Spaceplane

SpaceNut · 2015-11-07 21:34:23

Air is 80% nitrogen, 20% oxygen approximately and rockets use 100% oxygen for starters for the fuel ratio and not 20% so as we go further up in the atmosphere we keep losing on that ratio. So at the top end of say traveling at mach 10 we would need to begin to send in oxygen to over come the ever thinning air ratio to allow for the engine to continue to keep its efficency to propel the rocket into orbit.

GW Johnson · 2015-11-08 09:47:18

Sorry, Spacenut, it really doesn't work that simply (just injecting oxygen to make the fire hotter).

First thing to consider: with an airbreather, when inlet airflow is large, fuel flow is large. When airflow is small, fuel flow has to be small. What you hold constant is fuel/air ratio, not fuel flow.

Next: thrust is m(Ve-V) for an airbreather (of any kind), where m is massflow into the engine, which is density x captured streamtube area times V. V is flight speed, and Ve is exhaust jet speed. Adding heat to the stream is what increases Ve, creating thrust.

It has to be done at chamber pressure greater than ambient, so that nozzle expansion can be had. The increase in pressure over ambient is due to the ram effect in the inlet, and the action of any compressor machinery, everything else is a pressure loss. If there is a compressor, there is a turbine, which subtracts energy from the stream before nozzle expansion, in order to drive the compressor.

Ramjet is typically 3 to 6 times ambient pressure (scramjet is similar), turbine much higher at 10 to 20-something. Those factors persist at high altitudes, where ambient pressure is getting very low.

Adding oxygen to the combustor does two things (1) it upsets the carefully-designed massflow, momentum, and energy balances through the engine, causing it to malfunction, and (2) if you can forestall the the malfunction, it increases Ve by a modest percentage, because the fire is hotter.

That modest percentage increase in Ve in no way offsets the effect of the incredibly low density way up in the thin air. At 150,000 feet the density ratio to sea level standard is .001454. That's crudely a 700-fold direct reduction in thrust that a modest percentage increase in Ve (or Isp if you like to book-keep it that way) cannot hope to overcome.

The malfunction is worse than an unswallowed shock system, it is actual reverse flow. This is fire spitting out the inlet as well as the nozzle, no thrust, and flameout within seconds. Along with considerable engine damage from the reverse flow of flame out the inlet.

If you keep the oxygen additions very modest in an engine actually designed to accommodate them, you might be able to widen the combustor flameout limits somewhat at low air density. No one has ever actually done that to my knowledge, but it is a possibility, I'd hazard the guess.

All airbreathing combustors have a low chamber pressure level, below which the rich and lean mixture flameout limits converge. No operation is possible at all at pressures lower than that. Just an important but unfortunate empirical fact of life.

That's fundamentally why all airbreathers become useless if you try to fly too high with them. It's why scramjet is no holy grail for launch. It's why Skylon's design shifts to rocket at only Mach 5 way down around 100,000 feet. It's why cycle code predictions at such extreme conditions are just plain wrong (ramjet, scramjet, gas turbine, and any sort of combined-cycle engines), unless you know (in exquisite detail !!!) just exactly how to allow for such real-world thermo-physical/physical-chemical effects.

Up in thinner air, you simply must use a rocket, which except for the nozzle extension, is completely independent of ambient conditions. That's the real message here.

GW

Last edited by GW Johnson (2015-11-08 15:40:14)

SpaceNut · 2015-11-08 21:27:14

What we are talking about is a hybrid jet/rocket engine and Why BAE Could Lead the Next Commercial Space Launch Revolution with a SABRE engine.

Among the critical technology breakthroughs Reaction engineers claim is a heat exchanger that can cool incoming air as hot as 1,800 degrees Fahrenheit to subzero temperatures in a fraction of a second, allowing the engine to function at higher speeds than traditional air-breathing jet engines.

So condensing the airs oxygen to feed the rockets engine. About the same as turning on an onboard LOX supply when the scram jet engine doors are closed and turning it off so that the rocket engine can fire.

GW Johnson · 2015-11-09 08:59:19

SABRE isn't an airbreathing engine in the same sense as a ramjet, scramjet or gas turbine. It's a sort of a combined-cycle device, except that it is not really one of those either. It's really a variant on the 1960's idea of the liquid air cycle engine (LACE), which is really a rocket engine with machinery for scooping up and liquifying air added to it. It's fundamentally always a rocket, and not thrust-limited by altitude low density. Only the air scoop/liquifaction rate suffers the low density problem at altitude.

The bugaboo with that concept all these decades was totally-inadequate heat transfer technology for rapid liquifaction. If the Brits have really solved that difficulty (and I do hope they have!!!!), then this will be a significant advancement. I'd really like to see it fly.

I am worried about their airframe shape for HOTOL with its tip-mounted SABRE engines, though. The strong shock waves off the compression spikes are going to cut into the wing leading edges very severely above Mach 4 or 5. This may well be a fatal problem for reentry at Mach 25; so I wonder how hard they have really looked at their airframe concept relative to these problems.

Things like Avcoat and PICA will be inadequate under shock-wave impingement, that's why no reentry vehicle has ever had adjacent connected nacelles. They'll likely need several inches of silica-phenolic in the shock impingement zones, heavy as it is, and they'll likely need to replace it every flight.

GW

Antius · 2015-11-09 13:23:55

GW Johnson wrote:

SABRE isn't an airbreathing engine in the same sense as a ramjet, scramjet or gas turbine. It's a sort of a combined-cycle device, except that it is not really one of those either. It's really a variant on the 1960's idea of the liquid air cycle engine (LACE), which is really a rocket engine with machinery for scooping up and liquifying air added to it. It's fundamentally always a rocket, and not thrust-limited by altitude low density. Only the air scoop/liquifaction rate suffers the low density problem at altitude.
The bugaboo with that concept all these decades was totally-inadequate heat transfer technology for rapid liquifaction. If the Brits have really solved that difficulty (and I do hope they have!!!!), then this will be a significant advancement. I'd really like to see it fly.
I am worried about their airframe shape for HOTOL with its tip-mounted SABRE engines, though. The strong shock waves off the compression spikes are going to cut into the wing leading edges very severely above Mach 4 or 5. This may well be a fatal problem for reentry at Mach 25; so I wonder how hard they have really looked at their airframe concept relative to these problems.
Things like Avcoat and PICA will be inadequate under shock-wave impingement, that's why no reentry vehicle has ever had adjacent connected nacelles. They'll likely need several inches of silica-phenolic in the shock impingement zones, heavy as it is, and they'll likely need to replace it every flight.
GW

I don't see how this could possibly work. The engine has to physically stop an airstream moving at kilometres per second, cool it from +1000C to -200C and overcome latent heat of boiling, only to inject it into a rocket engine where it can burn to produce thrust. Where is the heat sink for cooling the incoming air? Even if the fuel is cryogenic it won't help very much with cooling. The airstream outweighs the fuel 20-1 at stoichiometric ratios. The drag imposed by the air on the intake would rival the thrust produced by the engine long before orbital speed were reached (exhaust velocity would be at least 40% lower in air fed mode as the combustion products weigh 3 times more).

