Nuclear jet engine

RobertDyck · 2014-12-22 22:20:45

I have raised the idea of a nuclear jet engine many times. Let's put it in a separate thread. GW Johnson is an aero engineer with experience with RAM jets. Very useful! However, I am going to propose something new. This is the same thing I've raised before, but let's separate the discussion into its own thread. And lets try to be real, but constructive. Yes, it does mean doing something new.

Let's start with the ideal before we compromise. Ideal is a single engine for take-off from wheel stop on the runway, to flying at high altitude a what NASA now calls hyperspeed. Supersonic is anything faster than the speed of sound. Trans-sonic means crossing the speed of sound. Hypersonic is faster than mach 5. The new term (a couple years ago) is hyperspeed. That is faster than mach 10.

This will require a multi-mode engine. As has been pointed out, a turbine can't go hypersonic. But a RAM jet requires mach 1+ just to ignite. So now we're talking about a changing engine channels for different operation at different speed. The J58 engine was used by the SR-71 Blackbird. It could take off from a runway, and fly up to a maximum speed of mach 3.6. GW Johnson reported that at mach 3.8 the compressor would start to melt. The official line from the US military is that it can fly mach 3+, but refuse to say exactly how fast. But SR-71 did fly once over a test range in the UK, and tabloid reporters were there with binoculars and a stop-watch. They timed it, and published it's top speed. This is an aircraft that first flew in 1966, but one lesson is don't fly your super secret plane over a test range you don't control. But here is a diagram of that engine.

That engine is a turbine inside a RAM jet. Another individual told me the Iroquois engine for the Canadian Avro Arrow was the reverse: turbofan with bypass air down the core. This cooled the engine when operating at high speed. It was late 1950s design, and the first to use inconel.

As GW Johnson points out, none of these designs bypass the compressor. For truly high speed, you need a RAM jet. That requires fully bypassing the compressor. How do you do that?

A nuclear engine does not burn anything. The heat of a nuclear reactor heats air. Thermal expansion creates thrust. First designs to do this were the NB-36. Engines for the test version did not power the aircraft. They tried to put the nuclear reactor in the aircraft body, with ducts from engine compressor to a heat exchanger in the reactor, then back to the engine and out through a turbine. The plumbing was best described as Rube Goldberg. An efficient design will place the reactor right in the engine, like Project Pluto.

One idea I proposed was to use the same glaze as the black HRSI tiles on the Space Shuttle. It was designed to reject heat in hypersonic air. Instead of a "combustion chamber" with flame cups, this engine would have glazed fins inside a "heat chamber".

Now the difficult part. How do we get this thing to work with supersonic air flow. That's equivalent to SCRAM jet, although technically it won't have "Supersonic Combustion" since no combustion. And how do we get operation with aircraft speed into the mach teens?

Another trick is "plasma magic". This was discovered a few years ago. It reduces friction on the airframe, allowing higher speed. A microwave laser (MASER) shines forward, converting the air the craft is about to fly through into weak plasma. Then the aircraft flies through that plasma. Independent analysis has shown it does work, but limited. It's only useful for a poorly streamlined aircraft. A well streamlined aircraft gains no advantage at all. Aero engineers have ignored it due to this, but it could be useful for a lifting body. Making this…

fly more like this…

Would plasma intake affect engine performance?

Of course, assume the nuclear reactor uses Americium-242m, not uranium. It needs low mass. And assume fuel elements are sealed, no fission fragments in exhaust.

Last edited by RobertDyck (2014-12-23 04:12:07)

Impaler · 2014-12-23 03:44:24

This is a good split off thread, I prefer these Nuclear engine concepts get their own threads. I'm assuming this is a Earth-surface to LEO vehicle intended to do the conventional space-pane job of shuttling modest mass cargo on a rapid commercial-jet like duty cycle.

I'm dubious of the such a complex engines and complex concepts generally, but assuming that Skylon concepts ultimately fly some day (a big if), then I think the natural solution is to essentially integrate nuclear rockets into them and do what they do and exploit heat-exchangers for a while and then transition to pure rocket at the point when air-injection becomes too thermally stressful on the engine and the air becomes too thin.

