Index Home About Blog
From: Bruce Dunn <>
Subject: Re: Cannon Launch? (Very cheap access to space)
Date: Fri, 26 Jul 1996 06:30:17 -0700

Using ordinary cannon propellants and a slightly modified naval cannon, 
muzzle velocities of 1500 to 2500 m/sec for projectiles of several 
hundred kg have been achieved back in the 1960s.  Using a cannon to 
launch solid rockets makes it possible to orbit payloads on the order of 
a few 10s of kilograms, with peak accelerations of no more than 5000 g. 
 I reproduce below some technical information project HARP using such 

Gun Launch for Orbital Vehicles

Bruce Dunn, January 1995

     In the 1960s,  project HARP (High Altitude Research Project), was 
run out of McGill University in Montreal, with U.S. Army funding.  
Project HARP involved the use of large guns to fire instrumented 
ballistic projectiles and rockets to high altitudes. The HARP program 
was terminated in approximately the mid 1960s before the group's 
ultimate goal of launching an orbital vehicle was achieved.  This 
message gives some otherwise hard to get information about project HARP 
from two older papers in the Canadian Aeronautics and Space Journal.   
Material in quotation marks is from the cited papers.

Paper 1:  Bull, G.V. (1964) Development of Gun Launched Vertical Probes 
for Upper Atmosphere Studies.  Canadian Aeronautics and Space Journal 
10:236-247. This paper was written to accompany a speech made by Bull in 
Toronto in May 1964. In the Introduction to the paper, Bull writes:

"During the past several years, both theoretical and experimental 
investigations have been undertaken to determine the applicability of 
guns to scientific studies of the ionosphere.  Such possibilities have 
intrigued ordnance workers for many years, but involve a complex mixing 
of advanced gunnery techniques, scientific experiment considerations and 
economics....In late 1961, with material support from the US Army, 
McGill University undertook the development of a 16 inch gun system.  In 
early 1962 this program came under full support of the US Army through 
the Army Research Office and the Ballistic Research Laboratories" 

     In a section on sub-calibre ballistic projectiles, Bull says:

"For example, in the case of a 16 inch naval gun which normally fires 
shells in the 3,000 lb. class at velocities of 2,800 fps, velocities as 
high as 6,000 fps can be obtained with shot weights of the order of 400 
lbs., the sub-caliber vehicle in this case having a ballistic 
coefficient considerably higher than the normal shell.  By re-design of 
the gun (i.e. extending the chamber and barrel) to optimize at this 
lighter shot weight, velocities approaching 7,000 fps are possible."

     A series of sub-calibre "Martlet 2" vehicles were built, which were 
sub-calibre and rode the barrel in a fall-away sabot.  Canted fins on 
the projectile maintained aerodynamic stability, and spun the projectile 
up so that it was stable once leaving the atmosphere.  These were fired 
at elevations of from 60 to 90 degrees from a 16 inch naval gun (on loan 
from the U.S.) which was located in Barbados.  The gun was bored out to 
16.5 inches and made into a smooth-bore cannon.  Altitudes of 
approximately 500,000 to 600,000 feet (100 miles, 160 km) were projected 
for this arrangement, and early trial reported in the reference cited 
went as high as 112 km.  Martlet vehicles carried instruments made from 
discrete solid-state electronics - they were potted in a mix of epoxy 
and sand (!) and the designers did not seem to have any real trouble 
getting the electronic to survive the launch acceleration which peaked 
at approximately 20,000 g.  Martlet vehicles also routinely carried a 
liquid mixture of trimethyl-aluminum and triethyl-aluminum to be 
released at high altitudes for ionosphere studies.  Another option was 
to carry sodium-thermite mixes which when ignited would release sodium 
vapor.  If projectiles of a similar weight were fired for range rather 
than height then ranges of up to 150 to 200 miles were calculated, 
depending on the ballistic coefficient. Shots from the gun were routine 
and relatively inexpensive.  Bull states:

"Normally, loading of the gun can be accomplished in under one half 
hour, allowing a firing rate of one an hour....Standard service 
propellant available as surplus (WM/.245) has been used, and the gun 
geometry has not been modified.  Firing programs are planned for the 
summer and fall of this year [1964] when the gun barrel will be extended 
and lighter sabots used with propellant designed to match the light 
projectiles, which should extend the Martlet 2A apogee to 200 km....The 
economics of the gun launched probe has been as predicted, with the 
Martlet 2A airframes loaded with TMA/TEA and a flare in the nose cone 
varying in price between $2500 and $3500, with gun launch costs 
(propellant and gun wear) included."

     After having discussed ballistic projectiles, Bull discusses 
gun-launched rockets:

"Gun fired artillery rockets have been developed extensively since World 
War II and normally must withstand barrel acceleration loads of the 
order of 30,000 g along with the rotational loads superposed by shell 
spin.  The performance of this type of rocket is only of marginal 
interest in the vertical probe application where non-spinning (from a 
stress viewpoint) vehicles are flown at acceleration levels of less than 
10,000 g and relatively very large rocket motors are desired with high 
mass fractions....In May of 1963, work was started on what was 
designated as the Martlet 3A rocket assist vehicle as part of the HARP 
program.  The objective of this activity was the development of a 16 
inch gun launched probe which would carry some 40 lbs. of payload to 
altitudes in the 500 km range."