Last edited by Antius (2015-11-09 13:26:45)

GW Johnson · 2015-11-09 14:11:35

Essentially stopping a supersonic or hypersonic airflow is exactly what the inlets of ramjets and gas turbine engines do. And yes, the inlet ram drag is on the order of at least half the nozzle thrust force. A Mach 3 gas turbine is doing this with air at about 0.9 km/sec, and a Mach 5 ramjet does it with air at 1.5 km/sec.

The only real problem arises above about Mach 4 or 5, when the captured decelerated air starts approaching flame temperatures. I show inlet total temperatures near 3000 F (near 1650 C) at Mach 6 100,000 foot conditions, for example. There's certainly no such thing as cooling air for vehicles trying to fly at these speeds.

The SABRE engine is liquid hydrogen fueled. I do not know how they separate the nitrogen or what they do with it, but they claim a breakthrough in heat exchanger technology that allows them to liquify the air as fast as they scoop it up, using the hydrogen as the heat sink. In rocket mode, it burns hydrogen and (largely) air-derived oxygen that it stored in tanks. The chamber pressure is typical rocket: fairly high.

The high heat exchange rate and heat exchanger icing risk are the two historical show-stoppers with that, but they claim to have solved those problems. What little ground test data I have seen points in the direction of verifying their claims, but they have a long way yet to go before they fly.

And no, they do not operate the air scoops above Mach 5 on ascent, or at all during descent. I think the HOTOL launch trajectory says they are climbing and accelerating at Mach 5 at somewhere around 100,000 ft when they cease scooping air and go plain rocket on the propellants they have on-board. Above that flight point there just isn't enough air to do anything with it that is useful. They leave the sensible atmosphere long before reaching anything approaching orbital speeds.

The idea is not to take off with a full load of LOX, but to scoop up a major portion of it as they ascend.

As I said in the other posting, the real dangers occur during reentry. How they protect their air scoop and machinery I have no idea, although a static gas column is the best heat shield by far. Perhaps they simply allow no flow through the inlet by closing it off, I don't know. But the inlet compression spike sticks out and is at risk, at least.

And as I said in the other post, the real vulnerability here is the wing leading edges, hit by shock impingement coming off those spikes. That is an inherent risk they face for deciding to use wingtip-mounted engines. It's far worse on reentry, but has become significant on ascent by about Mach 4-ish.

GW

Antius · 2015-11-18 06:34:39

Based upon the information provided by GW, I have investigated the options for a two-stage launch vehicle. The first stage is provided by a horizontal take-off ramjet aircraft, which burns jet fuel and is boosted to ~400mph by lox-jet fuel rockets within its wings, at which point ramjet thrust takes over. I have not subjected this stage to detailed analysis, but have assumed that the vehicle will release the upper stage at a speed of Mach 6 (2km/s) at a height of 100,000’ (30km).

The second stage will be fuelled by LOX-jet fuel. The additional delta-V required to reach orbit is 6.75km/s. Assuming acceleration takes place at 5g, gravity losses work out to be 1.25km/s, which is only slightly lower than the upper stage upward release velocity of 1.4km/s, assuming second stage release takes place at 45 degree angle. I have taken exhaust velocity to be vacuum exhaust velocity of 3500m/s. With these values, the mass ratio of the upper stage is 6.88. My assumption is that (initially at least) the upper stage will be expendable, which increases payload mass and reduces design complexity. A reusable upper stage can be developed later on if cost-benefit analysis shows it to be worthwhile. The traditional coupling mechanism between stages is explosive bolts. However, this raises the potential for single point failure across a large number of bolts and is not very compatible with lower stage reusability and rapid turn-around. Hydraulic clamps may be a better alternative if the mass penalty can be tolerated.

My initial preference for the expendable stage was maximum possible design simplification. This led me to favour pressure fed fuel feed systems, thus avoiding the need for turbo-pumps. The engine combustion chamber would work at a pressure of 20bar. At such low pressure, it could be ablative lined carbon steel, with the exit nozzle being carbon steel and preferably radiatively cooled. The low chamber pressure does not impact performance as the engine will be firing in a near vacuum. Tank pressures would be 30bar.

My calculations indicate that even using spherical pressurised tanks it was very difficult to reach the required mass ratio with safety factors of 3. To overcome this difficulty would require either the use of maraging steels, which cost ~$100/kg or the use of more than one stage. Alternatively, a lower safety factor could be used, but this would push up quality control costs. All of these appear to be counterproductive for an expendable stage, which needs to be cheap and easy to mass produce, rather like a bullet. However, the low chamber pressure would appear to allow the use of gas jet feed pumps, which have no moving parts and should be very cheap to produce. The gas can be provided by an electrically heated boiler. Lithium ion batteries appear to provide sufficient energy density to power the heaters without severely impacting the mass ratio. The use of jet pumps is possible only for engines with relatively low chamber pressures. Assuming tanks are pressurised to 3bar feed pressure and are constructed from alloy steel with yield stress of 600MPa, then the tank mass ratio reduces to ~2% (i.e. the empty mass of the tank is 2% full mass, ignoring valves).

Attitude control during the 135 second acceleration can be accomplished using some combination of nozzle gimbaling or cold gas (nitrogen) jets, as the required impulse is modest. The elimination of gambling would allow further simplification of the engine (i.e. fixed nozzle).

