Butane Badness




This is awesome: A butane powered mini-gun with interchangable barrels capable of firing a wide range of ammo.

I started playing with this several months ago. I had seen the film canister cannon and wondered about the possibilities. Initially I saw this: Pirate Cubicle Cannon. I then saw several other implementations all using a sprayed fuel, typically Binaca.

I thought that "autoloading" would be better than having to spray fuel into the canister and attach it. I thought that the fuel-air ratio was probably too rich, not too light in the Pirate Cannon but I planned on making a small hole in the film canister to relieve the pressure and allow fuel to flow in if it were indeed too light.

I built a prototype using butane as the fuel, piped in. There was no problem, just a _very_ rewarding *loud* report + fireball followed by insane laughter. Finally I calmed down and began thinking (ir)rationally.

As I thought about making a hole, my thoughts morphed from a small one to a larger one to a barrel: a-la the "ultimate spit-wad shooter".

I tried it and the performance was beyond anything I expected. I used a Bic Round-stik pen as the intitial barrel...hunting for ammo... would you know that Tic-Tacs fit really well in a Bic pen? Yeah and when the tic tac hurtled across the room, bounced of an angled door, down the hallway and ricocheted arround in the bathroom for a while I knew I had something.

I scoured the web looking for anything similar, I found one thing at the BleachSoft website. At instructables, I filtered through all the KNEX guns, marshmallow guns, airsoft guns, etc and turned up nothing. I have however recently seen several similar projects appearing here on instructables:
Q-Tip Gun
Small Dart Gun
Axe Rocket Launcher/ Mini Spud Gun
Mini Spud Gun, Classroom Nemesis
Mini Spud Gun
Now that my fear of being the sole corruptor of the DIY builders of the world has abated, and I will not share sole responsibility for the fallout, I present my work so far.

Step 1: Parts Overview

All parts should be easily found around the home or office. The Ronson lighters came from a local pharmacy. The candlelighter came from a dollar store.

1. Wood.
This could be 3/4" stock, 1" stock, Exotic hardwoods...anything you want. For office
expediency a flat ruler could be employed or better, one of the square cross-section yard
sticks. I'm using 3/4" here. It would be great to combine this: Wooden Luger with the project...mayabe in the future.

2. Lighter to be used as trigger and fuel control
- I used the illustrated Ronson lighter. Almost any would work. Make sure it is refillable.
You will be refilling it. Trust me. Choose one and get several of the same type -- they all
have little differences. Once you've used one type it will be easier in future devices.

3. Film Canisters.
- These are easily available and will be used as the combustion chamber and for
changing barrels for firing a variety of ammunition. Agian, you may want several. I have
not tried the translucent Fuji variety or the oval advantix ones.

4. Fender washer.
- Dimentions are not critical. You could use any round rigid flat piece of material.
This will support the back of the lid as it attaches to the body of the device.
I've used plastic scrap before. You want a rigid plataform that adheres well to the lid
and the device body.

5. Barrel.
- shown are a standard Bic pen and a black tube - this is the spindle from a CD spindle
pack. Other options are chapstik tube, pen barrels of other dimentions; basically any
straight piece of tube of a diameter that matches the ammo you want.

6. Candle lighter.
- From the dollar store. Bascially we are harvesting the wire and the gas tube from this.
I also use the piezo striker as it has the wire already soldered neatly to one of the leads.

7. Wire.
- Used to convey spark from piezo into chamber. Not needed if using wire from
candle lighter.

1. Saw - cutting wood to proper angles / lengths
2. Dremel or other carving tool for channeling the wood to accept the lighter
3. Drill or method of making long thin holes for gastube and wires
4. Soldering iron - for attaching / tinning wires
5. Hot melt glue gun - could use another adhesive if desired

That's it. Move on for the good stuff.

Step 2: Electronics

Now for the piezo setup.

Take apart the lighter and the candlelighter. Below are illustrations.

The idea is to divide the long wire on the candlelighter piezo lighter and attach it to the brass base of the same piezo element. The spark will then jump between the two wire endes. I use a pen spring to hold the wire in place. I have not attempted soldering to this type of piezo element. I have had no trouble with the reliability of this piezo either.

We will be replacing the native ronson piezo with the one we salvaged from the candlelighter. They are the same device as illustrated.

These wires will protrude from the finally almost fully reassembled lighter and be fed into the chamber.

Step 3: The Body

Now we cut the wood to the right size and angles. The excess wood is removed to make space for the lighter to nestle within. This is illustrated in the pictures.

In an office setting the wood could be removed by a patient person with an exacto blade. Or be slowly burned out with an element heated in the lighter flame.

This will vary for the shape / size of the lighter you choose.

You should play with this and do what feels right in your hand.

As in all good instructables, we'll be using hot melt glue but do not glue anything together yet.

