Go to my Home page.
Go to my Joy of High Tech page.

The Joy of High Tech


Rodford Edmiston

Being the occasionally interesting ramblings of a major-league technophile.

Please note that while I am an engineer (BSCE) and do my research, I am not a professional in this field. Do not take anything here as gospel; check the facts I give. And if you find a mistake, please let me know about it.

Air Breathing Propulsion

     Updated June, 2004.

     A project which hoped to produce a plane with a combination of good low speed acceleration and high top speed resulted in the Republic XF-84H, aka the Thunderscreech. The intent behind this project was to take a jet-propelled airframe known to be capable of supersonic flight (that is, a late-model turbojet-powered F-84) and modify it for use with a turboprop engine. This would give a high-speed plane with propeller acceleration and endurance. Note that there was never actually any intent for it to reach supersonic speeds using its propellers, though that claim appears in many references. (A bit of a digression, here. Other sources claim the F-84 airframe was chosen specifically because it was capable of supersonic flight, and therefore the plane was meant to fly supersonic using the propellers. The truth is that the plane was intended to have supersonic dash capability, using an afterburner. The props would be feathered, the afterburner lit and the plane flown on pure jet power. As it turned out, the program was such a disappointment the afterburner was never lit in flight. Which may be why some sources don't even know about it.)

     The test flights proved that it was a fast (though it never flew over 720 kph, due to handling problems) and quick-accelerating aircraft, among the fastest prop jobs of its day. It also, however, held another record, one hinted at by its nickname. It was arguably the loudest fighter plane ever built. The howl of the 6000 shaft horsepower (and that's without the afterburner) twin-turbine engine combined with the roar of the prop made standard ear protectors irrelevant. Worse, during run-ups the blade tips went faster than sound, reaching Mach 1.18. This produced a rapid series of sonic booms which could pummel ground crew members insensible in short order. Due to prop torque the plane was a major handful to fly at takeoff, and none too pleasant at other times. (The plane was originally to be equipped with twin contrarotating props, but these proved so problematic that this arrangement was changed to a single prop.) Also, as with any experimental plane, it had development problems. One pilot reported that 10 of his 11 flights were cut short by emergencies of varying degrees. Another made one flight and refused to get back in the cockpit. The plane - two were built, one of which survives - deserves an article all to itself, or perhaps extensive mention in an "it seemed like a good idea at the time" article.

     Propellers are airfoils, something the Wright brothers figured out pretty quickly, and which is one reason they were first into the air with controllable powered machines. To produce lift - which comes out as thrust in the airframe - propellers have to move air, and to do that they have to move through the air, swirling around and around, pushing the air they encounter back in a spiral pattern. However, in flight propellers are generally also moving linearly through the air, with the plane.

     Propeller thrust is generated by accelerating air. For generating thrust, accelerating a large mass of air a little bit is much better than accelerating a small mass a lot. (F = ma, and if "m" is larger "a" can be smaller. Since KE = 1/2mv2, a larger "m" means a smaller "v" and since the energy required is proportional linearly to the mass but to the square of the velocity this means that for the same thrust less energy is needed to accelerate a large mass a little than vice versa.) So large propellers are more efficient than small ones, at least at low speeds. However, to produce more thrust (such as is needed for higher speed) with the same propeller (assuming fixed pitch) requires turning it faster. Eventually the tips are moving so fast in their rotation that even with the plane sitting on the ramp they are going faster than the local speed of sound. And transonic flow creates huge amounts of drag.

     To postpone the problem, a propeller can change its pitch, "digging in" when more thrust is needed and flattening out for less, while keeping a steady RPM. These constant speed propellers generally are designed for a particular range of rotation rates, matching their characteristics to those of the engine (it's maximum efficiency RPM range, for instance) and the requirements of the plane. Of course, there are limits to how much the pitch can be increased and the propeller still do a good job. With the blades turned perpendicular to the slipstream ("feathered") a propeller produces no thrust at any RPM. Many propellers - fixed or adjustable - also vary pitch from root to tip.

