How to Make a Hydraulic Rocket Press

In my experience, there are three basic machines, which become necessary as one gets deeply into fireworking: a ball mill, a star roller, and a hydraulic press. Ball mills were discussed extensively in Fireworks Tips #91, and I showed some options for star rollers in #92. Now it’s time to look more deeply at rocket presses.

Commercial Rocket Presses

In past newsletters I have shown several of my hydraulic presses in action as I’ve made rockets, comets, pressed stars, black powder pucks, or fountains.

Some devices, like simple gerbs or black powder rocket motors, can be made by simply hand-ramming them with a rawhide mallet and a pounding post. Or they can be pressed with a hydraulic press.

Other devices such as whistle rockets or strobe rockets utilize more sensitive fuels, and require hydraulic pressing for their manufacture. Hand-ramming these motors is simply asking for a disaster.

Since I’m about to present projects showing how to make whistles, whistle rockets, and strobe rockets, I thought an introductory essay on hydraulic presses would be in order.

In the previous articles mentioned above I’ve shown my small Hobby Fireworks press. It’s a nice press and was not too expensive. It sits on top of my workbench and is light and portable. I can take it to club events like the PGI convention. Unfortunately, Hobby Fireworks has gone out of business.

Over the years I made some modifications to the press so that it suited my needs better. I replaced the 1/4-inch-thick steel plate on the top of the bottle jack with one that is 3/4-inch thick. The thinner one started bending a bit, and I want that plate to be perfectly flat.

I replaced the original 4-ton bottle-jack with a fast-action 6-ton one. The new jack can exert forces up to 12,000 pounds, which is about all I ever need, even when pressing large 4-inch comets and large rockets. Additionally, the new jack raises very quickly compared to standard-lift models.

The jack is available at Northern Tool, and currently it sells for $28, including a nice, collapsible lug wrench.

When I showed an advance copy of this article to my friend, Dan Creagan, he sent me a photo of a jack he had just found at WalMart, apparently identical to this one except it was painted black and had the Torin Brand name on it. It only cost Dan about $19.

I had the welding shop reinforce the press’s adjustable cross “bridge” so that it would withstand the additional force exerted by the 6-ton jack.

I drilled holes in the back of each horizontal leg-support so that I could bolt the press down to the workbench to secure it during use.

The pressure release knob had a hole drilled through it and a 3/16-inch rod-handle installed for fast and easy operation.

This press has gotten a lot of use over the years, and if I had welding capabilities and I wanted a nice little press, I’d duplicate this model.

There are H-frame shop-presses available at various suppliers. In the past I’ve referred to one such press sold by Greg Smith Equipment. It’s a floor standing unit, and weighs over 150 pounds, but it looks like a pretty nice press, has a good range of adjustability, comes with a pressure gauge, and sells for only slightly more than $200.

Simple Do-It-Yourself, Hydraulic Press

Note: My friend, Jim B, has a favorite saying; “For every 10 pyros, you’ll get a dozen different ideas on how to do any task in pyro.” The ideas I present in the following project are designed to simply get your creative juices flowing. I seriously doubt anyone will build a model that is exactly like mine. But maybe these ideas can point you in the right direction.

In addition to the commercially-manufactured options described above, I got to thinking about a sturdy, relatively lightweight, bench-top press which did not require much welding. I’ll describe one possibility that came to mind.

The setup and operation of the hydraulic bottle-jack is similar to the Hobby Fireworks press. I had six, 4×4-inch, 3/4-inch-thick, steel-plate shims cut at my local steel supply, to augment the adjustability of the press. Between the 5.5-inch travel of the bottle jack’s piston, and the 4.5-inches of adjustability the shims provide, the press has a total of 10 inches of travel between all the way down and all the way up.

This allows me to adjust the press’s top plate with the hex-nuts only one time per the particular device that I’m pressing. I never needed more than that 10 inches of adjustability for any of the devices I make. I’ve tripled-up the top hex-nuts because of the forces they endure during pressing. I don’t want to be stripping any threads.

The Press’s Main Frame:

The main frame of the press is made from four 3/4-inch x 36-inch threaded rods, nuts and washers from Home Depot. I had my local steel supply-house cut the bottom and top plates, which are 12-inch long pieces of 1-inch (thick) x 9-inch steel plate.

One-inch-thick steel plate is obviously very strong, and that strength is necessary to withstand the forces, which will be involved in pressing fireworks devices. I wouldn’t want to use steel plates that are thinner than the ones I used, and I also wouldn’t want to make them larger and spread the threaded rods out farther. This could lead to bending the plates.

The PVC plumbing pipe sections on the threaded rod uprights are there to keep me from cutting my knuckles on the threaded rod as I insert and remove devices in the press.

Installing a Blast-Shield

When I’m pressing whistle or strobe rocket motors, which use pressure sensitive, powerful fuels, the installation of a blast-shield is a good idea. The 1/2-inch thick plastic sheet will offer some protection just in case a motor “goes off” while it is being pressed.

The blast-shield is attached to the press and held in place with 3/8-inch eye-bolts, large fender washers, and hex nuts. Polycarbonate plastic such as Lexan is used in bullet-resistant windows, and serves well for blast-shields.

The other benefit of the PVC pipe on the threaded rods is to hold up the bottom blast-shield, eye-bolt supports.

Jack Return Springs

The two 7-inch springs, also from Home Depot, serve to return the bottle jack to its “down” position when the pressure-relief valve is opened. The top of each spring is attached to the 4×4-inch plate that is welded to the screw-out jack-post. The bottoms of the springs are attached to eye-bolts that are mounted in holes drilled through the bottom steel plate.

I had my welding-shop weld on small hex nuts for the top spring attachment at the same time they welded the plate to the jack post. This was the only welding required in this project.

Pressure Relief Valve Handle

To create an easily-operated handle for the jack’s pressure-relief valve, a hole was drilled through the end of the relief valve. A piece of 3/16-inch galvanized steel rod, bent in an L-shape was inserted through the hole, and the small end was pounded flat on a vise-anvil to hold it in place.

Holes in the Steel Plates

The holes in the top and bottom steel plates were drilled using a drill-press. That was the only large piece of equipment that was necessary in the fabrication of the rocket press. The threaded-rod holes were drilled at 9.5-inch centers, side to side, and 5.5-inch centers, front to back.

The jack is attached to the bottom plate with three, 5/16-inch bolts, which go up through the steel plate and into holes I drilled and tapped in the bottom of the jack. (Threading holes in metal is done with a tool called a “tap.”)

Two extra holes were drilled toward the back of the bottom steel plate, through which bolts will go to attach the press to my workbench. This will make the press nice and steady as I’m pressing rockets.

I used a hand-held grinder to smooth all the edges and corners of the steel plates. Then I primed and painted the plates using spray primer and finish paint.

Hydraulic Pressure Gauge

There are a few pressure-to-force (PtoF) hydraulic pressure gauges available to the pyro-hobbyist community. These gauges employ a one-square-inch-area piston, so they directly read out the number of pounds of force being applied to the item being pressed.

For example, if the PtoF gauge is reading 2000 psi, the actual force being applied to the tooling is 2000 pounds, the equivalent of 2000 pounds of concrete sitting on top of the tool.

An advantage to using one of these gauges is that you won’t need to install a pressure gauge on the press’s bottle jack.

I personally like to use a gauge that is actually installed in the bottom of the bottle-jack. Doing so enables me to eliminate one loose, movable component, like the PtoF gauge, when I’m aligning and pressing devices in the press.

Using a gauge on the jack, though, requires that the gauge’s reading be multiplied by the area of the jack’s piston, in order to determine the actual force being exerted by the jack. I’ll show what that means in a minute.

Installing a gauge on the bottle-jack presents what is probably the most challenging aspect of this project–drilling and tapping/threading the bottom of the bottle-jack, and installing a hydraulic pressure gauge. But, it’s good to know how to do this, even if a PtoF gauge is going to be used.

Installing a gauge on a bottle jack requires the partial disassembly of the jack, drilling a couple of holes, tapping/threading the hole where the gauge will be installed, cleaning debris out of the jack, and reassembling it.

One of the nice things about the bottle-jack I’m specifying in this project is that it is relatively easy to take apart and put back together.

First, the rubber drain/fill plug on the back of the jack’s cylindrical body is removed, and the oil that fills the jack is emptied into a clean pot. This oil can be filtered through a coffee filter and reused in the jack when it is reassembled.

When draining the oil, it helps to remove the pressure-relief valve. This valve has a 1/4″ steel ball bearing down in the hole into which it is screwed. Carefully set the ball and valve aside, and finish draining the oil. Pumping the lever assembly a few times works the rest of the oil out. Now is a good time to drill the hole in the pressure-relief valve and install the L-handle.

The lever-arm has a couple of steel pins, held in with spring-clips, and is easily disassembled. (You are making mental notes of how all this goes back together, right?)

It’s time to remove the large hex-nut at the top of the jack now. This requires that the base of the jack be held securely in a vise or a rocket press. (Waitaminnit, I’m making my rocket press! How can I hold the jack in my rocket press? I have 3 presses, and this will be my fourth.)

The hex-nut is then loosened with channel-lock-pliers or a large pipe-wrench. It may be necessary to whack the wrench with a rubber mallet or similar heavy object. The nut is screwed off when it is loose, and the central jack piston and outer jack shell-body can also be removed. The nut has a plastic O-ring gasket on it where it hits the main body, but this gasket is usually “glued” on with paint and does not need to be removed.

There is a “tapered” large rubber O-ring which sits in the groove that the shell-body came out of. Remove this O-ring. Remember that it was in there with the thin edge up, and the wide edge down.

Inside the jack, there will be a small, wire-mesh filter shoved in one of the holes in the base. Make a note of which hole it’s in, and then remove it. Actually, this is a good recommendation, which has never worked for me in real life. Each time I’ve disassembled a jack, the filter has dropped out before I get to notice where it was in the first place. I’m not sure how they get the darn thing to stay in during shipping and/or operation.

I’ll show in a moment how to determine which hole the filter ought to go back into when the jack is reassembled.

The screw-post will only unscrew so far as it extends out of the jack’s piston. It is not necessary to remove this screw-post all the way. The whole jack can be taken to the welding shop when the 4×4 plate is welded to the screw-post. If one wants to remove the post all the way, some filing/grinding is necessary to remove the small “indents” which have been knocked into the top of the cylinder to hold the post in place.

Now is a good time to measure and make note of the diameter of the bottom of the piston. In this case it measures 1.375 inches. Squaring half that diameter (the radius) and multiplying that by Pi (3.1416) yields an area of the bottom of the piston of 1.5 square inches. (3.1416 x .6875 x .6875 = 1.5 square inches)

Because of that, when my new, jack-mounted pressure gauge is reading, say, 1000 pounds-per-square-inch (psi), I’ll multiply that gauge’s reading by 1.5 to determine the actual amount of force the jack is exerting on the tooling in the press, which in this example would be 1500 pounds.

Once again holding the base of the jack in a vise or rocket press, I now carefully use a pipe wrench to loosen the jack’s inner cylinder. I apply the wrench right down at the bottom of that cylinder in order to avoid crushing or distorting the tube as I loosen it.

Once the inner cylinder has been removed, another plastic O-ring gasket can be seen inside the base where the cylinder bottoms out. This O-ring does not need to be removed. Notice that there is a top and a bottom to the inner cylinder. The top is beveled on the inside lip to make insertion of the piston easy. The bottom has a flattened edge, which bears on the O-ring seal.

The small lever-operated jacking piston/cylinder should also be removed at this time. There is a metal washer and a 1/4″ steel ball down in the base’s recess which should also be removed.

And, now, finally we’ve arrived at the final disassembly step. There is another 1/4-inch metal ball in the bottom recess of the base, held in with a plastic retainer. This can be seen in the base’s large recess in the photo above. The retainer is removed by prying it with a screwdriver, and the ball is also removed.

I’m keeping all the little parts in a clean paper cup to prevent me from losing them as I go along.

There is also an over-pressure, safety relief valve, covered by a plastic cap. This assembly, including the cap, screw-out post, spring, metal-mushroom, and very small metal ball, is all removed and placed in the paper cup.

I can just hear ya hollering, “Crikey, Ned, what the heck have you gotten me into?”

It’s really not as bad as it all sounds and looks. If you keep track of all the little parts, and remember how they all go back together, this can be fun. Really! There’s learnin’ happenin’ here.

At this point, for my own education, I spent a bit of time envisioning how the jack works when it is being operated. The small jacking-piston and cylinder create high pressure using the principle of mechanical leverage. The pressurized oil is forced through the small hole in the bottom of that recess and up past the ball/hole/retainer in the large base recess.

All those balls in this device simply act as valves, sitting in nicely machined recesses, and only allowing oil to flow in one direction, pushing the ball slightly out of its recess. Oil pushing from the other direction forces the ball against the machined seal and shuts off the flow.

As it is needed, more oil is “sucked” into the small base recess from the main reservoir between the outer jack body and the inner cylinder.

The pressure in the cylinder jacks the piston up a small amount. The process is repeated as the piston gradually is lifted.

If too much pressure is generated inside the main cylinder, the oil can push the small ball and spring in the over-pressure relief valve and allow the excess oil to escape back into the main oil reservoir between the outer jack body and the inner cylinder. This acts as a safety to prevent the jack from being over-pressurized and dangerously rupturing.

And finally, when we want the jack to retract and go down, the pressure-relief valve is loosened. This allows oil to move past the ball at the bottom of that valve, and back into the main reservoir.

Since the only hole through which oil moves out of the main reservoir is the one leading to the bottom of the jacking-cylinder’s small recess, that is the hole that the small filter will be replaced into (so it functions to remove debris from the oil as it circulates). I find which hole that is by blowing into it to make sure the air is coming out of the ball-blocked hole in the bottom of the small base recess.