Step 4: Gasflow

The gas will be conveyed in the clear tube we salvaged. We must fasten this to the lighter.
I used superglue for the but there are a few gotchas...

Firstly if you just don't have the tube, you can substitute by removing a piece of insulation from a wire and using this insulation as the tube. I have done this using a toothpick to flare out the end of the tube so it would fit over the brass gas outlet on the lighter ( see photos ).

The tube will flex as the gas is delivered because the valve is opened by moving the assembly upward. This means we will have to leave plenty of wiggle-room for our tube as it travels to the chamber.

There is a little sometimes yellow spongy disk surrounding the gas outlet on the lighter. If you get superglue in this which is almost inevitable, the valve may not fully close and you will lose all control of the ratios and lose your gas as well. It may be best to remove this. I have simply been careful and kept it compressed when glue got into in it (i.e. let it close as it normally would--don't hold the valve open as the glue cures).

The joint between the tube and the brass fitting is not going to be strong. It will be gastight, but not physically strong. You will need to be careful in the next steps to protect this.

Step 5: Film Canister Lid

Next we need to prepare the lid.

I sanded the fender washer and scratched up the film canister lid to improve adhesion. I used simple hotmelt glue which has worked very well.

Next we drill the appropriate holes and begin treading the wires / tubes.

Step 6: Final Assembly

This is the tricky part. You want to drill the holes to get you from the trigger area into the breech. As previously stated, make the hole for the gas tube ~3-4 tube diameters to allow wiggle room. Be careful here. It is easy to break the gas tube off the lighter at this stage.

Only after everything is positioned properly do I glue it in place. You can see I again sanded the washer where it would join the wooden body.

Once the wires are through, position them and the gas tube and cut to length. I like to tin the electrodes, but it is not essential. I try to position them in mid-chamber. The gas tube I bend 90 degrees and direct the flow against the upper wall of the chamber. I do this for two reasons. One is because I believe it will result in better gas mixing. The second is that when you have to open the chamber to vent combusted gasses it doesn't interfere as much -- it depends on the motion you use to seperate them.

Okay, so why have the images till now been showing a black ronson lighter and the final assembly shows a white one? Well...I got to this stage before discovering that the black lighter had a cracked base and permanent leak. So I continue with the white one.

Step 7: Barrels

The only trick here is to drill in the center of the film canister. This is not difficult, and may not be necessary. I'd like to play with some offset designs and have begun to do so already.

The smaller the barrel diameter, the greater the need for reinforcement of the end of the chamber. The barrel can wobble too much. Be creative here. I have used plastic scrap glued inside the barrel to essentially thicken the end wall. Also you want some mechanism to prevent your ammo from entering the chamber after it is muzzle-loaded.

I have used pens with various restrictions in them and have also used a copper wire through a diameter of the barrel. This works well while not restricting the combustion outflow like a circumferential stop would.

Step 8: Usage

Firstly, study, memorize, practice, recite, remember:

Gun safety:
1. Every gun is always loaded.
2. Never point a gun at something you do not wish to see destroyed.
3. Keep your finger off the trigger till your sights are on the target.
4. Be sure of your target.

I could tell stories illustrating how each of these have led to negative consequences when violated. Remember them. Practice them always. Read them again.

When I assemble the device I usually set the lighter gasflow lever setting to the middle. I find that three half pulls of the trigger sends in three puffs of butane which is about perfect. I wait or not, and pull fully the fourth time. Showtime.

The performance of this device will amaze you and astound your friends. I can shoot minimarshmallows through both sides of a cardboard box. Printer paper folded a-frame is penetrated through both sides by a tic-tac while remaining standing still. This usually works with two such a-frames i.e. penetrating 4 layers of paper.

I have better pictures...but there is a deadline to meet. Stay tuned.

Tic-tacs to the skin - OUCH. Marshmallows? Yikes.

The longer the barrel, the better the performance - within limits. I have not yet fully explored those limits. Stay tuned.

Step 9: Wrapup

Notes on Ammo

These work great. Chaptstik tube uses them native from the bag...minimarshmallows. I like to dry them out though till they are good and crusty. I carefully roll them to near barrel diameter and use the barrel like a mandrel to "extrude" them. This sets a perfect diameter and works great.

Already discussed, work great, become sticky with time...water is a product of combustion. sticky is not good. They're cheap, buy more.

You would be amazed at the performance of a short length of soda straw with wax vs hotmelt glue plugging the end. Add a sewing pin and you have nice darts.

Wax balls, Hot melt glue tubes, foam plugs, balls of yarn, peas, beans, spit-wads...have a blast. The range is from harmless short range stuff to fairly lethal. The marshmallow can seem nearly lethal with a long barrel. So many possibilities....have fun.

The Future
I like breechloading better than muzzle loading.

I like having the gasflow control seperate from the spark control.

