Section D: Propulsion Concepts List



This section lists propulsion concepts grouped into categories by type.

D.1 Structural Methods

D.1a Static Structures

Static structures have parts which are mostly fixed in relation to each other,
although the structure as a whole may move with respect to the ground.
Large structures are primarily governed in their design by the ratio of

strength to density, or specific strength. Other important properties in

certain cases include stiffness, temperature dependance of properties, and

resistance to decay from the surrounding environment.

Methods of movement on the structure include: (i) Standard elevator: (refer

to standard engineering references for design details) (ii) Inchworm type

winch: A small motor driven trolley pulls a length of cable behind it as it

climbs up the structure. It then hooks the cable to a fixed point on the

structure. The cargo elevator remains attached to the next lower point on the

structure during this time. The elevator then uses an on-board winch to reel

itself up from one attachment point to the next. This type of winch is

useful where continuous attachment track or full length elevator cable would

be too heavy. Requires independant power for winch. (iii) Fluid transfer in

pipes: For example, Dr. Dana Andrews has suggested pumping gas

generated on the Lunar surface up to the Lunar L2 point. A column of

Oxygen at .1 atmosphere at L2, and a temperature of 1000 K (a solar heated

pipe can be used to keep the gas hot) would have a pressure of 2310 atm

(234 MPa) at the bottom. Another approach is to have pumping stations

spaced along the tower.

1 Large Towers

Alternate Names:

Type: C.1b/B.2a (Potential Energy via Mechanical Traction)


Use of advanced aerospace materials makes possible the construction of

towers that are many kilometers tall. Such towers can be used as a

high altitude platform, as a launch platform for a propulsive vehicle, or a

support structure for an accelerator system. Structural design is a major


If a tall structure is being considered, the weight of the tower structure

becomes the driving issue, because it can end up being many times the

weight of the 'payload' the tower is supporting. If the 'payload' is at the

top of the tower, the structure just underneath only has to support the

payload's weight. The next piece of structure below that must support the

payload plus the top bit of structure, so it has to be a little bit beefier

(have a larger cross sectional area). Going down the structure, it has to

get stronger and stronger to support the greater weight above.

To put some numbers to the problem, let us take a plain carbon steel

structure (the type of steel used for ordinary building construction). It

has an allowable load of 125 MPa. To make the problem simple, assume we

are holding up a 1275 kg payload on top of the tower, which under one

gravity has a weight of 12,500 N. Therefore we need one square centimeter

of cross sectional area of steel to hold up the weight. Steel has a density

of 7800 kg per cubic meter. The top meter of the tower has a volume of

0.01x0.01x1.0= 0.0001 cubic meter. This has a mass of 0.78kg. So the

structure 1 meter down from the top has to support a mass of 1275.78 kg,

i.e. the payload plus the top meter of steel. The load has increased by

0.06%, so the cross sectional area also increases by 0.06%. The area

increases in a compound interest fashion at the rate of 0.06% per meter as

you go down the tower. Over the course of 1 km in height, the increase is

by a factor of 1.8433.

We define the scale height of a structure as the length over which the cross

sectional area increases by a factor of e (2.71828...). In the case we have

been using it is 1635 meters. The scale height can be found by dividing the

allowable load of the material by the density times the local acceleration

(one gravity in the case of the Earth):

h(scale) = load / (density x acceleration)

= 125 MPa / (7800 kg/m^3 x 9.80665 m/s^2) = 1635 meters

So a tower 4.9 km tall would have an area at the bottom e cubed (20.08)

times the area at the top, and the weight of steel would be e^3 - 1, or

19.08 times the payload weight.

Now, plain carbon steel is not a very good material to use if you want a

really big tower. Let us look at advanced carbon composites, such as is

used in modern aircraft and spacecraft. One specific formulation (Amoco

T300/ERL1906 if you must know) has a compressive strength of 1930 MPa

(280,000 psi). We derate this by half to get the allowable load. This is

the same as is done for the steel, where you only use 50% of the strength to

give you a safety margin. So we have 965 MPa (140,000 psi) as an

allowable load. The density is 1827 kg/m3 (0.066 lb/cu in.) Dividing we

have a scale height of 53,878 meters (176,800 feet, or 33.5 miles) If you

build several scale heights tall, you can see in theory you could build

structures hundreds of kilometers tall.

In a real structure the payload probably won't all be at the top. For the

bottom 20 kilometers or so wind loads, ice build-up, and other

environmental effects have to be accounted for. Above this height, atomic

oxygen can attack your carbon/epoxy structural material, so a protective

layer is needed. This adds weight so there will be some reduction in how

high you can build. But you still can build many times taller than anything

built so far.

These types of towers can be built 'from the top down' in order to avoid

construction work in a vacuum. In this process, the top section of the tower

is assembled at ground level. Jacks raise the section up by one section

length. The next section down is then installed underneath. The process is

repeated for the whole tower height, so all the construction work takes place

near ground level. Special anchoring provisions are required to stabilize

the tower while being built in this fashion.