And, keeping debris out of the whole jack is why I’ve completely disassembled it. After the next drilling and tapping steps are completed, all the parts will be completely cleaned before any reassembly. Small bits of metallic debris are the enemy of a properly functioning jack. They can become lodged in the various ball-valve assemblies and allow slow leakage, preventing optimal performance.

Drilling and Tapping the Jack-Base to Receive the Pressure Gauge

You’ll notice, when looking at the jack base, that all the existing holes and inner “channels” that the oil flows through are located in the right side of the base.

Conveniently, this jack’s base has a nice flat area on its left side, and plenty of room on the left-inside of the large recess where a hole can be drilled.

This is the point we’ve been heading toward. I want to drill a 3/16-inch hole down from the bottom-inside of that main base recess, but not all the way through the base. I drill this hole about 3/8-Inch deep.

I want to drill in from the left-outside of the base with the same 3/16-inch drill bit, until that hole hits the first hole that was drilled. I only want to drill as far as that first hole so that I don’t hit any of the other inner channels in the base.

The hole coming in from the side is drilled high enough from the bottom to allow the fittings I’m going to install later to clear the press’s base plate. I also plan that hole so that it is centered in the bottom “thickness” of the base, so that the strength of the remaining metal surrounding my new fitting is maximized.

Drilling this hole, centered up 1/2 inch from the bottom of the base, accomplished all the above goals. And it kept the metal thickness between the hole and the bottom of the base no less than 5/16 inch, which is needed to withstand the internal jack pressures.

These two holes are gradually deepened until they hit each other, and no further.

The two holes will form a new channel which will allow the pressurized oil inside the inner jack cylinder to reach the new gauge which will read out the same pressure that exists inside the cylinder.

Warning: The main power tool I’m using in this process is a drill-press. Like Norm Abrams says, “Read and understand the safety precautions concerning this tool before you use it.” I do this drilling at low speeds. I firmly hold the piece I am drilling with a clamp and/or other tools. This drill-press can be my best friend, or it can slice my hands open and/or break bones. Be careful.

The hole in the side of the base is enlarged with a 5/16-inch drill bit (Q drill bit) enlarging a section about 3/4-inch deep. This side hole (only) then has threads cut in it with a 1/8-inch-pipe-thread tap.

This is also a good time to drill and tap the bolt holes, in the flanges on the base, which will attach the jack to the press’s bottom steel plate.

There, the hard part, the machining, is done. I now clean all the debris, excess paint, and metal shavings off of all the parts in a pot of clean kerosene or paint thinner. I pay special attention to the base to make sure all the small metal debris has been washed off of it and out of all its holes and channels.

After the parts have dried, the bottle jack is reassembled in the reverse order in which it was taken apart. Before adding the oil back into it, I attach the new pressure gauge using hydraulic fittings and Teflon tape. My local hydraulic-fitting supply-house was able to supply the fittings that I needed, and which would handle the pressure the jack will be exerting.

These fittings were inexpensive, and it pays to use fittings certified for hydraulic pressure, rather than plumbing fittings which might rupture under that pressure.

The gauge sells for about $22 at McMaster-Carr. It is a 2.5-inch diameter dial, glycerin filled, 0-10,000 psi range, 1/4-inch pipe-thread bottom-connected, Model #4053K16.

But, the same supply-house where I bought the fittings, had a very similar gauge for only $16. I bought one for a spare while I was there.

I have temporarily hooked up gauges to lower-pressure jacks with iron pipe fittings. But those plumbing fittings are not rated for the 8000 psi that will be developed in this new jack when it is putting out the full 6 tons of force.

Remember that when the gauge reads 8000 psi, that reading is multiplied by 1.5 to determine the force that the jack is exerting. That means an 8000 psi reading equals 12,000 pounds of force, the maximum force this jack is rated for. That’s why I chose a gauge with a range of 0-10,000 psi.

The Teflon tape is carefully wrapped on the pipe threads, in the direction that will tighten the tape wraps as the male threads are screwed into the female fittings. 4-5 wraps of the tape are put on each threaded section. I’m careful not to overlap the tape down onto the end of the fittings, where bits could break off and clog the channels or valves in the jack.

After the gauge was installed and all the fittings tightened up, I filtered the oil through a coffee filter and filled the jack back up with the oil through the fill hole on the back of the jack’s body. I pumped the jack up and down a few times to work any trapped air out of the system. Then the jack was installed on the rocket press.

I topped the oil off with more, new hydraulic-jack oil until it started to run out of the fill-hole in the main jack body. Then I installed the rubber plug.

I put my Pressure-to-Force gauge on the jack-plate, and jacked the press up to various pressures. This was to make sure that, indeed, the PtoF gauge read 1.5 times what the gauge on the bottle jack was reading. I also removed the PtoF gauge, and jacked the press up to its maximum pressure and let it sit there for a while to make sure it wasn’t losing any pressure through leaks or badly sealed steel-ball valves.

Everything worked great, so I moved the press into its permanent location on my work bench and attached it there with bolts which go through the two extra holes in the back of the bottom steel plate, and on through the workbench top.

Conclusion

Great! My fourth press is now up and running. Why the heck do I need four presses? I think I’ll paint and clean up my old Hobby Fireworks press and see if I can find a new pyro who wants to give it a good home.

There’s one final thing I thought about this press when it was done: “Hey, I built that thing. I know every part of it, and if anything goes wrong with it, I know how to fix it.”

I bought an extra bottle jack while I was working on the project, and drilled and tapped it at the same time as the main one. That way I have replacement parts, or the whole complete replacement jack if necessary.

This all results in a good feeling, and I suppose that’s why I do all of this in the first place.

Stay Green, and now on to more, ahh, Pressing matters.

Ned

How to Make Flashing Fireworks Strobe Pots

Introduction

Close your eyes and listen to this music. What do you see when you do so?

Click here to listen to The Who

If you don’t see a large fireworks mine-shot, followed by a line of 30-second strobe pots, ending with another large mine-shot, then you really need to be subscribed to the Quilting-101 newsletter instead of this one on making homemade fireworks.

Man, that music gets me in the mood for strobes. The first mine-shot would grab the attention of any fireworks-display audience. Then the soft and subtle section of strobes would calm them down and get them ready for their emotions to build during the show.

Strobe pots are among the simplest of fireworks devices and are easy to make. They can really add some of that low-level variety to a pyro-display that so helps to keep an audience’s attention.

“Hey, here’s something different,” they’ll say to themselves as they stop, settle in, and start to pay attention.

How do these pyrotechnic “twinklers” work?

It is not necessary, of course, to have a scientific understanding of strobes in order to make them. Like baking a loaf of bread, chemistry is not necessary. All you need is a recipe, the right ingredients, and a feel for the proper ways to manipulate those ingredients.

But, for the scientifically minded, there are a few informative resources, which explore the strobe phenomenon in depth. In the 1979 edition of Pyrotechnica, Number 5, Robert Cardwell, the editor and publisher of the Pyrotechnica series, wrote an article, Strobe Light Pyrotechnic Compositions: A Review of Their Development and Use.

In this essay, Cardwell explores the historical development of strobing compositions and presents quite a few different formulas.

Dr. Takeo Shimizu, in Fireworks, the Art, Science and Technique (FAST), originally published in 1981, writes about “Twinklers,” which is how he refers to strobing stars. He presents an outline of the development of these strobing compositions, progressing during the second half of the 1900’s.

Specifically Shimizu writes, “In Germany, U. Krone and F. W. Wasmann suggested that a twinkle composition consists of two kinds of compositions mixed with each other, i.e. a smolder composition and a flash composition-Ammonium perchlorate smolders when it is mixed with a small quantity of magnesium. This can be used for the smolder composition. A mixture of magnesium and sulfate flashes when it is heated to a high temperature. This can be used as the flash composition.”

So, interestingly, a strobe composition is actually a mixture of these two types of comps, a smoldering one and a flashing one. When the mixture is lit, the first one begins to smolder. When the heat rises high enough, the flash comp ignites and emits a flash of light and heat. Then the mass returns to the smoldering state until the heat rises high enough to repeat the flash.

In some compositions, magnesium-aluminum (magnalium) is used instead of the magnesium. Magnesium requires a coating to prevent it from prematurely reacting with the oxidizer in the comp.

Additionally, sometimes barium nitrate or other oxidizers are used instead of ammonium perchlorate.

In 1987, John “Skip” Meinhart offered some details about his noteworthy strobing star formulas in Pyrotechnica XI. Except for Shimizu’s White formula, and Skip’s Pink formula, all the rest of the formulas use magnesium as the metallic fuel ingredient.

In the 1992 Pyrotechnica XIV edition Jennings-White explores Blue Strobe Light Pyrotechnic Compositions. Up until that point in time, blue strobes had not been explored in depth because of some unique problems associated with the chemical mixtures required to produce that color in a strobe.

All of this information ought to be able to keep you reading until late into the night if you are so inclined.

Making strobe pots

I won’t be focusing on making strobing stars in this project, but only simple, ground-effect strobe pots.

I’m also not going to be making any of the formulations, which contain magnesium. As I said, using that metal requires a special coating process because it does not form an oxidized protective layer on its own, as do aluminum or magnalium.

There seems to be some debate as to whether or not magnalium needs to be treated and coated when it is used in compositions containing ammonium perchlorate. Meinhart states, “I have had success using magnalium powders that have not been treated with potassium dichromate. In practice I have often used treated metal powders, but this does not always seem to be necessary.”

Whereas in Hardt’s Pyrotechnics, Barry Bush notes that the formulas he cites which contain magnalium or magnesium in combination with ammonium perchlorate do “require the metal powders used to be treated with potassium dichromate.” Shimizu also specifies treated magnalium, and details the methods of treatment in FAST.

Shimizu does state that if there is any reaction between magnalium and ammonium perchlorate, which would be encouraged by the presence of water, it would only be a slow reaction in which the metal is affected gradually.

I have used untreated magnalium in these formulas, with no problems. One sign of an unwanted reaction would be the heating-up of the composition as I’m working with it, so I always pay attention to see if that is occurring. I avoid adding any water to such a composition. I also don’t store these devices for long periods of time, which could produce a slow reaction of the ingredients, especially in the presence of moisture.

So, I think I’ll make simple white and pink strobe pots. The white formula is the most commonly cited one:

White Strobe Composition

Chemical	Percentage	16 Ounces	450 grams
Ammonium perchlorate	0.57	9.15 ounces	257.1 grams
Magnalium*	0.24	3.8 ounces	107.1 grams
Barium sulfate	0.14	2.3 ounces	64.3 grams
Potassium dichromate	0.05	0.75 ounces	21.5 grams

* Shimizu specifies 80-mesh, whereas other sources specify 100-200-mesh. The mesh of the metal is known to vary the flash rate of the strobe, so some experimentation is in order. Initially, I’ll be using 200-mesh magnalium, Skylighter #CH2072.

Barry Bush has an interesting note in Pyrotechnics concerning this formula. This formula “may be given a faster frequency by replacing the barium sulfate with anhydrous magnesium sulfate. The resultant fast strobe is sometimes called a “shimmer effect.” I’ll have to try this sometime in aerial-shell strobe-stars, since it is an effect I have admired in commercial shells.

Additionally, the flame created by this “white” composition is brilliant, but it does have a very slight green tint caused by the barium. Barium normally produces very green flames with the addition of a chlorine donor such as parlon or saran. Another experiment would be to include small amounts of these chlorine donors to shift the color of the white strobe pots to green.

The pink strobe pot composition is as follows:

Meinhart Pink Strobe Composition

Chemical	Percentage	16 ounces	450 grams
Ammonium perchlorate	0.57	9.15 ounces	257.2 grams
Magnalium, 200 mesh	0.15	2.45 ounces	68.6 grams
Strontium sulfate	0.11	1.85 ounces	51.4 grams
Strontium carbonate	0.08	1.2 ounces	34.3 grams
Parlon	0.04	0.6 ounces	17.1 grams
Potassium dichromate	0.05	0.75 ounces	21.4 grams

All the chemicals (except the magnalium, which I don’t put through fine screens) are fine enough to pass through a 100-mesh screen. If they are not, they are milled individually in a blade-type coffee mill.

Note: Ammonium perchlorate does not play well with potassium nitrate. The combination forms ammonium nitrate, which is very hygroscopic, attracting moisture out of the air like crazy, rendering any mixture or composition containing it wet and useless. Don’t grind either of these chemicals in a coffee mill which has been used on the other chemical, unless the mill has been thoroughly cleaned with soap and water.

Warning: Potassium dichromate is toxic and a known carcinogen. A good respirator and rubber gloves are required when working with this chemical, and when using it in pyrotechnic compositions. Don’t breathe this stuff or get it on your skin.

All the chemicals for a given formula are weighed out individually and are passed through a 20-mesh screen 3 times to thoroughly mix them.

Then the composition is mixed with enough nitrocellulose (N/C) lacquer (Skylighter #CH8198P) to create thick putty, similar to Play-Do. I did not dilute the lacquer, but used it right out of the can, as-is. The one-pound batches required 3 ounces, by weight, of the lacquer.

I started mixing the composition in a plastic tub with a paint stir-stick, and finished by kneading it with gloved hands.

The dough is then pushed with gloved fingers into paper tubes to create strobe pots. I start this process by pushing the tube into the composition-putty to get the filling started.

Large pots can be made with 1.5-inch ID tubes, cut into 1.5-inch long sections. Or, smaller pots can be made with 3/4-inch ID tubes, cut into one-inch long sections, or even longer. While thicker-walled parallel tubes, like rocket tubes can be used, strobe-pot tubes do not need to be super-strong, so spiral-wound tubes like Skylighter #TU2142 or TU2053 can be used.