     Complicating things, a propeller - being an airfoil - has a pitch (equivalent to a wing's angle of attack) at which it is most efficient, producing the most thrust for the least drag. As a propeller moves forward through the air the effective angle of incidence between propeller and air reduces. Changing the pitch helps here, too. However, as airspeed increases a point is reached where changing the pitch can no longer compensate.

     A propeller therefore becomes less and less efficient with increasing airspeed, so that thrust decreases while the drag of the propeller - and the airplane as a whole - increases. Reducing propeller diameter allows a higher forward speed, since a smaller propeller can produce the same thrust with a lower tip speed. However, since smaller propellers are less efficient this means using more power for the same thrust. This produces a situation of declining returns; to go faster you not only need the power to overcome increasing drag and decreasing propeller airfoil efficiency, but to make up for the reduced efficiency of the smaller propeller. There are ways around this, such as sweeping the blade tips (or the whole blade) to delay transonic shock, and adding more blades to make up for the reduced diameter. The Russian Tu-95 Bear bomber/reconnaissance plane uses four 12,000 horsepower turboprop engines with high-revving, contra-rotating, transonic props. (Note that some of the thrust comes from the turbine exhaust, as was also true with the XF-84H.) It is currently the fastest propeller plane in the world, so fast (with a dash speed of around 885 kph) that it actually needs its swept wings. American pilots intercepting these planes report experiencing huge noise near the plane, including multiple rapid small sonic booms similar to those produced by the Thunderscreech.

     However, in spite of such tricks as are described above, the top speed for propellers seems to be around 900 kilometers per hour. Above that, they produce too much drag. Much above that and the propeller acts like a solid disk! The closest any prop-driven plane has come to breaking the sound barrier was a British research plane which lost its propeller in a dive. Radar clocked it at about Mach 0.9 on its way to the ground.

     Jet engines work better at high speeds. The fans in the front are for compressing air and pushing it into a combustion chamber, with the thrust coming from the chemically-heated air being exhausted out the back as a high-velocity jet. (I'm talking about pure turbojets, here. Turbofan engines can be thought of as turboprops with small propellers inside ducts.) However, as the aircraft moves faster, the air entering the duct will eventually exceed the speed of sound. Supersonic air hitting rapidly spinning blades creates some very bad vibes. Even before that, transonic drag at the inlet rises sharply, and the shockwaves produced as subsonic slipstream air goes supersonic at the inlet lip can cause major problems outside as well as inside. So turbine-powered planes designed to exceed Mach 1 have some means of slowing the air to below the speed of sound before it reaches the compressor, and of preventing inlet lip shockwaves from forming or otherwise preventing them from disturbing the airflow at the inlet. This may be as simple as using a clean inlet followed by a long duct. Examples can be seen in several early supersonic jet fighters, such as the F-100 Super Sabre, where the inlet is simply a sharp-edged hole in the nose. However, as speeds increase even more, this becomes insufficient.

     Shock ramps ahead of the inlet can be used to deliberately generate shock waves, controlling shape and location. This controlled shockwave reduces inlet-induced drag and not only slows incoming air but compresses it, aiding the compressor section of the engine. These work well up to about Mach 6, where thermal effects become major considerations at inlets. However, turbojets don't work at such speeds, having operational problems as speed and altitude increase. The practical top speed for turbojets is around Mach 4.

     Note that high-speed turbojet or turbofan aircraft generally need afterburners, since turbojets of reasonable size and weight simply don't have enough thrust on their own to propel a plane much past Mach 2. (The Concord cruises at just over Mach 2 without afterburners, but uses them on takeoff and climb and acceleration to cruise.) With an afterburner, exhaust from the turbine section is sent into a specially-shaped duct, where more fuel is added and ignited, more heat is produced, and more thrust generated. In many ways an afterburner is like a ramjet, described next.