The tallest existing structure is a TV antenna which is 655m (2150 ft ) tall.

Some engineering/ architectural studies on very large towers have been

done. No attempts to build anything over 1000 meters tall are known. This

concept should be within current technology for structural materials,

although it may require an advance in construction techniques.

1a Unguyed Mast

In this approach the base of the tower needs to be 1/10 to 1/20 of the tower

height to provide stability. In the lower part of the tower, wind loads will

require the base to spread at a greater slope than the upper part, which only

depends on buckling for its necessary base width. This approach assumes

that most of the loads on the tower act vertically, as in an elevator riding

up and down the tower height.

1b Guyed Mast

If the loads are substantially sideways the tower mast may be stabilized by a

set of guy wires that spread out at a 30-45 degree angle.

1c Series of Towers

A very long, tall structure, such as a 300 km long electromagnetic

accelerator, may use a series of towers as supports.


2 Tethers

Alternate Names: Beanstalks, Jacob's Ladder, Space Bridge,

Geosynchronous Towers



Tensile members in orbit store and transfer momentum to vehicles. The

tethers may be gravity-gradient stabilized or rotating endwise. A ground-to-

geosynchronous cable is not feasible with today's structural materials.

Tethers, of which a geosynchronous cable is a special case, obey an

exponential mass-ratio-to-payload-weight relation similar to that for

chemical rockets. It is possible, with existing materials, to build tethers

which will provide several km/s of delta v. In a launch system application,

an orbiting tether can be set rotating so that the lower end travels slower

than orbital velocity. A launch vehicle could rendezvous with the tether,

drop a payload, then release. Since only the payload remains in orbit, the

propulsion system on the tether only has to provide momentum to add to the

payload; the launch vehicle never has to take itself to orbital velocity.

In this case the tether acts as a 'momentum bank', lending velocity to the

launch vehicle temporarily while the payload is unloaded.

Tethers are the generalization of the 'beanstalk' or geosynchronous tower

concept. In the original concept, a cable is placed so that it hangs vertically

over the equator, and is in a 24 hour orbit. It thus appears to hang

vertically over one spot on the Earth. The task of reaching Earth orbit then

reduces to a very long (35,000 km) elevator ride. Unfortunately for the

original idea, tensile strengths approaching 2 million pounds per square inch

(12.5 GPa) are required for reasonable designs.

Tethers generalize on the original concept by (1) allowing any length, (2)

allowing any orbital period, (3) allowing any swinging or rotating states,

and, (4) allowing multiple tethers to be connected in various geometries.

One simple case would be a tether vertically oriented in earth orbit, spanning

the altitudes from 300km to 2000km. A cargo could be carried on an

elevator over this altitude range. While it is not as elegant as the

geosynchronous case, it is constructable with existing materials.

Material strength to density ratio is the critical criterion for designing

tethers. To build a minimum mass tether, one wishes to taper it's cross

section by a factor of e per scale length. The scale length is the length

at which under one gravity, the weight of a constant section cable equals

the tensile strength (i.e. just breaks). While the gravitational field

around a planet is non-uniform, the 'depth' of the gravity well is equal

to the surface gravity times the radius of the planet. The following

table shows the taper factors derived for each gravity well given materials

available at different times:


Taper Factors Required For Various Gravity Wells and Technology Levels |


Gravity Depth ---------------- Time Period --------------

Well (g-km) 1960s 1970s 1980s 1990s 2000s

---------- ------ ----- ----- ----- ----- -----

Moon's 287 21 3.1 2.5 2.1 1.9

Mars' 1289 7.8E5 160 58 28 17

1/2 Earth's 3190 3.8E14 2.7E5 2.3E4 4000 1060

Earth's 6375 1.4E29 7.2E10 5.1E8 1.5E7 1.1E6


Material Fiber- Kevlar Carbon Carbon Adv.

glass Carbon

Tensile Str. (MPa) 2410 3625 5650 6895 8273

Density (kg/m^3) 2580 1450 1810 1827 1840

Scale length (km@1g) 95 255 318 385 458




2a Orbital Hanging Tether

2b Orbital Rotating Tether

2c Terrestrial Tether

One vehicle pulls another without direct mechanical attachment. Allows

modification of one vehicle without reconfiguration of joined pair. Allows

one type of vehicle to pull another. Reduces loads on lead vehicle by

lift-to- drag ratio.


[D1] Ebisch, K. E. "Skyhook: Another Space Construction Project",

American Journal of Physics, v 50 no 5 pp 467-69, 1982.

[D2] Carroll, J. A. "Tether Space Propulsion", AIAA paper 86-1389, 1986.