Large diameter strobe pots would be appropriate for large displays and venues. Smaller ones are nice in backyard size shows. Varying the length of the paper tube will adjust the total burn-time of the strobe pots, so their duration can be dialed in for specific uses.

For this project, I think I’ll make mostly 3/4-inch ID by 1-inch long strobes to determine how well they are working and how long they burn, plus a few other sizes to see how they perform, too.

Once the composition has been stuffed into the paper tubes, they are placed on their sides and set aside on a tray to dry out in the open air. N/C lacquer releases acetone and other highly flammable solvents as it dries, and I don’t want these vapors collecting in my shop as this occurs.

Toward the end of the tube filling, the remaining strobe-putty started to dry out and became difficult to consolidate into the tube. I added just a touch of acetone to the composition to re-dampen it.

It took 3 or 4 days for these pots to dry completely. When I tried to burn them before they were completely dry, they did not burn with a regular strobing-action, but with a more continuous flame.

Priming the strobes

The dry pots will light well if they are ignited with a piece of Visco-fuse or with a propane torch. But if I want them to ignite reliably with a fast-fuse or quickmatch line of fuse, then I need to prime them.

A black powder prime containing potassium nitrate cannot be used on these compositions because of the incompatibilities between the nitrate and the ammonium perchlorate.

In FAST, Dr. Shimizu lists a different prime specifically for this use.

Ignition Composition for Twinklers

Chemical	Percentage	16-Ounces	450 Grams
Potassium perchlorate	0.74	11.85 ounces	333 grams
Red gum	0.12	1.9 ounces	54 grams
Charcoal, airfloat	0.06	0.95 ounce	27 grams
Potassium dichromate	0.05	0.8 ounce	22.5 grams
Aluminum*, flake 100-325	0.03	0.5 ounce	13.5 grams

*Skylighter #CH0174 aluminum would fit the bill in this prime.

After making sure all the individual chemicals (except the aluminum) will pass through a 100-mesh screen, I weighed them out individually and mixed them together by passing them through the 20-mesh screen three times.

I weighed out 1 ounce of the dry strobe prime composition, and added 1 ounce (by weight) of the nitrocellulose lacquer. This created a wet prime comp that had a consistency between that of honey and peanut butter.

I used a wood stick to apply this wet prime to one end of each strobe pot, and quickly pushed that wet end into some dry strobe composition for the final prime layer.

I perform one final operation to finish the individual strobe pots. I hot-glue a paper disc onto the bottom of each pot. This prevents sparks and/or slag from dropping and igniting the bottom of a twinkler prematurely as the pot burns. It also facilitates mounting the pots to a board when a show is being set up.

Mounting and fusing strobe pots for use in a fireworks display

Once the individual strobe pots have been completed, they can be mounted to a board and fused for easy installation out in the field prior to a fireworks show.

To do this, I simply hot-glue the pots to a board at the desired spacing. I find a spacing of 4 feet on-center to work well. Then a run of quickmatch or tape-covered fast-fuse
is used to fuse all the pots together. A “window” is opened up in the quickmatch-pipe, and the bared black-match is taped on top of the strobe pot with 3 wraps of masking tape.

The quickmatch can be ignited by a piece of Visco-fuse, or an electric igniter can be employed, per the information in How to Make Electric Matches and Wiring Fireworks and Firing Systems in a Fireworks Display.

Results

I burned white and pink strobes made with the 200-mesh magnalium. The white one burned for 15 seconds with a very fast strobe rate of about 10 flashes per second. The pink one actually looked red, burned for 23 seconds, and flashed about 4 times per second.

Warning: These strobe pots burn with an extremely brilliant flame and light. It is best to avoid looking directly at them to prevent eye damage. Placing the pots where their light can reflect off of a structure or trees makes their effect visible without having to look directly at them.

Click here to see a video of the white and red strobes.

I liked the performance of the pink/red strobe, but the white one flashed too rapidly for my taste.

So, I made a new batch of each color using 60-mesh magnalium. I know that using a larger granulation of the metal will slow down the burn time and also its strobe frequency.

Burning these new strobes produced the following results:

White strobe with 60-mesh magnalium, burned for 25 seconds, and flashed 1.5 times per second. I found this to be a very pleasing strobe frequency.

Red strobe with 60-mesh magnalium burned for 27 seconds, flashed at a rate that varied from slow to fast. This pot just couldn’t seem to find a groove and settle into it.

Check out the video of these two types of strobe pots:

Of the four variations I prefer the white strobe pots made with the 60-mesh magnalium, and the pink/red twinklers made with the 200-mesh magnalium.

Although I made mostly 1-inch long twinklers, I also made some larger ones. Two-inch long ones, made in the 3/4-inch ID tubes, burned as follows:

• 2-inch white strobe with 60-mesh magnalium burned for 40 seconds with about 2.5 flashes per second,
• 2-inch long red strobe with 200-mesh magnalium burned for 40 seconds with flashes varying from slow to fast again.

And, last but not least, I rigged up some white strobe pots using 60-mesh magnalium on a board and accompanied them with the music I linked to right at the beginning of this article. You can get the idea of what I had in mind in the first place as a nifty addition to a fireworks display. Click the video below of the three white strobe pots accompanying Who’s Won’t Get Fooled Again.

I do like what these simple, low-level ground devices can contribute to a fireworks show.

Enjoy,

Ned

How to Make a Model Ammonium Dichromate Volcano

A while back I received an email from Harry:

“Have you ever made an ammonium dichromate volcano? They are really popular with parents and teachers and look great daytime or night.”

I responded: “Never learned how to make a volcano. Led a deprived childhood. No way they’re doin’ anything like that in science classes nowadays.”

Making a volcano would, of course, be exactly the sort of school project which would capture the imagination of so many young students.

Harry shot back: “Neat little effect. The ammonium dichromate burns by itself, but you can add other fuels to make more lifelike ‘lava.’ For instance, charcoal 10 or 20-mesh. I imagine there’s some flake titanium, ferro-titanium, etc. that would work as well. You could provide volcano formulas for high school projects.”

Well, I was intrigued. I’ve made 16-inch ball shells and 36-inch girandolas, but never an ammonium-dichromate volcano. I couldn’t let that stand for long. So I asked Harry to send me a few tubs of the ammonium dichromate, and I put it on my to-do list.

In the meantime, I did a little research on the Internet, and also found a couple of paragraphs about the volcano in the Skylighter Project Plans pages.

Warning: So, first off, a bit of a warning: Ammonium dichromate is rated as a hazardous toxic chemical on its MSDS. One should not inhale its dust, but that’s not too tough to avoid since it comes as relatively large orange crystals, not a dust that gets easily airborne.

One should not ingest it. Well, Duh!

One should avoid skin contact. Wear rubber gloves if you’re gonna touch the stuff (which is not necessary to create one of these volcanoes). Avoid eye contact. The stuff can cause cancer.

So, exercise appropriate caution with this chemical. If you’re going to be making a volcano like this for a science fair, perform all actions and experiments with it outdoors, and when the volcano is “erupting” don’t allow anyone to breathe the small amount of smoke and ash, which rises off of it.

Use common sense. I know, I know. Common sense is not in common use nowadays to protect folks from harm. Since everything even remotely dangerous is becoming illegal, folks are increasingly being born with no common sense chip. But, use common sense anyway. There, you see? You got me started.

The scientific stuff: science for kids – old and young

I’m not much of a chemist, but I did find this information interesting, and some other folks might as well.

Ammonium dichromate is sometimes referred to as Vesuvian Fire due to its use in the creation of these small volcano replicas.

Ammonium dichromate’s formula is (NH₄)₂Cr₂O₇.

When it burns, it decomposes according to the following equation:

(NH₄)₂Cr₂O₇ (solid) -> Cr₂O₃
(solid) + N₂ (gas) + 4H₂O (gas)

The gaseous byproducts are simple nitrogen gas and steam, so they are innocuous.

The solid product is chromium (III) oxide. What remains after the volcano burns is this solid, grayish-green ash. The chromium in it is toxic and possibly carcinogenic, so care should be exercised when handling it and disposing of it. Some of this ash flies up into the air when the volcano is burning, so you don’t want anyone close enough to it to breathe the stuff.

The chromium oxide can reportedly be used in a thermite reaction to produce elemental chromium metal, so I’m going to save the chromium oxide ash to try to use it in such a reaction, since I am planning an article on thermite reactions.

A volcano project

I decided to go simple, and make a “volcano” out of a can and heavy-duty aluminum foil. I formed a lip at the bottom of the aluminum foil “cone” to catch the ash as it formed.

I lit the top of the ammonium dichromate with a propane torch, and it burned and created a “volcano” effect, but at times the flame went deep into the orange crystals, down the sides of the can, and propelled some of the crystals up and out of the can, unburnt. I was not completely pleased with the effect that was produced.

So I decided to try a shallow, tuna-fish can, sitting on top of the pop can.

This volcano burned a little better than the first one, but still the flame worked its way down the sides of the can, and the ammonium dichromate did not burn evenly. I still was not satisfied.

I also noticed that igniting the ammonium dichromate with the propane torch was not all that easy. The pressure of the propane blew the crystals out of the way of the flame, and although some of them eventually lit and initiated the desired continuing reaction, the pile did not ignite right at the top of the heap.

So, I decided to try something really simple. I made a “tray” out of flat aluminum foil with the edges turned up to catch the ash. I poured a cone of about a half-pound of the ammonium dichromate in the middle of the foil tray, and inserted a 6-inch length of Visco-fuse into the middle of the pile of orange crystals.

Once the Visco fuse burnt down into the cone of crystals, they ignited and a nice volcano-action formed and burned until only ash remained.

Well, then I thought that the way to get this baby really looking like a volcano, but still burn like it did in this last test, was to form the aluminum foil over the cans, but not cut the opening. I just piled the ammonium dichromate on top of this “mountain” on the flat section and inserted another piece of Visco.

I was really pleased with how this volcano looked and performed.

I think it almost looks like the real thing with glowing, flaming lava flowing down the side of the Mt. St. Gorski.

Making model volcanoes this way uses about 1/4 pound of the crystals per volcano.

Nighttime volcanoes

I thought I’d make a few other versions of the ammonium dichromate volcano at night.

For the first one, I repeated the procedure I followed in the last daytime test. This looked pretty cool as the “lava” burned and flowed down the sides of the “mountain.”

Honestly, I like watching the volcano better in the daytime when I can see the greenish ash being formed and flowing down the sides of the hill.

For the second nighttime volcano, I did the same as above except I stirred in a tablespoon of 36-mesh charcoal to see what sort of effect that would produce.

As this volcano burned, a lot of orange, charcoal sparks rose up into the air above it, creating an effect that was different than the standard, ammonium dichromate-only volcano.

For the third nighttime test, I mixed in about a tablespoonful of medium flake aluminum, and for the fourth and final test I added about a teaspoon of fine spherical titanium.

The addition of these metal fuels seemed to slow down the burning reaction, and did not produce the silver-white sparks, that I was half-expecting. There was an interesting, different sort-of molten flowing effect, as the reaction seemed to almost melt the ammonium dichromate and metal together, causing molten “lava” to flow down the mountain.

As a last offering to the pyro-gods, I simply poured about a half-pound of the ammonium dichromate onto a bare patch of ground and lit it with the propane torch.

This caused the little volcano crystals to ignite from the sides and burn toward the center, and this was probably my favorite effect of the night. (After it, I had to get a shovel and completely clean up the green ash from the area.)

Conclusion

This was a fun little science project. I saved some of the ammonium dichromate for when my grandkids visit, and I’ll introduce them to something they’ll probably never have the chance to see in science class. And there’s enough left to show them how to make a school project volcano just in case that time comes.

I learned something today working on this model volcano, and that’s never a bad thing.

Take care and have fun,

Ned

How to Make Chinese Sky Lanterns

Can mere mortals make sky lanterns?

I gotta admit, I’ve seen sky lanterns around for several years. I’ve seen ‘em advertised right here by Skylighter. Have seen them at displays and club events. But, honestly I never gave them much thought, and I wasn’t much tempted to buy any.

But, then I saw the latest Skylighter ad for them, and I thought, “I wonder how sky lanterns are made.” Ya see, one of the things I enjoy most is learning how something is made. I watch This Old House and The New Yankee Workshop, just to see how they do stuff, even though I’ve been doing that kind of work myself for a living for 30 years.

So, I had Harry send some of the flying lanterns to me. By the time they arrived on my doorstep, I was excited to see them. I opened the package up, looked at the treetops to make sure there wasn’t too much wind blowing, and I hustled my wife and granddaughter outside to fire the first one up.

I enjoyed finding out how to light the paper lantern, and how to let it inflate and launch it. And, the three of us really did have fun watching it lift off, and then gazing at it for minutes until it flew out of sight and we couldn’t see it any more. Then I quickly got another one out and launched it as well.

My son and his family came over last Sunday, and I knew I just had to demonstrate these new toys for my grandsons. The photos below say it all.

So, now, back to my original quest: How to make sky lanterns.

Note: I’m gonna tell you how I ended up successfully making these homemade paper hot-air balloons. But, as with any pyro project, I learned some lessons the hard way, and I had some significant failures. I’ll note these as I go along in this hot paper tale.

Sky Lantern Autopsy

A little reverse engineering revealed:
Weight of a flying lantern	2.7 ounces
Thin bamboo hoop at bottom	45.75″ circumference, 14.56″ diameter
Weight of hoop	0.3 ounces (with a little paper and glue still hanging onto it)

Bamboo is about 1/16″ x 1/8″

X of thin wire tied to hoop

X of wire is “woven” through waxed cloth and paper “burner”

The burner is composed of a 2.5″ x 17″ piece of wax-impregnated cloth, folded in Z-fashion back on itself 5 times.

