View Full Version : Question to ponder; Pneumatic Cannon

03-08-2004, 01:13 AM
The key with these things is being able to dump ALL the air in the tank as quickly as possible, as close to all at once as you can get. I've been pondering what the best valve is to use to do this. On mythbusters, they used a big butterfly valve that looked pretty good.
An interesting thing to note, one guy that competes at the Pumpkin Chuckin events, was on tv with his giant cannon, 2nd Ammendment. He said that he can get the same distance with 50 psi as he can with 200psi by timing the release of the air.
So Im only thinking about 100 psi max here, any suggestions about what valves to use? It seams that the valves are a closely guarded secret.

G.A. Ewen
03-08-2004, 01:25 AM
I would say that a ball valve that is the same size as the bore on the cannon would be the fastest. A quarter of a turn and it's full open.
Maybe not practical for a pukin gun but if you want to toss something the size of tennis balls they should work fine.

Forrest Addy
03-08-2004, 01:51 AM
Use a rupture disk. Consider the time the projo is in the barrel Vs time to open the valve.

Rupture disks open in a millisecond or two. You can make you own from scored plastic.

Mike W
03-08-2004, 02:16 AM
I am watching the chicken show right now. They called it a butterfly valve. It has a flat disc that opens with a quarter turn.

Mike W
03-08-2004, 02:20 AM
50 psi and 120 psi gave the same speed with a chicken. 20 psi also broke the windshield which wasn't rated for bird strikes. One test hit a speed of 132 mph.

03-08-2004, 02:36 AM
That ball valve sounds good if it opens fast enough. How about a solinoid actuated
lever on the ball valve and a pressure
regulator on the air supply side.The solinoid would open it fast.

03-08-2004, 02:38 AM
What you should be asking yourself, is why don't I have a multiple airchamber gun that can ACCELERATE the projectile down the barrel by constantly increasing the airpressure? The reason it makes no difference now is because the inital impulse is doing all the work! In the case of a rifle the propellant progressively accelerates it by burning until it reaching its maximum CUP - by this time the bullet has reached maximum velocity in the barrel and is beginning to slow down before it even exits the barrel (in the case of complete powder combustion).

In the case of the pumkin, it has be accelerated to its maximum velocity in the barrel by the inital impulse of air, if additional high pressure air is added after the first impulse further accelleration will take place. The opening of these valves would have to be millisecds behind each other to steadily increase pressure behind the pumkin as the barrel volume increases as the pumkin travels up the barrel.

[This message has been edited by Thrud (edited 03-08-2004).]

03-08-2004, 03:43 AM
Note that the maximum speed of a projectile is
the speed of sound in whatever is shoving it down the barrel...

Why not use a dump valve? Something like
mcmastercarr's 6646K23 might work....

A 3/4" lawn sprinkler valve might work pretty well,

= Bart

J Thornton
03-08-2004, 05:15 AM
In one of the text books I had there was a forging machine that operated on air pressure. It used a system that relied on differential air pressure. it had 2 connected chambers one high and one low pressure The piston was in the low pressure chamber. The pressure on area of the piston on the low pressure side was used to seal a hole in the divider. The machine was triggered by increasing the pressure on the high side enough to over ride the balance. Then the high pressure was applied to the entire area of the piston instead of just the area sealing the hole. That might be quicker than a ball or rotary valve.

03-08-2004, 06:08 AM

Wasn't it a Super-cannon that got the canadian guy kidnapped by the Mossad? Bringing up old history?

Not just accellerate, but to accellerate without rupturing the projectile. A piston type arrangement is what they use to launch people in the circus. A poly piston that fits the bore and is latched in, you pump it up then release the piston. Doesn't a bb gun work the same way? What a big bee bee..

Pneumatics? How about Hydrogen peroxide? Steam, now there is some expansion for you. Launches aircraft off carriers everyday.


Bill Cook
03-08-2004, 06:13 AM
I come up with a comparitively large, light weight, hinged trap door.

Firing would require opening a latch that holds the door closed against the air charge. The door would open completely.

The compressed air chamber and valve assembly would be part of the breech - right up close to the projectile - to get the most out of it.

The door would open against a hole in the side (or top or bottom) of the breech. This would:

A) Tend to hold the door open.
A shock absorbing device may be required to keep the door from bouncing on the atmospheric hole.
Or maybe saayy a traveling weight that would delay and tune the bounce.