As I said in the other thread you have to transition to rocket eventually so you have to have a fuel line into the nuclear core to deliver liquid hydrogen for that great 900+ ISP, and if you have such a line then by all means COMBUST that hydrogen with the air-steam it will be a better ISP then running it through the nuclear core because your getting FREE molecular mass from the air stream and FREE energy release. Combustion heat is entirely additive to what ever heat the reactants already had so if the engine can withstand the temperatures theirs no reason you can't put nuclear heat into your hydrogen and air and then let them combust together.

On the other hand taking in only air and simply blowing it out the back is technically INFINITE ISP as you expended no propellent at all, but it's less total thrust because your using high molecular mass and only the nuclear heat source. Maybe you want to start the engine on hydrogen in a conventional jet manor so you have take off thrust, then take away the hydrogen and just heat air in a nuclear ram jet, then gradually add the hydrogen back in to make up for diminishing air density and as a coolant. The more transitioning you can do by changing the flow rate of the hydrogen rather then physically changing the engine configuration the better.

Maybe you have two separate nuclear cores, one for hydrogen one for air, presumably this can be done by just having every other channel through the core be for one or the other so a flow of either one is effective at taking up the heat output but they are not mixing while in the core, essentially it's a nuclear injection manifold on a conventional rocket engine and it injects into a proper combustion chamber. Material physics will almost certainly be the limiting factor in all these concepts.

Maybe right from take off run the H2 into the core and have hot hydrogen coming out the back of the core and THEN being burned by the ingested air stream as a kind of afterburner. That will probably be incredibly hot and require that the cold liquid hydrogen be used to cool the afterburner (maybe that hydrogen then can't go through the core and has to be feed into the after-burning combustion itself). The ISP is probably incredibly high to make up for the head-ache though. Think of it as a NTR engine in the place of the fuel injector in a conventional jet engine, and kind of the reverse of Skylon in which the hydrogen cools the incoming air so it can go into a chemical combustion chamber, instead just let the combustible rocket exhaust burn like a normal jet engine would burn it's fuel in a tube.

As your air flow gets thinner and hotter you use your hydrogen to cool the incoming air-stream and turbine more and the after-burner less, then finally when it's not worth it any more the forward spike shuts off entirely to spare the turbine system and the after-burner stops doing any actual burning unless your willing to bring oxygen along, but that would lower ISP so don't bother. The vehicles carries ONLY LH2 as fuel and all of the O2 take off weight in a Skylon line vehicle can just be more payload mass assuming your nuclear core doesn't eat that all up, in addition the higher ISP after rocket transition means you probably save a lot of hydrogen propellent as well on the way to orbit and thus more payload mass fraction.

Having a rocket take off might allow you to skip the turbine entirely which would be a great simplification. The propellent used by the vehicle to accelerate down the run way and take off is almost free from an engineering standpoint, it can be mass above the take off weight because you haven't taken off yet!

Last edited by Impaler (2014-12-23 03:57:32)

RobertDyck · 2014-12-23 14:07:37

I have found many people keep changing the design, then complain that changed design doesn't work. Why not sticking with the requirement? That is: air breathing nuclear jet engine. No rocket mode at all. No propellant.

Designing for too many different modes creates a complex engine that ends up massing more. Adding rocket mode is just too complicated. Assume a separate nuclear thermal rocket for the final push to orbit. Separate reactor, separate heat exchanger.

With greater propellant mass, less propellant acceleration, but you push against greater mass so more thrust. They cancel. To put it another way, higher molecular weight means heat produces less gas expansion, so less pressure. But it also means less flow rate, so more thrust from the same propellant. Either way, they cancel. So higher molecular weight does not mean lower thrust. However, specific impulse is directly proportional to exhaust velocity. Chemical rockets favour low molecular weight because higher Isp. Ion engines favour high molecular weight because propellant remains under influence of the static field longer, resulting in more thrust before it exits the chamber. Notice NASA ion engines use Xenon gas, the heaviest noble gas that isn't radioactive.