When the projectile fails to leave the barrel---it becomes a film canister cannon. The pressure builds till the canister seperates from the lid. This is LOUD = ouch my ears, especially indoors. Can this be fixed while still having a pressure release mechanism for safety?

I would like some muzzle velocity measurements. I am exploring high-speed photography among other things for this and I would like some nice pictures.

Basically, this is the beginning, stay tuned for more.

Feel free to offer suggestions / improvements / post your own designs.



    • Paint Challenge

      Paint Challenge
    • Planter Challenge

      Planter Challenge
    • Sensors Contest

      Sensors Contest

    203 Discussions


    4 years ago

    So I'm not the only one. I've been tinkering with breach load and .68 paintballs. How have you got on?

    1 reply

    7 years ago on Introduction

    I have been working on a project that involves an Airsoft gun powered by this system but using shells, really just individual chambers, to fire. I just witnessed the raw power of butane, by having a PVC cap shot into my wall. please, don't be dumb like me.


    8 years ago on Step 9

    Wanna make this more interesting you should make a crude blow back loading system so it atleast reloads by gravity, you know like opening a trap door or something but probably will be tricky


    Reply 12 years ago on Introduction

    This is an interesting idea that I also had in the back of my mind. It may acutally be simpler than I imagined it. Thanks, I'll check it out.


    Reply 8 years ago on Introduction

    Another way to find speed is to make and cast some ballistics gel (i think there is an Instructable for this), and fore into it at point blank range. Then, i think, by using some fancy math that im sure you can find somewhere, you should be able to find out how fast its going.


    Reply 11 years ago on Step 9

    how bout a defined shutter time and a scale in the background like 100/S shutter and a metre rule as close to projectile... take start of the blur to the end of the blur on the ruler to find out the distance travelled in 100/S x 100 gives you cm/s and leaving it undivided gives you M/s


    Reply 11 years ago on Introduction

    Alright, I have been experimenting with an ~30 fps video camera. I have collected quite a bit of data but have not completed the analysis. What I have found so far using a marshmallow as the projectile: 1. It is difficult to see in flight against a well lit dark background 2. It can travel 2m in a single frame 3. Frame rate is constant but exposure or "on" time varies with ambient light 4. This means "off" time varies as well 5. one must determine the "on" to "off" ratio for accurate calculations 6. it would be nice to have a "slow" constant velocity object to calibrate this Roughly, I can estimate that we're seeing 91 m/s or ~300 f/s. Stay tuned, detailed analysis to follow along with better performance pics...video?


    Reply 11 years ago on Introduction

    1, 2. Try a banded background - black and white stripes of known width (1cm?) to allow accurate measurement of distances, plus increasing the chance of spotting the 'mallow.

    At 30fps, 2m in a frame = 60m/s, not 91m/s.

    60m/s = 134mph = nearly 200f/s (not 300, but still impressive).

    3, 4, 5, 6:

    Make or buy a clock with a smoothly-moving sweep-hand (as used by crash-test centres) or a simple "bob" pendulum of known length (and hence calculable period).


    Reply 11 years ago on Introduction

    The banded background sounds interesting I'll have to try that. As it is, I've included a measuring tape in the video which I then use to calibrate the image processing software. As for velocity calculations: At 30fps you would be correct if each frame was exposed for that entire time. I was interested to discover that this is not usually the case. Time is necessary to record each frame. Beyond that ambient light can change dramatically--this is dealt with by varying the exposure time of each collected frame. The maximum exposure time would be 1/30th of a second, but under bright lighting conditions this could be significantly less. The meaning of all this is that the data collection rate is 30fps but the actual exposure time of each frame will be somewhat less than that. I'm estimating that this could 50-60% with the setup I'm using. Hence the time scale of the recorded distance per frame is less than 1/30th of a seocnd. I do like the pendulum idea. The need for a known velocity is not so great as the need for a fixed velocity. This will allow easy calculation of the exposure time of each frame as a ratio of the collection rate of 30fps.


    Reply 11 years ago on Step 9

    yeah I used the 100/S as a way of getting a direct measurement, timing would be everything and prefocusing is a must but the video idea might provide more reliable figures say: (travel over 10 frames)/30 = velocity in M/s also using ten frames cuts down on error by a huge amount.

    These are good ideas. Thanks for your patience with this. I have been spending time elsewhere of late, but I am interested in collecting this data. I will try out these suggestions - there are a lot of potential pitfalls, but planning to avoid them will take time. I may get lucky just by trying.

    yeah a decent camera with a reliable framerate or shutter timing is the main thing to get over then timing the firing and capture so video is a better bet


    Reply 8 years ago on Introduction

    An Airsoft bb does not actually fit very tightly in a bic pen tube. Not a big deal, but annoying, as it may roll out.


    11 years ago on Introduction

    Some FYI for the high speed photography,you can buy glow it the dark airsoft BB's which might make tracking the projectile a little easier.