B) Allow air to be sucked in when the charge pressure in the unit drops below atmospheric pressure. The amount of air sucked through this hole would give some insite when modifying barrel length, charge pressure, and/or projectile properties.

C) Without it, the decreasing pressure may "pull" the cannon forward after the recoil (if it matters).

D) Allow access to the valve area and the back of the projectile.

E) Allow air to escape when ramming that VW down the barrel.


J Thornton
03-08-2004, 07:20 AM
The Canadian was Gerald Bull and he was murdered in Brussels. Interesting write-up here:


03-08-2004, 09:42 AM

[This message has been edited by pgmrdan (edited 03-12-2004).]

Forrest Addy
03-08-2004, 11:19 AM
Listen to the no-load discharge to get an idea of how fast the launch valve on these pumpkin guns is opening. There should be a POW! All the ones I've heard on TV go CHUFF! The CH...! part indicates to me choked inlet air and the video of the air gage subsiding to zero in a couple of seconds bears this out. The gage should snap to zero.

Butterfly and ball valves, unless opened instantaneously, choke the air so you get reduced accelleration in the first part of the barrel. Butterfly valves and ball valves have significant mass in their working parts.

This is apparent in the Mythbuster episode where they were testing aircraft windshields with 100 MPH chickens. Increasing reservoir pressure had little effect on muzzle velocity. The valve was opening too slow. They were opening it by whacking the handle with a 2 x 4.

Look at quick cylinder exhaust valves for air cylinders from Schrader Air. They make them up to 3" IPS which is too small for a pumpkin gun. You need a valve and supply line from the air reservoir having half or more the effective area of the barrel.


(right column of images 5 down; its a simple thing - three parts and four screws)

Quick exhaust valves designed for use in air cylinders to open rapidly direct to the atmosphere on release of supply air pressure so you don't have to wait for the pressure to bleed off through the connecting line and valve before reverse motion of the piston rod commences. Thanks to their low mass design they open almost instantaneously - explosively it could be said if you've ever been head down close to one that exhausted unmuffled into your ear. I think I still have a bump in my head and it's been years.

They can be used for transfer functions are well like air from a reservoir to the cannon breech. Furthermore they're mechanically very simple being nothing more than a three port housing and a diaphragm and can be scaled up to pumpkin sized.

The beauty of these valves is the active element is a low-mass rubber diaphragm. As I said they scale up easily. Spend $15 and buy a small one and use it to cobble up one having an effective port diameter equivalent to 8" IPS pipe. Remember BIG, HUGE ports. BIG connection lines.

03-08-2004, 11:41 AM
THanks all for the help, Quick Dump valve definetly sounds good to me. The size gun I've been thinking about would be for Tennis ball or potato size loads, use it for a coastal cannon at the lake.
I've done this before with PVC spud guns, fun shooting it, then watching the lake down range to see where it lands.
I also have a little OOpic programmable robot controller that I could easily use to do thruds idea of multiple chambers going off milliseconds apart, even a control for limiting airpressure inside the chamber for precise measurements, perhaps even a gunnery computer. Hmmm...

03-08-2004, 11:54 AM
How about using one valve (or several valves) different lengths of pipe connecting to openings in the side of the barrel? The delay in the plumbing would get the pressure pulses to the side openings just after the projectile passed.

A sliding gate valve could be used. Open it with a large solenoid like this one with over 40 pounds pull at 12 volts, and it's rated at 24 volts DC.



Paul Alciatore
03-08-2004, 11:56 AM
<font face="Verdana, Arial" size="2">Originally posted by ibewgypsie:


A piston type arrangement is what they use to launch people in the circus. A poly piston that fits the bore and is latched in, you pump it up then release the piston. Doesn't a bb gun work the same way? What a big bee bee..




I am thinking somewhat along this way also. If the problem is to open the valve fast enough to have an almost instantenous build up of pressure and this seems to be almost impossible, then why fight it. Just do it another way.

What you need is a well sealed projectile and a lock on it to hold it in place while the valve opens. ONLY when the chamber has built up to full pressure and volume, then release the projectile and let fly. This avoids any delay in the valve altogether.

Paul A.