[D3] Penzo, P.A. and Mayer., H.L. "Tethers and Asteroids for Artificial

Gravity Assist in the Solar System" Journal of Spacecraft and Rockets, Jan-

Feb 1986. (Details how a spacecraft with a kevlar tether of the same mass

can change its velocity by up to slightly less than 1 km/sec. if it is

travelling under that velocity wrt a suitable asteroid.)

[D4] Baracat, William A., Applications of Tethers in Space: Workshop

Proceedings Vols 1 and 2. (Proceedings of a workshop held in Venice,

Italy, Octover 15-17, 1985) NASA Conference Publication 2422, 1986.

[D5] Anderson, J. L. "Tether Technology - Conference Summary",

American Institute of Astronautics and Aeronautics paper 88-0533, 1988.

[D6] Penzo, Paul A. and Ammann, Paul W. Tethers in Space Handbook,

2nd Edition, NASA Office of Advanced Program Development, 1989.

(NTIS N92-19248/3)

3 Aerostat

Alternate Names: High altitude balloon



One approach to minimizing drag and gravity losses is to carry a vehicle

aloft with a high altitude balloon. Research balloons have carried ton-class

payloads in the range of 15-30 km high, which is above the bulk of the





4 Low-Density Tunnel

4a Light Gas Tunnel

Alternate Names:



One or more light gas balloons are strung along the path of a vehicle or

projectile. The gas has a lower density than air. The formula for drag is

0.5*C(d)*Rho*A*v^2, where Rho is the density. Thus the lower density will

lower drag.




4b Evacuated Tunnel

Alternate Names:


Description: An evacuated tunnel is supported up through the atmosphere

(as by one or more towers). A launch system such as an electromagnetic

accelerator fires a projectile up through the tunnel. Drag losses are

minimized within the tunnel, and are low in the remaining part of the

atmosphere which must be traversed. If the top end requires some means of

keeping air from flowing in and filling the tunnel - such as a hatch that

remains closed until the accelerator is about to fire.




D.1b Dynamic Structures

Static structures rely on the strength of materials to hold themselves up.

Dynamic structures rely on the forces generated by rapidly moving parts to

hold up the structure. The advantage of this approach is it can support

structures beyond the limits of material strengths. The disadvantage is that

if the machinery that controls the moving parts fails, the structure falls


5 Fountain/Mass Driver

Alternate Names:



An electromagnetic accelerator provides a stream of masses moving up

vertically. A series of coils decelerates the masses as they go up, then

accelerates them back down again, at a few gravities. When they reach

bottom, the accelerator slows them down and throws them back up again, at

hundreds of gravities. Thus the accelerator is many times shorther than the

fountain height. The reaction of the coils to the acceleration of the fountain

of masses provides a lifting force that can support a structure. The lifting

force is distributed along where the coils are located. This can be along the

length of a tower, or concentrated at the top, with the stream of masses in

free-flight most of the way.




6 Launch Loop

Alternate Names:



A strip or sections of a strip are maintained at super-orbital velocities. They

are constrained by magnetic forces to support a structure, while being

prevented from leaving orbit. A vehicle rides the strip, using magnetic

braking against the strip's motion to accelerate. Several concepts using

super-orbital velocity structures have been proposed. One is known as the

'launch loop'. In this concept a segmented metal ribbon is accelerated to

more than orbital velocity at low Earth orbit. The ribbon is restrained from

rising to higher apogees by a series of cables suspended from magnetically

levitated hardware supported by the ribbons. The ribbon is guided to

ground level in an evacuated tube, and turned 180 degrees using magnets on

the ground. A vehicle going to orbit rides an elevator to a station where the

cable moves horizontally at altitude. The vehicle accelerates using magnetic

drag against the ribbon, then releases when it achieves orbital velocity.





7 Multi-Stage Tethers

Alternate Names:



A multi-stage tether has more than one tether, with the tethers in relative

motion. For example, a vertically hanging tether in Earth orbit can have a

rotating tether at it's lower end. The advantage of such an arrangement is to

lower the mass ratio of tether to payload compared to a single tether. The

mass ratio of a rotating tether is approximately proportional to exp(tip

velocity squared). If two tethers each supply half the tip velocity, then the

ratio becomes exp(2(tip velocity/2)squared), which is a smaller total mass


Another feature of a multi-stage tether is that the tip velocity vector of the

two stages add. Since one rotates with respect to the other, the sum of the

vectors changes over time. Given suitable choices of tip velocities and

angular rates, one can receive and send payloads with arbitrary speed and

direction up to the sum of the two vectors.




D.2 Guns and Accelerators

D.2a Mechanical Accelerators

8 Leveraged Catapult

Alternate Names:



A leveraged catapult uses a relatively large or heavy driver to accelerate a

smaller payload at several gravities by mechanical means. Devices such a

multiple sheave pulley or a gear train convert a large force moving slowly to

a small force moving fast, and transmit the force along a cable.

