New Mars Forums

Official discussion forum of The Mars Society and MarsNews.com

You are not logged in.

Announcement

Announcement: We've recently made changes to our user database and have removed inactive and spam users. If you can not login, please re-register.

#1 2015-10-23 17:41:32

RobertDyck
Moderator
From: Winnipeg, Canada
Registered: 2002-08-20
Posts: 5,667
Website

Spaceplane

A question from " Martian Politics and Economy" - "Corporate Government":

martienne wrote:

Tom's suggestion that people should pay their own trip to Mars is crazy. Unless something very revolutionary happens, we are talking about hundreds of millions of dollars in todays value, for one ticket!

As Tom said, one solution is what SpaceX is currently working on: Mars Colonial Transport. That idea is a conventional rocket, but stages are reusable. Elon Musk wants to drop the ticket price to $500,000. Robert Zubrin had stated in his book "The Case for Mars" published 1998, that if you take the price Pilgrim's paid for the Mayflower and adjust it to today's currency, it's somewhere around $300,000 to $500,000 each. I guess Elon read that. Elon is trying to make it happen.

Since this deals with transport to space, I keep this separate. You want something revolutionary? I have posted this idea before, but we are rehashing how to get there. My idea is a multi-engine nuclear space-plane to go from Earth surface to Low Earth Orbit. Use a space station as a bus terminal, initially ISS but once serious colonization begins we'll need a much bigger one, and a station dedicated for that purpose, not a research station. Then a reusable interplanetary ship that goes from Low Earth orbit to high Mars orbit, and back. Using aerocapture at each planet, and refuelling before departing again. And a reusable Mars shuttle; I originally said the Mars shuttle would be based on DC-XA, but SpaceX is now doing work on a reusable rocket, so base it on Falcon 9 reusable.

The radical thing here is the space-plane. I would use a lifting body, something similar to HL-20 and DreamChaser. Engine for take-off and landing would use conventional jet fuel. The engine would be based on the engine for SR-71 Blackbird, able to take off and fly a maximum of mach 3.6. That engine had multiple modes of operation, changing air flow as speed increased, but it did not have full-bypass to convert the turbojet engine into a full ram-jet. I suggest a new engine that would. The purpose of this engine is to take off and reach mach 5. Here's the air flow of the J58 engine, used on SR-71:
800px-SR71_J58_Engine_Airflow_Patterns.svg.png

Once the aircraft reaches mach 5, then a SCRAM jet would take over. GW Johnson is an actual jet engine engineer, and critiqued my idea often. He has very useful input. Based on his feedback, I suggested a separate SCRAM jet engine. I suspect Lockheed-Martin read our debates, because they released this idea:
1400599103878.jpg
Notice airflow for SCRAM jet operation completely bypasses the turbojet. But I want to take it even further. I want a nuclear SCRAM jet. The Air Force worked on a nuclear engine for the B-36, but had several design problems in 1955-1957. They did build the NB-36H, but it was only a proof-of-concept aircraft, the nuclear engine didn't provide propulsion. Many engineers have said that it could be done with today's technology. Well, I want to do it. But not as a subsonic bomber, I want a hypersonic space shuttle. Use the same glaze as the black heat shield tiles of the Space Shuttle, but not for heat shield, use that glaze for radiator fins of the reactor inside the SCRAM jet engine itself. Technically there wouldn't be any combustion, so not exactly "Supersonic Combustion RAM" jet, but still supersonic air flow. Ideally this engine should propel the aircraft from mach 5 to mach 20 in the upper stratosphere.

Then yet another engine: solid core nuclear thermal rocket. NERVA was such an engine, but the engine mass was too great. The air force worked on a much lower mass engine called Timberwind. It was developed as part of Ronald Regan's Strategic Defence Initiative (SDI), known to the media as Star Wars. Timberwind had a number of innovations, the greatest was to use Americium-242m instead of Uranium. That permits reactor critical mass with much lower mass of fissile material. You don't need to carry multiple years of nuclear fuel when the rocket will only operate for 20 minutes. Timberwind was designed for a launch once missile, so wasn't designed to be restartable. NERVA was. Incorporating some NERVA technology with Timberwind innovations would provide an engine that's actually practical. And this isomer can be used for the nuclear SCRAM jet engine as well.

This space-plane must be small. An aircraft skin has to enclose all propellant tanks, and a heat shield for re-entry must protect the skin. To keep the propellant tank small, use steam distilled water for the NTR rocket, not liquid hydrogen. That has the advantage that you don't have to deal with cryogenic liquid hydrogen. But most importantly, density of LH2 is so low that tank size is huge. Notice the Saturn V used kerosene/LOX for the first stage, and LH2/LOX for second and third stages. Actually it was RP-1, but that's highly refined kerosene. Atlas V uses RP-1/LOX for its first stage, and LH2/LOX for the upper stage. It works. I am suggesting kerosene based jet propellant for the turbojet engine, then a nuclear SCRAM jet that uses air as propellant, so no carried propellant at all. The engine for the final push into space with a solid core NTR, with water as propellant. This means all propellants will be storable, meaning liquid at room temperature. The main NTR rocket could be used for orbital manoeuvres, and RCS thrusters could use the same propellant mix as the Shuttle or Orion: that is MMH/N2O4. Thruster quads for the Apollo service module used that same propellant mix, although its main engine used something else.

Windshield of this space-plane would be alumino-oxynitride (AlNO), which is a transparent aluminum ceramic. That was developed in the 1980s under contract for the US army for windows of tanks. It's intended to stop bullets. This would be highly resistant to micrometeorite strikes, so wouldn't suffer from pitting. Space Shuttle had to have its windshield hand ground before each flight. This would prevent that.
(The movie "Jurassic World" claimed their "hamster balls" were made of "aluminum oxynitride glass". Sounds like script writers are reading this forum too.)

NASA Ames Research Centre developed an advanced thermal blanket, never used on Shuttle but was used on X-37. DurAFRSI. It's a quilt of Nextel-440 fabric, with Saffil fibre fill, the surface has a metal wire mesh of Inconel-617 sewn on with Nextel-440 threads, and finally foil of Inconel-617 is brazed onto the wire mesh. This provides a smooth metal skin for the aircraft. It's more durable than the AFRSI quilt used for white areas of the Shuttle. Nextel and Saffil are both ceramic fibres; this quilt can handle higher temperatures than AFRSI, but not quite as high as the black tiles on Shuttle.

One final technology is something called "plasma magic". Most engineers don't like the word "magic", so someone should come up with a better name for this. It uses a microwave laser aimed ahead of the aircraft. It turns the air the aircraft is about to fly through into a weak plasma. It reduces aerodynamic drag. Detail study shows it only works on poorly streamlined aircraft. For this reason many engineers have dismissed it. They argue why bother with this when a well designed aircraft can fly just as well without it. But notice the key point: it reduces drag for poorly streamlined aircraft. A lifting body is designed to be an optimal re-entry vehicle, it isn't well streamlined. It does fly hypersonic, but has a lot of drag. For an entry vehicle, that's Ok because you want to slow down for landing. But for launch, it's a problem. Just compare the body shape of X-43A (designed for mach 10 flight) with HL-20 (designed for re-entry).
X-43A:
220px-X43a2_nasa_scramjet.jpg
HL-20:
260px-HL20_mockup.jpeg 220px-HL-20_wax_model.jpeg
HL-20 is much more squat. So a lifting body would benefit from "plasma magic" during launch. Or since this would use jet engines, perhaps I should say during atmospheric flight.