In between each Z-fold of the waxed cloth are five 2.25″ x 3″ pieces of thin, coarse-fiber paper, about the weight of 40# kraft paper.

(The cloth/paper burner smells “nice,” sort of like the scent of roses. Maybe the young Chinese lady who made this one was wearing some fragrant perfume.)

Note: I have also seen sky lantern burners, which resemble fiber-reinforced blocks of wax. I don’t know how they are made and have not tried to duplicate them.

The “bag” of the balloon is made up of white tissue paper. This paper has been treated with a fire retardant; it does not catch fire when touched by a flame. It just scorches a bit.

The tissue paper bag weighs 1.5 ounces.

There are four panels (called gores in the ballooning world) that make up the balloon, and they are glued together and to the bamboo hoop.

Here’s a sketch of one of the gores, with a bit added to the edges to allow for gluing.

I flew one of these fire-lanterns in my high-tech, windless wind-tunnel (actually my garage, emptied of any gas cans or other combustibles). I tethered the Chinese lantern to a weight with some very thin, light string as it burned. The fuel pack burned for 4.5 minutes.

I added 0.05-ounce pieces of wire, one at a time, to the bottom wire “X” of the balloon as it “flew,” and found it would carry 0.25 ounce of payload before starting to sag toward the ground garage floor. That 0.25 ounce, added to the paper-lantern’s original weight of 2.7 ounces, added up to a maximum flying weight of 2.95 ounces for a lantern with this internal volume.

It is the internal volume of a paper hot-air balloon, and the heated air it can contain as well as how much that air is heated, which determines its maximum carrying capacity.

Note: The information above is important. These flying lanterns are delicately balanced for flight, and they are just light enough to allow them to fly. If they were much heavier, they would not leave the ground. Due to some circumstances, which I describe below, my first balloon ended up weighing 3.9 ounces, and obviously would not have flown.

So, let’s make a paper sky lantern

I went up to my local Hallmark store and bought some nice tissue paper in different colors.

Some checking online produced some leads on products designed to flameproof paper. One company, Universal Fire-Shield (www.firechemicals.com), sells a product, Universal Paper Shield P-3000, designed to fireproof paper products. I ordered some. They sell a quart spray bottle for about $30, including shipping. This would be enough to treat about a dozen paper hot-air balloons.

I hung 4 pieces of the red tissue paper on a clothesline, and sprayed them with the Paper Shield until they were saturated. After they were dry, I tried to burn a little piece of the paper, and it only scorched like the original fire-lantern paper, but it would not burn.

Note: On my third attempt to make one of these paper lanterns, I decided to try to spray the untreated bottom half of the balloon after it was assembled, in order to skip the step described above. I hung it up, slightly inflated it with my heat-gun, and started spraying it. The tissue paper soon started to weaken, sag, and tear, ruining the balloon. Crud! Another lesson learned.

I decided that the process of hanging the panels like laundry works best. The upper corners of the sheets will be cut off when the gores are cut out, so I don’t spray those areas because they get weak when they are wet, and allow the clothespins to tear through the paper and sometimes that has the paper to tear loose from the string.

Warning: Paper Shield has an acid in it, and it will damage a concrete garage floor slightly. It’s best to have a plastic drop cloth under the clothesline to protect the floor. Don’t breathe its fumes or get it on your skin.

The red tissue will be the bottom of the paper hot-air balloon, which will be the only part that gets exposed to the flame. The top of the balloon will be blue paper, and it does not need to be treated.

I glued a piece of the red paper to a piece of the blue with thin stripes of Elmer’s, and I did that four times for the four panels, and let the panels dry.

Note: The Elmer’s glue tends to wet the tissue paper, bleeds through, and tries to stick to the other stuff around it. In an attempt to avoid this problem, I used hot glue when building my first balloon. This worked nicely during construction, but when I fired that baby up, the hot-glued seams at the top of the balloon let go completely. I did not think the internal temperature would get high enough to cause this problem. I was wrong. (My wife, Molly, told me later on that she wondered about me using the hot glue, and that she thought it would melt when the burner was lit. Oh, well.) The hot glue, which is significantly heavier than the dried Elmer’s, also contributed to the excess weight of the first model.

After gluing the red and blue sheets together and letting them dry, I cut the four panels out with scissors, using a kraft paper template that I made based on the sketch above. Folding the kraft paper in half lengthwise, and then every 6″ made the pattern transfer easy.

Then it was time to glue the lantern gores together to form the Thai-lantern’s “bag.” I laid one of the gores, unfolded and open, on the worktable with the inside up. (The outside of the gore is simply the side that I think looks best.) Then I laid a gore, outside up, on top of the first gore, weighed the two down, and glued the right sides together with a thin stripe of Elmer’s. The table had waxed paper on it so that any glue that seeps through wouldn’t stick to it.

I then folded the top gore’s loose side over on itself, so that half’s inside was facing up, inserted some waxed paper between the glued side and this loose side, and laid the third gore on top of that one. I glued those two loose edges together, inserted more waxed paper, folded the loose half of the third gore over, laid the fourth and final panel on top, and glued those loose edges.

Then the last step was to fold the loose half of the top, fourth gore, over on itself, and fold the loose half of the bottom, first gore over onto it, and glue the loose halves together. (This all sounds much more complicated than it actually is. Once you try it, it’ll all make sense.)

Then I pulled the glued edges up and off all the waxed papers, propped the panels apart from each other and from the table, and allowed the seams to dry.

Another Failure Note: On my first attempt to build one of these, I tried gluing the bag together with the panels inside out. I let the seams dry, and then I tried to turn the bag right side out, so that the seams would be hidden. This probably would have worked OK, but the fireproofed red tissue paper was somewhat brittle because of the fire-treatment, and as I tried to turn it inside out, it started to tear and crack at the bends and creases. I had to try to repair these tears with clear packing tape, which increased the lantern’s weight, and made it ugly, and not something to be proud of. I decided to simply allow the seams to be on the outside of the bag in future models, and avoid the “turning inside-out” step.

Making the sky lantern’s bamboo hoop

Home Depot had some 1″ diameter bamboo poles in their lawn and garden department. I bought one and carefully split it into thin strips. I took one of the strips and smoothed it with sandpaper and a razor knife until it was about the dimension of the original lantern’s bamboo. I only sanded the “interior” side of the bamboo because I did not want to weaken the smooth, exterior side of it. Then I glued it into a hoop with the same circumference of the original.

Another source of good bamboo strips is from a “Tiki-Torch” pole. These are often split to obtain bamboo strips for girandola frames, wheel frames, and the like. (Or you can steal some green bamboo from Harry Gilliam’s bamboo-infested front yard.)

Making the Sky Lantern Burner

I had some blue, industrial paper-towels, and I decided to melt some grocery-store canning-wax, and impregnate the towels with the wax in an attempt to duplicate the waxed fabric that I found in the original fire-lantern’s burner.

Warning: Canning (paraffin) wax is very flammable, and should only be melted over low heat in a double boiler. It should never be exposed to open flames or high heat.

I took a strip of this waxed paper towel, and burned it alongside a strip of the waxed material from the original lantern’s burner. Both samples burned identically and for the same amount of time.

I had some coarse, recycled kraft paper, and cut it into rectangles to match the original burner’s paper layers. Then I cut some strips of the waxed paper towel to match the original burner, and sandwiched 5 pieces of the kraft paper in between each layer of the waxed paper-towel.

Then I stacked the layers of the burner together, punched 4 holes through all the layers with an awl, and threaded two pieces of wire through the holes.

The ends of the wire were then wrapped around the bamboo hoop and twisted tightly to secure the ends. The hoop was then carefully glued into the end of the lantern’s tissue paper bag, and the glue was allowed to dry.

A simple alternative burner can be made by simply installing a plain X of wire on the bamboo hoop. Then strips of cotton-ball like material can be saturated with rubbing alcohol, draped over the center of the wire X, and ignited when launching the balloon. But keep in mind, you have to use this method right away; the alcohol fuel evaporates, and these have no “shelf life.”

All that remained, then, was the installation of the “FAA” aircraft identification numbers, and a flight in the “test chamber.”

This paper hot-air balloon’s final weight was 3.0 ounces. At about 50-degrees F in my garage, it was able to carry a payload of .55 ounces of the wire pieces before it started to sag toward the ground.

After this testing, some soot and moisture condensed on the inside of the lantern. I was able to allow it to dry out, attach a new burner, and fly it outdoors for real. It was about 40-degrees F outside on the evening that we flew it, and it really took off for the heavens very quickly.

Note: The flying lantern in on the right, Molly is on the left. The fact that Molly’s attire matched the balloon she was launching was entirely coincidental. That’s my story and I’m sticking to it.

Conclusions

This was a fun and educational project. I learned a lot about what to do and what not to do, when making one of these little hot-air balloons. This project was not as easy as it might look.

Based on my experience with flying the paper lanterns tethered in my garage, I think it would be fun to fly one outdoors in windless conditions, tethered by a short wire leader and a roll of light thread. It could be flown like a kite, reeled back in when the burner burns out, reloaded and flown again.

This project resulted in me having a high amount of respect for the folks overseas who turn these things out by the thousands, which fly successfully every time. I’m continually amazed by the low cost of a device like this, sold by Skylighter, compared with the time and materials I invested in producing a successful one.

It was fun to make these lanterns, and I enjoy knowing how to do it. If I had to come up with a bunch of ‘em for an event like a wedding, holiday, or memorial service, I sure wouldn’t be making them, though.

Happy Flying, and Stay Green,

Ned

Making a Firework Star Pattern Shell

A firework shell which bursts with a ring pattern, a smiley-face, or a star pattern can be a unique and creative addition to a fireworks display. Suddenly, after a procession of fairly typical full, spherical shell bursts, a simple ring of stars, or a display of four or five of them fired simultaneously, changes the focus of attention of the audience. “Hey, here’s something different,” they’ll think to themselves.

Pattern shells have some distinct advantages and disadvantages to their construction. They don’t use nearly the quantity of firework stars that a fully loaded shell would use, so if I have a few stars of a particular size and color, they might come in useful in a pattern shell. Patterns can be chosen to coincide with a particular theme in a show, with blue stars in a patriotic section, or pink hearts in a romantic interlude.

On the other hand, it will be hit-or-miss when it comes to the pattern’s orientation in the sky when the shell bursts. The smiley-face may display upside-down, or the ring may be seen on edge by a portion of the audience, looking more like a simple line in the sky. For this reason, most display designers choose to fire 4, 5, or 6 of the same or similar patterns at the same time. That will usually result in the audience in a particular location seeing at least 1 or 2 of them in the desired orientation.

If 6 ring-pattern shells of different colors are fired at once, the audience at one end of the field may see, say, the blue and red ones as true rings, and imagine all of them being the same shape.

Ring shells can use simple color stars, which leave no tail behind them, as in the photo above, or tailed stars can be employed, as below.

This Escher print, “Vuurwerk,” is on the cover of Pyrotechnica XI. It shows a pattern I would expect a ring shell of slow-burning, silver-tailed stars to display. It would have to be oriented so that the ring broke “flat” in order to display the “parasol” of stars just right.

A small rising comet tail produces the “handle” to the umbrella.

An advantage to using patterns such as rings, stars, squares or triangles is that they can break in many directions that still have them look correct, as long as they don’t break on-edge to the viewer. A smiley-face has to break in just the right direction to be recognizable.

The star-in-a-ring pattern shown below would look correct if it was rotated any number of degrees clockwise or counter-clockwise. It would also look fine if it was flipped 180 degrees front to back. The only way it would not show up well is if it broke on-edge to the viewer.

My friend, Mike B., made the heart-pattern shell shown below. While it did not break on-edge to the audience, unfortunately it did break almost upside-down. The fortunate thing about hearts is that they look good in almost any orientation, and the audience can make out what they are supposed to be representing.

I want to make a blue star-pattern shell. I don’t want to make my stars much smaller than 3/8-inch in diameter, so that they burn long enough to allow the pattern to show up. Additionally, ball shells break more symmetrically than cylinder shells. For these reasons, I’ve settled on assembling an 8-inch ball shell for this project. With 3/8-inch stars, a smaller shell simply wouldn’t allow the use of enough stars to create a nice star pattern.

The general construction techniques I’ll be using when assembling and finishing this shell were detailed in Fireworks Shells in 2-1/2 Days – Part 2, Part 3
and Part 4. I’ll be using 1/4-inch time-fuse in this shell, though, instead of a spolette. The use of time-fuse was explained in Really Nice 4″ Platic Ball Firework Shells.

The first thing I did was draw a pattern of the stars that would fit the inside diameter of one of my 8-inch shell casings, which has an ID of 7.25 inches. 360 degrees divided by 5 gave me 72 degrees between each of the points of the 5-pointed star, which I measured out with my protractor.

My 3/8-inch pumped stars actually end up being about 7/16-inch in diameter once they are primed, so I drew lines of that size star on my pattern. Precision in these initial planning stages, right through the actual construction of the shell, will result in a more precise star-pattern in the sky when the shell bursts.

I took a piece of tissue paper, cut a circle out of it about 1/2-inch larger in radius than my drawing above, and traced the star pattern onto it.

Then I made some blue stars. Firework Shells in 2-1/2 Days – Part 2 and Part 3 included instructions for making and priming pumped stars.

Although I didn’t need a large number of these stars, it was important that all the stars were consistent in size. For this reason I used a 3/8-inch star plate to make a pound of the Shimizu Blue star composition included in the table of formulas in 14 Great Cut Star Formulas.