Overall a simple, mass produced upper stage would appear to be workable. Assuming a mass ratio of 2 for the 1st stage and a 10% empty mass for the 2nd stage, the overall mass ratio would be 44. If total take-off mass is 1000 tonnes, payload to orbit would be 22.7 tonnes. To deliver payload at a cost of $100/kg, the total launch cost must be no greater than $2.3million. This is a tall order, and requires that the upper stage be manufactured for not much more than $1million each. It also requires that the vehicle is able to use conventional facilities (airports) with minimal additional ground infrastructure (hence the choice of propellants).

Antius · 2015-11-18 07:45:25

Just realised I forgot to account for the Earth's spin. That knocks 500m/s of the orbital speed velocity change and reduces overal mass ratio to 32. Quite a significant improvement.

RobertDyck · 2015-11-18 08:12:56

I am very disappointed. GW, you were not successful with X-30, so claim no one can do it. What ever happened to "If first you don't succeed, try try again?" You are actively impeding any attempt. This isn't "lessons learned", this is bitterness.

You two have violated a very simple engineering principle. You have violated the engineering requirements. The requirements for this discussion thread are single stage to orbit. No drop off stages, no drop off external tank. No drop off anything.

Dan Goldin is a former NASA administrator. He pointed out when commercial air liners fly, only the fuel doesn't come back. They don't drop off parts of the aircraft. No commercial airline could afford to operate if aircraft dropped off major parts of the plane. Space travel requires the same thing. This discussion started as a response to question by martienne about cost of ticket. She said "Unless something very revolutionary happens, we are talking about hundreds of millions of dollars in todays value, for one ticket!" The point of this discussion is to show how to build an SSTO: Single Stage To Orbit.

We don't want to build a propeller aircraft. We don't want a horse-drawn buggy. Multi-stage rockets are just that. They are crude attempts to compensate for the limits of chemical rockets. The solution is something other than chemical rocket for propulsion. We also have the problem that today's launch vehicles evolved from military technology. The military built ballistic missiles as weapons. You don't reuse ammunition. So military just can't think in terms of a commercially viable vehicle. You have to stop thinking of a launch vehicle as ammunition.

Terraformer · 2015-11-18 11:23:31

See, it's quite simple really...

https://www.youtube.com/watch?v=BKorP55Aqvg

Antius · 2015-11-18 12:38:16

I will respond in more detail later.

I sympathise with your aspirations, but absent an entirely new propulsion concept, the SSTO is not going to deliver the sort of low cost performance you anticipate. The engineering requirements for an H2/O2 fuelled SSTO are extreme - mass ratio is slim (<10%) and payload ratio worse still. That pushes you towards the state of the art in every area, ultra high chamber pressure engines, ultra-lightweight cryogenic tanks, reentry heat shields that weigh no more than 1% of take off mass and of course hydrogen fuel, with all the problems that brings. A chemical SSTO would have similar economics to the shuttle - it's capital cost and maintenance schedule would be comparable, it's payload fraction even worse.

Last edited by Antius (2015-11-18 12:52:57)

GW Johnson · 2015-11-18 13:54:31

Thread title is "spaceplane", not "SSTO". Under that title, multistage systems are indeed kosher.

Desire to do something does NOT trump physics and chemistry. With chemical or even solid core nuclear energy sources, the mass ratio requirements are too high and the payload fraction too low for a true SSTO aircraft. Not with the technologies and materials that we actually have. That's just physics.

And don't forget that in the thin air all airbreathers have low frontal thrust density, because the combustion chamber pressures are inherently low, because they are only a ratio to ambient air pressure. To accelerate or climb, (thrust minus drag)/weight must be significantly greater than 0. Physics.

What killed X-30 was low frontal thrust density at over 100,000 feet altitudes, plus the utterly-appalling drag losses of trying to fly Mach 10+ that low in the atmosphere (drag losses far, far larger than the orbital speed requirement). It was mostly physics, not so much technology, that caused the failure of that program. (That and incredibly inept management and political decision-making right up front when the project started.) Although, scramjet technology was unready then, just like it still is today.

Nothing about that fatal physics problem has changed since. Nor have the technologies, when you get right down to what we actually have available.

If you want to go chemical, then you really need to go TSTO (or even 3 stages) to reach orbit at feasible mass ratios and payload fractions. Doesn't matter whether you launch vertically or horizontally.

From the standpoint of drag loss reduction, vertical launch is more efficient, typically only needing a 10% increase in delta-vee over the endpoint requirement velocity. But you are typically looking at leaving sensible air at only about 2000 ft/sec (700 m/s) on a near-vertical path at around 100,000 ft (30 km) altitude.

Horizontal takeoff trajectories have exponentially-higher drag losses, but allow you to take advantage of airbreathers for their higher Isp. But it has to have air to breathe! Which forces you to stay low as you accelerate. Where the drag and heating get enormous as you speed up. That's a real iffy tail-chase, because of physics. Solution spaces go to zero size very quickly as you attempt more demanding design requirements.

From a reusability standpoint, it does not have to be SSTO. You can fly a carrier airplane back to base like any other airplane. No different than a bomber, fundamentally. The second stage can be an airplane (spaceplane), or it can be a capsule-plus-rocket stage.

Mass ratio is more feasible if the staging speed is closer to 10,000 ft/sec than 6000 ft/sec. This requirement is tougher to meet with a spaceplane second stage (because so much more inert mass must be delivered on-orbit) than it is a capsule-plus-rocket-stage. Nasty little fact of life, but there it is.

I do notice that no one has ever recovered a stage from orbit, nor is anyone contemplating making the attempt yet. The heat protection for entry is heavy, and the decelerating airloads on the stage are enormous, requiring a far stronger structure which is far heavier. Not to mention the precision attitude control, and the final touchdown scheme (chutes or whatever). These things are all unfavorable for attempting a solution to that problem.

I also notice that there has never, ever been a reentry vehicle that had adjacent nacelles. That's because of shock-impingement heating. It's even worse than stagnation point heating. So, a small plane carried on the back of a big one makes less and less sense as speeds exceed about Mach 4. The impinging shock waves cut the adjacent structures to pieces. To stage at speeds in the Mach 6 to 10 range, your second stage will need to be an internally-carried store, and you will have to pull up out of the air to open the doors and release it.

Be careful in your estimates about what can be done with airbreathers, because neither thrust nor specific impulse are even remotely constant as flight conditions change. And there are other less-obvious limits on these technologies.