Air flow for hypersonic jet engine intake requires a shock wave initiated far head of the intake. That's the main reason the nose of X-43A and SR-72 are both so long. But could a microwave laser (maser) initiate a shock wave ahead of the aircraft to create a virtual nose? This would allow a more squat aircraft, like HL-20. A lifting body instead of wings is lighter for the same volume of propellant tanks. And smaller aircraft skin surface means less surface to cover with heat shield, so lighter heat shield. So perhaps this would have under-wing engines like SR-72, but lifting body like HL-20, and instead of fixed wings like HL-20 (based on BOR-4), it would have variable-dihedral wings like Spiral.
SR-72:
300px-Lockheed_Martin_SR-72_concept.png

Spiral, also known as MiG-105:
photore3.gif epos6m.jpg epos2.gif model11.gif

Last edited by RobertDyck (2015-10-30 01:12:59)

Offline

#2 2015-10-23 18:40:44

RobertDyck
Moderator
From: Winnipeg, Canada
Registered: 2002-08-20
Posts: 5,667
Website

Re: Spaceplane

Here's an interesting idea. NERVA engineers have suggested a version that could use the reactor to generate electric power as well as thrust. If we have substantial electric power available, could we use a electric propulsion for manoeuvring thrusters? The MR-510 arcjet thruster was formerly manufactured by General Dynamics, now by Aerojet Rocketdyne, produces 570 to 600 second Isp (Beginning Of Life). Thrust is only 258-222 mN (mlbf), power consumption 2 kW.

A more interesting option is ESEX (Electric Propulsion Space Experiment). It uses ammonia propellant. Isp=800s, thrust 2.00 N (0.40 lbf), 26 kW.

Each engine of an Apollo CSM thruster quad was an R-4D thruster, producing thrust = 440 N (100 lbf), Isp=312s. Shuttle primary RCS thrusters produced thrust 3.87 kN (870 lbf). I don't know what Isp was for RCS, but OMS had Isp=316s.

So, would an ammonia arcjet thruster produce 440 N (100 lbf), consume 5,720 MW? That's 5.7 megawatts! Oops. It would be nice to have propellant that's non-toxic, non-hypergolic, liquid at room temperature, cheap and easy to obtain. And specific impulse over double that of current RCS thrusters. But that power consumption is not reasonable.

Offline

#3 2015-10-24 17:31:47

Tom Kalbfus
Banned
Registered: 2006-08-16
Posts: 4,401

Re: Spaceplane

Sounds familiar, doesn't it? I think I've seen this movie before.

I have posted this idea before, but we are rehashing how to get get there. My idea is a multi-engine nuclear space-plane to go from Earth surface to Low Earth Orbit.

15ae3379c7823c1281d5383c6cca7c5a.jpg

Use a space station as a bus terminal, initially ISS but once serious colonization begins we'll need a much bigger one, and a station dedicated for that purpose, not a research station.

20110925-011821.jpg

Then a reusable interplanetary ship that goes from Low Earth orbit to high Mars orbit, and back. Using aerocapture at each planet, and refuelling before departing again.

aries1b.jpg
2001.jpg

Offline

#4 2015-10-24 23:27:00

RobertDyck
Moderator
From: Winnipeg, Canada
Registered: 2002-08-20
Posts: 5,667
Website

Re: Spaceplane

You have a real obsession with that movie. NASA initially tried to design Apollo to land the CSM on the Moon, but found it was so heavy that even a Saturn V couldn't lift it. They had a real problem until one engineer showed them that splitting the vehicle into Lunar Module and mothership would dramatically reduce total mass. This architecture does the same. If you want reusable, then you need one that will go from Earth surface to LEO (shuttle), and a completely different vehicle for interplanetary travel. And a different vehicle again for the Mars shuttle.

An Earth shuttle will never look like Aries Ib. I looked up the name of that thing. The engine looks roughly like an aerospike, but not enough to work. It has a tiny compound delta wing. In fact, I wouldn't even call it that; it's a trapezoidal or diamond wing with chine. A real-life fighter jet to use a trapezoidal wing was the F-104 starfighter. And it was contemporary with that film. The F-104 had various nicknames:

Wikipedia wrote:

"The Flying Coffin" from the translation of the common German public name of Fliegender Sarg. The F-104 was also called Witwenmacher ("Widowmaker"), or Erdnagel ("ground nail") – the official military term for a tent peg. The Pakistani AF name was Badmash ("Hooligan"), while among Italian pilots its spiky design earned it the nickname Spillone ("Hatpin"), along with Bara volante ("Flying coffin"). In the Canadian Forces, the aircraft were sometimes referred to, in jest, as the Lawn Dart, the Aluminium Death Tube, and the Flying Phallus.

And Discovery is way too big. It's not supposed to be a cruise ship in space, or a navy battleship. The Discovery from that movie was supposedly 5,440 tonnes. An Oliver Hazard Perry-class frigate has displacement of 4,200 metric tonnes, crew compliment of 176.

For a science/exploration mission, think space Winnebago. If you want a movie reference, think Space Balls.
freezeframe022b.jpg
I've said before, the Mars Direct habitat has an upper floor with as much floor area as a 60-foot Class A motorhome with slide-outs. The lower floor will be filled with landing rockets, propellant tanks tanks, RCS thrusters, airlock, life support equipment, battery the size of a bar fridge to power life support on Mars at night (no light for solar arrays), and garage for the rover. That garage will be the size of a single car garage, and every corner around the rover packed with surface science equipment and an inflatable greenhouse. Once on Mars all that will be deployed, leaving an empty single car garage for lab/workshop. And once erected and inflated, the greenhouse will be the size of a double car garage. But in transit to Mars, none of that space will be usable. I posted this image before; this is the hab from the movie "Mission to Mars".
2011-review-mars.jpg

I also said before that when the Shuttle was operational, before the Challenger accident, people thought this was the beginning of commercial space flight. One airline pilot's union started classes to teach their pilots how to fly the Shuttle. They were worried that conventional airliners would be obsolete, so pilots had to learn to fly the Shuttle for job security. At that time I drew a "back of the napkin" design for a passenger module. This would fit in the Shuttle cargo bay. Two decks; the upper deck would have 13 rows of seats, with one isle and 2 seats on each side of the isle. No window, but an LCD display on the back of each seat. Each seat the size of an economy class airline seat, with overhead storage, and storage under your own seat instead of the seat in front. Two decks; the lower floor would have just as many seats. The lower deck would not have overhead storage, instead seats would be on a raised platform due to curvature of the cylindrical module. There would be storage under the platform, accessible from side a door at the isle. So each passenger of the lower deck would have storage under his/her own seat, and as much storage under that platform as an overhead bin. Obviously no luggage compartment, only two pieces of carry-on. One carry-on under your seat, the other in the overhead bin. The mid-deck of the Shuttle has one washroom, this module would have one on each deck. And the module would have an integrated airlock with APAS docking adapter, for docking with ISS. The mid-deck has 3 seats, stewardesses would sit there. And the mid-deck has food storage and kitchen facilities. The flight deck has 4 seats: pilot and co-pilot, then two back seats. Either two more stewardesses, or mission specialists. Life support for this passenger module would be tucked in spaces between ribs of the cargo bay. So this would carry 104 passengers.

But unfortunately Challenger blew up. I think people don't realize just how much that traumatized everyone. That killed all expectations of the Shuttle becoming "Ares Ib" from the movie "2001: A Space Odyssey".