GW Johnson has complained that an engine can only operate at one speed, or one very narrow speed range. By that argument, the J58 engine could never work, the SR-71 Blackbird would require many separate engines. Yet, the engineers did get it to work. Let's design one that works.

And yes, air at very high altitude is very thin. And flying at very high speed produces a lot of compression. That's why these two go together. Now speed at lot altitude, high speed at high altitude. That means intake compression where air is thin.

And let's not repeat dumb mistakes. NASA had a design study at one time to develop a vehicle using a Rocket Based Combined Cycle (RBCC) engine that would start in RAM jet mode, transition to SCRAM, then transition to pure rocket. The dumb part was a catapult to throw it from stop to mach 1. The RAM jet would ignite at mach 1, so the catapult had to throw it to that speed for ignition. That meant trans-sonic transition close to the ground. That raises all sorts of problems: shock wave ground reflection damaging the vehicle, and ground damage from the shock wave.

Yes, SCRAM jets that burn hydrocarbon fuel, or even hydrogen, have a heat problem. Intake compression heats air, but additional heat from burning fuel does not add enough heat for significant expansion. No gas expansion, no thrust. But a nuclear engine can avoid that. Just use brute force: increase air temperature to an extreme. Difference in gas temperature causes expansion, and thrust. So going nuclear solves that problem. The catch is whether you can do it without melting a solid core nuclear reactor.

GW Johnson · 2014-12-23 15:36:48

OK, just to understand what you are dealing with at "hyperspeed" as I see it termed above: at Mach 10 in the stratosphere, the inlet total temperature computed by ideal-gas methods is some 6131 deg R, same as 3406 K. At temperatures like that, it isn't even air anymore, it is becoming plasma. If you slow this down to subsonic speed inside your airframe, the static (or thermodynamic) temperature of the "air" is actually very close to that value. Plasma recombines in an expansion nozzle, but this adds no thrust. There’s no point to running plasma through such a nozzle.

My question is then: why would you ever want to incur that? Why not just rocket out into vacuum where there is no temperature, and no heating. Leave the air at lower speeds before it ever gets that bad.

That being said, I spoke the other day with retired colonel Buzz Carpenter, a former SR-71 pilot. He told me they never flew that plane as fast as 3.6 Mach. Their operational target was 3.2 Mach. Sometimes a surge would send them faster, but they would quickly pull up to reduce speed by energy management, and never exceeded about Mach 3.3. And, service ceiling was near 85,000 feet.

The diagrams above are not quite accurate for the J-79. The bleed directly to the afterburner duct was taken by four discrete tubes from stage 4 of the compressor, and never exceeded about 25% of the inlet ingested airflow. It was in fact NOT "a turbine within a ramjet", although a real air turbo-rocket would be. It was a turbine equipped with an afterburner, and with an added partial air bypass to that afterburner.

The J-79 with its blow-in doors had some features resembling an air ejector pump sitting static or at low subsonic speeds. But only partly. At modest transonic and supersonic speeds, it was simply a low-bypass gas turbine. To go faster, they had to bypass air to the afterburner in order to make the inlet and engine massflows match up. And, they had to run afterburner just to go that fast. So, they bypassed the excess inlet air directly to the afterburner, but only after some compression.

The only other Mach 3 engines were the low (essentially zero) bypass engines of the XB-70 and the Mig-25. Those were also afterburning, but did not have the air bypass tubes of the J-79's. About Mach 3.0 was "tops" with the B-70. I know a lot less about the Mig. But it didn't exceed Mach 3 by very much, and its engine life was very short by US standards, or so I am told by those who should know. With external stores under the wings, the Mig was placarded at M2.8 max. That I do know.

I never ever said "an engine operates at one speed". But they do have speed ranges that you have to stay within. Gas turbine is about 0-3 Mach, give or take. Ramjet with external compression features is about Mach 1.8 to 6, demonstrated by actual missile designs self-accelerating in flight tests. Scramjet starts at Mach 4, and has been demonstrated to M5 with hydrocarbons, and not quite 10 with hydrogen. But by demonstrated, I mean a burn, not actual vehicle acceleration capability. And the test flight success with scramjet is only roughly 50% of attempts made. It is not ready for application, not by a long shot.