For example, Antius has a ramjet concept in his post above that takes over at 400 mph and yet thrusts greater than drag at Mach 6. Sorry, the technology simply doesn't work that way. Physics does not allow it.

Ramjets capable of thrusting at high subsonic speed have nose-mounted pitot/normal shock inlets and convergent-only nozzles that don't choke until flight speed reaches about Mach 1.1. These designs have thrust that peaks about Mach 3, Isp peaking at around 900-1000 sec at Mach 1.5, and thrust usually is less than almost any imaginable airframe drag by about Mach 2 to 2.5, at the very outside 3.

The high speed ramjet designs capable of thrust = drag near Mach 6 (and only on a VERY clean airframe!!!) have external compression features to their nose-mounted inlets, a choked convergent-divergent nozzle, and a min takeover speed in the Mach 1.8 to 2.5 range. Isp peaks just above shock-on-lip speed (near Mach 3.5 to 4), thrust rises from low to max at shock-on-lip speed (just about Mach 3 for this type of application), and slowly decreases as speed increases. Peak Isp is around 1300-1500 sec. At Mach 6 it's down to around 400 sec. At Mach 2, thrust is around 25% what it is at Mach 3-4.

I might add that thermal protection inside and outside becomes THE dominant design issue if you try to fly faster than about Mach 4. The roots of this technology came from missiles, which were ablatively-protected one-shot devices. That kind of refurbishment is very slow and expensive.

You will NOT be able to air-cool them, either. No matter what type of air scoop you use, at Mach 6 the scooped-up air temperatures will be near 3000 F (for reference, meltpoints of steels are 2935 F, or a little less for the stainless grades). Flame hot air, and that's before you add fuel and burn.

Some sort of low-density (and therefore fragile) ceramic might serve for the insulators of the inlet duct and combustor. But you will have to backside-cool them to make it work: you cannot radiate-away heat from a surface located on the inside of your vehicle. Inconvenient physics again. You can only take advantage of low thermal conductivity by maintaining a heat flow through the material, and that heat has to go somewhere.

The coolant massflow may greatly exceed the engine fuel flow. Depends on how thick you make the low-density insulation. Thicker reduces backside cooling requirements, but raises the surface temperature inside the combustor much closer to the flame temperature. At Mach 6 conditions, flame temperatures will be near 5000 F. (Ionization is reducing the thrust you can get out of the nozzle, too.)

Physics is a bitch, ain't it?

I did an experimental low density ceramic insulator like this decades ago, but not at conditions this extreme. It would handle 3200 F, and so would work in the inlet duct we are talking about here, just as I built it back then. I am looking at some other ceramics to handle a 5000 F flame temperature in the combustor. Will it work? I think so, but I simply don't yet know. No one else does, either.

You don't do this with fire bricks or the new "super-ceramics". Those are high-density, high conductivity, and the backside cooling requirements are truly enormous, even compared to rocket engine flow rates. That's the wrong physics to apply to inlets and combustors. Works for nose tips and leading edges, though.

Characteristics of various airbreathers:

Gas turbine: operates from 0 to about Mach 3.3 max, somewhat low frontal thrust density, very heavy. Limited by (1) turbine inlet temperature < 2200 F for exotic military technologies, (2) compressor stage catastrophic overheat failures above about Mach 3.3 to 3.5, (3) extreme difficulty matching engine air flow demand to inlet scoop massflow characteristics above Mach 2.5, even with variable geometry (which is also very heavy).

Low-speed range ramjet: Mach 0.7 to at most Mach 3, peak Isp 900-1000 sec at about Mach 1.5, nearer 400 at Mach 0.7 and Mach 3. Low frontal thrust density, peaking about Mach 3. Frontal thrust density varies very nearly directly with atmospheric ambient pressure. Very lightweight hardware.

High-Speed Ramjet: Mach 2 to Mach 6 possible with a fixed-geometry nose inlet, shock-on-lip about Mach 3. Peak Isp around 1300-1500 sec about Mach 3-3.5. Thrust rises to peak at shock-on-lip Mach 3, holds near-constant to about Mach 4, decreases slowly to Mach 6 (as drag rises quadratically). Isp near 400 sec at Mach 6. Same frontal thrust variation with altitude as low-speed ramjet, just the numbers are a little higher because chamber pressures are a little higher. Still low. Very lightweight hardware, although slightly heavier than low speed designs, because it is stressed more.

Scramjet: technologically unready-to-apply; min takeover speed Mach 4. With hydrocarbon fuels, peak speed may be in the range Mach 8 to 10, we just do not know yet. With hydrogen, peak speed might be as high as Mach 15-ish, we just do not know yet. Isp is no better than ramjet, really. Neither is frontal thrust density. "Best" values simply occur at higher design speeds. Same frontal thrust density variation with altitude as the other airbreathers. Hardware tends to be heavier than high-speed ramjet, because it is stressed much more.

Effects of using nuclear instead of chemical energy release in any of the airbreathers: somewhat-higher Isp, much heavier inert weights, no changes to thrust levels. Risk of radiation in the exhaust.

GW

RobertDyck · 2015-11-18 14:14:18

Antius wrote:

The engineering requirements for an H2/O2 fuelled SSTO are extreme - mass ratio is slim (<10%) and payload ratio worse still.

That's why I said don't use chemical rockets. Instead most of the speed, and most altitude out of the atmosphere are achieved by a nuclear jet engine. That means absolutely no propellant carried what so ever. The purpose of the kerosene fuelled jet engine is to achieve speed necessary to ignite the nuclear jet engine, and because I expect airports will be squeamish about a nuclear engine. The nuclear thermal rocket is for the final push into orbit, and orbital manoeuvres. For an SSTO, the propellant tank must be small because it must be enclosed in a heat shield. Expect the heat shield will be greater mass than the tank itself. That's why I said distilled water instead of liquid hydrogen. With an NTR, that still provides greater Isp than the best LH2/LOX rocket engine. And that means fuel carried on-board is just jet fuel (jet propellant) and water, so not cryogenic, not hypergolic, and not toxic. That greatly simplifies handling.