But this discussion is about a radical new technology. A nuclear jet engine uses air as propellant, so the vehicle carries no propellant at all. You can't get any better. I had considered a Rocket Based Combined Cycle (RBCC) engine, but GW Johnson convinced me that too many modes of operation would be impractical. Despite the fact Boeing received funding from the US military to work on one. Besides, airports may have an issue with nuclear engines. So this craft would use a turbine engine to get up to speed where a SCRAM jet can ignite. The turbojet engine based on the one for SR-71 Blackbird, because that was a 1960s aircraft that could fly at mach 3.6. Full bypass of the compressor, operating as pure RAM jet, should allow operation to mach 5.

Offline

#5 2015-10-25 10:08:06

GW Johnson
Member
From: McGregor, Texas USA
Registered: 2011-12-04
Posts: 3,492
Website

Re: Spaceplane

As RobertDyck noted above a "real" air turbo ramjet engine can close off the air path through the turbomachinery core.  This requires 100% bypass from ahead of the compressor face direct to the afterburner duct.  That way,  the AB duct really can be operated as a ramjet,  with potential speed capability to the Mach 5 or 6 range. 

The J-58's that pushed the SR-71 were not that description.  The air bypass came off stage 4 of the compressor,  and was only 0 to 25% max of the engine airflow.  There were 6 tubes that carried it aft to the AB duct.  All or most of the airflow still went through the turbomachinery core,  with temperature limits imposed by both the turbine blading and the compressor blading.  Not to mention the massflow matching difficulties from zero to max Mach between engine and inlet. 

I talked to Col Buzz Carpenter,  who flew SR-71's.  He said they never deliberately flew it beyond Mach 3.3,  and when it did exceed speed,  they pulled up and slowed it with energy management within seconds,  to avoid overheat damage.  The bird's most common extreme flight condition was M3.2 to M3.3 at about 85,000 feet.  It almost no climb rate up that high,  a fault of all airbreathing engines of any type and any energy source,  simply because ambient pressure is so low. 

I like the fundamental notion RobertDyck proposes,  for a nuclear-engined spaceplane.  But I'd caution that even a nuclear scramjet will have no climb rate in the thin air.  We tried a chemical scramjet for X-30,  but found you had to fly re-entry in reverse (an extremely-wasteful thing to do) to get any frontal thrust density out of it up high in the thin air. 

We tested a NERVA nuclear thermal rocket under Project Rover in Nevada to the point where we nearly eliminated radioactive debris in its exhaust plume.  The weight of the core was awfully large.  The potential of that design seems to be Isp in the 900-1000 sec range at an engine thrust/weight of about 3.5 to 4.  The Timberwind design addressed the core weight to improve thrust to weight,  at the expense of engine life.  It was never proven,  though,  that it would work without a radioactive exhaust stream.  There's still lots of room for development there.

We also ground-tested a nuclear ramjet engine under Project Pluto in Nevada.  This was a M3 low altitude design for an "eternally-flying" cruise missile.  It suffered greatly from core erosion leading to an extremely radioactive exhaust,  and had a core support structure that operated 10 F away from its meltpoint,  leading to short life (as experimentally tested).  These were direct-connect ground tests at Jackass Flats,  the same site where the NERVA rocket facility was also located.  The nuclear ramjet cruise missile was cancelled because of these tremendous engineering difficulties to be overcome,  the advent of new-at-that-time ICBM's,  and because the plume radiation and the strong trailing shock wave would have killed more people on the way to target than the megaton-range warhead could have killed at the target. 

Those considerable engineering problems with the nuclear airbreather (which would be horrendously greater with supersonic flow inside the engine) vs the nuclear thermal rocket suggest that maybe you just want the nuclear rocket pushing your spaceplane.  That would be a nearer term development with a lot fewer intractable engineering problems to solve.  That's not to say you cannot build a nuclear scramjet;  it will simply be extremely difficult,  expensive,  and time-consuming on a scale of decades.

I kind of doubt that solid core nuclear rocket could do the spaceplane job single stage (except at an uneconomically-low payload fraction) because of "only" 1000 sec Isp.  A gas core design might work a lot better at around 1500-2000 sec.  I think there was some lab bench work done toward a "nuclear lightbulb" engine that would contain its gas core completely,  but no real tests were ever done. Predictions were for Isp in the 1300-1500 sec ballpark,  as I recall.  The open-cycle gas core had higher Isp potential,  but inherently had a somewhat-radioactive exhaust.   

To get to orbit from level supersonic flight at high altitude requires a pullup and a lot of frontal thrust density that no airbreather has.  I think you might be better off pulling up and using a rocket at that point,  nuclear or otherwise.  For single stage,  it had better be nuclear.  If chemical could do it at high payload fraction,  we'd have already done it. 

GW

Last edited by GW Johnson (2015-10-25 10:43:24)


GW Johnson
McGregor,  Texas

"There is nothing as expensive as a dead crew,  especially one dead from a bad management decision"

Offline

#6 2015-10-25 11:00:45

SpaceNut
Administrator
From: New Hampshire
Registered: 2004-07-22
Posts: 15,037

Re: Spaceplane

Some tools for non expert scam/Ram jet engine to make use and that includes me as I know nothing.....

[url=https://spaceflightsystems.grc.nasa.gov/education/rocket/Lessons/machC_ans.html]Beginner's Guide to Rockets
Mach Number[/url]

So what is a Mach number and why it changes with altitude https://en.wikipedia.org/wiki/Mach_number which has to due with https://en.wikipedia.org/wiki/Density_of_air and when you actor in the altitude you now have https://en.wikipedia.org/wiki/Density_altitude which then we apply to what we need to achieve https://en.wikipedia.org/wiki/Orbital_speed

I think that the x series crafts have been taking some hydrogen onboard for the boost in speed while carrying no oxidizer as they are getting that from the atmosphere.

Offline

#7 2015-10-25 11:21:20

Tom Kalbfus
Banned
Registered: 2006-08-16
Posts: 4,401

Re: Spaceplane

GW Johnson wrote:

As RobertDyck noted above a "real" air turbo ramjet engine can close off the air path through the turbomachinery core.  This requires 100% bypass from ahead of the compressor face direct to the afterburner duct.  That way,  the AB duct really can be operated as a ramjet,  with potential speed capability to the Mach 5 or 6 range. 

The J-58's that pushed the SR-71 were not that description.  The air bypass came off stage 4 of the compressor,  and was only 0 to 25% max of the engine airflow.  There were 6 tubes that carried it aft to the AB duct.  All or most of the airflow still went through the turbomachinery core,  with temperature limits imposed by both the turbine blading and the compressor blading.  Not to mention the massflow matching difficulties from zero to max Mach between engine and inlet. 

I talked to Col Buzz Carpenter,  who flew SR-71's.  He said they never deliberately flew it beyond Mach 3.3,  and when it did exceed speed,  they pulled up and slowed it with energy management within seconds,  to avoid overheat damage.  The bird's most common extreme flight condition was M3.2 to M3.3 at about 85,000 feet.  It almost no climb rate up that high,  a fault of all airbreathing engines of any type and any energy source,  simply because ambient pressure is so low. 

I like the fundamental notion RobertDyck proposes,  for a nuclear-engined spaceplane.  But I'd caution that even a nuclear scramjet will have no climb rate in the thin air.  We tried a chemical scramjet for X-30,  but found you had to fly re-entry in reverse (an extremely-wasteful thing to do) to get any frontal thrust density out of it up high in the thin air. 