Shimizu Blue Star Formula	Percentage	16-ounce batch	450-gram batch
Potassium Perchlorate	0.61	9.75 ounces	274.5 grams
Copper Carbonate	0.12	1.9 ounces	54 grams
Parlon	0.13	2.1 ounces	58.5 grams
Red Gum	0.09	1.45 ounces	40.5 grams
Dextrin	0.05	0.8 ounces	22.5 grams

I dampened this star composition with an additional 10% water, and pumped and dried the stars. I primed them with the black powder “meal prime” which is also in that star formula table cited above. I add an additional 5% of 200-mesh magnalium to the prime, which improves the ignition of perchlorate stars.

As I said above, the shell was constructed in the standard fashion, except for the details below.

Once I had the time-fuse and passfire-tube installed in the shell casing, I hot-glued a 1.5-inch wide tissue paper ring inside each hemisphere at the equator. These bands served the purpose of locking the shell’s contents into the hemispheres later on when I closed the shell.

Then I filled the fused hemisphere with black-powder-coated rice hulls, folded the tissue-paper band over onto the hulls, and hot-glued a tissue-paper disc onto the whole shebang to cover and seal it. As I loaded the hemi with the coated hulls, I packed them tightly one layer at a time to make sure the casing was solidly filled. I also filled the hemi slightly higher than the equator. This half of the shell held 29.4 ounces (825 grams) of the coated hulls.

Then I filled the un-fused hemi with coated rice hulls up to within about 3/8 inch of the rim. I made sure the rice hulls were tightly packed and very level. This filling was loosely capped off with the tissue paper disc which had the star-pattern traced on it.

Starting with the points of the star, blue stars were lightly hot-glued onto the tissue pattern. These stars only had a small dot of hot glue put on them where they touched the pattern. Just before the shell bursts, the tissue paper disintegrates and the stars are free to fly out in the star shape.

Then I filled in around the stars with more black-powder-coated hulls, tightly filling all the voids and bringing the level of the rice hulls slightly above the rim of the casing. This hemi actually took about 35 ounces (1000 grams) of the coated rice hulls, for a total of about 4 pounds (1800 grams) in the whole shell. This was all capped with another hot-glued disc of tissue paper.

Because the tissue paper rings and discs were glued to the shell casing hemispheres, it was easy to flip one of the hemis over onto the other and close the shell up, ready for pasting, lifting and leadering.

With the blue stars sandwiched between the layers of tissue paper, with the rice hulls really packed in tightly and the hemis overfilled and slowly tapped and brought together, the star pattern was held firmly in place.

I’ve developed a nifty trick for bringing the stuffed hemis together at the equator. I use 4 strap-clamps, available at Home Depot or stores which cater to woodworkers.

As the clamps are slowly tightened, tapping the shell with a solid, heavy rod brought the two halves together and solidly packed the contents. Then the joint was closed with strips of masking tape. This method is so much easier than “laying” on the shell while tapping it in order to close it.

Warning: I use a non-sparking, aluminum rod for tapping on the shell. But, the metal strap-clamp parts are not non-sparking. I’m working around relatively exposed black powder on rice hulls during this process. I’m very careful to avoid smacking the metal clamp ratchets, which could potentially cause sparks.

Then I pasted the shell, allowed it to dry, and lifted and leadered it. A small rising comet tail was attached to direct the viewer’s eye toward where the shell will break.

When I shot this shell, it did indeed break a bit on its “side” relative to the camera, as shown in the photo below. There were viewers down and to the left of the shell-burst, and they said that the star really looked nice, big, and symmetrical.

Oh, well, maybe I’ll get to see it next time.

Here’s a link to a video of the shell in action.

I enjoy making pattern shells. They offer a unique challenge in shell construction, and use less of the chemicals that go into stars. More black-powder-coated rice hulls are used than in a typical chrysanthemum or peony aerial shell, but these are the less expensive ingredients.

I think an audience enjoys the variety that these pattern shells bring to a display.

The next time I make a shell like the one in this project, I think I’ll add a red ring around the blue star pattern so the sky is filled a bit more when the shell bursts.

I’m working on my version of a way to at least have aerial shells burst with their equators level with the earth. This will allow rings, star patterns, etc, to display well for anyone underneath them. The method is one that I’ve heard about over the years, but have never seen, where a rope is attached to the bottom of the shell to produce drag on the shell’s way up. This keeps the shell oriented with its “bottom” down on the way up.

I plan on shortening the shell’s time fuse delay so that the shell bursts before apogee while it is still oriented correctly. I’ll keep you posted on the progress in this project.

Have fun and stay green,

Ned

Making Homemade Fireworks Mines

There is a beautiful photo on the cover of Dr. Takeo Shimizu’s Fireworks, The Art, Science and Technique (FAST). At the top of the picture is a huge, double-petaled, fireworks-shell starburst, with several smaller star-flowers between it and the ground. And, at the very bottom of the shot is a spray of stars shooting upward from the ground where the shells were fired: a mine.

Similarly, the back cover of Hardt’s Pyrotechnics shows a rainbow of 5 colored mine-shots, with corresponding colors of small and large starbursts over each one.

Fireworks mines, or star-mines, can be used in conjunction with aerial shells to create such an effect, filling in the low sky as the shell bursts fill the middle and high sky.

These devices are also very effective during a fireworks display when they are fired as a mine-run: a series of firework mines fired in sequence from various spots on the field in harmony with beats of music in the soundtrack.

A single mine, a mine-run, or a wall of simultaneously fired mines all serve to bring the audience’s attention back down to the ground, providing contrast and variety during a fireworks display.

For precise timing of electrically fired mines, putting the electric match right down in the lift powder eliminates the objectionably noticeable, split-second delay, which occurs when a length of quickmatch leader is ignited by the electric match.

If I am going to put the e-match into the lift powder, I still attach a couple of inches of quickmatch to the electric match, with the e-match’s protective shroud in place. This ensures that quite a bit of flame will be introduced into the lift powder, and it further protects the e-match from premature ignition due to shock or friction.

One of the clubs I belong to, The Bluegrass Pyrotechnics Guild, has assembled and fired literally thousands of 4-inch mines over the years for our displays at the PGI annual conventions. These devices can be very easily constructed, and they provide a nice hand-made touch to a fireworks show.

In this article I’ll detail two methods of constructing 4-inch firework mines. But, these methods can be applied to any size mine, from 1-3/4-inch through 6-inch models. Simply by varying the size of the stars, the materials used in the assembly, the amount of lift powder, and the size of the mortar from which they are fired, mines can be tailored in size to any effect and any venue.

Each 4-inch mine will use about one pound of stars. I like to use 1/2-inch stars in 4-inch shells and mines, and in mines, I prefer a fast-burning star, which burns out before it reaches apogee and begins to fall back to the ground.

For this project I’ll use two variations of one type of star, commonly known as Gold Spider Web (Ofca’s Volume 5, Mastering Cut Stars the Easy Way) or Chrysanthemum 6 (Shimizu’s FAST). This is a very fast-burning charcoal streamer star which is easy to make, and which produces bright streaks of sparks in the sky.

So the final working formula was:

Chrysanthemum 6	Ratio	16-ounce batch
Potassium nitrate	0.55	8.8 ounces
Charcoal, airfloat	0.33	5.3 ounces
Sulfur	0.07	1.1 ounces
Dextrin	0.05	0.8 ounces

The individual chemicals are screened through a 100-mesh screen, and any that will not pass the screen are milled in a coffee grinder
until they will pass. (I have a grinder that I only use on oxidizers like the potassium nitrate, and one for fuels such as charcoal. Except for when I am ball-milling black powder, I never mill complete compositions, only individual chemicals.)

After I screen the chemicals individually, I mix them through the 100-mesh screen, and then shake them in a closed plastic tub. 1.3 ounces of water is added (+8%) to the composition, and it is thoroughly shaken in the closed tub once again. The dampened pyrotechnic composition is then pushed through a 20-mesh kitchen colander to completely incorporate the moisture into the composition.

I use my 1/2-inch star plate to press 1/2-inch diameter by 1/2-inch long stars, as detailed in Fireworks Shells in 2-1/2 Days: Part 2, which I dry overnight in the drying chamber.

The stars are then primed, as shown in Fireworks Shells in 2-1/2 Days: Part 3, and dried overnight again.

I make two, 16-ounce batches like this, and one 16-ounce batch of the same composition with 3.2 ounces (+20%) of 10-185 mesh spherical titanium
(CH3001) added to the composition after all the 100-mesh screening is done. I never put any metals through my fine screens. The addition of this titanium will produce bright silver-white sparks as the stars burn, creating a silver-streamer star.

So, with these three, one-pound batches of stars, I can now proceed to make three, 4-inch mines.

There are some extremely simple ways to assemble mines. “Bag Mines” can be as simple as a paper bag with some lift powder in a baggie on the end of a fuse-leader inserted into the bottom of it. Then some stars are poured in, the neck of the bag is tied closed and she’s ready to fire.

In Introductory Practical Pyrotechnics, Tom Perigrin shows how to make a simple mine by installing a baggie of lift on a quickmatch leader, sliding that into a cardboard-tube mortar. Then dropping in some stars, bundled in some tissue paper, into the tube.

These simple mines are functional, but they’ll typically not fire all the stars out the top of the mortar together and in a straight column of fire. Because there are some stars to the sides of the lift powder, they get fired a bit sideways and bounce around a bit in the mortar before they exit. Some of these stars can “ploof” out of the gun and drop onto the ground.

So, I like the following slight variations on the simple-mine theme. They are more effective at getting all the stars lit and out of the gun in a straight-up spray of stars.

Typically, firework mines use the same amount of lift black powder, and the same amount of stars, as a cylindrical star shell of the same diameter. I like to refer to the charts on Pages 138-140 of The Best of AFN II to determine these amounts.

For a single-break, 4-inch cylindrical shell, 1.5 – 2 ounces of commercial 2FA black powder is recommended for the lift powder. I’ll use 1.75 ounces in these mines.

I have made three 24-inch lengths of homemade quickmatch for the mine leaders. Similar lengths of commercial quickmatch (GN3001), or super-fast paper fuse (GN1205) (wrapped in aluminum foil duct tape), can also be used as mine leaders.

I then weigh out the three 1.75-ounce loads of 2FA lift powder, and put them in individual plastic baggies. The one-inch of black match protruding from the quickmatch leaders is folded back on itself, and then inserted into a baggie of lift. Masking tape is used to seal the baggie around the leader, and the extra plastic trimmed off with scissors. Then a final wrap of masking tape seals the baggie/tape to the quickmatch.

If I’m only going to make a few mines, and I want them to be top-quality, I build them this way.

First I make a “piston” to go between the stars and the black powder lift charge. Here’s a photo-essay of that process. It uses two 4-inch kraft discs (DK3401 or DK3500), a 2-inch length of 3/4-inch ID tube (TU2053 or TU1065 or TU1068), and a 2-inch length of 3-3/8-inch or 3-1/2-inch OD tube (TU2300).

Places like HarborFreight.com sell the sets of hollow gasket punches for ridiculously low prices. I much prefer to make the holes in discs with the punches instead of with a drill.

This piston will allow the lift-gasses to pass through it, igniting the stars, and will evenly push the stars out of the mortar and into the air in a tight column.

The center holes in the discs are 1/2-inch diameter and are large enough to allow the quickmatch leader to be threaded through them. The other holes are 3/8-inch diameter, keeping these holes smaller than the size of the stars that I am using.

Now, I position the piston on an 18×24-inch piece of kraft paper as shown. The paper is rolled up around the piston and glued to itself with hot-glue. Then the end with the baggie of lift powder is gathered together, tied with string in a clove-hitch knot, and trimmed with scissors. This is my mine casing.

Kraft paper is available from stores like Staples or Costco (white butcher paper), on-line from places like ULine.com, or paper grocery bags can be used.

Now, it’s a simple matter of dumping the pound of stars into the open end of the mine casing, gathering the kraft paper around the leader, and tying it securely in place.

When filling a mine like this, I like to keep the depth of the stars equal to the diameter of the mine. So 3-4 inches of stars in this mine is my goal.

A piece of Visco fuse is inserted and secured into the end of the leader, which is then S-folded. A safety cap can be installed over the fuse, and a band of masking tape sticky-side-out, followed by a band of tape sticky-side-in, secures the bundle until the mine is to be loaded into a mortar.

Here are shots of the two mines made this way. They had different types of charcoal in the stars, commercial airfloat and homemade poplar airfloat, but I didn’t see any significant difference between the two.

As you can see, the piston pushed all the stars out evenly, and very high. Nice, bright, fast and dramatic mines.

The above “high-quality” method produces very nice, traditional firework mines, but it is a bit time-consuming. When the Bluegrass guild wants to construct several hundred mines in an afternoon, we’ve developed the following system. It is quick and easy, and while it does not allow the use of quite as many stars, and the quality of the mine-display is not quite as nice, it is a very effective method for producing large quantities of mines quickly.

We start with the same lift-bag/leader configuration as shown above.

Two disposable plastic drinking cups are used for each mine. For these 4-inch mines, cups that are 3-5/8-inches in diameter at the top are ideal.

One of the cups has its bottom perforated with a hot, pencil-tip soldering iron. I’ve tried using a drill for this operation, but often the cup-bottom will crack when doing so.

The mine leader is inserted through the center hole in the cup. That whole assembly is placed into the second, intact cup, and the perforated cup is filled with stars. In this case, it held 12 ounces of the stars.

Then the assembled mine is sealed with four, 9-inch strips of aluminum-foil duct-tape. The leader is completed with Visco fuse and S-folding as shown with the paper-cased mine in the Method #1 above.

This mine was made with the silver-streamer stars, which contained spherical titanium. While it did not contain the same quantity of stars as the previous two mines, and it did not fire quite as high as they did, it was still a very nice, bright mine.

Here is a link to a video of the three mines in action. They were all impressive, firing 30-40 feet into the air, and producing unique effects.

While these are simple devices, they can be crafted with care, and will add variety to a fireworks display.