Of course RCS propellant is an issue, the best we have so far is MMH/N2O4, used for thruster quads of the Apollo service module, and Shuttle, and now Orion. I think Dragon uses the same fuel mix. MMH is hypergolic, N2O4 is toxic, so that complicates handling. I have a paper published in 2013 about something called AF-M315E. The paper claimed it's TRL5 with demonstration engine at 11.5 hours firing. The paper pitched a mission called Green Propellant Infusion Mission (GPIM) Technology Demonstration Mission. Part of their pitch says: "The culmination of this program will be high-performance, green AF-M315E propulsion system technology at TRL 7+, with components demonstrated to TRL 9, ready for direct infusion to a wide range of applications for the space user community."

AF-M315E offers higher performance than hydrazine, yields 12% higher Isp (257 vs. 235 sec), and is 45% more dense (1.47 vs. 1.00 g/cc), affecting both reduced propellant and tank mass.
...
With its lower minimum temperature threshold, AF-M315E yields an additional advantage of mitigating operational concerns related to long-duration system thermal management. Whereas hydrazine space tanks and lines must be heated at all times to prevent freezing, AF-M315E cannot freeze (it has a glass transition). During long coast periods an AF-M315E propulsion system may be allowed to fall to very low temperatures and later reheated for operation without risk of line rupture by phase-change-induced expansion.

Aerojet Rocketdyne has 1N (GR-1) and 22N (GR-22) advanced monopropellant thrusters. All this sounds good until you realize the bipropellant used on Shuttle have Isp=316 seconds for OMS, 289 for RCS. Aerojet Rocketdyne compared their new thing to monopropellant.

Last edited by RobertDyck (2015-11-18 14:57:28)

RobertDyck · 2015-11-18 22:06:46

Just looked up "plasma magic". Apparently it originated in Soviet Russia. Some western analysts gave it that name to discredit the idea. I can only find limited reports of analysis. One analysis says at supersonic speed the reduction in drag is limited, that energy necessary could be just as well added to propulsion. But I can't find any serious analysis at hypersonic speed. Then there's the idea of running a jet engine at such a high temperature it's at the limit of materials, so adding more energy there is just not an option. And there's the fact this would be a lifting body, optimized as a re-entry vehicle, not optimally streamlined. So reduction of drag would be highly beneficial. And plasma is multi-dimensional. And there are reports that the plasma pushes the shock wave further in front of the aircraft, another benefit for a the intake of a hypersonic aircraft.

SpaceNut · 2015-11-18 22:45:08

coolers ,compressors and heat exchangers...

http://mice.iit.edu/mnp/MICE0161.pdf

http://lss.fnal.gov/conf/C720919/p653.pdf

http://docs.lib.purdue.edu/cgi/viewcont … changer%22

Antius · 2015-11-19 12:32:02

Some excellent posts and useful information. The need to accelerate to Mach 2 (using LOX/kerosene rocket propulsion) before activating the high speed ramjet is an acceptable mass penalty (MR for that stage works out at ~1.27). To estimate an overall MR would require working out MR for each stage of the flight and multiplying them together. I will attempt this later, though I don't expect it to be particularly accurate with the resources I have.

The ultimate launch cost will be a combination of development cost, capital cost, operating cost and maintenance cost. The best (cheapest) solution will therefore be very specific to the operating environment. As launch volumes will be low to begin with and you will likely be short of cash as a start-up, it is wise to base the launch concept on proven technologies that can be engineered without extensive R&D and operated from existing facilities. Hence my preference for an expendable LOX/Kerosene upper stage to begin with and a ramjet lower stage that is confidently rooted in existing technology. As launch volumes increase and your company earns more cash, a reusable upper stage can be developed without having to re-engineer the lower stage. The end result is a fully reusable system, but developed increments as you can afford.

GW Johnson · 2015-11-19 13:27:45

Returning to the original topic “spaceplane”, combined with RobertDyck’s preferred focus on a SSTO approach, I did a little bounding calculation to show what kind of propulsive energy might be required to provide operating characteristics like an airliner. I assumed the following: (1) vertical ballistic launch so that delta vee for the ascent propulsion is but 10% higher than LEO velocity, (2) the weight statement characteristics of a Boeing 747-100, (3) some sort of rocket propulsion so that the rocket equation applies, and (4) that exhaust velocity divided by the standard acceleration of gravity approximates engine average Isp.

For the B-747-100 I quickly found these characteristics on the internet: empty weight 358,000 lb, max takeoff weight 750,000 lb, max fuel capacity 48,445 US gal (approximately 330,900 lb of any of the commercial jet fuels), max cargo capacity (as a freighter) 247,800 lb. This aircraft was powered by four JT-9D engines of 46,500 lbth each, and required about 10,200 feet of runway. Its cruise speed was 0.84 Mach, max 0.89 Mach. You cannot carry max cargo and max fuel simultaneously, that violates max takeoff weight. You can only have up to 392,000 lb of fuel plus cargo, that’s all.

Wempty/Wmax=358,000/750,000=.4773 is the inert mass fraction for this vehicle. That corresponds to an airframe designed to last decades and/or about 40,000 landing/takeoff cycles. If you want long life and airline-like maintenance, you had better be considering large inert fractions like that! Structure that survives a long time costs weight, there is no way around that ugly little fact of life! Besides, a spaceplane faces far-harsher loads than a subsonic airliner, and I did not make any adjustment for that, or for the fact that we cannot fly a simple aluminum airframe to and from space. This thing will need a heat shield. And a lot of its structure will get warm. Aluminum is slush at 350 F, organic composites are debris at 250 F. Steels are good to 1000-1500 F depending on how expensive an alloy. Titanium is no better than steel, and often lacks forming properties.

Further, we will need at least 800,000 lbth to take off vertically for a ballistic ascent. 1,000,000 lbth would be a lot more attractive for efficient rocket ascent at 750,000 lb initial weight. The original airliner had only 186,000 lbth at takeoff. Remember, airbreathers have low frontal thrust density!!! You will never achieve a practical design trying to take off vertically with airbreathers.

Max payload corresponds to a fuel fraction of 144,200/750,000=0.1923 and a payload fraction of 247,800/750,000=0.3304. Note that 0.4773+0.1923+0.3304=1.0000, satisfying gross weight. The mass ratio is MR=1.23808, corresponding to an average engine Isp=4002 sec, average over the entire ascent! The only rocket propulsion concept that could fill this bill would be open-cycle gas core nuclear thermal, and only if regenerative cooling would work (it would not, the max Isp for open cycle gas core without a gigantic radiator was estimated to be 2000-2500 sec years ago). The only other known concept capable of doing this job is nuclear pulse propulsion. Open cycle gas core and pulse propulsion both shed radioactive plumes.