We tested a NERVA nuclear thermal rocket under Project Rover in Nevada to the point where we nearly eliminated radioactive debris in its exhaust plume.  The weight of the core was awfully large.  The potential of that design seems to be Isp in the 900-1000 sec range at an engine thrust/weight of about 3.5 to 4.  The Timberwind design addressed the core weight to improve thrust to weight,  at the expense of engine life.  It was never proven,  though,  that it would work without a radioactive exhaust stream.  There's still lots of room for development there.

We also ground-tested a nuclear ramjet engine under Project Pluto in Nevada.  This was a M3 low altitude design for an "eternally-flying" cruise missile.  It suffered greatly from core erosion leading to an extremely radioactive exhaust,  and had a core support structure that operated 10 F away from its meltpoint,  leading to short life (as experimentally tested).  These were direct-connect ground tests at Jackass Flats,  the same site where the NERVA rocket facility was also located.  The nuclear ramjet cruise missile was cancelled because of these tremendous engineering difficulties to be overcome,  the advent of new-at-that-time ICBM's,  and because the plume radiation and the strong trailing shock wave would have killed more people on the way to target than the megaton-range warhead could have killed at the target. 

Those considerable engineering problems with the nuclear airbreather (which would be horrendously greater with supersonic flow inside the engine) vs the nuclear thermal rocket suggest that maybe you just want the nuclear rocket pushing your spaceplane.  That would be a nearer term development with a lot fewer intractable engineering problems to solve.  That's not to say you cannot build a nuclear scramjet;  it will simply be extremely difficult,  expensive,  and time-consuming on a scale of decades.

I kind of doubt that solid core nuclear rocket could do the spaceplane job single stage (except at an uneconomically-low payload fraction) because of "only" 1000 sec Isp.  A gas core design might work a lot better at around 1500-2000 sec.  I think there was some lab bench work done toward a "nuclear lightbulb" engine that would contain its gas core completely,  but no real tests were ever done. Predictions were for Isp in the 1300-1500 sec ballpark,  as I recall.  The open-cycle gas core had higher Isp potential,  but inherently had a somewhat-radioactive exhaust.   

To get to orbit from level supersonic flight at high altitude requires a pullup and a lot of frontal thrust density that no airbreather has.  I think you might be better off pulling up and using a rocket at that point,  nuclear or otherwise.  For single stage,  it had better be nuclear.  If chemical could do it at high payload fraction,  we'd have already done it. 

GW

Suppose Lockheed comes through with a working miniature fusion reactor, could that be used as part of a nuclear rocket or a scramjet to get us into orbit?

Offline

#8 2015-11-03 06:14:20

Antius
Member
From: Cumbria, UK
Registered: 2007-05-22
Posts: 1,003

Re: Spaceplane

I am always stuck for time when it comes to posting on these boards.

As a nuclear engineer who has worked extensively on thermal hydraulic analysis, I think the idea of an earth-launch nuclear scram-jet is unworkable.  As a direct cycle concept the cladding must be thick enough to prevent any fission product leakage.  At the same time, it must accomodate huge heat transfer rates and work at very high temperatures to achieve a high enough ISP.  A nuclear-assisted rocket (i.e. with a reactor preheating cryogenic fuels) might be more workable as you can then exploit huge thermal gradients and achieve high heat transfer rates at achievable clad temperatures.  Maybe that could boost the performance of an H2/O2 SSTO.  But the accident risk will always play on peoples minds.

On other worlds, where gravity is lower and there is less concern over radioactive contamination (i.e. Titan and the gas giants) the conclusions might be different, but I cannot see how this would work on Earth politically.  Keep in mind as well that the cost of nuclear ownership adds significantly to any propulsion project here on Earth.

Offline

#9 2015-11-03 10:03:27

RobertDyck
Moderator
From: Winnipeg, Canada
Registered: 2002-08-20
Posts: 5,667
Website

Re: Spaceplane

Forget politics, and forget government extortion costs. Focus on the technology. Project Pluto worked, but fission products were permitted to spew out exhaust. I suggested a heat sink with fins that have the same glaze as black heat shield tiles on Shuttle. The reason is rapid heat transfer into hypersonic air stream.

I talked to one nuclear physicist working on the Maple reactor in Canada. He mentioned a problem with heat transfer. I mentioned computer heat sinks use aluminum to transfer heat, but he said aluminum poisons the nuclear reaction. Ok. What do you use as a reactor safe heat sink?

Offline

#10 2015-11-03 13:38:40

Antius
Member
From: Cumbria, UK
Registered: 2007-05-22
Posts: 1,003

Re: Spaceplane

The heat transfer coefficient will be some function of Reynolds number which is in turn a function of flow velocity.  So it will increase as a function of velocity.  Not sure how things would work at super sonic speeds.  I will take a look in my heat transfer book when I get the chance.

Aluminium has a low neutron capture cross-section so won't poison the reaction.  The UK used it as cladding in early natural uranium reactors which are highly intolerant of any neutron leakage.  It has excellent thermal conductivity, but loses half its strength at 300C.

Offline

#11 2015-11-03 15:07:07

GW Johnson
Member
From: McGregor, Texas USA
Registered: 2011-12-04
Posts: 3,492
Website

Re: Spaceplane

I repeat from post 5 above:

"I like the fundamental notion RobertDyck proposes,  for a nuclear-engined spaceplane.  But I'd caution that even a nuclear scramjet will have no climb rate in the thin air.  We tried a chemical scramjet for X-30,  but found you had to fly re-entry in reverse (an extremely-wasteful thing to do) to get any frontal thrust density out of it up high in the thin air."


So,  my first question is,  what is the point of flying on a scramjet hypersonic (M>5) down in the air?  That's reentry in reverse at q ~ 2000-5000 psf.  Wasteful in the extreme. 

My second question is,  what's the point of a scramjet (or any other airbreather) way up the thin air (100,000-200,000 feet) when they haven't the thrust to produce a nonnegative climb rate?

A reasonably-efficient trajectory has you leaving the air that is thick enough for real airbreather thrust (80,000-110,000 feet) at no more than about Mach 5.  Your speed is closer to about Mach 2 with conventional vertically-launched rockets.  Who needs a scramjet for this?  Nobody.

Mach 5 needs only a high-speed design ramjet (with external shock features on its inlet) that lights at about Mach 1.8-to-2.  Mach 2 can use a low-speed design with a simple pitot normal-shock inlet that lights high subsonic.  Ramjet has been a ready-to-use technology for decades.  It powered Talos and Bomarc and Sea Dart,  and the SA-6 and SA-4,  and it now powers Sunburn and Yakhonts/Brahmos,  plus another anti-shipping missile who name I can never recall. 

Why not put both kinds of engines on your vehicle,  and burn them both sequentially and in parallel?  Take off on rockets,  light the ramjets when feasible and take some or all of your thrust from them for a while,  the relight the rockets and burn rocket + ramjet in parallel as the air thins while you climb.  From about M5 to 6 at 120,000+ feet shut the airbreathers down and continue flying rocket alone.  The payload fraction is low but nonzero this way with chemical propellants,  a bit larger with nuclear engines,  but still not "Buck Rogers" style. 

Project Pluto was a nuclear thermal ramjet intended to fly at Mach 3.  It was successfully tested direct-connect on the ground,  but nothing ever beyond that.  Which means it was never tested with a real supersonic inlet. 

Unlike NERVA,  they never really solved the core erosion and core structural support erosion problems.  I think they were using materials like Inconel for the supports,  but memory may be faulty here.  There was virtually no strength left at only about 10 F from melting,  and the oxygen in the air was very reactive with the materials. 