Have fun with them, and remember, “A mine is a terrible thing to waste.”

(I hope that line at least elicited a groan or two, if not a chuckle. My wife, Molly, only gave me that blank “wife’s” stare, and that curled-up lip, when I said it to her.)

Until next time, Enjoy!
Ned

Spectacular Glitter Tailed Firework Rocket

What do I mean by a “spectacular” black powder rocket?

By this term, I am thinking of a great looking firework rocket, with a unique tail as it ascends, followed by a long-lasting, eye-catching heading. Maybe I have in mind a rocket that I wouldn’t know how to improve on. (Of course, I am playing with two-pound versions of this baby already, since bigger is almost always better, or at least different.)

Here is a link to a video of the glitter rocket we are about to make, to get your juices flowing, and so that you don’t have to read all the how-to information before you get to see what it is we’re trying to accomplish.

You’ll notice that the rocket does not fly into the stratosphere. It stays relatively low, allowing the audience to watch the graceful, arching glitter trail, and then to be close enough to the sparking horsetail header finish to be able to really appreciate it. This was all done by design, and I wanted this to be a really satisfying “fireworks rocket”, not a high-powered machine.

I described end-burning black powder rocket-and-girandola motors, in a past post, but this project will focus on core-burning rockets.

In Fireworks Tips #65, John Werner discusses the construction of core-burning, black powder sky rockets, specifically 1/2-inch ID (4 ounce) models. John is a master pyro craftsman, and his articles are well-written, detailed, and complete. This one is no exception.

Toward the end of his article, after describing the rocket’s construction in detail, John includes a “Troubleshooting” section, some options for “Modifications and Enhancements,” and answers to some “Frequently Asked Questions.”

I don’t believe in constantly reinventing the wheel, so I’ll simply use John’s article as a foundation and base of reference for this one. I will be constructing 3/4-inch ID, one-pound, black powder rockets in this project.

When I wrote my first article for Skylighter I focused on making clay nozzle and bulkhead mix, and I will be using those mixes in this project.

I discussed black powder techniques in Making & Testing High-Powered Black Powder
, and that BP will be used in the shell headers for these rockets.

I will be cutting and treating tubes, Skylighter #TU1065, as I described in Cutting and Hardening Fireworks Tubes.

I discussed making small plastic can firework shells and those small shells will be used as the headings on the rockets made for this article, with some minor modifications in technique.

I showed how to pump stars and make blackmatch in Firework Shells in 2-1/2 Days – Part 2, and how to prime the stars in Firework Shells in 2-1/2 Days – Part 3, and I’ll make stars and blackmatch using those techniques, with a special star formula for these rocket headings. (Flying fish fuse could also be used in this project with a nice effect.)

Quickmatch pipe was illustrated in Firework Shells in 2 1/2 Days – Part 4, and I’ll be using some of that this time around, too.

And, last but not least, I’ll be showing how to attach some glitter comets, to achieve an extraordinary rising tail with these rockets.

Special techniques for finishing the rockets will be described so that the finished product will be safe and look nice even before it is fired.

So, you can see that all the skills and techniques that have been described in the past years start to build on themselves and come together in this stunningly beautiful rocket I’m about to describe.

You have your homework. Fire up your printer, and get the above mentioned articles in front of you. Study them, and assemble your materials. Then we’ll get to work.

Note: One thing you’ll hear from experienced fireworkers is, “Always take good notes about your experiments and projects, and keep them in a good notebook for future reference. A year from now you won’t be able to clearly remember which of those experiments was the one that worked so well if you don’t take good notes.”

One of the extremely beneficial things about writing these Skylighter articles is that they then serve as my notes, when I might have been too lazy to follow the above advice otherwise. I also have access to the notes of others like John Werner. If you print out the articles, and then annotate them with your own modifications, they can serve as your notes as well.

In Firework Shells in 2-1/2 Days – Part 2, Part 3, and Part 4, I described the process of making “Nice Shells in 2 1/2 Days” at a fireworks event such as a local club gathering or the annual PGI convention.

I think I’ll approach this rocket project in the same way, wherein a fireworker travels to a pyro event with absolutely no complete pyrotechnic materials, and makes everything from scratch at the event, and only in the quantities needed for this project. This eliminates any worry about licenses, storage, or transportation.

In Fireworks Tips #111, Harry included a shot of him and me at the most recent PGI convention out in Gillette, Wyoming. One of the great things about the convention is the opportunity to work alongside others as fireworks devices of all sorts are constructed.

Speaking of kids at the PGI convention, the Junior Pyros there always plan and execute one of the best shows of the week. The next generation of fireworkers is nurtured and brought along slowly and safely. Where would I be now if I’d started in all of this at the age of 15 instead of 35?

So, let’s imagine we are bringing some materials and supplies to the PGI convention, and we are going to build a few of these fine rockets. I’ll actually scale this project so that 10 of these babies can be constructed in a two-day period.

I will arrive on site, and begin building on Friday morning, with the goal of flying some rockets Saturday night. I’ll want to plan my activities according to a time-line.

Before I even leave for the event, I accomplish a few things on the project:

• Print out all how-to articles and formulae.
• Unwind 40 feet of 16-ply cotton string with which to make black match.
  
  
  Untwisted 16-Ply Cotton String

• Clean all my rocket tooling and lubricate with Sailkote (by Team McLube, available online or at sailboat distributors). This is a great lubrication product, which allows the tooling and spindle to release very easily from the rocket motor and tube. I also spray it on my star plates, comet pumps, and the aluminum rod I roll my match-pipe on.
  
  
  SailKote Spray Lubricant

• Treat and cut my rocket tubes to 7-1/2 inches.
  
  
  Ten 3/4-Inch ID, 7-1/2-Inch Long, Rocket Tubes

• Roll twelve, 18-inch lengths of paper quickmatch pipe
  
  
  Paper Match Pipe for Quickmatch

• Rip rocket sticks on my table saw. I like to rip sticks out of clear poplar from Home Depot. A 5-foot piece of 1×3 will yield 18 sticks, 5/16-inch square. It is also possible to use round wood dowels, but I much prefer square sticks. 5/16-inch square sticks are nice because I can rip a 5/16-inch wide by 3/4-inch strip off my 1×3, and when I rip that strip in half, my 1/8-inch thick sawblade leaves two 5/16 x 5/16 square sticks.
  
  
  Bundle of Freshly Cut Rocket Sticks

Tasks to accomplish on Friday:

• Making 40 feet of blackmatch
• Making 10.5 ounces of black powder for header-burst, and for star and comet priming
• Mixing, dampening, screening, and drying the rocket fuel
• Pumping and priming stars
• Pumping and priming glitter comets

That ought to be a good day’s work, and after drying things overnight in the drying chamber, I ought to be able to assemble some rockets on Saturday.

I first set up my work station; I have a pop-up tent, work tables, and a chair. I’ll need extension cords and a generator if no electricity is available where I’ll be working. I might need a light if I get delayed too much on my timeline as I chat with pyro-pals. Some plastic sheeting can come in handy in the case of a sudden rain shower.

I have my drying chamber in which to dry various products as they are produced. It’ll be especially handy to have an electric outlet somewhere, even if it’s not right at my work station, so that the dryer can be plugged in there and left running all night long.

I have my tool box with my miscellaneous hand tools, my measuring cups and spoons, my little digital scales, some funnels, my miter box and saw for cutting tubes on site, and the various supplies which will be mentioned as I go along.

I’ve brought along some food and a cooler of drinks, so I don’t have to pause for long in the midst of my pyro activities.

Note: As I go along describing this project, I’ll be applying materials and methods that I have described in the articles above. I will not be re-citing those articles in the text below. I’ll trust you have the references available and are familiar with them.

I take the 40 feet of my untwisted string and make blackmatch out of it. It’s a nice, warm, sunny and breezy day, so I string it up between two trees, tied onto nails, to dry. Before it gets dark, I’ll cut it into 18-inch lengths and put it on a screen in the drying chamber.

For this batch of blackmatch, I used a formula with a slight variation. 15 ounces of potassium nitrate, 3 ounces of commercial airfloat charcoal, 2 ounces of sulfur, 0.8 ounce of dextrin, and 0.2 ounce of CMC (Skylighter #CH8080). This batch took about 14 ounces of water, added carefully and stirred in with a paint stirring stick, avoiding getting the slurry too wet.

Replacing one-fifth of the dextrin in the original formula with the CMC produces a nice, smooth black powder slurry, which does not separate as I use it, and which produces a nice, smooth coating on the cotton string.

I run 7.5 ounces of potassium nitrate, 1.5 ounces of airfloat charcoal, 1 ounce of sulfur, and 0.5 ounce of dextrin through my 100 mesh screen and onto a piece of kraft paper. I know from experience whether or not the individual chemicals will pass the 100 mesh screen, and if one won’t, I’ll run it through my blade-coffee-grinder
first to pulverize it. I never run pyrotechnic mixtures through the grinder, only individual chemicals.

Then I run the mixture through the 40 mesh screen in order to intimately mix it.

I weigh 3 ounces of this mixture into a plastic tub, add enough water to it to make a putty out of it, and grate it through the 4 mesh screen. This will become black powder granules to be used to burst the shell headers. I have determined that it will take 0.3 ounce of this powder for each of the ten rocket headers.

This granulated black powder, made with dextrin as the binder and not ball-milled, is called “polverone” rough powder (see Pyrotechnica IX, Traditional Cylinder Shell Construction, by A. Fulcanelli), rather than a hot black powder. I don’t want a hot powder; I simply want to ignite the stars in the headers, and pop the headers open, allowing the lit stars to cascade down in a “horsetail” effect.

The remaining 7.5 ounces of the powder is set aside to be used for priming the stars and comets.

John Steinberg, Kurt Medlin, Steve Majdali, and Brent Anderson teach a “Black Powder Rockets for Beginners” seminar several times throughout the PGI convention each year. It’s a wonderful, hands-on seminar in which the basics of black powder rockets are taught, and folks actually get to make a rocket of their own to take out to the rocket range and fly.

They use a rocket fuel formula which creates a pretty rocket tail during flight, has enough power to lift a nice heading, yet is foolproof enough that even beginners can ram a rocket and have it fly every time.

Each of the ten motors in this project will use 2 to 2.5 ounces of the fuel, so I’ll make up a 24-ounce batch of it.

Basic Black Powder Rocket Fuel	Ratio	24-ounce batch
Potassium nitrate	0.6	14.4 ounces
Airfloat charcoal	0.1	2.4 ounces
80-mesh charcoal	0.18	4.3 ounces
36-mesh charcoal	0.02	0.5 ounces
Sulfur	0.10	2.4 ounces

The potassium nitrate, airfloat charcoal, and sulfur are all screened through the 100 mesh screen. Then the coarse charcoals are added to the mix and it is all screened through the 40 mesh screen 3 times, and shaken in a closed tub to completely mix it.

Then enough water is sprayed onto the powder to dampen it to the consistency of brown sugar: just slightly damp, so it will barely stay together in a ball when squeezed in a fist. The water is worked in with gloved hands, and the fuel is pushed through the 16 mesh kitchen colander screen several times to fully integrate the moisture.

The fuel is then pushed through the 8 mesh kitchen colander screen and onto kraft-paper-lined screens to sit out in the sun and breeze during the day. I bring it in to dry in the drying chamber overnight.

The dry, mixed powder, can be used as-is before granulation, but wetting and granulating it significantly decreases dust during the motor ramming and probably increases the power of the fuel.

I want a slow-burning star which will form a “horsetail” when ejected from the rocket headings. One of my favorite stars is Willow Diadem, which is a long-burning, charcoal, Willow star, with the inclusion of some Ferro-Titanium and Titanium for metallic sparks in the stars’ tails as they fall.

I’m going to use 2 ounces of the stars in each of the 10 rockets, for a total of 20 ounces of stars.

Willow Diadem Stars	Ratio	20 ounce batch
Airfloat charcoal	0.66	7.8 ounces
Potassium nitrate	0.53	6.25 ounces
Sulfur	0.18	2.15 ounces
Dextrin	0.12	1.4 ounces
Ferro-titanium 30-60 mesh	0.08	0.9 ounces
Ferro-titanium 40-325 mesh	0.08	0.9 ounces
Titanium, medium spherical	0.05	0.6 ounces

Note: The original formula, as it has been passed around, specifies the metals as fine FeTi, medium FeTi, and medium Ti sponge.

This composition is made into 3/8-inch pumped stars and primed on one end. To prime them, I simply spritz them with a little water as they all stand on end in a tray, dust on a little of the BP prime that was set aside earlier, and spritz them lightly again. Then I tumble them a bit and place them on a drying screen.

The plastic can shells that I’ll be using for the rocket headers are 2 inches in diameter, so I want to press 1/2-inch thick comets that same diameter. Each of these will use 1.5 ounces of D1 glitter composition, so I want to mix up 15 ounces of it.

D1 Glitter	Ratio	15-ounce batch
Potassium nitrate	0.53	7.95 ounces
Sulfur	0.18	2.7 ounces
Airfloat charcoal	0.11	1.65 ounces
Aluminum, 325 mesh atomized	0.07	1.05 ounces
Sodium bicarbonate	0.07	1.05 ounces
Dextrin	0.04	0.60 ounces

This composition is mixed, dampened, and pressed into the comets, and then primed.

Everything that was in the drying chamber overnight is dry as a bone this morning. It’s time to start making some rockets.

I’m going to assemble the headers first. These small shells will be identical to the 2-inch plastic can shells I described in How to Make 2-Inch Firework Cylinder Shells, with a couple of exceptions.

Note: Similar rocket headers can be assembled using kraft paper, paper discs, chipboard or manila folders, and glue if plastic shells are prohibited at a particular shoot site.