Max fuel loadout corresponds to a fuel fraction of 330,900/750,000=0.4412, and a payload fraction of 61,100/750,000=0.0815. It’s the same inert, for 0.4773+0.4412+0.0815=1.0000, and a MR=1.78955. That corresponds to an average rocket engine Isp = 1469 sec, again the average over the entire ascent! That’s easily within range of a regeneratively-cooled gas core nuclear thermal engine (and certainly nuclear pulse propulsion), and possibly within reach of a closed-cycle “nuclear light bulb” gas core engine. I saw various estimates for those years ago ranging from 1300 to 1500 sec. They would have a clean exhaust.

If you lower the trajectory, and go to horizontal launch to make airbreathers feasible, you pretty much have to at least double the delta-vee to around 50-60,000 fps, which is going to drive the impulse requirements even higher. It could be even worse than that, if you attempt to fly a q=5000 psf reentry-in-reverse, the way they chose to do on the X-30 project. (I did NOT work on that one, by-the-way.)

A low-bypass gas turbine engine has typical TSFC=1.0 lb/lb-hr, corresponding to Isp=3600 sec. But it only operates from takeoff to Mach 3.3. If you average that with say 900 sec for a solid core nuclear rocket, you get something near 1224 sec averaged over 27,500 ft/sec. But you cannot takeoff vertically with the gas turbine, so you must increase the delta-vee drastically, which lowers your average Isp.

Assume that 60,000 ft/sec is “realistic” for HTO/HL and a nuclear ramjet of some kind. 0-3300 fps on gas turbine at Isp=3600, 3300-9000 fps at about 2000 sec on the ramjet (just a guess), and 9000-60,000 fps on a solid core nuke rocket at a generous 1000 sec. That averages Isp over the ascent at about 2700 sec. RobertDyck’s spaceplane might possibly be made to work with gas turbine takeoff, nuclear ramjet, and solid nuke rocket propulsion. If the delta-vee can be kept down to 60,000 fps. If the solid-core nuke ramjet and a solid core nuke will both fit within the inert weight allowance (I really doubt one core can do both jobs, the geometries are just too incompatible). And if a nuke ramjet will really delivery an average of 2000 sec Isp from Mach 3.3 all the way to Mach 9 (I really doubt that will be doable, either). That’s a long list of big “ifs”.

If you do it vertical ballistic launch with a nuclear lightbulb engine, the only “if” is the nuclear light bulb engine that averages 1500 sec Isp. Only one “if”. That’s just a better probability of success.

From this, I conclude that if you want a SSTO spaceplane with characteristics that truly resemble an airliner, you ought to look at vertical ballistic launch, and you had better spend some very serious technology development funds up-front on closed-cycle gas core nuclear rocket propulsion. I’m pretty sure from this bounding calculation that the winged SSTO vehicle concept would work, but only if the propulsion can be made to work! It would likely be near 45% inert structure (actually including on-orbit and de-orbit propulsion, plus some undefined propulsion for ”go-around” at landing), 45% ascent fuels/propellants, and 10% dead-head cargo payload.

It had better not have maintenance-intensive shuttle tiles for its heat shield! But Avcoat ain’t gonna cut the mustard for this. And I’m not sure PICA-X would either. Both are ablatives and would have to be replaced every ride or two ($$$$$). Low density ceramic might be the way to go, but not the way we did it on shuttle or X-37B ($$$$$).

GW

Antius · 2015-11-20 09:03:40

The closed cycle light-bulb would definitely be the preferable option for a reusable nuclear space launch vehicle.

I once carried out a thermal hydraulic study of a solid core nuclear SSTO as a minor degree project. It was very difficult to produce a design that maintained enough heat transfer to give good enough power to weight ratio, whilst also meeting the requirements for high exhaust temperatures, which are essential to ISP, within the melting points of metallic cladding. Other claddings, such as carbides and graphite, tend to be more permeable to fission products and have poorer thermal conductivity. Radiation exposure to crew is also significant. I estimated that gamma-ray doses from the engine would have been around 100mSv (10rem) per flight. Not such a problem for cargo on an automated vehicle, but would only work for people if they made only one flight per lifetime or could use payload to provide extra shielding. Due to low material strength at such temperatures, I opted for a tube-in-duct core design, with fuel embedded within a graphite moderator block and propellant ducts passing through the block. The high temperature would otherwise have made the assemblies too soft to stand up to the forces of propellant flow. This provided the necessary bracing, but added a lot of mass to the core. The fuel assemblies must also be vented to prevent excessive fission gas pressure from over-stressing the cladding, which would otherwise occur after just a few flights.

The decay heat removal task was eased somewhat by carefully timing the engine shutdown before propellant exhaustion and allowing the final few percent of delta-V to be achieved using residual decay heat. The remaining decay heat could be removed by liquid sodium channels passing through the graphite engine block and carrying heat to radiators on the outside of the vehicle.

The lightbulb would appear to side-step a lot of the killer problems of the solid core rocket. The increased power density makes shielding much easier and the higher temperatures reduce propellant mass fraction.

One of the biggest advantages of the lightbulb is the fact that the core can be purged of fission products after each flight, potentially before re-entry. This is key to minimising the nuclear risk. Assuming 30% efficiency for the rocket engine, some 150MJ of nuclear heat must be expended for each kg of vehicle mass that reaches orbit. For a 100t craft, that works out to be 15,000GJ (4.2GWh). Each fission yields 3x10-11J of harvestable energy, so a total of 5E23 fissions are needed to reach orbit. That’s a little less than 1mol of 235U, about 0.2kg of 235U per flight. That’s a small amount compared to the inventory of a large nuclear reactor here on Earth, which might contain up to a couple of tonnes of fission products at the end of core life. But the frequency of accidents is much higher for a launch vehicle, so purging allows the radiological consequences of a launch failure to be kept much lower, and therefore public risk to be more tolerable. If the burn-up of the hexafluoride gas can be increased to 10%, then fission product separation becomes unnecessary and the fuel gas can simply be dumped after a single orbital flight. That would allow the vehicle to re-enter the atmosphere without a radioactive payload. If the fission gas can be stored in orbit for 300 years, its activity will decline to a level no greater than uranium ore. So any orbital storage facility must maintain orbit for that long.