The mechanical advantage produces more than one gravity of acceleration.

This concept may be one of the simplest to implement on a small scale.

Despite the seeming simplicity of the concept,

velocities of several km/s are possible, which would greatly reduce the size

of a rocket needed to provide the balance of the velocity to orbit.

The performance of this concept reaches a limit due to the weight, drag, and

heating of the cable attached to the payload and the magnitude of the driving

force, which is divided by the leverage ratio to yield the force on the




8a Drop Weight

A falling mass is connected to a vehicle by a high strength cable running over

a multiple-sheave pulley, or cable reels with different diameters or connected

with a gearing ratio. Two types of location are possibilities - river gorges

and mountain peaks. Locations such as the Grand Canyon and the

Columbia River gorge have lots of vertical relief for the drop weight. At

these locations the weight can consist of a large fabric bag filled with water

from the river at the bottom. The bag can be emptied before hitting bottom.

This reduces the weight that has to be stopped by a braking system.

For mountain peak locations, the drop weight runs down a set of rails and is

stopped by running into a body of water or running up an opposing hillside

plus possibly wheel braking. The mountain location may be preferred

because of the greater launch altitude.

This example assumes a mountain with a solid weight sliding on rails:

We assume that a 15,000 kg cryogenic rocket using RL-10 engines is

being thrown. An acceleration of 60 m/s^2 is tolerable by humans for

the 20 seconds required to reach 1200 m/s assuming the human is in

good health and properly supported. The linear path traversed would

be 12,000 m (7.5 miles) at constant acceleration. The tow cable

pulls with 900 kN (202,000 lb) of force on the rocket.

High strength carbon fiber is available with tensile strengths of

up to 6.9 GPa (1,000 ksi). Allowing for a factor of safety and a braided

overwrap surrounding the fiber bundle (to protect the carbon fibers

from abrasion), we assume a working stress of 3.45 GPa (500 ksi).

The cross sectional area required is then 0.00026 m^2, or a circular

cable 0.0182 m (0.72 inch) in diameter.

With a density of 1840 kg/m^3, the cable mass is 0.48 kg/m. For a

12,000 m long cable, the mass would be 5,760 kg without allowing for

taper. Since the cable has to be accelerated also, the leading end,

closer to the drive mechanism, has to be able to apply sufficient force

to accelerate the rocket and all the cable in between the rocket and

leading end.

To a first approximation, the leading end must be 40% larger in cross

section than the trailing end to account for the acceleration of the

cable itself. The mean weight of the cable is then 0.576 kg/m, giving

a cable mass of 7,000 kg.

Something has to be done with the cable after it finishes the job of

accelerating the vehicle. It can be taken care of by looping the cable

back from the drive mechanism to the starting point (where the vehicle

is at the start). At the completion of a launch, the loop of cable is

moving at 1200 m/s, and is gradually allowed to come to a stop.

The return portion of the cable has to be able to accelerate the part of

the cable past the drive mechanism at 6 g's. Given a 7000 kg lead section,

we get a mass of 3200 kg. Thus the total mass accelerated at 6 g's is

25,200 kg, and the total accelerating force is 1.512 MN.

Given a 45 degree slope on a mountain, we assume that the drop weight

accelerates down at 3 m/s^2. Thus the gear ratio is 20:1, and the

force of the drop weight on the drive system must be 30.24 MN. Since

a free-falling weight on this slope would accelerate at 7 m/s, there is

a 4 N/kg retarding force due to the drive system. Therefore the drop

weight must be 7,560 tons. If it is a 4 meter square block of steel

(about the cross section of a railroad car), it will weigh 124.8 tons

per meter, and hence must be 60.6 meters long. The rails must be

12,000 m/20 long, or 600 meters long.

8b Locomotive Driver

A set of railroad locomotives provides the motive force, which is multiplied

by a gear mechanism to a higher speed. Example: launching a 20,000 lb

vehicle at 3 g's to 1100 m/s:

* Need 20 km straight run of rail.

* Rail cars needed:

- 1 tank car

- 1 car special purpose to carry glider

- 2-3 cars with tow rope guides

- 1 car pulley system

- 30 locomotives in tandem

We assume the locomotive top speed is about 27 m/s, therefore a 40:1 gear

ratio will provide the desired speed at the vehicle. Locomotive traction

averages 80,000 lb/engine, or 2000 lb per engine when reduced through the

gear ratio. The gear-down mechanism and launch cable drum are mounted

on a flatbed rail car. This car can be anchored to a foundation on either side

of the railroad track to hold it in place when the combined pull of the

locomotives is exerted. The starting traction of 30 locomotives is 1800

tons. Since the couplings between engines are probably not designed for

this load, a set of steel cables on both sides of the locomotives are used to

transmit the traction force from each engine to the gear mechanism. The

vehicle is attached to the anchored rail car by a high-strength cable which is

20 km long. At 3 g's it takes this distance to accelerate to the desired speed.