03-08-2004, 12:42 PM
As winchman says, a sliding gate valve should work. They are readily available for the sewage dump on RVs. I have no idea how much pressure they can take but they don't cost much to try.

03-08-2004, 08:24 PM
While working on my hydrogen peroxide rocket I found a site that the "NUT" used a 2,000 psi bottle to rocket propel a bicycle.

The video showed him opening the valve and the bicycle taking off faster then he could hold on launching the bicycle.

Later development brought forth a gate type slower opening valve where the thrust was more controllable.

Sorry, I lost hte site..


Jason J
03-08-2004, 09:06 PM
RE: Dr Gerald Bull.

Once many years ago I spent an evening sipping a "Blue" or two with one of Bull's right hand men. This was after the Space Research Co had been shut down, but before Bull went to work on the Jerusalem Gun. The whole project was funded by the CIA to produce a super artillery piece rather than a "super gun". Bull had developed an artillery piece that could put a projectile in orbit. Think of the implications of that. Intercontinental artillery. The CIA freaked at the implications of what Bull had done and Space Research closed within days and Bull ended up in jail for doing what he was paid to do. The Israelis had no choice but to stop him, and he knew what was in store for him if he continued. One of the reasons we were so worried about the Iraqui artillery in GW I was that all of their guns were of Bull's design and would shoot MUCH farther than anything we had or have today.

03-08-2004, 10:34 PM
Pure speculation on my part but Paul's "release with full pressure" is a fine idea. but a pumpkin (or any weak object) will press through the release mechanism.

Forrest's rupture disk would be my direction of investigation/design. Use F=MA to find acceleration and keep in mind that with a small reservoir, the force with drop quickly as the volume expands. So what you want is a LARGE reservoir, short barrel (once the force drops to point where no acceleration is taking x place, the projectile should be clear of the tube.

For rapid fire, maybe a slide in gate with a rupture disk for each shot/ and a simple pin through the breech end to start the rupture and make the disk of thinnest possible material so it ruptures fast.

Can't remember the details .but some company made a fast air release valve used to blast grain loose in silos. Written up in "plant Engineering" back the 1970's. Gate valve, slide valve did exactly what Addy says- release so slowly that no blast occurred. They DID NOT use a rupture disk per se but the mechanism was a wide poppet valve with a release trigger. They said rupture disk was best and their valve was a poor but practical compromise needed to allow unattended "blasts".

I would say the volume of air in the reservoir should be at LEAST three times the volume of the barrel. If the volume in the rear is equal to the barrel volume, the pressure will drop from 100 psi to 50 psi just due to expansion so your acceleration is dropping as the projectile travels and will be halved at the muzzle (if rear volume equals front).

. A little math is needed before construction (IMHO). Start with air pressure available, then size and weight of the projectile. Size determines tube size, air pressure gives force (thus acceleration), but (not to stutter ) force falls off as air expands. Some where in those Numbers is a good starting point for valve speeds, and reservoir size.

03-08-2004, 10:58 PM
Well, if the Release is slower than instant, a longer barrel would be needed for excelleration right?
Hmm, I've just been pondering this as a future project, got plenty of time to think about it. I was just thinking about taking a compressed air tank, cutting the front of it off, welding on a plate to mount the dump valve, and a barrel off that. Going to use liquid expanding foam as a sabot.

03-08-2004, 11:32 PM
On Gerald Bull, last I heard his son was so cheesed off at the Mossad for killing his father, he headed off to Iraq to finish his father's work. Wonder what ever became of him?

Bull based his designs on work the Germans had done during WW II under the code name "High Pressure Pump." NASA has been doing some research on a similar setup using LPG of all things as the propellant.

Jason J
03-09-2004, 03:12 AM
Bull's work wasn't based on the German "London Gun" but he did research it early on. Little of value was available after the war apparently. Bull's own pet project which got him in so much trouble was an attempt to put satellites in low earth orbit via his gun. The CIA originally was looking for a way to spy satilites in orbit very quickly with no fanfare and untracable by the normal spy satellites that look for rocket launches. The problem with that is that Pakastan or Libya or Iraq for instance could drop small nukes on any place on earth and no one would know where they were coming from. The Jerusalem Gun could have dropped huge shells on Israel w/o warning and very cheaply, if it had been completed. Space Research still has all of the high speed photgraphy and data locked up, last I heard. Undoubtedly NASA has access to it.