Most aircraft are extremely volume-limited in their fundamental designs. The specific gravity of kerosene is about 0.8. Its heating value is about 18,000 BTU/lb (about 42 MJ/kg). Hydrogen's heating value is not quite 3 times higher at 51,000 BTU/lb, but its liquid density is about 11 times lower at specific gravity 0.071. What that means is that you can store about 3.7 times the fuel energy in the same volume-limited space with kerosene that you can with hydrogen. A few % difference in energy conversion efficiencies does not materially affect that picture.

So, why would I ever consider hydrogen for fuel in an airbreathing space plane? Or even a rocket-powered spaceplane?

Just thought I'd give some pertinent data and ask some pertinent questions.

GW

Impaler · 2014-12-23 15:52:31

Having a separate nuclear core as a rocket and a separate one for air-breathing might work if the mass fractions are do able, mass fraction has always been the killer of multiple engine vehicles. It's been shown quite conclusively that a conventional jet engine can't even pay for it's own mass and that of the wings when attached to a conventional rocket, the air-breathing engine would need to do a LOT of the total Delta-V to make it worthwhile to have, I'm guessing it needs to take you into double digit Mach to be worth it, that would still leave more then half of total Delta-V on the rocket engine mind you. Rockets are lighter then air-breathing engine so the configuration of two wing mounted air-breathers and one rocket in the tail it perhaps not too much heavier then the configuration of two wing mounted hybrids, we will move the wing forward to balance the weight.

Still I think their are significant advantages to hybridizing the system, you have access to the cooling of the hydrogen which can cool first the incoming air stream letting you run that air through the air-breathing core for much longer, and then later it can cool the decay heat of the core. My understanding is that any Nuclear core will continue to produce substantial heat even after it's control rods are inserted and the chain reaction has come to a stop because daughter isotopes produced just moments earlier are spontaneously going through several short lived isotope decay chains and release more heat, the decay heat output is on the order of ~10% of full power immediately after killing the chain-reaction and then decays logarithmically essentially forever. In a conventional NTR some hydrogen has to be kept in reserve after the main burn just to flow through the engine and prevent it from melting down. The Pluto concept just dunked the vehicle into the deep ocean to let it melt-down underneath a few miles of water, so taking that concept and trying to make it reusable requires some coolant flow.

Remember that as speed increases the engines that are optimum for that speed actually get simpler, rockets are simpler then SCAM-jets, which are simpler then RAM-jets which are simpler then Jet turbines. The difficulty is all in the 'transformer' aspects of the engine that can assume all the configurations, eliminating the most complex configuration the Jet makes the problem easier which is why NASA's hybrid concepts tend to something that can't produce thrust at stand-still and they use a rocket to get up to speed, the beauty of a rocket is it works at ANY speed. The less moving parts the simpler it all gets, that SR-71 engine goes through 5 total flow configurations and is maxed out by Mach 3.8, I don't think we can make something go through more then 3-4 configurations and still be practical. If rocket is our final configuration then that leaves 3 earlier configurations max, and the first configuration probably needs to be rocket or rocket-hybridized in some way to give maximum take off thrust. The ideal solution lets you transition the engine by just opening or closing one air-inlet and throttling the flow of fuel and directing it from one section of the engine to another, but that's probably not achievable, a few ports and air-flow control surfaces like spikes and bypass louvers will be needed in any practical design.

The configuration I'm imagining is essentially Nuclear-Skylon, from the front to back you have an air pre-cooler which feeds into the nuclear core manifold with the air going through half the tubes in the core, the core manifold then dumps directly into the combustion chamber of a rocket engine, the combustion chamber feeds into a bell nozzle that has louvers to admit air that has bypassed everything prior to this, the bell then continues as a tubular after-burner in which hydrogen is injected. The Hydrogen flows through one valve that directs it either to the pre-cooler or to the core manifold, the hydrogen flow from the pre-cooler flows directly to the bell-nozzle-afterburner to cool it and is then burned their.