The support and heat protection problems are easier in a subsonic reactor chamber compared to a supersonic one.  While supersonic film coefficients are higher,  everything else including simple air loads pushing the reactor off its supports is much worse. 

Besides,  the lowest feasible takeover speed for scramjet is Mach 4.  If you leave usable air at Mach 6,  you only got about 2000 ft/sec delta-vee out of it.  Like I said,  with supersonic ramjet,  takeover is closer to Mach 2,  for twice the delta vee at about 4000 ft/sec out of it. 

Why bother with scramjet when ramjet works better and is already technologically ready?  Scramjet certainly is not ready for application yet.  X-43A was 2 for 3,  and X-51A was 2 for 4.  Neither accelerated at all in airbreather mode. 

GW

Last edited by GW Johnson (2015-11-03 15:12:33)


GW Johnson
McGregor,  Texas

"There is nothing as expensive as a dead crew,  especially one dead from a bad management decision"

Offline

#12 2015-11-03 15:25:10

RobertDyck
Moderator
From: Winnipeg, Canada
Registered: 2002-08-20
Posts: 5,667
Website

Re: Spaceplane

GW Johnson wrote:

We tried a chemical scramjet for X-30,  but found you had to fly re-entry in reverse (an extremely-wasteful thing to do) to get any frontal thrust density out of it up high in the thin air.

We have no idea what you mean. We ignored this because, what is it?

You said you like the idea, but then question part of the basic premise. You don't want to fire a rocket through air. Operating through air with oxygen, but not using that oxygen, instead carry oxygen in tanks? That's wasteful. A jet engine is not wasteful. So you want to gain as much altitude and speed as possible before switching to a rocket motor. The Shuttle re-entered atmosphere at mach 25. Get as close to that as possible before the final push into space.

You talk about a RAM jet engine operating to mach 5. Yes, that's useful and part of it. But remember the goal is mach 25, so mach 5 is only a small fraction.

A nuclear jet engine operates with no carried propellant what so ever. In that sense, Isp is infinite.

GW Johnson wrote:

Scramjet certainly is not ready for application yet.

The key word there is "yet". So let's fix that. After all, I'm interested in developing new things. If a thing is already finished, there's nothing for me to do. So anything that's finished is useless.

Offline

#13 2015-11-03 15:48:31

GW Johnson
Member
From: McGregor, Texas USA
Registered: 2011-12-04
Posts: 3,492
Website

Re: Spaceplane

Hi RobertDyck:

My own personal awareness has scramjet work being done since about the early 1960's.  To the best of my knowledge,  the only successful burning tests so far were X-43A and X-51A. 

NASA'a X-43A used hydrogen fuel,  did nothing to balance or utilize waste heat,  and did boost-to-speed with a rocket for about a 3 second burn at Mach 7 and another just short of Mach 10.  Neither test accelerated,  showing zero thrust margin over drag,  which is a zero climb rate capability. 

USAF's X-51A was a more remarkable feat in my opinion,  because it used a liquid hydrocarbon fuel (JP-7 I believe).  They got a 3 minute burn both times,  on a boost-to-speed profile at only Mach 5,  both times.  They had no acceleration in the scramjet.  But,  they did try to utilize some of the waste heat. 

Back in 1980,  one (of several) bird I worked on was ASALM-PTV,  a demo prototype for a supersonic cruise missile.  It was a plain kerosene ramjet,  with Mach 2.5 takeover and Mach 4 cruise at 80,000 feet.  It did a Mach 5 dive onto target. 

We tested it 7 times in flight,  with 6.5 of those 100% successful,  way back then.  The first flight was only half-successful,  since it did not do what we intended.  That one had a throttle runaway problem,  and left the range at low altitude at Mach 6.  Took us 3 or 4 days to find it,  stuck like a big dart in a farmer's field,  10 miles off the base where we tested. 

As for your question about reentry in reverse:  the drag is very high when you do that,  precisely because it is a high dynamic pressure trajectory.  But,  that's also the only way to get high thrust out of an airbreather of any type (high dynamic pressure). 

Regardless,  the hydrogen scramjet that they were looking at on X-30 still could not generate thrust equal to drag,  much less thrust greater than drag,  on that class of trajectories.  Your climb angle relates to (thrust - drag)/weight.  Thrust must exceed drag or you cannot climb.  What would be the point of thrusting to Mach 25 down in the air,  when you've lost most of it to drag by the time you leave the air?  That's what happens on reentry in reverse.

GW


GW Johnson
McGregor,  Texas

"There is nothing as expensive as a dead crew,  especially one dead from a bad management decision"

Offline

#14 2015-11-03 16:43:16

RobertDyck
Moderator
From: Winnipeg, Canada
Registered: 2002-08-20
Posts: 5,667
Website

Re: Spaceplane

One major problem with SCRAM jets, is that compression heating is so high that burning fuel does not increase temperature significantly. Gas expansion is directly proportional to temperature above absolute zero, in metric measured in Kelvin. I'm sure you know that, I'm stating the obvious for those reading our discussion. But a nuclear engine can bypass that, can increase gas temperature inside the engine much hotter. The only problem then is operating the reactor and engine at extreme temperature.

Offline

#15 2015-11-03 16:50:45

Tom Kalbfus
Banned
Registered: 2006-08-16
Posts: 4,401

Re: Spaceplane

RobertDyck wrote:

One major problem with SCRAM jets, is that compression heating is so high that burning fuel does not increase temperature significantly. Gas expansion is directly proportional to temperature above absolute zero, in metric measured in Kelvin. I'm sure you know that, I'm stating the obvious for those reading our discussion. But a nuclear engine can bypass that, can increase gas temperature inside the engine much hotter. The only problem then is operating the reactor and engine at extreme temperature.

One little detail, what if it crashes? Lets say we built a fleet of nuclear scramjets and lets say we used them to launch nuclear waste from commercial reactors into space. Would you think that was a good idea?

Offline

#16 2015-11-03 17:29:09

RobertDyck
Moderator
From: Winnipeg, Canada
Registered: 2002-08-20
Posts: 5,667
Website

Re: Spaceplane

Tom Kalbfus wrote:

One little detail, what if it crashes? Lets say we built a fleet of nuclear scramjets and lets say we used them to launch nuclear waste from commercial reactors into space. Would you think that was a good idea?

You don't launch waste into space. That's just dumb. Space launch is far too expensive for waste disposal. First, reprocess. Remove fission fragments from unspent fuel, so the fuel can be reused. Fission fragments will decay. And why not use heat from decaying fission fragments for something?

Offline

#17 2015-11-04 06:05:31

Antius
Member
From: Cumbria, UK
Registered: 2007-05-22
Posts: 1,003

Re: Spaceplane

I would suggest the use of ducted fuel assemblies (i.e. cylindrical tubes in a solid block) through which the supersonic airstream in passes.  This is the only arrangement capable of standing up to vibration from the turbulent air stream which would shake normal fuel assemblies apart.  Additionally, the forces on the duct cladding are entirely compressive and you don’t need to worry about loss of tensile cladding strength at very high temperatures.