Two or three rail cars are spaced out along the 20 km with erectable towers

with a pulley wheel on top, to guide the cable and keep it off the ground

during the initial acceleration. The vehicle has glider type wings attached

that will generate lift as it gains speed, so the vehicle will climb once it

reaches 100 m/s or so. When the vehicle reaches the desired speed, the

cable is released and the vehicle continues to climb under the glider's lift.

Eventually the glider drops the vehicle, which proceeds under rocket power.

A small prototype:

Single Locomotive driver. 1250 lb rocket @ 4 g peak. Final velocity = 700

m/s. Accel time = 17.5 sec distance = 0.5at2 = 6.1 km. Engine traction =

80,000 lb average @ 25:1 gear ratio.

8c Jet Driver

This is similar to the locomotive case, but the gear ratio is lower since the

jet can reach a higher speed on a take-off run. Example: an F-15 can tow

40,000 lb rope tension if near empty. @10:1 gear ratio can accelerate 1000

lb object @ 4 g's. Aircraft top speed on deck = 300 m/s. Object top speed

in theory would be 3000 m/s. In practice would be limited by aerodynamics

and cable heating (perhaps to 1500 m/s? limit is not well understood)


9 Rotary Sling

Alternate Names: Centrifuge Catapult



In principle, this is a sling or bolo scaled up and using aerospace materials.

A drive arm is driven in rotation by some means. A cable with the payload

attached to the end is played out gradually as the system comes up to speed.

The drive arm leads the cable slightly so the cable and payload see a torque

that continues to accelerate them. When the desired payload velocity is

reached, the payload releases and flies off. The cable is then retracted and

the drive arm slows down. When it stops, another payload is attached.

In a vacuum, such as on the Lunar surface, this is theoretically a very

efficient system, as the sling can be driven by an electric motor and the

mechanical losses can be held to a low value. Some method of recovering

the energy of the arm and cable (such as by transferring it to a second

system by using the motor as a generator), can lead to efficiencies over 60%

in theory.

On Earth such a system is hindered by air drag. One method of reducing

drag is to attach a shaped fairing to the cable material, so as to lower

drag compared to a circular cable. Another is to mount the drive arm on the

top of a large tower, so the cable is not moving in dense air. A third is to

generate lift along the cable or at the payload, so the rapidly moving part of

the cable, near the payload, is at a high altitude, where there is less drag.

Example #1: Single cable

- Assume v(tip) = 3000 m/s and a(tip) = 10000 m/s2

- Then r = 9000 meters.

- For 1000 kg projectile at end, cable tenssion is 1 MN. If carbon fiber at

3400 MPa design stress, then cable area is 1/3400 m2, or about 3 cm2.

- Cable weight is 0.6 kg/meter, adds 600 N//meter at tip, or 0.06%/meter.

Over 9 km, area

increases by factor of 6.

- Accelerating force to spin up in 1 hour iis 1 m/s/s or 1000 N.

- Drag force/meter @ sea level = ~0.5 x draag force/meter at tip

= 0.5Cd rho A v2. Cd = 0.04 for shaped airfoil. rho =1.225 kg/m3 . A =

0.01 m2.

v= 3000 m/s. Drag = 2205 N/m x 0.5 factor = 1102 N/m.

- Want drag<= spin up force, thus want ddrag < 0.11 N/m.

- Therefore want air density at 10-4 x sea level = 240,000 ft = 73 km high.

- Thus put 9000 meter cable at top of 73 kmm tall tower with drive motor to

spin up cable.

Example #2: Two Stage Centrifuge Catapult

- High g's small payload catapult.

- Assume 3 km/s/stage

- 33 g's in 1st stage and 67 g's in second stage.

(this example is incomplete)




D.2b Artillery

10 Solid Propellant Charge

Alternate Names:


Description:Explosive vaporizes behind projectile in barrel. Gas pressure

accelerates projectile to high velocity. Conventional artillery reaches speeds

of around 1000 m/s.

Status: Artillery has a long history and extensive use. The High Altitude

Research Probe project attached two naval gun barrels in series and used

relatively light shells to reach higher muzzle velocities than conventional




Verne, Jules, "From the Earth to the Moon".

11 Liquid Propellant Charge

Alternate Names:


Description: Similar to conventional solid propellant artillery except liquid

propellants are metered into the chamber, then ignited. Liquid propellants

have been studied because they produce lighter molecular weight

combustion products, which leads to higher muzzle velocities, and because

bulk liquids can be stored more compactly than shells, and require less

handling equipment to load.




12 Gaseous Charge

12a Fuel-Oxidizer Charge

Alternate Names:


Description: Similar to conventional artillery except gaseous propellants are

metered into the chamber. This is essentially what happens in the cylinder

of a car engine, as a point of reference.