The launch configuration is with ingested air going through the pre-cooler but with little or no no cooling initially being used because the uncompressed air doesn't need it. Hydrogen and uncooled air instead flows through the core and and burn in the rocket combustion chamber, the air probably needs something to force it in to the core, maybe a pump. Then as the vehicle climbs it reduces hydrogen flow to a minimum to conserve propellent, it gets as close as practical to just being an air-thermal engine. As thinning air starts to reduce thrust the hydrogen flow is re-introduced by into the pre-cooler this time with all combustion occurring in the after-burner, the core is still just air-thermal but is running on pre-cooled air so the air can still cool the core. Finally the flow of hydrogen is shunted back to the core as the last dregs of air are left and the core goes to hydrogen cooling and were back to plain NTR with maybe a bit of hydrogen going to cool the bell nozzle-afterburner and be expelled into it too just mix into the flow. During the final phase the core is perhaps being shut down gradually and our decay heat can be used for doing some final orbit trim or raising/circularizing the orbit a bit. The total configuration set is just four which seems manageable.

Now I'm sure GW will have a better understanding of if this is even remotely practical and may have a better configuration. Remember I'm assuming that Skylon engines end up working, reaching orbit and being practical and that in a search for MORE performance out of them we add nuclear power and have solved all the radiation problems both preventing radioactive exhaust and shielding the passengers of the vehicle from the engines themselves.

RobertDyck · 2014-12-23 17:43:22

Ok. No hydrogen. This engine is air breathing. And at high speed, air velocity through the engine will also be supersonic. Strictly speaking not "SCRAM" because there won't be any combustion, but same principle.

Last edited by RobertDyck (2014-12-24 02:48:40)

GW Johnson · 2014-12-24 09:35:04

There's nothing about the laws of physics that says you can't build a nuclear-heated turbojet, ramjet, or scramjet. But, as they found out during Project Rover, it's not very easy to do. The device that flew in the NB-36 was never a propulsive device, just an air-cooled heat source. For that, they developed a containment case that could safely contain a reactor core in a 500 mph crash into solid rock. It was successful, but quite heavy. That's about as far as that project got toward its goal of a nuclear turbojet.

Project Pluto was the nuclear-heated ramjet. Ramjet, BTW, is far simpler than scramjet, especially in terms of what you are required to deal with technologically. Pluto was a fixed-geometry air passage with a reactor inside to heat the air. Except for the reactor, that's actually very lightweight. It had a supersonic-type belly inlet with external ramp compression before capture, and an S-duct to the feed plenum going into the reactor. Post-reactor, a short plenum fed the C-D nozzle (which in a ramjet the throat is near 65% of plenum cross section, and the expansion ratio is near only 1.3, quite vastly different from typical rockets).

Pluto's biggest problem was solid core operating temperature: they were about 10 F (5C) away from melting-out the core supports every time they ran the thing on the ground in Nevada. But the air approach speeds were very subsonic and so the pressure drop going through the core passages was low. This kept air loads on the core low while allowing the core to be "dense" geometrically to achieve critical mass.

Nobody has ever tried nuclear-heated scramjet. The duct geometry (especially the inlet) would be incompatible with ramjet. So also would the exit nozzle geometry be completely incompatible with ramjet, much less any imaginable rocket geometry. There really isn't much potential for any combined cycle designs with rocket here.

Now, the air speeds approaching the reactor are supersonic, so that the air loads going through the reactor become quite enormous, leading to a really serious core retention problem. I'm not saying it cannot be done, but it's a more difficult thermo-structural problem than anything I have ever heard of, including gas-core NTR. Such a design would also require much more open passages through the reactor, leading to real troubles establishing a geometry capable of reaching critical mass.

Again, I'm not saying it cannot be done; I am saying it is supremely difficult. Might be a problem not worth solving. I don't know.