Finding materials for the duct lining is a real challenge, as metals would be subject to corrosion in the super-heated airstream and most would melt at the temperatures that you need to be working at (~2000C).  Stainless steel melts at 1400-1600C.  Zirconium melts at 2125C, but corrodes terribly at temperatures above 1000C.  Platinum group metals are corrosion resistant and have high enough melting points, but are expensive and generally have high neutron cross-section.  Some sort of mechanical alloy with titanium oxide embedded within a platinum group amalgam might provide a good protection against erosion (sublimation) at these temperatures and minimise the amount of platinum group metals in the engine.  If you can avoid all unnecessary moderation and keep the neutron spectrum hard, then your engine will be more tolerant of neutron absorbing materials.  In gas cooled fast reactors, the moderating effect of the gas coolant is generally quite modest even for low atomic number coolants, because the density of the coolant is low.  Hence a GCFR has a harder neutron spectrum than a sodium fast reactor, even though the former uses helium or CO2 as its coolant.

I did some rough heat transfer calculations for circular a ducted fuel assembly for your proposed air breathing engine.  I assumed that the ‘average’ air temperature in the engine was 1000K (less at the intake, more at the outlet) and that the average clad-coolant temperature difference was 1000K (i.e. a 2000K clad temperature).  For a duct 0.3m wide and 2m long, the heat transfer coefficient works out as 600W/m2K.  For the surface area of the duct and assumed temperature gradients this gives you a maximum thermal power of 1.3MW per duct.  If you clustered these assemblies into a cylindrical reactor, then a 1GW core would have a diameter of 8.3m.  If the ducts have an individual thickness of 0.5cm and are mostly composed of uranium nitride, then the mass of the engine would be 145 tonnes, giving a power density of 6.9kW/kg for the core alone.

One other problem I can see is that air is not a homogenous coolant.  Density tends to vary and moisture content varies.  If you fly your nuclear ramjet through an area of high moisture (i.e. a cloud) the moderating power of the coolant would suddenly increase resulting in a huge power surge, et la Chernobyl.  The easiest way around this problem is to employ a booster stage to bring your ramjet above the cloud layer before you allow air to enter the core.  This would appear to be necessary anyhow, as the ramjet will not be effective until its airspeed is well above the speed of sound.  The booster stage could either be a separate rocket or for a more integrated solution you could blow a stored propellant into your core operating it as a pure rocket until you reach appropriate height and speed.

To avoid a buildup of fission products within your core (which are radioactive, exert pressure on the cladding, generate decay heat and are neutron absorbing poisons) it would be advantageous to vent the fuel assemblies whilst the reactor is operating.  At typical fuel operating temperatures, most of the products will be gaseous.  You could vent assemblies into a graphite flask which can then be ejected when you reach Earth orbit.

I cannot imagine that using a nuclear fuelled ramjet would save you any money.  For one thing, much if not all of the mass savings avoided by not using chemical fuel would be eaten up by the shielding and the mass of the core.  Handling and maintaining a radioactive aircraft on the ground would add all sorts of additional costs that you would not have with a chemical rocket.  In developing the craft you would essentially be developing an entirely new nuclear reactor.  If that weren’t enough, the burn-up of the nuclear fuel with each launch would be considerable.  You may not get many uses out of your core before you had to replace it.

Offline

#18 2015-11-04 13:01:52

RobertDyck
Moderator
From: Winnipeg, Canada
Registered: 2002-08-20
Posts: 5,667
Website

Re: Spaceplane

Thanks! This sounds like a great start. I couple points.

You said the nuclear ramjet would have to operate above the clouds. Ok. The idea was to use a conventional jet engine for take-off, gain altitude and speed, before starting the nuclear ramjet. I talked about an engine based on the engine for SR-71 Blackbird. Specifications on Wikipedia say the service ceiling for Blackbird was 85,000 feet. That's with a well streamlined aircraft body, not a lifting body, but without the engine improvements we talked about. Thanks to GW Johnson for that. But this means we should have no problem getting above clouds before starting the nuclear engine.

Secondly, I said don't use uranium at all. Use Americium-242m. The reason is you want 100% of the nuclear fuel to be fissile isotopes, not just a fraction. And this isomer has a much smaller critical mass than uranium; smaller than pure ²³³U. (Extended ASCII font has superscripts for ² and ³, but no other digits. So I can only write it correctly for that one isotope.)

Vent fission products into a graphite flask? Brilliant!

Offline

#19 2015-11-04 14:06:37

Antius
Member
From: Cumbria, UK
Registered: 2007-05-22
Posts: 1,003

Re: Spaceplane

I may have made a mistake calculating the Reynolds number.  If so, effective power density may be somewhat higher.

Most of the power density problem stems from the fact that heat transfer requires a series of conductive steps.  First it must transfer from fuel to clad, then through the clad, then across the convective boundary layer into the bulk fluid.  A nuclear light using gaseous fuel sidesteps the need for these rate limiting processes by releasing all energy as ultraviolet light, which then heats the working fluid directly.  You could even 'dope' the incoming airstream to ensure good absorption.  The fuel is uranium hexaflouride gas.  Question: does americium have a gaseous halide compound?

As a gaseous fuelled reactor, the lightbulb would be much easier to detoxify after each launch, as the Am/U fluoride will be much heavier than any of the fission product halide.  This could be done in orbit using a small centrifuge.  Decay heat removal should be easy as well as the fused silica lightbulb will radiate heat at temperatures up to 1000C  allowing heat to be removed by liquid sodium lines around the inside of the engine chamber.  Refuelling the lightbulb should involve injecting fresh hex fuel.  If the neutron spectrum is sufficiently hard then it won't be necessary to remove higher actinides.

Last edited by Antius (2015-11-04 14:51:26)

Offline

#20 2015-11-04 15:10:44

RobertDyck
Moderator
From: Winnipeg, Canada
Registered: 2002-08-20
Posts: 5,667
Website

Re: Spaceplane

Antius wrote:

Question: does americium have a gaseous halide compound?

Short answer: no. Well, only one. According to a web periodic table, AmF3 melting temperature is 1395°C. No boiling temperature listed. AmF4 melting temperature is unknown. Americium diiodide AmI2 melts at 700°C, Americium triiodide AmI3 melts at 950°C. No boiling point listed for either. AmO2 will decompose at 1000°C, but no melting temperature listed for either AmO or AmO3. No hydrides are listed at all. The only compound with a boiling temperature is Americium trichloride, AmCl3. It melts at 715°C, boiling point 850°C.

Why do you want a gas core nuclear thermal engine? Isn't solid core more safe? If fission fragments are gas, isn't phase change a simple means to separate fuel from fission fragments?

Offline

#21 2015-11-05 05:49:48

Antius
Member
From: Cumbria, UK
Registered: 2007-05-22
Posts: 1,003

Re: Spaceplane

Phase change should be an effective means of separating most of the fragments, though not all (i.e. noble gases).

A rocket engine is a heat transfer nightmare.  Trying to design a solid core nuclear thermal engine that will remain solid at the power density and exhaust temperatures neccesary for Earth-orbit launch, is probably a hopeless task.  In the event of an inflight accdent, it is doubtful that core melt could be avoided and an accident during reentry would burn the core up.

A GCR would appear to allow higher power density and allows fission products to be removed after each use.  On balance it would therefore appear to be safer and would certainly be more economically effective.  The silica reactor vessel should remain stable in flight so long as it is kept under compression - it is certainly invulnerable to oxidation.

As the fuel is gaseous I would also expect the reactor to include strong negative reactivity addition with temperature.  As power surges, the gaseous fuel could be allowed to expand out of the reflected zone of reactor vessel.  Hence, power surges would tend to dampen themselves.  This would make the GCR much more tolerant to changes in the moisture level of the incoming airstream, allowing it to work in airbreathing mode from ground launch.

