically reduced costs to space imminent with hypersonic flight?

#11 2024-06-14 11:33:15

Re: Are radically reduced costs to space imminent with hypersonic flight?

#12 2024-06-14 15:30:44

Re: Are radically reduced costs to space imminent with hypersonic flight?

#13 2024-06-14 23:01:32

Re: Are radically reduced costs to space imminent with hypersonic flight?

#14 2024-06-15 09:09:25

Re: Are radically reduced costs to space imminent with hypersonic flight?

#15 2024-06-15 09:45:08

Re: Are radically reduced costs to space imminent with hypersonic flight?

#16 2024-06-15 12:12:22

Re: Are radically reduced costs to space imminent with hypersonic flight?

#17 2024-06-15 14:46:59

Re: Are radically reduced costs to space imminent with hypersonic flight?

#18 2024-06-16 20:09:47

Re: Are radically reduced costs to space imminent with hypersonic flight?

#19 2024-06-16 23:05:40

Re: Are radically reduced costs to space imminent with hypersonic flight?

#20 2024-06-17 08:59:17

Re: Are radically reduced costs to space imminent with hypersonic flight?

#21 2024-06-17 10:46:06

Re: Are radically reduced costs to space imminent with hypersonic flight?

#22 2024-06-17 15:12:52

Re: Are radically reduced costs to space imminent with hypersonic flight?

#23 2024-06-17 18:05:11

Re: Are radically reduced costs to space imminent with hypersonic flight?

#24 2024-06-18 08:29:44

Re: Are radically reduced costs to space imminent with hypersonic flight?

#25 2024-06-19 16:18:49

Re: Are radically reduced costs to space imminent with hypersonic flight?

Board footer