GW

RobertDyck · 2014-12-24 12:00:10

For Earth, one option is a traditional turbojet for take-off and landing. That would use normal jet fuel. But once supersonic, ignite a nuclear RAM jet. Then accelerate to nuclear heated SCRAM jet. So this nuclear engine would go from just over mach 1, to mach teens. This would require 3 separate engines on the vehicle, which is already getting complicated. Designing an aircraft with inlets for two separate jet engines is already difficult. But for Earth application, it would solve the problem of operation from a runway at a commercial airport: nuclear engine not operating.

The third engine would be NTR using distilled water as propellant. I once calculated Isp, forget how I did that now. Starting with Am-242m for low mass nuclear reactor, operating at the temperature of the 1991 re-design of NERVA so it doesn't melt itself, but larger molecular weight of water and stored at room temperature instead of cryogenic propellant. The result was just barely higher Isp than SSME. But with dramatically more dense propellant, the fuel tank was small enough to be contained internally in an SSTO shuttle. The catch to keep propellant tank small was to achieve most of the speed for orbit through the nuclear jet engine. That means mach teens before igniting the NTR. Starting the NTR at mach 8 just isn't good enough.

So the hard part of this is the nuclear jet engine. Going from RAM to SCRAM. Yea, it's hard. Big speed range: say mach 1.1 to mach 17, just to pick figures out of thin air. The military SCRAM jet guys claimed at one point they could achieve mach 17. If we don't get exactly that number, then mach teens. And I picked 1.1 because it should be the lowest speed to start a RAM jet. Again, picked out of thin air. But we want to maximize the nuclear jet because it doesn't use propellant. Minimize engines that require propellant.

Doing this for Venus becomes much more difficult. Venus requires VTOL in 92 bar atmosphere at +462°C mean surface temperature, with 96.55% CO2, 3.5% N2, and significant corrosive carbonyl sulphide (COS). But let's forget Venus for now. Focus on an engine for an SSTO shuttle for Earth. As you said, nuclear-heated SCRAM is difficult enough.

Impaler · 2014-12-24 21:47:46

If your going to a dense propellent that you have no intent to burn for your NTR, wouldn't strait Methane be better? Methane has more hydrogen in it per unit weight then Water, so it should logically have higher ISP when it is not mixed with Oxidizer. The heat of the nuclear core disassociates the carbon and hydrogen so your blowing out hydrogen and carbon separately and getting the benefit of hydrogen's atomic weight, hopefully this dose not create terrible carbon coak in the core which would probably be a fatal flaw.

If Methane doesn't work then their may be other simple hydrogen rich compounds that might work and would give higher ISP then water alone while still being dense an non-cryogenic, maybe Ammonia? If you can rediscover your ISP calculations on these guys try them out if you have not already and tell us what you get.

Also your 3 engine concept convention turbine->nuclear-ram/scram->NTR seems like it would significantly benefit from a carrier aircraft first stage. That moves the turbine engine off the orbital vehicle AND probably lets you cut down hugely on wing mass too cause the 2nd stage vehicle can have wings specialized just for the high speed environment AND the wings are not holding two engine sets. Carrier aircraft has always been very marginal when your putting a rocket on the back because rockets accelerate fine from the ground and the plane is just giving a small performance boost, but once your 2nd stage is ram or scram you really NEED something to give mach 1+ initial acceleration and you don't want to carry that thing with you to orbit cause conventional turbines jet engines are HEAVY.

Last edited by Impaler (2014-12-24 22:09:13)

GW Johnson · 2014-12-27 13:19:09

There's actually two design ranges for ramjet. Both are very simple devices, and usually fixed geometry, and extremely lightweight.

One is something I call "low speed range design", which feature a simple pitot/normal shock inlet, and a convergent-only nozzle that often runs unchoked. Nozzle area is pretty near 65% of combustor flow area. You have to size the pitot scoop correctly. These can light at subsonic speeds, start producing useful thrust (at greater than solid rocket Isp) around Mach 0.5 to 0.7, and peak out at about Mach 2.5 to 3, depending upon the drag of the airframe they are installed in. They are rarely used above flight speeds of Mach 2, and peak Isp occurs somewhere near Mach 1.1-or-so. That peak Isp with kerosene or gasoline fuels is near 900 sec if well designed.