Offline

#22 2015-11-05 14:15:27

RobertDyck
Moderator
From: Winnipeg, Canada
Registered: 2002-08-20
Posts: 5,667
Website

Re: Spaceplane

Well, I said operate it at high temperature. So how do you separate fuel from fission fragments? But remember, this isn't ground launched. The nuclear jet engine (not rocket) is started at hypersonic speed (mach 5+) and high altitude: well into the stratosphere.

Offline

#23 2015-11-06 11:28:50

GW Johnson
Member
From: McGregor, Texas USA
Registered: 2011-12-04
Posts: 3,492
Website

Re: Spaceplane

Whether you launch it vertically or horizontally,  a winged lifting craft with any sort of airbreathing propulsion will have to accelerate and climb,  pulling over horizontal to reach its highest speed at an altitude where it still has enough frontal thrust density so that thrust exceeds drag by a small margin,  so that it does not take "forever" to reach speed.

With practical and realistic airframe designs,  that pullover altitude is going to be somewhere between about 60,000 and 80,000 ft.  With a really clean design,  maybe 100,000 ft.  That's for M5 +/- design speeds.  Scramjet,  unready-for-application as it is,  is no "savior" allowing operation at 150,000+ ft.  It's subject to the same frontal thrust density limitations as ramjet or gas turbine propulsion.  That is the real lesson of the failed X-30 program. 

From that cruise point,  whatever it is (say M5 or 6 at 80,000 feet),  you have to pull up on rocket thrust to about 45 degree angle to either launch a second stage payload ballistically to orbit,  or you accelerate your entire vehicle ballistically to orbit.  It's ballistic because the air around 100,000 ft + is going to be too thin for lift to equal weight,  and generating lift induces more drag. 

Orbit speed is 25,000 ft/sec,  for eastward low orbits.  That plus a big allowance for drag and gravity losses is the delta-vee required of your propulsion.  Vertical launch rockets typically see a loss allowance of 10% or so.  Flying reentry in reverse,  that loss allowance can exceed 100%,  even 200%,  depending upon what dynamic pressure you choose to fly at. 

For the sake of argument,  say the loss allowance for our spaceplane is 50%,  for 37,000 fps required.  Now,  how much of that can come from your airbreather,  if it is no longer useful once you pull up for the final ballistic ascent?  Turbine is good from 0 to no more than about 3300 fps (air overheat troubles in the inlet) for a 3300 fps delta.  Ramjet with a simple pitot inlet is good from about 700 fps to no more than about 2500 fps,  for an 1800 fps delta.  Ramjet with a spike or ramp supersonic inlet is good from about 2000 fps to no more than about 6000 fps,  for a 4000 fps delta.  The rest of the delta vee has to come from rocket propulsion,  period. 

Now,  it's still hard to predict what scramjet will be able to do when it gets developed.  X-43 functioned once for 3 sec at M10 but had no thrust margin and did not accelerate.  X-51 functioned twice for 3 minutes at M5,  but had no thrust margin and did not accelerate.  OK,  a mature design will be able to accelerate,  and takes over at M4 as a minimum,  about 4000 fps.  Assume for the sake of argument that M10 is indeed feasible,  that 10,000 fps,  for a delta vee in scramjet of 6000 fps. 

Heat transfer problems with a M10 airframe at 80-110,000 ft will make that airframe much heavier than the ramjet airframe that only reaches 6000 fps.  It starts looking like entry heat protection.  That prospect alone makes me wonder whether scramjet will ever be useful for anything but 1-shot missile work. 

What all this tells me that you want a two-stage airplane system,  not a single stage system.  From a payload bay in the aircraftlike first stage,  you launch to orbit a clean projectile second stage.  The carrier plane never has to do anything but reach Mach "xx" at somewhere around 60-100,000 ft.  It must add rocket briefly in order to pull up to launch,  there is absolutely no way around that requirement.  Studies show that carrier launch like that prioritizes as (1) speed,  (2) angle,  and (3) altitude,  in that order. 

To reduce system size and cost,  what you want is (1) max fraction of delta vee to be airbreather,  and (2) min inert weight possible relative to final payload.  Combined cycle is no answer to this problem,  none of those schemes has ever flown.  Trying to build a ramjet that converts to scramjet runs smack into totally-incompatible geometries for both the combustor/nozzle and the inlet.  Combined-cycle rocket with any sort of airbreather also runs smack into totally incompatible geometries for all components.  That's historically been a recipe for thrust/efficiency losses and enormous weight ()and cost) gain. 

To reiterate:  delta vees and max vees for the airbreathers are 3300 fps delta for 3300 fps max gas turbine (if built like the engines on the SR-71 or the XB-70),  3000-4000 fps delta for 5000-6000 fps max for the supersonic-inlet ramjet,  and if we are really,  really lucky,  6000 fps delta for 10,000 fps max with scramjet.  If we're not so lucky,  scramjet will look more like 4000 fps delta for 8000 fps max.  If it ends up looking more like X-51,  it could easily be only 3000 fps delta for 7000 fps max,  or worse. 

Again,  scramjet is no holy grail or saviour for spaceplane propulsion. The likely numbers say that loud and clear,  as just described.  Nuke energy won't make much difference to those outcomes,  only to the Isp and inert weight at which you get those outcomes.  You'll gain both Isp and inert weight,  and probably some radiation in the exhaust. 

Personally,  I think you'd be better off with a nuclear ramjet,  or even stay chemical.  Go for Mach 5 to 6 at 80,000-100,000 feet.  Then stage.  Send only the payload to orbit,  not the whole plane.  That's the lesson of the shuttle program. 

GW

Last edited by GW Johnson (2015-11-06 16:58:15)


GW Johnson
McGregor,  Texas

"There is nothing as expensive as a dead crew,  especially one dead from a bad management decision"

Offline

#24 2015-11-07 08:35:20

Antius
Member
From: Cumbria, UK
Registered: 2007-05-22
Posts: 1,003

Re: Spaceplane

I think GW has hit the nail on the head.  Even a gas turbine driven lower stage with a d-v of 1km/s offers a lot of value in terms of reducing the engineering cost of an upper stage.  A ground launch SSTO is extremely difficult, not just because of the high d-v and all that that means in terms of mass ratio, but because the booster engines firing at sea level have to meet an entirely different set of requirements to upper stage engines firing in a vacuum.  They need high chamber pressure and the poor expansion ratio ruins exhaust velocity to such an extent that high performance fuels offer little advantage.  The engineering complexity of an SSTO would ruin its economics in the same way it did for the shuttle.

Even a poor performance lower stage is a game changer.  The upper stage engines will be firing in a vacuum from start to finish.  This boosts their effective exhaust velocity and the need for a high chamber pressure and expandable exit nozzles disappears.   That means cheap pressure fed engines like the TRW engines instead of space shuttle main engines.  It reduces d-v from 9km/s to 8km/s, which allows the use of LNG as upper stage fuel instead of liquid hydrogen.  It means tanks and structure with reduced insulation made from high strength aluminium or steel alloys instead of carbon fibre.   It allows aerodynamics to be ignored for the upper stage meaning lighter spherical tanks.  A ramjet with 2km/s d-v would be of even more value, but would only really pay for itself if it allowed comparable cost reductions or could improve mass ratio of the upper stage.  The only thing that really matters is how cheaply you can deliver payload when all revenues are divided by all costs.  This is why the Russians still have launch capability and the US does not.