03-09-2004, 12:56 PM
Here is a quick drawing I did of one design I dreamed up of this morning. Requires a lot of welding, I'd think I would buy a Tig welder before tackling this.
Very simple design, not quite sure how much force would be required to open up the dump valve in front by pulling on the silver rod.
50 PSI in the tank, I guess it has a few inches of surface area, I hope that does not translate into over 100lbs of force.

[This message has been edited by BillH (edited 03-09-2004).]

03-09-2004, 01:23 PM
Bill, Put a large spring on the "pull rod" (a truck valve spring would surely do)to assist in pulling the valve back. Add a trigger to release the spring and I think you have it. The force on the valve would be pressure in the large chamber times the area of the valve seat. Should give quick dumping. Use an old CO2 fire extinguisher for chamber. Long skinny rod with "O"rings would lessen the need for accurate seating of the valve. a tapered valve seat and valve would get the lined up well enough so the pressure would seat the valve. not much machine or weld work to do at all. oxy/act would do fine (IMHO).

Go for it and report back http://bbs.homeshopmachinist.net//smile.gif

Paul Alciatore
03-09-2004, 06:12 PM
Sorry, I forgot you were shooting pumpkins. I like the large pressurized chamber idea and the sliding gate valve with a large spring to open it really fast.

Now lets see, get a really large tank for the chamber (one of then really big RR tank cars perhaps) and about 200 or 300 feet of pipe and space shuttle look out. The great pumpkin is in orbit Charlie Brown.

Paul A.

Peter S
03-09-2004, 06:20 PM
I looked at making a spud gun a couple of years ago, there are lots of websites with plans and ideas (eg 'snipers' spud gun, a guy who makes and sells rifled plastic tubing for barrels, also "Spudzooka", The SE2000 (Spud Excellerator 2000, etc etc!) Once you start looking, you will find lots of links.

One idea for the dump valve is to use air pressure to hold it both closed and open. Forget about pulling rods, and trying to seal them to outside etc.

Using your drawing, you make the dump valve a piston that slides in its own cylinder (or the dump valve becomes a piston). This (small)cylinder is charged with air from the main chamber via a non-return valve. Thus the valve is held closed against the end of the barrel by air pressure.

When you want to fire, you operate a small pilot valve (eg an air gun trigger) which lowers the pressure in the dump valve cylinder. Suddenly the dump valve is no longer balanced, the air pressure in the main chamber slams it back, thus opening it rapidly, with no hard work required except pulling a little trigger.

The dump valve needs to be sized right in relation to its cylinder so the imbalance takes place.

The barrel can be extended right inside the main air chamber, so the dump valve cylinder is attached to the opposite end to the barrel opening. The non return valve, trigger etc can all be piped externally with nylon.
Also, the main air chamber can be increased in length for more capacity, the barrel still remains the same length, it can continue inside the main chamber.

Alternatively, I saw various simple dump valves using a simple rubber diaphram and plastic fittings. The diaphram ideas (ie a disc of rubber) look very simple and effective.

A guy I work with has a gas powered potato delivery system, the combustion chamber is made from a air-con compressor housing, aluminium barrel. Charge with air freshener, ignite with BBQ ignitor "Whomph" -piece of citrus dissapears over back wall somewhere into industrial land.

Fishermen use them to fire long lines off the beach.

[This message has been edited by Peter S (edited 03-09-2004).]

03-09-2004, 08:26 PM
Yes, I was studying that balanced dump valve idea, that will be my next revision.
I am not launching pumpkins, the barrel I'm thinking about is 2" OD.
THese are just idea's, for future projects.
I've made a PVC spud gun before, few of them actually, using hair spray as propellent, from what I gather, the Pneumatic ones are much more powerful.

03-09-2004, 09:00 PM
Ok, here is my balanced Dump Valve design, Can use a solenoid for the bleed/fire valve for saftey.

03-09-2004, 10:28 PM
Hey, Bill, you might want to complete your arsenal with a gauss rifle. (Go here for details: http://www.scitoys.com/tech_tv/ ) You could use it to shoot your pumpkins out of the air! You could pumpkin chunk and skeet shoot all at the same time! http://bbs.homeshopmachinist.net//cool.gif

03-11-2004, 05:16 AM
I would add a spring to the piston to force it open faster - never rely on simplicity to do the job right.