The fusing on the bottom of the shell will be quickmatch instead of time-fuse. I want the headers to pop at the same moment the rocket fuel burns out and passes fire out the top of the motor and to the header.

I cut a 1/2-inch long piece of the paper match pipe, insert two 2-inch pieces of the blackmatch (cut with anvil cutters or a razor blade), pinch the pipe around the match and fold over the excess paper pipe. This fuse gets glued (hot-glue on the outside only), into the hole I previously drilled in the bottom of the plastic can (not the hole in the shell cap).

Then 0.3 ounces of black powder is poured into the shell, the tissue paper disc inserted, and then the stars are added.

I use a coping saw to remove the fuse-hole protrusion from the plastic casing cap, and then I use PVC plumbing cement or thickened methylene chloride to glue the cap on the shell. The capped shell can be tapped with the handle of my awl, and as many more stars as possible can be introduced through the hole in the cap.

Next, a 1-1/2-inch paper disk is hot-glued into the recess of the shell’s cap; then a 2-inch disk is glued on; and finally the comet is glued on.

And that completes the assembly of a rocket header.

My work table has been cleaned up from the previous processes, and I’ve laid out my materials for ramming the rocket motors. Nozzle clay mixture, bulkhead clay mix, rocket tubes, rocket tooling, pounding post, rawhide mallet, 1/2-tablespoon measuring spoon, funnel, rocket fuel and paper cup.

The pounding post I actually use is 24-inches long. I like to work with a small amount of rocket fuel in a paper cup, keeping the tub of fuel closed rather than having it sitting open with all that fuel exposed.

All my tooling drifts have rubber o-rings on them to further minimize dust, which has already been reduced by granulating the fuel. The drifts also have ‘do-not-pass’ marks on them so they don’t hit or get stuck on the spindle.

When hand-ramming these motors with the specified tubes, a tube support-sleeve is not necessary. Care must be used with the ramming/hammering so that I achieve good, consistent consolidation, without over-stressing or splitting the tube.

Note: I suppose the first real question that popped into my head back when I first assembled the materials to ram rockets was, “How will that dry clay and rocket fuel stick together with just pounding on it? Don’t I need to moisten it or something?”

The answer is, “Nope.” The dry powders will consolidate into a solid grain simply with the pressure of the ramming process. Pretty amazing!

Below is the kind of drawing I sketch up for each of my types of rockets once I dial them in. The sketches enable me to come very close to duplicating my results each time I make up another rocket of that type.

Of course, the details of the sketch will vary according to the exact tooling, tubes, and materials that are being used. But, if I keep those the same from batch-to-batch, the sketch becomes very useful.

A tube is placed on the spindle, a flat 1/2-tablespoonful (0.35 ounce) of the nozzle mix is placed in the tube, along with the nozzle forming drift. Then, 8-10 solid whacks with the rawhide mallet will form the nozzle. I try to ram the tube hard enough to create a very small bulge in the area of the nozzle, without it being seriously deformed or split. This seats the nozzle more securely and reduces the chances of it being blown out.

I have determined that this amount of pounding is equal to 1000 pounds of force being applied to the hollow drift, which is equal to about 3000 psi of pressure being applied to the composition. These are the figures I would use if I was using a hydraulic press to press these motors. With 1000 pounds of force, and with the solid drift, about 2300 psi would be applied to the composition.

When I begin to ram the fuel grain in the motor, I weigh out the amount of fuel that will be rammed with each drift, according to the specifications I’ve noted on the sketch, and put it in the paper cup. Then I scoop the fuel out of the cup, one increment (1/2 tablespoonful) at a time, until that amount of fuel has been rammed. Then I know it’s time to switch to the next drift, and I weigh out the total amount of fuel to be rammed with that drift.

This keeps me from having to constantly be counting the number of increments I have rammed, or guessing if it’s time to switch to the next drift.

The delay fuel at the top of the motor, approximately 0.3 ounce of it, is weighed out, and I add a flat 1/8 teaspoonful of fine spherical titanium to that fuel and swirl it around in the paper cup to mix the metal in. Then I ram those increments of delay fuel into the motor, being careful to only bring the fuel grain up to between 1 and 1-1/16 inches above the spindle.

This amount of delay fuel gives me the proper delay between the powerful thrust fuel burn, and the heading burst. When I was dialing the motors in, it is this amount of delay fuel that I adjusted to get the heading to burst at just the desired point in the rocket’s flight.

Once the delay fuel has been rammed up to the correct height, I plunge the open end of the motor into the bulkhead clay to pack the void full of that clay. Then I scrape off the excess clay, and ram the bulkhead.

Using a 1/4-inch drill bit, I carefully hand-twist-drill a hole in the bulkhead, just barely into the fuel grain. This creates the passfire hole which will transfer fire from the motor to the heading when all the motor’s fuel has been spent.

Twisting the motor off the spindle reveals a nicely formed central cavity. The motor is now ready for the final rocket assembly.

Whew! The hard part of this project is done. Now for the easy part.

I trim the black-match sticking out of one of the headers so that it is just long enough to go all the way to the bottom of a motor passfire hole when the header is pressed onto the motor. Then I hot-glue the header to the motor, and reinforce the assembly with some strapping tape.

Now I hot-glue and tape a 5-foot rocket stick to each motor. I glue the sticks on so that any bow in the stick curves it under the center of the motor. Instead of the stick curving out and away from the motor, or to the left or the right, I want it curving back in toward the centerline of the motor. I bevel the end of the stick with my anvil cutters to minimize drag (as if that big, clunky, flat-ended comet is aerodynamic).

It is now time to fuse these babies. I take two pieces of the black-match I made, and slide them into one of the paper match pipes. One of the pieces of match protrudes from the pipe 2 inches, and the other one only 1 inch. The pipe is crimped around the match for about 4 inches, and that end of the quickmatch is inserted into the rocket motor as far as it will go.

Then I bend the quickmatch leader up alongside the motor, and secure it to the motor and header with masking tape.

The quickmatch is folded over to the center of the comet, and bent upward. Using the awl, a hole is pierced in the match pipe, and an additional 4-inch length of black-match is inserted into the pipe and bent onto the top of the comet. This end of the match will pass fire to the comet while the other end ignites the motor.

I want the comet to ignite just slightly before the rocket motor does, so that the comet is really emitting its fire as the rocket starts to ascend.

The quickmatch and black-match are held in place with lengths of masking tape, and the quickmatch is cut with the anvil cutters to a length of 3 inches above the comet.

An 8-inch square of light tissue paper or wrapping paper is wrapped around the comet and header, and hot-glued to itself only. This creates a loose wrap which will protect the comet from unwanted fire, but will burn and fall away as the rocket ignites and ascends. This wrapper is tied around the quickmatch leader.

A piece of visco fuse is secured into the end of the leader, and a reusable safety cap, made of more match pipe, is installed to protect all the fusing from sparks (and to make the whole deal look a little more finished and pretty.)

It’s funny, but over the years I’ve seen other pyros who get to the point where they want to put as much effort into making the outside of their devices as attractive as they hope the performance of the inside will be.

Two days. I started with nothing but a few chemicals, some materials, and a few basic tools. Now I have a finished product that I can watch the video of, and look at the photos of, and I can step back and say, “Yep, I’m proud of that.” That’s why I got into all of this in the first place.

I’m gonna go back and watch that rocket video a few more times.

I hope you can follow these tips, create some of your own tricks, and come up with something you can be really proud of. Perhaps we’ll see the results at the next PGI convention.

‘Til then, Stay Green,

Ned

Using a Coffee/Spice Grinder to Pulverize Potassium Nitrate

If one does not have a ball mill there is another option for grinding coarse potassium nitrate into a free flowing, fine powder. Coffee and spice grinders work well for grinding small batches of individual chemicals.

Even though I have a ball mill, there are times when the coffee grinders come in handy for pulverizing smaller batches of chemicals. I have some Parlon, most of which will pass through a 40-mesh screen, but which has some larger particles as well. I’ll take those larger bits and run them through the coffee grinder in order to reduce them to smaller particles.

Warning: Dedicate one grinder for use on oxidizers, and another one for use on fuels such as charcoal. We don’t want fires or explosions when we’re grinding chemicals. Never grind complete or mixed compositions such as black powder in a coffee grinder.

I have found two kinds of coffee grinders: blade-grinders and burr-mills. Don’t get a burr-mill; they don’t work as well as blade-grinders. The blade-grinders have a stainless steel blender type blade that spins at high speeds in the bottom of the material cup, chopping the material into small bits in the process.

I have purchased many of the smaller, less expensive, blade-type coffee grinders. But here’s the warning: they really don’t last too long if you mill chemicals for a minute or two at a time. To use them, mill your chemicals in pulses of a few seconds at a time. I’ve found that shaking them while pulse-grinding gives me the fastest results.

The Kitchenaid blade mill has a larger hopper, and a larger, more powerful motor, and is rated to be used often. I’m hoping that it will last longer than the $13 WalMart models I’ve been using.

I put a half-cup, 4.6 ounces, of 12-mesh potassium nitrate into its hopper, pressed down on its lid to start it, and pulse-milled the powder for just under a minute, shaking the grinder now and then in the process.

Quite a bit of fine powder started to accumulate on the inside top of the clear lid as it milled. I dumped the ground chemical onto my 100-mesh screen, and used a fine paint brush to clean off any that was clinging to the inside of the hopper or the lid.

About three-fourths of this milled powder would pass through the 100 mesh screen, and I set aside that which wouldn’t to be ground again with the next batch.

Granular potassium nitrate can be dried if necessary, and ground easily with a ball mill or with a coffee blade mill, so that it passes through a 100-mesh screen and is ready to be used in pyrotechnic compositions.

How to Make 2-Inch Firework Cylinder Shells

After using flying fish fuse
to make a reloadable star-gun cake, I thought it might be nice to go a little further and make some simple, small shells using that same fuse. I’ll also make some star shells using the same methods.

These little, quick-and-easy shells are perfect for testing stars in a star shell, and also for simple shells for a backyard display. You can even attach a gold glitter rising tails
to the top of them.

With a little assembly line, a bunch of these small shells can be quickly made for launch the same day.

I’ll be starting with Skylighter #PL1020 plastic aerial fireworks shell casings. These casings come with a 3/32-inch hole in the cap to accommodate visco time fuse. One could also use PL1022 casings, which come with a 7/32-inch hole in the cap. (I won’t be using the fuse hole in the cap, so the size of the hole makes no difference.)

I’ll also be using 1/4-inch Chinese time fuse, GN2010, and super-fast paper firecracker fuse, GN1205. This same paper fuse can be used as a quickmatch leader by wrapping it with peel-and-stick aluminum foil duct-tape, or standard GN3001 quickmatch may be used.

I will not be inserting time fuse in the caps of these shells, but through 1/4-inch holes drilled in the bottom of the plastic cans. This makes the assembly that I’m about to describe really easy.

In Really Nice 4″ Plastic Ball Firework Shells, I describe 1/4-inch time fuse, how to determine its burn rate, and how to split and cross-match it.

The roll of time fuse that I’m currently using burns at a rate of 2.2 seconds per inch. For these little shells I want a 1.5-second time delay, which is about 5/8-inch of time fuse. I also add one inch to that length because I’ll be splitting each end 1/2-inch for the cross-matching, so I cut 1-5/8-inch lengths of the time fuses for these six shells.

I split and cross-match one end of each fuse with short pieces of black-match out of the fast paper firecracker fuse or out of quickmatch.

After making sure the time fuses fit through the holes in the plastic cans, I put a ring of hot-glue around the mid-point of one fuse at a time, and insert the fuses into the plastic casings. I pull each fuse through until the cross-matching on the fuse is almost against the bottom of the can, keeping the cross-matching just out of the inside hot-glue.

Then I add another bead of hot-glue around the outside of the time fuse to really seal the lift gasses out of the shell when it is launched.

I fill the holes and recesses in the casing caps with hot-glue now.

In an early blog post I demonstrated making and testing high-powered black powder. This article included several different ways of making and granulating black powder. In the following step, either granulated BP or BP coated on rice hulls may be used. Either commercial or homemade black powder would be suitable.

I fill the bottom of the shell can with BP coated rice hulls up to the level of the tip of the time fuse.

Then I cut a circle out of cheap tissue paper and press it down on top of the black powder charge. A small paper cup, with the sides slit with scissors, is the perfect size for pushing the tissue paper down into the plastic can.

Then it is a simple matter of filling the shell casings with stars or flying fish fuse, up to a level which will allow the caps to be glued on.

Note: In some of these photos, I’ve temporarily taped the fused end of the shells that I’m working on onto the open end of empty casings so that the shell will stand upright during construction.

Next simply glue on the shell caps with PVC plumbing cement or thickened methylene chloride. I also further secure the cap on with a layer of masking tape.

The outside end of the time fuse is now split with a razor blade, and cross-matching is inserted and tied in place.

I’ll be shooting these babies out of a 2-1/2-inch HDPE mortar that I own. It’s the smallest one that I have that these shells will fit into, and they will be a slightly loose fit.

I weigh out 1/2-ounce of 2FA lift powder for each shell. Fg or FFg commercial sporting black powder could also be used.

I place the lift powder in a plastic baggie, insert a piece of quickmatch leader with some of the black match bared, and tape the baggie closed around the quickmatch leader. The excess baggie plastic is trimmed off with scissors.

I hot-glue the leader to the side of the shell, and cover the lift powder and time-fuse with aluminum-foil duct tape.

Then I tape some lengths of visco safety fuse into the leaders, and the shells are ready to be loaded into mortars and fired.

My goal with this type of shell construction was the creation of a sort of “sky-mine” effect, where the shell contents are propelled out of the open end of the shell casing. I did not want the whole shell casing splitting open and bursting like a typical aerial shell. I thought the sky-mine effect would look better, especially with the flying fish fuse.