     The ramjet is a very simple heat engine. Air comes in, energy is added in the form of heat, and it expands and goes out the back faster. Ramjets can work up to about Mach 6 or, with inlet cooling to aid in compression and thermodynamic efficiency, around Mach 8. The problem is that plain ramjets don't produce thrust unless air is entering the front at a fair clip. Which means you need some way to get the engine - and the plane it's connected to - up to high speed before it will operate. For most ramjet designs, "high speed" means past Mach 1, and maybe Mach 2, to get enough compression from forward movement through the air for the ramjet to work. And even a ramjet must have subsonic air inside; supersonic airflow would disrupt the combustion pattern in the flame holders, literally blowing out the fire.

     Airbreathing propulsion units which start in one mode of operation and switch to one or more others are know as combined cycle engines. The most practical way to get very high speed in an airbreathing craft is to use a combined cycle of some sort. The turboramjet is one type of combined cycle engine, and is described in the next paragraph. A number of schemes have been proposed for getting ramjets going fast enough to work, and a few of them tested. A couple have actually flown a considerable amount of time. The most used of these is the turboramjet. (Note that the US has fielded at least one missile with a rocket booster and a ramjet sustainer. The Russians have fielded several.)

     In a turboramjet engine a turbojet with afterburner is used to get to minimum ramjet speed and altitude. The incoming air is then partially or completely bypassed around the turbine portion and ducted into the afterburner, so it functions as a true ramjet. This is the type of engine flown on the Blackbird family of aircraft. With the Blackbirds, as speed increases, more and more of the ram-effect compressed air bypasses the turbine section of the engine and goes directly to the afterburner. At Mach 3+ cruise most of the thrust comes from the afterburners acting as ramjets, and most of the air necessary for the combustion which produces this thrust comes from ram effect at the inlet. (Many references erroneously state that at cruise most of the thrust come from the inlets, which is a misunderstanding of a statement made by Kelly Johnson about how Blackbird engines work at cruise. Inlets are actually a source of drag, not thrust.)

     That the turboramjet concept works well is demonstrated by the fact that the Blackbird family of aircraft has officially been the fastest operational manned airbreathing vehicles for nearly 40 years. (If the Aurora program did produce something faster, it either didn't reach operational status, or was not operational for long, or is a drone with no crew on board.) However, even with exotic materials and special fuels, the axial turboramjet is limited to a top speed of around Mach 4, due partly to the drag from the cross-sectional area needed for the ducting. Other variations on the turboramjet have higher potential top speeds, all the way up to the maximum a ramjet can achieve.

     The Blackbird aircraft use a primarily annular - or axial - system. The engines are mounted in the wings, the inlets projecting forward past the leading edge. In the center of the inlet is a movable cone (or spike), which translates back and forth to keep the shockwave in the proper position, as required by the speed, altitude, air temperature and so forth. In turboramjet mode, much of the incoming air is sent around the turbine section through large ducts. Other types of turboramjet use a parallel configuration, which has the turbine completely out of the flow during high speed flight. Ducts and movable dampers channel the air through the turbojet at low speeds, its exhaust going into the AB/RJ. At high speeds the dampers shift the air flow straight through the AB/RJ, so that it operates as a pure ramjet. However, these and similar systems are heavy and complicated, and would probably only be practical in vehicles much larger than the Blackbird.

     Another solution is the airturboramjet, which has a one- or two-stage compressor section driven by a small turbine, followed by the AB/RJ. At low speeds it acts like a very high bypass turbofan with an afterburner. At high speeds the turbine idles, with most or all of the compression provided by ram effect. Lightweight and simple, the two main drawbacks are low fuel efficiency at transonic and low supersonic speeds and some rather extreme materials requirements.