     The Martlet 3A and later 3B rocket vehicles were sub-calibre and 
used various solid propellants in various configurations. The main 
problem with gun launched rockets is supporting the solid propellant 
during the launch acceleration so that it does not collapse into the 
internal cavities molded into the propellant grain, and a lot of 
development work was performed to investigate the performance of various 
solid propellant grains.  From their knowledge of the performance of the 
16 inch gun system and general information about the specific impulse 
and mass fraction of solid fuel rockets, it was calculated that it would 
be fairly easy to put a payload into orbit using the HARP gun and a 
multistage solid fuel rocket.  

Orbital Launch Vehicle Characteristics from Figure 31 in the Bull paper:

Total launch weight:     2000 lbs
Stage 1 weight:          1440 lbs
Stage 2 weight:           403 lbs
Stage 3 weight:           117 lbs
Payload:                   40 lbs

Muzzle velocity          4500 fps
Mass fraction             0.8
Specific impulse          300 sec (vacuum)

     The first and second stages were to be fired at relatively low 
altitude, but clear of the atmosphere.  The third stage was to 
circularize the orbit, and would be fired horizontally at orbital 
altitude.  Such a vehicle was never built before the program was shut 
down, although motors of the first stage size were developed.  The HARP 
group was also involved in exploring the possibilities of launching 
liquid fueled rockets from the gun.  These could be thin-shelled as long 
as they had no gas spaces in them (you can accelerate a balloon full of 
water at any g force you like, as long is it is fully supported during 
the acceleration).

Paper 2: Eyre, F.W. (1966) The Development of Large Bore Gun Launched 
Rockets. Canadian Aeronautics and Space Journal 12:143-149.

"The concept of a rocket launched from a gun is not new.  It will 
suffice to affirm in this paper that the gun launched artillery rocket 
was in full development during the Second World War and this 
investigation still continues.  Like so much work in allied fields, a 
great deal of what has been done and is being done is classified and 
cannot here be repeated....The conventional solid propellant gun, firing 
meaningful projectiles, currently appears able to develop a maximum 
muzzle velocity of some 6000 to 9000 fps.  Allowing an 80% recovery of 
muzzle kinetic energy as potential energy, this corresponds to a ceiling 
for sounding work of some 800,000 to 1,000,000 ft. (say 160 to 200 
statute miles).  Significant improvements beyond this level must come 
either from use of a different type of gun or from rocket boost during 
vehicle flight, which is here considered."

"Figure 3 shows muzzle velocity vs. shot weight for the Barbados gun. 

"Assumed conditions:  Max. pressure 60000 psi
                      Fixed charge, 1000 lbs M8M propellant
                      Web size optimized."

[some approximate data points from Figure 3 graph, and from Figure 4 
showing acceleration vs. shot weight]

Shot weight  Muzzle velocity  Max. acceleration
 500  lbs          7700 fps       13,000 g
1000  lbs          6400 fps        9,000 g
1500  lbs          5700 fps        6,500 g
2000  lbs          5200 fps        5,000 g

	Eyre then goes into a long technical discussion related to how 
to support propellants of various types in a solid fuel rocket during 
the gun acceleration.  Perhaps the neatest concept is to simply fill all 
empty spaces in the rocket with a fluid which then can support the 
propellant grain hydrostatically during launch (sort of a rocket 
water-bed).  The rocket is then accelerated using some form of pusher 
plate, which seals the liquid in.  The plate drops away after launch, 
and the fluid is then vented or drained before ignition.  With regard to 
practicality and performance, Eyre writes:

"It has transpired in design studies that although structural problems 
do arise due to the acceleration loads, and additional problems are 
posed by the necessity to use a folding stabilizer assembly, mass 
fractions almost as high as conventional rockets can be achieved and the 
design problems are partially alleviated by an all supersonic flight 
regime.....Given this condition the advantage of the gun can be seen in 
that a typical vehicle of mass fraction 0.8 would have an apogee of 176 
miles used conventionally, 257 miles at 1000 fps launch, 342 miles at 
2000 fps, 435 miles at 3000 fps, 529 miles at 4000 fps and so on."

     Eyre then discusses the fabrication of a full-scale, full bore (16 
inch) motor with a weight of 1450 lbs., designated the Martlet 4A and 
designed for the Barbados gun.  At the time of writing of the paper, it 
does not appear as if this had yet been test launched - I do not know 
how far the program was carried before it was cancelled.

"Current work is directed towards development and application of a thin 
plastic wear resistant coating [they were worried about excessive wear 
on the rocket casing], and launching of 16 inch motors to investigate 
scale factor effects.  At the time of writing  [1966] full bore Aerojet 
General Corp. grains are awaiting launch. ... At the present time a 
heavy test program is about to commence with many agencies participating 
and for the most part full scale hardware ready for launch."

     In summary, up until the time of writing of the later of the two 
quoted papers in the mid 1960s, HARP under Dr. Bull appeared to have 
been highly successful using a surplus 16 inch naval cannon in firing 
projectiles to high altitudes and in firing solid fueled rockets..  Bull 
has been called the most brilliant gun designer of this century. His 
comment on vehicle design for guns of different scales is interesting:

"Obviously since launch weight (ie payload) is increasing roughly as the 
cube of the scale, while peak accelerations are decreasing linearly, the 
larger the gun the simpler the vehicle engineering problem."

Index Home About Blog