Status: Used as the driver for the Livermore gas gun (fuel-air mix drives 1

ton piston, which in turn compresses hydrogen working gas).



12b Scramjet Gun

Alternate Names: Ram Accelerators



Fuel/oxidizer mixture present in barrel is burned as projectile travels up

barrel. If projectile shape resembles two cones base to base, as in an inside-

out scramjet, the gas is compressed between the projectile body and barrel

wall. The combustion occurs behind the point of peak compression, and

produces more pressure on the aft body than the compression on the fore-

body. This pressure difference provides a net force accelerating the


One attraction of this concept is that a high acceleration launch can occur

without the need for the projectile to use onboard propellants. If the

projectile has a inlet/nozzle shape (hollow in the middle) it might continue

accelerating in the atmosphere by injecting fuel into the air-only incoming

flow, extending the performance beyond what a gun alone can do. Another

attraction of this concept is the simplicity of the launcher, which is a

simple tube capable of withstanding the internal pressure generated during


Status: Research being performed at the University of Washington under

Prof. Adam Bruckner. Research gun in basement of building.



[ D7] A. Hertzberg, A.P. Bruckner, and D.W. Bogdanoff, "The Ram

Accelerator: A New Chemical Method of Accelerating Projectiles to

Ultrahigh Velocities" , AIAA Journal, Vol. 26, No. 2, February, 1988.

(The seminal reference.)

[ D8] P. Kaloupis and A.P. Bruckner, "The Ram Accelerator: A

Chemically Driven Mass Launcher" , AIAA Paper 88-2968,

AIAA/ASME/SAE/ASEE 24th Joint Propulsion Conference, July 11-13,

1988, Boston, MA. (Applications to surface-to-orbit launching.)

[ D9] Breck W. Henderson, "Ram Accelerator Demonstrates Potential for

Hypervelocity Research, Light Launch," , Aviation Week & Space

Technology, September 30, 1991, pp. 50-51.

[ D10] J.W. Humphreys and T.H. Sobota, "Beyond Rockets: the

Scramaccelerator" , Aerospace America , Vol. 29, June, 1991, pp. 18-21.

13 Rocket Fed Gun

Alternate Names:


Description: Rocket engine at chamber end of gun produces hot gas to

accelerate projectile. In a conventional gun, all the gas is formed at once as

the charge goes off. In this concept the gas is produced by a rocket type

engine and fills the barrel with gas as the projectile runs down it. Compared

to a conventional gun, the peak pressure is lower, so the barrel is lighter.




D.2c Light Gas Gun

Light gas guns are designed to reach higher muzzle velocities than

combustion guns. They do this by using hot hydrogen (or sometimes

helium) as the working gas. These have a lower molecular weight, and

therefore a higher speed of sound. Guns are strongly limited by the speed

of sound of the gas they use. The drawback to light gas guns is that the gas

does not generate high pressures and temperatures by itself (as do

combustion byproducts). Therefore some external means are required to

produce the gas conditions desired.

14 Pressure Tank Storage

Alternate Names:


Description: The gas is stored in a chamber, then adiabatically expanded

in a barrel, doing work against a projectile.




[D11] Taylor, R. A. "A Space Debris Simulation Facility for Spacecraft

Materials Evaluation", SAMPE Quarterly , v 18 no 2 pp 28-34, 1987.

15 Underwater Storage

Alternate Names:



In a gas gun on land the amount of structural meterial in the gun is governed

by the tensile strength of the barrel and chamber. In an underwater gun, an

evacuated barrel is under compression by water pressure. The gas pressure

in the gun can now be the external water pressure plus the pressure the

barrel wall can withstand in tension, which is up to twice as high as the land


Other features of an underwater gun are the ability to store gas with very

little pressure containment (the storage tank can be in equilibrium with the

surrounding water), and the ability to point the gun in different directions

and elevations.

The underwater gas gun consists of a gas storage chamber at some depth in

a fluid, in this case the ocean, a long barrel connected to a chamber at one

end and held at the surface by a floating platform at the other end, plus some

supporting equipment.

The chamber is a made of structural material such as steel. An inlet pipe

allows filling of the chamber with a compressed gas. A valve is mounted

on the inlet pipe. An outlet pipe of larger diameter than the inlet pipe

connects to the gun barrel. An outlet valve is mounted on the outlet pipe.

This valve may be divided into two parts: a fast opening and closing part,

and a tight sealing part. The interior of the chamber is lined with

insulation. The inner surface of the insulation is covered by a refractory

liner, such as tungsten. An electrical lead is connected to a heating element

inside the chamber.Ê

An inert gas such as argon perfuses the insulation. The inert gas protects

the chamber structure from exposure to hot hydrogen, and has a lower

thermal conductivity. An inert gas fill/drain line is connected to the volume

between the chamber wall and the liner. A pressure actuated relief valve

connects the chamber with a volume of cold gas. This cold gas is

surrounded by a flexible membrane such as rubber coated fiberglass cloth.