     "Ramrocket" is the general name for a family of related concepts. These are ducted rockets, with a small rocket engine inside (what else) an aerodynamic duct, which has an AB/RJ at the back. The exhaust plume from the rocket engine will force air through the duct, allowing the afterburner to function even under static conditions. That, plus thrust from the rocket, is enough to get the vehicle moving. The faster it moves, the more ram effect forces additional air through the duct, and the more like a ramjet the engine operates. Past about Mach 2 the rocket shuts down, and the engine goes into ramjet mode.

     Ramrockets are even more fuel-inefficient than airturboramjets at low speeds. However, they have the advantage that once the ramjet mode has reached its upper speed limit the rocket can be reignited and the vehicle pushed even faster, perhaps to orbital speeds. Static and low-speed thrust are also quite impressive.

     When working with extreme propulsive modes like these, efficiency of design is important. The overall engine and various components must present maximum performance for minimum penalty in weight and drag. For instance, the struts holding the turbine pod in place in the airturboramjet engine can double as stator vanes, and triple as fuel pipe and control line conduits. The housing of the pod can be shaped to act as a compressive body, something needed in the front part of a ramjet. One clever variation on the ducted rocket is the strutjet, in which the rocket engine is linear, built into the trailing edge of one or more struts which cross the duct. This, however, requires that a separate compressive body be added.

     All of these systems (with the exception of the ramrocket) are limited to the top speed of a ramjet, Mach 6 to Mach 8. While that is quite fast, the Earth is pretty large. To get from one side to the other more quickly requires either a suborbital vehicle, or an airbreathing engine (rockets being too inefficient for long-range atmospheric flight) which can fly faster than Mach 8.

     The scramjet (or supersonic-combustion ramjet) doesn't require incoming air to be slowed below the speed of sound. While compression is still required, the air can stream through at supersonic speeds, which reduces internal drag. By careful design of the internal shape of the duct and the use of very fast-burning fuels - usually hydrogen, though methane and other low molecular weight fuels have also been tried - the flame front is maintained even with a supersonic flow. The theoretical upper limit of a scramjet is somewhere above Mach 25, which is orbital speed! However, this engine must have air moving through it at supersonic speeds before it begins to operate. And in this case, "supersonic" probably means above Mach 4! So the situation is like that of the ramjet, only worse. So far, only a handful of scramjets are known to have actually functioned in flight. The first was an Australian rocket-boosted test in 2002. This came after an earlier launch in 2001 failed. The second was NASA's X-43A test drone scramjet vehicle, the second of which successfully accelerated in about 10 seconds of powered hypersonic flight. It is now considered the fastest air-breathing vehicle.

     Complicated designs splicing turbojets, ramjets and scramjets together in one combined cycle engine, or putting separate scramjets on a vehicle which also has turboramjet or rocket engines, have been suggested. Again, they would require a large vehicle to be practical. More recently, NASA researchers have developed the ramscram engine. By varying where fuel is injected, this can switch between subsonic and supersonic combustion without using heavy, complicated ducts or moving ramps to change the airflow; they simply change where the fire is. Attach some of these to a properly designed aerospacecraft, with the underside ahead of them shaped to act as a compressive section, use afterburning turbojets to reach ramjet speed, and the result is a vehicle with air-breathing engines which can achieve orbital speeds!

     One of the more unusual proposals for single stage to orbit craft is to have a scramjet-propelled vehicle roll upside down as it approaches orbital speed and continue to accelerate, using aerodynamic lift to hold it inside the atmosphere until it has enough velocity for orbit. Then it would roll upright, shut off the engines and coast to LEO. A small rocket engine would be used for circularizing the orbit, orbital maneuvering and the deorbit burn. The math checks. If we can build the engines and airframe, it would work.

     Who knows... the next 20 years may see that long-held dream of actually flying from the ground into space and back come true.

     This work is Copyright 2010 Rodford Edmiston Smith. Anyone wishing to reprint this material must have permission from the author, who can be contacted at: stickmaker@usa.net