The other is probably more like what folks would want to use for ideas typical of these forums: what I call "high-speed range design". These have external compression features like spikes or ramps just ahead of the capture cowling, features that generate shock waves that need to fall correctly upon other adjacent structures. That's part of proper design. Inlet capture area has to be properly matched to ramjet throat, for the specific characteristics of the inlet components.

The nozzle is always choked, has a throat near 65% of the combustor area, and a mild supersonic expansion ratio near about 1.3. Peak Isp is usually near 2.5 at near 1300 sec with hydrocarbon fuels running full rich, and max thrust about Mach 3-ish. Takeover speeds are usually between Mach 1.8 and 2.5, usually quite near 2.

There are no attached-shock solutions for the inlet compression features at speeds like Mach 1.4 to 1.5, which means there is no air ingestion there, and no thrust. Thrust decreases gradually with Mach as you speed up from the max thrust point, although airframe drag is increasing like a power function. Max speed depends more on airframe drag than anything else, and usually falls between Mach 4 and Mach 6, where the thrust and drag curves vs Mach cross.

You can replace chemical heating with nuclear heating by some means, which will remove the 65% throat area limit (a combustion stability item). Flameholding features can also be eliminated, which reduces internal losses. This resizes all the components, of course. The main problem is the same as is faced in rocket engines: once the heated gas ionizes significantly, you cannot convert further heat energy additions into velocity with a C-D nozzle. Recombination does not add velocity to the stream. This is true of chemical or nuclear systems.

GW

GW Johnson · 2014-12-27 13:33:49

Scramjet geometry is quite different from ramjet, in spite of some inlet components whose appearance is externally similar. The shock wave geometry is different at higher speeds and forces the external compression features to fall further ahead of the cowl. So, even the similar-looking part is really quite different.

In a scramet, there is no subsonic diffuser and duct with a terminal shock that moves from one diffuser area to another to match backpressures. Scramjets must have an isolator duct with a weak shock train to provide stability, or they tend to blow up. The kind of thing tested as a short dummy unit on the X-15 decades ago won't work: no isolator duct. This kind of inlet passage functioning and shape is nothing at all like the subsonic combustion ramjet inlet.

There is no convergence to a ramjet throat. In fact the basic combustor is slightly divergent, and the expansion nozzle is more strongly divergent. This kind of combustor and nozzle geometry is nothing at all like that of subsonic combustion ramjet.

The most difficult problem has been fuel injection. Injecting a fuel stream into a supersonic flow always causes a shock wave around the injected stream, and this is very susceptible to a full shockdown to subsonic flow. That event has all kinds of fatal outcomes, if it happens.

As I have said earlier, the min takeover speed for stable scramjet operation has long been known to be right at Mach 4. So far with hydrocarbon fuels (X-51) we have demonstrated stable burn for a couple of minutes at Mach 5, but without demonstrated airbreathing acceleration. With hydrogen (X-43), this is just under Mach 10, and only for a few seconds.

Inside these devices, the convective heating is just awful, even in the inlet ahead of the burn. To be successful as a propulsive device, this massive heat flow has to be recycled back into the combustion release. As near as I can tell, no one has done that yet, not to the significant degree that is required. The tests have been "heat sinkers", by and large.

GW

Last edited by GW Johnson (2014-12-27 13:36:55)

New Mars Forums

Announcement

#1 2014-12-22 22:20:45

Nuclear jet engine

#2 2014-12-23 03:44:24

Re: Nuclear jet engine

#3 2014-12-23 14:07:37

Re: Nuclear jet engine

#4 2014-12-23 15:36:48

Re: Nuclear jet engine

#5 2014-12-23 15:52:31

Re: Nuclear jet engine

#6 2014-12-23 17:43:22

Re: Nuclear jet engine

#7 2014-12-24 09:35:04

Re: Nuclear jet engine

#8 2014-12-24 12:00:10

Re: Nuclear jet engine

#9 2014-12-24 21:47:46

Re: Nuclear jet engine

#10 2014-12-27 13:19:09

Re: Nuclear jet engine

#11 2014-12-27 13:33:49

Re: Nuclear jet engine

Board footer