If I had wanted a more traditional shell starburst, I would have mixed the black-powder-rice-hull burst with the stars throughout the shell, and perhaps would have used some of the slow-flash booster.

The sky-mine type of burst was achieved with the shells made in this project, as evidenced by the plastic parts when they were retrieved from the field after firing.

Here’s a link to a video of three of these little shells: one made with lemon crackling flying fish, one with red crackling flying fish, and one made with variegated (multi-colored) stars.

Note: I tried a shell with falling leaves fuse, and while some of the fuse pieces lit, many did not. I’ve heard that some of these “special effects” fuses, in order to ensure good ignition, need to be primed by dipping their ends in nitrocellulose lacquer and then in fine black powder. I did not try that, though, and will leave it up to the reader’s experimentation.

It is easy to construct a peanut shell with these small plastic shell casings. (A peanut shell is actually a shell that breaks twice, with a delay between breaks. It is called a peanut, because some shells made this way are constructed of two small, spherical shells. When taped together they resemble a peanut.)

First I construct two of the shells, in this case using multi-colored stars, using the same construction as outlined above, except that I want a one-second delay on the first shell, and a two-second delay on the second shell.

To accomplish this, I cut one piece of fuse 1-7/16-inches long, and the other piece 1-7/8-inches long. After splitting and cross-matching, this will produce the two delays I am looking for.

After I construct the two shells, I hot-glue 1-1/2-inch chipboard discs into the recesses in the ends of the shells to make those ends flat and flush.

Then, I hot-glue the two shells end-to-end, and cross-match the two time fuses. I also reinforce the joint between the shells with two turns of masking tape.

The shell is lifted and leadered as the shells above were, except I increased the amount of lift powder to 0.65 ounce.

The quickmatch is hot glued to the side of the shell, and next to the top time fuse. I open up a small “window” in the side of the quickmatch so that fire transfers to the top time fuse when the leader lights.

Then it’s simply a matter of using aluminum foil tape to cover the bottom of the shell and the passfire “window” at the top.

Inserting some visco safety fuse finishes off the shell, and she’s ready for the firing line.

Here’s a link to a video of this peanut shell.

Note: This type of shell, where the time fuses of both shells are ignited at the same time during the shell’s lift out of the mortar, is called a peanut shell, or a “piled shell.” A “multiple-break” shell is one in which each successive shell takes fire from the burst of the preceding one.

Have fun with these simple shells,

Ned

How to Make Gold Glitter Comets

One of my favorite effects is a nice gold glitter comet.

This is also one of the easiest and most impressive beginner pyro projects. Make some homemade black powder and one of these simple projectiles, and you are ready to impress the folks around you. And you made it all yourself!

This is also the simplest and most effective rising effect to put on my aerial fireworks shells. The shell is launched out of the mortar and leaves a beautiful glittering gold tail as it ascends skyward. Just as the comet tail burns out, the shell bursts. The rising effect effectively doubles the display time of the shell, and fills the sky all the way from the ground to the starburst. A tail also helps to point the spectators’ eyes at the exact spot where the shell is about to break.

Some master rocketeers put these comets on top of their rocket headings. The comet is ignited at the same time as the rocket, and leaves a beautiful glitter tail as the rocket ascends. I’ll be detailing this method in a future post.

It is also very easy to pop a bunch of these little comets out using a half-inch star-plate, and put them into a small ball shell like the post Really Nice 4″ Plastic Ball Firework Shells. The combination of some color stars and these glitter comets makes a beautiful starburst.

Note: The difference between stars and comets is a subtle one. Typically comets are fired individually, and stars are shot out of a device in a cluster.

I have a favorite gold glitter formula which I have been using for years in both stand-alone comets and as shell tails. Anytime I fire something made with this formula someone is sure to ask me what it was and how they can make it, too. This glitter is a slightly modified version of the Gold Twinklers found in Ofca’s Mastering Cut Stars, and in Weingart’s Pyrotechnics.

This formula is relatively expensive though, because of the chemicals it uses. There is a much less expensive gold glitter formulation which does not use chemicals which cost as much, but which also produces a beautiful effect. This glitter is a slightly modified version of one called D1.

I’ll be using both of these formulae in this project.

Besides the formulated glitter compound, one tool is essential for pumping comets: the comet pump.

The black individual comet pump and star plate shown in the photo are treated aluminum. The other pumps shown are aluminum, brass, and homemade, PVC-pipe-and-wood pumps.

It’s simple and inexpensive to make a 3/4-inch or 1-inch homemade comet pump as shown above. Start by going to Home Depot and getting the correct size oak dowel, a length of the corresponding size of PVC plumbing pipe, and 3 hose clamps which fit the outside of the pipe. (You can buy ready-made comet pumps from Skylighter. Skylighter pumps are rugged brass or aluminum and will typically last a lifetime. They are faster and easier to use than homemade comet pumps.)

Then cut a 6-inch length of the dowel, and a 5-inch length of the pipe, preferably with either a hand miter box or a power one to insure good, square cuts.

Using a hacksaw, slice about halfway up one side of the pipe, and remove enough of that slice of pipe so that it fits the dowel snugly at the sliced end when the gap is closed.

Sand the rough edges of the pipe and dowel, and make sure one end of the dowel is nice and square and smooth. Either seal this end with polyurethane, or cover it with a disc of aluminum-foil duct tape.

The Gold Twinkler formula is as follows:

Component	Percentage	Ounces
Black powder meal	0.68	5 ounces
Atomized aluminum	0.08	0.6 ounces	(I’m using Skylighter #CH0103)
Antimony trisulfide	0.08	0.6 ounces	(either dark-pyro or chinese-needle)
Sodium oxalate	0.11	0.8 ounces
Dextrin	0.05	0.4 ounces
Total batch weight:		7.4 ounces

The D1 formula is:

Component	Percentage	Ounces
Black powder meal	0.71	5 ounces
Sulfur	0.11	0.8 ounces
Atomized aluminum	0.07	0.5 ounces	(same aluminum as above)
Sodium bicarbonate	0.07	0.5 ounces
Dextrin	0.04	0.3 ounces
Total batch weight:		7.1 ounces

I’m planning on making one batch of each formula to compare with each other. Therefore I need a total of 10 ounces of the homemade, black powder meal. This will include:

To make the BP meal, I screen the potassium nitrate through a 100 mesh screen, and then screen all the chemicals together twice through the same screen to thoroughly mix them together.

Then I add 1/2 cup of denatured alcohol to the dry chemicals to form a damp ball of putty, which I screen through my 1/4-inch screen onto kraft paper to dry overnight.

Note: Alcohol fumes are combustible. I dry these granules outdoors to prevent the fumes from collecting and igniting.

When the black powder granules are dry, I screen them again through a 12 mesh screen or a wire-mesh kitchen colander. I then have a black powder meal which ranges from fine dust up through 12 mesh granules.

To complete the compositions, I split my meal powder batch into two, 5 ounce halves. I then weigh out the rest of my individual ingredients. I don’t screen the aluminum or antimony trisulfide, but I do screen the rest of the ingredients for each batch through my 100 mesh screen.

Then I put all the ingredients for each batch into a plastic tub, attach the lid, and shake vigorously to thoroughly mix the ingredients.

Using a small, trigger-operated, garden spray-bottle, I add just enough water to knock the dust down and start to make the composition not quite as free-flowing. I work the water into the powder with gloved hands and by capping the plastic tub and shaking it. Each batch took 0.35 ounces of the water, which is about 5% by weight.

Note: It is a good idea to used bottled, distilled water to dampen compositions containing aluminum and potassium nitrate. This helps to prevent reactions between the two chemicals. One person’s tap water might be fine to use, and another’s might cause problems.

Now it’s time to make some comets. I place my comet pump sleeve on my aluminum ramming puck after making sure that the hose clamps are tightened. I place a funnel in the mouth of the sleeve and introduce a weighed amount of the glitter composition into the sleeve.

Then I put the comet pump ram into the sleeve, place the whole shebang on my 6×6x36 ramming post, and I whack the ram with 8-12 blows with my rawhide mallet. At a certain point, the comet will start to feel solidly consolidated.

It’s just a matter of slightly loosening the hose clamps, and gently ejecting the comets from the pump sleeve with the ram. I then dry them for a couple of days in a well ventilated, warm area, or overnight in my drying chamber.

One of the things I want to record is how much composition it takes to form different length comets with the 3/4-inch and the 1-inch pumps. Those results are as follows for both formulae:

1-inch comets
0.4 ounce	1/2-inch long
0.5 ounce	5/8-inch long
0.6 ounce	3/4-inch long
0.7 ounce	1-inch long

3/4-inch comets
0.2 ounce	7/16-inch long
0.3 ounce	5/8-inch long
0.35 ounce	3/4-inch long
0.4 ounce	13/16-inch long

For stand alone comets, I’ll press them as long as they are in diameter. For rising shell tails, I’ll press them long enough so that they burn out just as the shell breaks (duration of shell timing fuse). I’ll be determining the burn time of each length comet in a minute.

Many folks would say that these comets do not need any priming because they are mostly made of BP meal, which ignites very well all on its own.

But, often pumped stars and comets have a very smooth surface, and I’ve learned the hard way to avoid assuming they’ll light without priming. They might, and they might not. So I prime everything.

Component	Percentage	Ounces
Potassium nitrate	0.75	7.5 ounces
Airfloat charcoal	0.15	1.5 ounces
Sulfur	0.10	1 ounce
Dextrin	+0.05	0.5 ounce
Total batch weight:		10.5 ounces

I screen the potassium nitrate through the 100 mesh screen, and then screen all the chemicals together through the 100 mesh screen twice to thoroughly incorporate them. Then I put them into a plastic tub, with a lid, and shake them a bit to really mix them well.

Depending on how many comets I plan on priming, I’ll take a few tablespoons-full of the dry prime comp, put it in a paper cup, and add enough water to make a thick syrup, like honey. After stirring this a bit with a wooden stick, I use a brush to coat one end of each comet. Then I dunk that end into some FFg sporting grade black powder, or some more of the homemade black powder meal. What I want is a rough, granular surface that will more easily take fire.

I allow the primed comets to dry overnight outdoors, or for a couple of hours in the drying box.

It is easy to hot-glue one of these comets onto a plastic or paper shell or header. Just put a healthy blob of the glue onto the bottom of the comet, and press it onto the device. Then apply more glue which laps up onto the side of the comet, and helps hold it in place during lift.

A more traditional way of installing rising tails on paper ball shells is to wrap the comet with a couple of turns of thin pasted kraft paper or moistened gummed tape. Have half of the strip lap onto the side of the comet, and half hanging off the bottom of it. Slice the overhang paper with scissors about every half inch and fold out the tabs. Apply Elmer’s or wood glue to the bottom of the comet and to the tabs, and press in place on the top of the shell.

I like to cover the shell’s rising tail with a disc of tissue paper, tied on with a bit of string. This dresses the shell up, and keeps the comet’s prime layer from rubbing against anything during transport. These comet tails will ignite when the shell’s lift gasses flow around them before the shell leaves the mortar.

To test fire the 3/4-inch and 1-inch comets made with the two different formulae, I shot them out of a star gun which has 7/8-inch and 1-1/8-inch tubes. I also tested some of the 1-inch comets out of a small paper mortar made with base #PL3002 and tube #TU2123.

Using commercial FFFg black powder, I had to use a flat 1/4 teaspoonful for the 3/4-inch comets, and a flat 1/2 teaspoonful for the one inchers.

With my homemade red-gum granulated BP, I had to use a heaping 1/4 teaspoonful, and a heaping 1/2 teaspoonful respectively.

I installed 3-inches of visco fuse, the BP lift powder, and then dropped the comets in. If I was making these babies for a display, and they were going to be boxed and transported, I’d use a layer of tissue paper between the comet and the BP, and a layer of tissue pressed in above the comet to hold everything in place until firing.

Both of the formulae resulted in beautiful comets, and the one-inchers would make a very nice addition to any display. I have to say I like the Gold Twinkler a bit better than the D1. The GT creates very golden, long hanging, large glitter, whereas the D1’s glitter is a bit more pale, and does not hang quite as long.

But either one is very beautiful, and the economics of the D1 formula make it quite attractive to produce.

In order to have a rising tail on a shell that lasts as long as the shell’s ascent before burst, I measured the burn times of various lengths of comets with the star gun and a stopwatch. I wrapped the comets with aluminum foil duct tape to simulate the amount of the comet surface that would be exposed and burning if it was attached to a shell.

The burn times were as follows, along with the size of the shell to use them on:

1/2-inch long	2.5 to 3 seconds	3-inch to 6-inch shells
5/8-inch long	4.3 seconds	8-inch to 10-inch shells
3/4-inch long	4.5 seconds	8-inch to 10-inch shells
1-inch long	5 seconds.	10-inch shells

For tails on 12-inch shells, I’d use 1.25-inch to 1.5-inch long comets.

Shell rising comet tails can vary from 3/4-inch to 2.5-inch in diameter or larger, depending on the size of the shell.

The one thing I’d add about these beautiful gold glitter comets is that my wife, Molly, who is not passionate about fireworks–especially really loud ones–has always loved gold glitter effects. That’s reason enough for me to use a lot of gold glitter in my fireworks.

It is easy to make a brilliant Silver Titanium Spark comet using the methods described above.

Component	Percentage	Ounces
Black powder meal	0.68	5 ounces
Spherical titanium	0.27	2 ounces
Dextrin	0.05	0.4 ounces
Total batch weight:		7.4 ounces

Fine Ti will give a short, bushy tail. Coarse Ti will produce a longer tail filled with larger sparks.

These titanium comets produce an effect which contrasts nicely with the glitter ones.

Stay Green,

Ned

« Previous Page — Next Page »