In operation, the gas inside the chamber, the inert gas, and the

water outside the chamber are all at substantially the same pressure. Thus

the outer structural wall does not have to withstand large

pressure differences from inside to outside. One part of the chamber wall

is movable, as in a sliding piston, to allow variation in the

chamber volume. The gas in the chamber is preferably hot, so as to

provide the highest muzzle velocity for the gun. When the gun is operated,

this gas is released into the gun barrel. In order to preserve the small

pressure difference across the wall of the chamber, either the chamber

volume must decrease or gas from an adjacent cold gas bladder must replace

the hot gas as it is expelled. This arrangement prevents ocean water from

contacting the chamber walls or hot gas. In the case of the sliding

piston, the membrane collapses, with the gas formerly within it moving in

behind the piston. In the alternate case, the membrane also collapses, with

the gas formerly within it moving through a valve into the chamber.

The chamber has an exit valve which leads to the gun barrel. It also has gas

supply lines feeding the interior of the chamber and the volume between the

chamber walls. These lines are connected to regulators which maintain

nearly equal gas pressures, which in turn are nearly equal to the ocean

pressure. This allows the chamber to be moved to the surface for

maintenance, and to be placed at different depths for providing different

firing pressures or different gun elevations.

The muzzle of the gun is at the ocean surface, so elevation of the gun can be

achieved by changing the depth of the chamber end. Since the gun as a

whole is floating in the ocean, it can be pointed in any direction. Some

means for heating the gas stored in the chamber is needed, such as an

electric resistance heater. At the muzzle end of the gun, a tube surrounds

the barrel, with a substantial volume in beween the two. There are

passages through the wall of the barrel that allow the gas to diffuse into the

tube rather than out the end of the gun, thus conserving the gas.

At the muzzle of the gun is a valve which can rapidly open, and an ejector

pump which prevents air from entering the barrel. In operation, the ejector

pump starts before the gun is fired, with the valve shut. The valve is

opened, then the gun is fired. In this way, the projectile encounters only

near vacuum within the barrel, followed by air.




16 Thermal Bed Heated

Alternate Names:



Hot gas is generated by flowing hydrogen through a chamber which

contains refractory oxide particles. The particles are heated slowly (roughly

1 hour time period) by some type of heater near the center of the chamber.

This sets up a temperature gradient, so the exterior of the chamber is

relatively cool, and can thus be made of ordinary steels. When the

hydrogen flows through the chamber, the large surface area of the particles

allows very high heat transfer rates - so the heat in the chamber can be

extracted in a fraction of a second.


- To match Livermore hypersonic gun, want 440 MJ projectile energy.

- Assume losses and initial gas pressure arre equal contributors.

- Want aluminum oxide to drop 500K in operaation. At 1 J/K/gm, need 80

kg of

aluminum oxide. In the form of grinding wheels, we are looking at about

$600 of

materials. 25 wheels, 30 cm diam x 2.5 cm thick.

- If pressure is 3000 psi (20.7 MPa), then 10 cm diam, 5 kg projectile will


162,600 N force, or 32,470 m/s2. To reach 4 km/s requires 0.123 sec,

distance = 250 m.

Allow 500 m for losses. Alignment = 1/6 mm along length.

- Final gas temp average 2000 K. Barrel vollume = 4 m3. Storage tank =

0.65 m3.

- Storage tank diam = 50 cm. Length = 3.3 mm.

- Chamber = 32 cm diam ID x 65 cm long.

Status: A small research gun of this type has been built at

Brookhaven Natl. Lab.



17 Particle Bed Reactor Heated

Alternate Names:



Hot gas is generated by flowing through particle bed type reactor. Gas

expands against projectile, accelerating it. Light gas guns have been

operated to above orbital velocity, and 1 kg projectiles have been

accelerated to over half orbital velocity. This type of gun rapidly becomes

less efficient above the speed of sound of the gas. As a consequence the

working fluid is usually hot hydrogen. Conventional gas guns have used

powder charge driven pistons to compress and heat the gas. This is not

expected to be practical on the scale needed to launch useful payloads to

orbit. One way to heat the gas is to pass it through a small particle bed

nuclear reactor. This type of reactor produces a great deal of heat in a small

volume, since the small particles of nuclear fuel have a large surface/volume

ratio and can efficiently transfer the heat to working fluid. This uses the

benefits of nuclear power for space launch, without the drawbacks of a

flying reactor.




18 Electric Discharge Heated

Alternate Names:



Gas is heated by electric discharge, then pushes against projectile in barrel.

The limiting factor for a light gas gun is the speed of sound in the gas. One

way to heat the gas to much higher temperatures is an electric discharge

within the gas.