Risk can be reduced further by launching a light bulb from a remote area where a launch failure would keep fallout away from populations. A sea launch may be a viable option. As launch volumes increase, sea launch would allow for much larger launch vehicles and economies of scale.

I am a proponent of nuclear power, but I cannot escape the impression that the development and operational costs of a nuclear launch vehicle would ruin its economics. The international hysteria surrounding the Fukushima accident should give you a fairly good idea as to the opposition you would face in trying to commercialise a flying nuclear reactor. But maybe I am mistaken. The risks could be shown to be acceptable and the public may accept them in exchange for space travel that they could afford. Large aeroplanes crash catastrophically about once in every million flights. If your vehicle has 10 times that accident frequency and flies 100 times per year, its failure frequency is once every 1000 years. If radiological consequences are only 1% those of a worst credible nuclear power reactor accident, then the radiological risk of a nuclear launch vehicle would be about the same as a power reactor. A higher frequency of failure, but lower consequence.

SpaceNut · 2015-11-20 20:37:08

I actually interviewed to work at this company http://www.fibermaterialsinc.com/ that was doing design work for Nasa on the Pica material and on Aerogel as well but did not get the job due to commute distance...

In January 2006, NASA's Stardust sample return capsule, equipped with a PICA heat shield, set the record for the fastest reentry speed of a spacecraft into Earth's atmosphere - experiencing 12.9 kilometers per second (28,900 miles per hour).

PICA is a modern TPS [Thermal protection systems] material and has the advantages of low density (much lighter than carbon phenolic) coupled with efficient ablative capability at high heat flux. Stardust's heat shield (0.81 m base diameter) was manufactured from a single monolithic piece sized to withstand a nominal peak heating rate of 1200 W/cm^2.

The Dragon capsule will enter the Earth's atmosphere at around 7 kilometers per second (15,660 miles per hour), heating the exterior of the shield to up to 1850 degrees Celsius. However, just a few inches of the PICA-X material will keep the interior of the capsule at room temperature.

A sample of PICA-X heat shield material subjected to temperatures of up to 1850 degrees Celsius (3360 degrees Fahrenheit), at the Arc Jet Complex at NASA Ames Research Center, Moffett Field, California. The NASA-originated PICA material holds the record for high-speed reentry into the Earth's atmosphere. The SpaceX-developed and manufactured PICA-X variants meet or exceed the performance of the original material, and will protect the Dragon spacecraft on its return to Earth.

The Dragon 1 spacecraft initially used PICA-X version 1 and was later equipped with version 2. The Dragon V2 spacecraft uses PICA-X version 3. SpaceX has indicated that each new version of PICA-X primarily improves upon heat shielding capacity rather than the manufacturing cost.

A Perspective on the Design and Development of the SpaceX Dragon Spacecraft Heatshield

GW Johnson · 2015-11-21 11:30:56

I've nothing against PICA-X, as Spacex developed it for Dragon to do a free return from Mars. A trajectory like that has a nominal entry speed in the vicinity of 50,000 ft/sec (15 km/sec). But at those conditions, it's a one-shot ablative. That same heat shield could likely be flown twice back from the moon at 36,000 ft/sec (11 km/sec), and several times from earth orbit. But for an orbital craft, you still must replace it every several trips. It's an ablative, it slowly burns away.

About 3 years ago, I gave a "gee whiz something I did long ago might serve as a heat shield" paper at a Society convention in Boulder, Colorado. The material was something I cooked up from alumino-silicate pipe insulation paste as a matrix, and commercial aircraft engine nacelle fire curtain cloth as a reinforcing fiber. It was the ceramic equivalent of fiberglass, and it had a pleasingly-low density and thermal conductivity. If not done as a composite (paste without fabric), it was as fatally-structurally weak as the NASA shuttle tiles that inspired it.

I used it as a simple low-conductivity liner inside a small combustor, where it did a very good job. As a composite, it withstood many accumulated hours of burn, near a hundred heatup/cooldown cycles, and dozens of excursions into a very violent rich blowout instability in my little combustor.

I did have a little surface melt damage, as my full rich flame temperatures were about 3500 F (1900 C). The meltpoint of these materials is right at 3350 F (1850 C). This material experiences cracking from shrinkage and solid phase change effects at temperatures that exceed 2300 F (1260 C). As a composite, the propagation and growth of these cracks are quite limited, and such surface cracks didn't affect my combustor liner application. The liner was still intact and quite serviceable, when my little combustor project was done decades ago.

I looked at this same material recently as a potential refractory heat shield material; that was what my convention paper was all about. The operating concept was re-radiation of convectively-gained heat back to space, with the conduction of heat through the material essentially cut off. That's how shuttle tile functioned (I told you I was inspired by shuttle tile, now didn't I?). The difference was my composite had substantially-more robust structural properties. The fabric reinforcement also afforded a second means of retention beyond just adhesive bonding. No more lost tiles, plus, larger panels are possible.

As a heat shield, surface cracks are a structural vulnerability under such extreme fluid shear conditions. So, my analyses explored entry conditions such that surface temperatures never exceeded 2000-2200 F (1100-1200 C). That puts restrictions on descent angle at entry and on ballistic coefficient (or its effective value for winged vehicles). The surface must be black, but the manufacturer of the paste said I could add carbon black to the surface coating that I used decades ago.

There are other zirconia materials that have much higher meltpoints that might be candidates, used in conjunction with my material, as a second outer layer. The zirconias are denser, more fragile, and harder to work with, and do not combine with a reinforcing fabric in the same way as the aluminosilicates did. But surface meltpoints are near 4700 F (2600 C).

Bolt-on panels of some refractory ceramic composite shield material, especially with redundant retention (mechanical plus adhesive bond) might offer a path to the heat shield we need for any sort of spaceplane or capsule between surface and orbit. I'm not at all worried about joints between panels. That was proven not a problem with the Gemini capsule that was re-flown in 1969 for the USAF MOL program, with a hatch cut in its heat shield (the Gemini B configuration).