I can remember looking into the possibility of a super-critical water lower stage.  This would have been a spherical carbon-steel tank filled with water heated to 320C at a pressure of 22MPa.  At the bottom of the tank would be a single globe valve, discharging the super-heated water into a fixed steel nozzle where it would expand into steam.  The effective exhaust velocity would have been about 700m/s and the maximum d-v would have been about the same.  The tank would have delivered the upper stage to the stratosphere (10km) with a residual velocity of about 700fps.  As the entire journey of the tank takes place it the lower atmosphere, it would not need engine-gimbaling, simple aerodynamic surfaces could control its attitude, which only needs to remain vertical for the duration of the flight.  The advantage of using such a poor performance lower stage is its extreme simplicity and low cost.  It eliminates most of the engineering complexity that would otherwise have to be built into an SSTO.

I envisaged that the tank would have been used for sea launch and would need to be engineered to be robust enough for repeated impact on the sea surface.  Fresh water would have been used as propellant and the tank would be charged before launch using electric heaters.

The upper stage would burn LNG (methane)-LOX in a single pressure fed engine with exhaust velocity 3500m/s.  Total mass ratio to LEO (for the upper stage) would be about 10.7, which is much easier to achieve without bulky super-insulated hydrogen tanks.

Offline

#25 2015-11-07 11:11:07

GW Johnson
Member
From: McGregor, Texas USA
Registered: 2011-12-04
Posts: 3,492
Website

Re: Spaceplane

Hi Antius:

Thanks for the compliment!

The advantage of plain ramjet over all other airbreathers (including scramjet) is its super light weight.  Like all airbreathers it has the disadvantage of low frontal thrust density,  and needs to fill the aircraft fuselage (or equivalent if podded) to do any good.  This is an inherent effect of the low pressure in the thermodynamic cycle.  Scramjet suffers the same thing,  but is also a bit heavier (at least so far).  Gas turbine has much higher pressures,  but is very much heavier. 

You can piggy-back your second stage payload,  and you can use podded engines if you like,  as long as your terminal stage speed does not exceed about Mach 4-ish (4000 fps) in air thick enough for lift equal to weight.  Faster than that,  you cannot have adjacent structures,  because the shock impingement heating is going to cut things to pieces,  just like on the Mach 6 X-15 flight that replaced the ventral fin with a scramjet test article.  The published “piggyback” concepts for the new XS-1 effort fail in this respect.  Target flight speed for that effort is Mach 10 +/-. 

If your first stage flies faster than about Mach 4, you will have to carry your second stage as an internal store,  and open that bay only after you pull up sharply on rocket and reach thinner air.  (You still have to execute the pull-up for slower piggyback configurations.)  The hardest part about either scenario is supersonic stores release problems that cause collision.  This is what killed an SR-71 crew trying to launch a D-21 drone from a ventral piggyback carry,  flying supersonic in the neighborhood of Mach 2 or so.  Not that it cannot be done,  but it is definitely not a trivial design task. 

Spaceships One and Two get away with “piggyback” because they are well-subsonic.  Shuttle got away with “piggyback” because (1) it left sensible air at only about Mach 2 (thus avoiding shock impingement heating amplification),  and (2) it did the booster separations nearer 200,000 ft where the air was so thin its forces could not exert a lot of force on the discarded boosters.  Even so, they used the thrust of small solid motors to force the boosters to move in the desired directions. 

I have my doubts about HOTOL and its tip-mounted SABRE engines reaching Mach 5 in air thick enough for those engines to still be airbreathing,  precisely because of the shock impingement heating amplification issue.  The engine inlet spike shocks will impinge on the wing leading edges.  Without some heavy duty (and likely sacrificial) heat protection,  those spike shocks could cut the wings apart above about Mach 4.  Unless they come back essentially dead broadside (beyond the technological reach of any reasonably-light structure) their shock impingement-on-wing-LE problem will be insoluble during reentry. 

This didn’t happen to the SR-71 simply because it didn’t fly so fast:  its speed limit was about Mach 3.3,  to avoid all sorts of overheat damage in the engines themselves.  That’s also why its skins could be titanium alloy,  instead of the Inconel-X they had to use on X-15.  Recovery temperatures in the boundary layer are a lot lower at Mach 3.3 than Mach 4,  5,  or 6. 

Using the success of XCOR’s two early rocket airplanes as a guide (plus the excellent work on their soon-to-fly Lynx),  I would suggest the booster plane be powered by rocket at takeoff,  even for horizontal takeoff (it’s the only option for vertical).  Gas turbine is just too heavy for a lot less thrust capability.  Accelerate off the runway to supersonic speed as you gently climb,  and continue accelerating as fast as you can,  to ramjet takeover at about Mach 2 for a high speed ramjet inlet design.  You can even do a bit of parallel rocket-ramjet burn as you take over,  just like Bomarc did. 

Hide these rockets in your wing fillets,  and let the ramjet occupy the bulk of the fuselage.  Package your payload forward of the combustor,  where you can shape the subsonic duct around your payload bay.  Use a nose or chin inlet for the lowest possible airframe drag. 

You may have to use two different ramjet fuels:  something more volatile down at Mach 2 takeover where the inlet air isn’t so hot,  switching to kerosene at higher speeds where the inlet air is hot enough to vaporize it in a practical time and distance.  Your rockets can use the same kerosene,  plus LOX.  You’ll need them again for the staging pull-up maneuver.  Plus,  flight safety demands that you land with enough rocket fuel left over to get at least one “go around”. 

I get nice-looking thrust and impulse characteristics in a fixed geometry from Mach 2-ish takeover all the way to Mach 6,  using a simple 1-D cycle analysis based on adiabatic ideal gas assumptions,  with a Mach 3 shock-on-lip inlet.  Those analysis assumptions start breaking down due to ionization/nonideal gas effects at Mach 5-ish,  and I do not believe the predictions at all above Mach 6,  due to both the nonideal gas effects and increasingly nonadiabatic behavior. 

It’s these effects that force the scramjet designers to use real CFD codes for their predictions.  However,  these codes are only as realistic as the models they contain for all the real-world thermophysical and physical-chemistry effects.  (I get to do all those things outside my simple cycle code with ramjet.)  If those things are not correctly modeled,  then the CFD code predictions are still worthless.  Which outcome explains in part why scramjet is still so very “iffy” and experimental,  even after more than a half century of work. 

Usually somewhere between Mach 5 and Mach 6,  with realistic airframe drag (even with very clean designs),  drag becomes bigger than thrust,  anyway.  Your typical missile with side-mounted inlets is draggy enough to hit thrust equal drag somewhere between Mach 3 and 4.  Meteor (about Mach 3.3) and Kh-31 (Mach 4 at most) are examples.  So was SA-6 (about Mach 2.8).  I notice the Russians have switched to half-circle-type side spike inlets to avoid shock impingement heating on the adjacent missile body that comes with full-round side inlets. 

I’d consider carrying the boost LOX in recoverable drop tanks to cut weight in climb.  I’d also consider adding extra launch thrust (when the weight is highest) by adding small solid rocket motors as “JATO bottles”.   Both solutions are well-proven,  and demonstrated as cheap enough.  They will not dominate system costs the way the giant supposedly-recoverable Shuttle SRB’s did. 

One of the problems with ocean splashdown is that at impact speeds over about 20,  maybe 25,  mph you might just as well hit concrete as hit water.  The forces are just about the same.  Ask any water skier. 

GW

Last edited by GW Johnson (2015-11-07 14:18:58)


GW Johnson
McGregor,  Texas

"There is nothing as expensive as a dead crew,  especially one dead from a bad management decision"

Offline

Board footer

Powered by FluxBB