19 Nuclear Charge Heated

Alternate Names:


Description: Similar to artillery, except explosive in chamber is atomic

bomb. This concept makes sense in a situation where very large payloads

need to be launched. A large underground chamber is excavated, and filled

with hydrogen gas as the working fluid. A large barrel leads off the

chamber upward at an angle. A crossbar is set into the barrel near the

chamber, and the projectile is attached to the crossbar with a bolt that is

designed to fail at a pre-determined stress. This restrains the projectile

until the operating pressure is reached. A small atomic bomb is suspended

in the chamber and detonated to create lots of hot hydrogen in a very short





20 Combustion Driven Piston

Alternate Names:


Description: This is a type of two-stage gas gun. A cylindrical chamber

contains a piston. On the back side of the piston high pressure gas is

generated by combustion. This can be gunpowder or a fuel-air mixture. On

the front side of the piston is the working gas, which is usually hydrogen.

The hydrogen is compressed and heated until a valve or seal is opened.

Then the working gas accelerates the projectile.

Status: This type of gas gun is the most common that has been built.

They were first constructed in the 1960's or earlier. The largest gun of this

type is a Lawrence Livermore Laboratory, where it is being used to test

scramjet components at 2.4 km/s (Mach 8) (Dec. 1993). It has a 4

inchx150ft barrel and a larger diameter, 300 ft long chamber.



[D12 ] Aviation Week & Space Technology, July 23, 1990.

[D13 ] "World's Largest Light Gas Gun Nears Completion at Livermore."

Aviation Week & Space Technology, August 10,1992.

21 Gravity Driven Piston

Alternate Names:



A sliding or falling mass is used to compress gas in a chamber. The gas is

then expanded in a barrel. An alternate method of compressing and heating

the working fluid in a light gas gun is a rapidly moving, massive piston. If

the gun is built on the side of a mountain, the energy for launch is stored as

potential energy in the piston. The piston floats on an air or lubricated

bearing and slides down the mountain to a cylinder. The cylinder leads to a

barrel containing the projectile, which accelerates upward.




D.2d Electric Accelerators

Electric accelerators typically require high peak power for a short period of

time. Hence inexpensive energy storage is very important for these

concepts. Two places to look for inexpensive energy storage are (1)

Magnetic fusion experiments, and (2) Inductive energy stores. The latter

falls into subcategories: cooled normal conductors, and superconductors.

22 Railgun

Alternate Names: Electromagentic Gun



High current electricity supplied by rails is shorted through plasma arc.

Plasma is accelerated by reaction against magnetic field produced by

current. Plasma pushes projectile. A railgun uses magnetic forces to

accelerate payloads. Typically two parallel conducting rails are bridged by a

plasma arc. The plasma is accelerated downÊthe gun by the arrangement of

currents and fields. Given suitable power supplies, it can be considered for

earth launch systems at lower accelerations than those proposed for weapon



This device was under intensive development for the Strategic Defense

Initiative. A large gun was built at Eglin AFB in Florida and used a bank of

thousands of car batteries wired in parallel as a power supply. Prototype

railguns achieved high velocities, but the high currents produced rail




[D14] Robinson, C. A. "Defense Department Developing Orbital Guns",

Aviation Week and Space Technology, v 121 no 12 pp 69-70, 1984.

[D15] Bauer, D. P. et al "Application of Electromagnetic Accelerators to

Space Propulsion" IEEE Trans. Magnetics vol MAG-18 no 1 pp 170-5,

Jan. 1982.

23 Coilgun

Alternate Names: Mass Driver Launcher



Series of coils forming gun react with coil(s) on projectile magnetically,

producing thrust. Popularly known as a 'mass driver', this concept uses

magnetic attraction between two current carrying coils to accelerate a

projectile. The concept has been developed in connection with launching

lunar materials for space manufacturing.Ê Accelerator designs with high

efficiency (>90%) and high muzzle velocitiesÊ (>8 km/s) have been

proposed. This potentially leads to a transportationÊ system whose

operating costs consist mostly of electricity, or $0.28/lb. Laboratory

versions of electromagnetic accelerators have reached 1800 gravities

acceleration. Accelerations in the range of ~100 gravities are sufficient for

cargo launch from the surface of the earth.




[D16] Nagatomo, Makoto; Kyotani, Yoshihiro "Feasibility Study on

Linear-Motor-Assisted Take-Off (LMATO) Of Winged Launch Vehicle",

Acta Astronautica, v 15 no 11 pp 851-857, 1987.

[D17] Kolm, H.; Mongeau, P. "Alternative Launching Medium", IEEE

Spectrum, v 19 no 4 pp 30-36, 1982.

[D18] Kolm, H. "An Electromagnetic 'Slingshot' for Space Propulsion",

Spaceworld pp 9-14, Feb. 1978.