guest blogger john hare
Air breathing engines have many problems competing with rockets on acceleration missions. High on the list is the inlet that slows the incoming air and uses that kinetic energy to compress the air before sending it on to the engine. Air breathing engines really really don’t like supersonic, or even very high subsonic air at the engine face. The inlet efficiency is crucial to any high speed air breathing vehicle, and it weighs as much as the engine in many cases. Listed efficiencies for tested units seem to be from 90+% subsonic to <40% at higher mach numbers that are still well below scamjet range. Inlet designers seem to useÂ the term pressure recovery to describe efficiency.
The inlet efficiency issue seems to be not understood to many ABE fans. I have seen suggestions that the boundary layer just be used in the engine to avoid the whole issue. At mach 2, a given inlet might compress air by a factor of four. If the engine behind this inlet is eating 1,000 cubic feet of air per second, then sucking in the boundary layer air would give the engine one fourth of the mass per second as using the inlet. At sea level, that is roughly 70 lbs/sec vs 280 lbs/sec for the same engine. A second issue is that increasing chamber pressure by a factor of four does very good things to the Isp, like doubling it or so. There can easily be a factor of eightÂ thrust difference with a good inlet vs no inlet at mach 2.
Â As the mach numbers climb, inlet efficiency becomes more critical yet. When potential compression ratio is 100, the difference between a 60% and 30% efficient inlet become the difference between acceleration and deceleration in a vehicle that typically has little excess thrust to begin with. This is where ramjets can fail. They need an inlet with compression ramps that adjust to the mach number in the best possible manner without busting the mass budget. Though turborockets are somewhat less sensitive, they are still dependant on a clean, consistent, well compressed airflow if they are going to perform their mission.
Inlet compression at supersonic speeds consists of having surfaces that create controlled shockwavesÂ that compress theÂ incoming air. Every mach number, and fraction of one, have a best angle for the compression shock, and differing numbers of shocks increasing with mach number. At mach 1.3, a single normal shock is sufficient to slow the airÂ Â to something usable while at mach 2, you need at least one oblique shock before a normal shock leading into the diffuser that settles the air before the engine. There is a limit to how much each shock slows and compresses the airflow. As the mach numbers climb, you want more oblique shocks of smaller angle to the airflow to recover as much pressure as possible. By mach 5, you want a dozen or more if you can get them, each at a very few degrees to the airflow that was turned by the shock before it. Isentropic ramps are a smooth curve that theoretically gives an infinite number of very small shocks for the best possible performance. The problem is that the best angle of each mini shock changes with mach number which prevents just building a simple solid inlet and using the inside of it for something useful like a fuel tank. People in the business and some references refer to the difference between drag and thrust at high machÂ as a very small difference in some very large numbers.
The normal method of dealing with varying mach numbers is to have ramps that shift angles scheduled with the mach number to give as much efficiency as possible. Most of these ramps are curved in one direction and flat in the direction normal to the airflow. They are not perfect for every mach number, just the best compromise for the expected use of the vehicle. The structure to keep these moving ramps intact and in place in high mach conditions, and the mechanisms to handle them,Â can get heavy fast if you are after an acceleration mission.
Then there is the boundary layer that builds up in the adverse pressure gradient of the inlet ramps. If it builds up too much in the wrong manner, it can trip the whole shock system and virtually eliminate pressure recovery. Either the layer must be bled off and dumped or reenergized toward the pressures of the rest of the stream. Elaborate systems are employed to bleed the layers to avoid turbulence, but not so much that good compressed air is being dumped. Figuring out the best methods of handling this boundary layer problem is a job for a full time team.
Then there is the temperature problem. As air is compressed, it heats up. By mach 4 or so, turbomachinery becomes a liability just on the heat issues alone. Turbine inlet temperature become excessive on turbojets even with low compression ratios about here, while the turborocket can get a couple of more mach numbers because the turbine doesn’t see the compressed air. Precooling is the answer most suggested for this mach range, and is a whole nother engineering field again. Physical precoolers can get fussy about excessive moisture in the air freezing and foreign object damage is like a rock through your car radiator. Not to mention they get heavy fast. Mass injection is another solution with possibilities. Liquid oxygen or waterÂ is injected into the stream in front of the compressor to cool the incoming airflow to something useful to the engine. At some point though, the inlet air is so hot that cooling mass flow resembles rocket mass flow for the same mach number, and the rocket isn’t dealing with all that heat and other problems of the ABE.
While I don’t explain everything well, and probably got a thing or three wrong so far, I think it is obvious why rockets are the preferred acceleration engine. All the problems above are with engine types that quit somewhere below mach 6 when a rocket is just getting started. Anybody advocating air breathing engines needs to be honest about the issues they have for spaceflight. Many are not, handwaving problems they don’t understand, that the rocketmen they talk to do know about. When ignorant of the field they are advocating, they destroy their own credibility, and damage that of sincere workers. At the moment though, we can run these things past somebody that is honest, competent and open on the subject, John Bossard at http://plasmawind.typepad.com
This is one more of my tech thoughts, this time on solving some of the issues above in a manner that works for spaceflight.
Â A turbine spins to drive a compressor with a combustor between them to heat and energize the mass flow in a turbojet. An inverse cycle uses the incoming airflow to drive a turbine that powers a compressor before fuel is added. As far as I can tell, the cycle has never been demonstrated to anybodies satisfaction, they have to put combustors in there somewhere.
I propose designing a turbine wheel to operate as the inlet with the rpms increasing with flight speed to vary the angle that the incoming air sees.The supersonic air hitting the turbine wheel is turned though an isentropic curve with the initial angle determined by the relative mach numbers of the turbine and airflow. This eliminates moving ramps to schedule inlet angles. The centrifugal action of the blade on the air re energizes the boundary layer to similar values as the rest of the incoming stream. This eliminates the complex bleed systems. To keep the turbine from over speeding, it is connected to a compressor stage with a stator group between them. This both controls the turbine rpm and improves the pressure recovery of the total inlet.
This idea would take a serious research effort to reach hardware if it is feasible in the first place. If it is feasible, it will make for a much lighter inlet, with good efficiency across several mach numbers, and a much shorter operational length than the current systems. It could make the difference between feasible and not for a marginally capable engine system.
If that system will solve the shock angle and bleed air problems, it still does nothing about the severe thermal environment at high mach numbers. I suggest regenerative cooling of the rotating inletÂ by injectingÂ large amounts of very cold water. Â The water at less than 40 degrees f flows through the inlet turbine, stator, and compressor blades and out into the inlet airflow. Inject so much cold water that the temperature of the airflow drops under 100 degrees f even behind the inlet compressor. Then recover the water in a vortex separator and send it to a cooler before reusing the same water.
In the lower atmosphere at the lower speeds, the cooling water can even increase in mass by collecting moisture from the air and drying it out with the cold air. There are no heavy, complex, andÂ finicky physical precoolers, and the mass injection losses are quite moderate until Â late in the flight when expending excess mass is acceptable. Relatively small amounts of hydrogen will cool large amounts of water with splorge cooling before being used to drive the turborocket. Late in the flight when lower turborocket rpms are acceptable, no LOX is needed for the turborocket with the gaseous hydrogen driving the turbine after cooling the water.
If something like this allows a turborocket first stage to accelerate an upper stage to mach 7 and 120,000+ feet, then even a dense propellant upper will have a mass ratio to orbit of well under 6. Economics of the market could decide on the viability of a system like this, I think favorably.