GW

Last edited by GW Johnson (2015-11-21 11:33:39)

SpaceNut · 2015-11-21 22:26:45

Sounds simular to the shuttle repair tile goop that was tested but it would seem that a layer of fiber between each application of it to form the shape that is needed would really be the cake....

http://www.space.com/2619-spacewalkers- … nique.html

NOAX, which is short for non-oxide adhesive experimental, is a sticky black substance that carries the initial consistency of peanut butter before it is worked into place in orbit, NASA officials said. The space agency described the material as a pre-ceramic polymer that is impregnated with carbon silicon carbide powder.
Engineers designed NOAX as a coating, crack and gouge filler for the black panels that protect Discovery's wing leading edges and nose cap. Made of a carbon composite called reinforced carbon carbon (RCC), the panels endure the hottest temperatures when the shuttle reenters the Earth's atmosphere.

http://www.coiceramics.com/specialtyproducts.html

GW Johnson · 2015-11-26 16:21:34

I did some more work and cleaned up my spaceplane bounding analysis. It's now posted over at http://exrocketman.blogspot.com, dated today (11-26-15). Conclusions didn't change, but it's well-supported with data.

In order to patch the kind of hole that killed the Columbia and her crew, the NOAX goo would have to be combined with some sort of substrate strong enough to hold the goo up against leading-edge stagnation pressures. That could be a simple heavy screen wire mesh with ordinary fabric enclosing it. Cut and bend the mesh to fit, smear some goo around the hole to hold the mesh in place, then overcoat the patch with a coat of the goo.

This kind of patching could have been done in one EVA to save Columbia and her crew. Being able to patch damage like that was what was originally promised, but never delivered, apparently.

GW

RobertDyck · 2015-12-15 17:04:13

Looking at propellant for a commercial spaceplane. I suggested using jet fuel for take-off/landing, and acceleration to speed required to start the next engine. Then air as propellant for a nuclear jet engine to gain speed and altitude. Then steam distilled water as propellant for a nuclear thermal rocket for final push into orbit. Nuclear engines require respect, but as long as the reactor remains sealed, vehicle processing is simple. And I suggested Americium-242m as the nuclear fuel. Americium-241 is the isotope in smoke detectors, so it's safe to handle as well. Plutonium is extremely poisonous, Uranium metal is toxic, Uranium oxide less so but must not be ingested, however Americium is in smoke detectors sold for use in homes. Carried propellant is either normal jet fuel, or water. That just leaves RCS thrusters for manoeuvring in space. One option was the "green propellant" being developed by NASA. AF-M315E is monopropellant with Isp=257 seconds, which really doesn't compare well to N2O4/MMH bipropellant. What could we use as an easy to handle propellant?

What about LOX/propane? It has Isp=361.9 seconds in vacuum with 100:1 nozzle-to-throat area ratio, fuel density 582 kg/m^3, bulk density (fuel + LOX) 905 kg/m^3 at Near Boiling Point (231°K). At 100°K which is just 10° above freezing, fuel density is 782 kg/m^3, bulk density 1014 kg/m^3. Combustion temperature 3734°K. As comparison, methane Isp=368.3 seconds in vacuum, fuel density 423 kg/m^3 at NBP, bulk density 801 kg/m^3, combustion temperature 3589°K. Propane is liquid at room temperature at very mild pressure, which is why it's used for backyard BBQ. Pressure for a propane tank at +25°C is just 10 bar (1 MPa). Of course the tank requires a safety margin above that, but we're talking about the pressure of a backyard propane tank at room temperature. That's easy to handle.

Shuttle RCS thrusters and Apollo service module thruster quads used N2O4/MMH. Apollo used R-4D, Isp=312 seconds, expansion ratio 164:1. So LOX/propane has a higher Isp. The issue is keeping LOX cold without boil-off. Is that reasonable?

RobertDyck · 2016-01-03 19:44:20

Superfast Military Aircraft Hit Mach 20 Before Ocean Crash, DARPA Says

A superfast unmanned military plane traveled at 20 times the speed of sound and managed to control itself for three minutes before crashing into the Pacific Ocean in a recent test, military officials said.
The prototype Falcon Hypersonic Technology Vehicle 2 (HTV-2), billed as the fastest aircraft ever built, splashed down in the Pacific earlier than planned on Aug. 11 shortly after launching from California's Vandenberg Air Force Base on its second-ever test flight.
The HTV-2 experienced some sort of anomaly, prompting the vehicle's autonomous flight safety system to guide it to a controlled splashdown, according to the Defense Advanced Research Projects Agency (DARPA), which oversaw the flight.

louis · 2016-01-03 19:55:29

Bit unfortunate if we have two separate spacecraft named Falcon...and same goes for the Dragon!

Sadly our spacecraft developers have limited imaginations when it comes to naming.

New Mars Forums

Announcement

#26 2015-11-07 21:34:23

Re: Spaceplane

#27 2015-11-08 09:47:18

Re: Spaceplane

#28 2015-11-08 21:27:14

Re: Spaceplane

#29 2015-11-09 08:59:19

Re: Spaceplane

#30 2015-11-09 13:23:55

Re: Spaceplane

#31 2015-11-09 14:11:35

Re: Spaceplane

#32 2015-11-18 06:34:39

Re: Spaceplane

#33 2015-11-18 07:45:25

Re: Spaceplane

#34 2015-11-18 08:12:56

Re: Spaceplane

#35 2015-11-18 11:23:31

Re: Spaceplane

#36 2015-11-18 12:38:16

Re: Spaceplane

#37 2015-11-18 13:54:31

Re: Spaceplane

#38 2015-11-18 14:14:18

Re: Spaceplane

#39 2015-11-18 22:06:46

Re: Spaceplane

#40 2015-11-18 22:45:08

Re: Spaceplane

#41 2015-11-19 12:32:02

Re: Spaceplane

#42 2015-11-19 13:27:45

Re: Spaceplane

#43 2015-11-20 09:03:40

Re: Spaceplane

#44 2015-11-20 20:37:08

Re: Spaceplane

#45 2015-11-21 11:30:56

Re: Spaceplane

#46 2015-11-21 22:26:45

Re: Spaceplane

#47 2015-11-26 16:21:34

Re: Spaceplane

#48 2015-12-15 17:04:13

Re: Spaceplane

#49 2016-01-03 19:44:20

Re: Spaceplane

#50 2016-01-03 19:55:29

Re: Spaceplane

Board footer