The igniters that I used for my early rocket motors (circa 1972) were quite simple, consisting of a pair of electrical wires, with a short length of nichrome wire soldered to the two leads. Since this concept does not guarantee ignition unless the nichrome wire is actually contacting the propellant grain, later igniters had a small "ball" of propellant cast onto the nichrome wire. This method of ignition was more reliable, however, there were still other drawbacks affecting the reliability. For example, if the power supply was, for some reason low (or if it was cold, such as during a winter launch), the propellant ball on the igniter would char, but not ignite. As well, since the nichrome wire is brittle, breakage would sometimes occur. There was another more significant drawback. I found that, after ignition, the motor would burn for several seconds to "build up pressure" before suddenly thrusting with full force. There was clearly propellant being wasted during this phase of the burn. This occurred because of the burnrate characteristic of most rocket propellants. The rate of burning is strongly dependant upon pressure. At atmospheric pressure, the propellant burns slowly, and only starts to burn rapidly once sufficient pressure has built up in the rocket motor. Only then does the motor produce appreciable thrust.
To solve this problem of wasted grain burning, and therefore reduced performance, I then tried installation of a "burst diaphragm" in the motor. This is a thin sheet of brass that blocks the nozzle throat, and allows pressure to build up in the motor to achieve efficient burning. At that point, the diaphragm should burst and be ejected out ot the nozzle. I experimented with such a concept, but this did not work satisfactorily. In fact, such an arrangement caused a motor failure on one of my flights (Flight C-18, July 1980), The problem was in sizing the diaphragm to burst (reliably) at exactly the right pressure, which was not simple to achieve, at least not so with my narrow-throated motors. I abandoned this concept, and tried an alternative approach--to devise an igniter that would more reliably and effectively ignite the propellant, and would serve an additional function--to simultaneously pressurize the motor. A Black Powder charge would achieve this. Black Powder is easy to ignite, and by placing the charge in a sealed cartridge, it would burn nearly instantly, releasing a sizeable volume of hot gases and molten combustion products. This served to pressurize the motor while very effectively lighting the grain. The motor then achieves full thrust almost immediately.
Over several flights and static tests, this pyrotechnic igniter proved to be highly reliable and significantly boosted the performance of the rocket motors.
This is an illustration of what is referred to as a pyrotechnic igniter which was used for igniting the B-200, C-400 and similar sugar-propellant rocket motors. It consists of a length of polyethylene plastic drinking straw filled as shown with a charge of Ignition Powder, and as such, is often referred to as a "straw igniter".
The igniter is sealed at both ends with polyethylene "hot-melt glue". The nichrome (nickel-chromium, high resistance) wire serves as the heating filament (bridgewire), and is soldered to the ends of the copper wire leads using solder. Nichrome wire is quite inexpensive (about $0.20 /ft.) . Alternatively, a strand (or two) of coarse "steel wool" may be used in place of nichrome wire, or even a strand of fine copper "speaker" wire.
Nichrome wire can be tricky to solder. An easy and secure way to attach the nichrome wire to the electrical lead wires is as illustrated below:
For added safety during storage, the bared ends of the wire leads are shunted, (twisted together) to eliminate the possibility of inadvertent current flow through the igniter
The electrical power source for my firing system had consisted of four 1.25 V nicad batteries (see Launch Support System), but I then switched to a pair of rechargeable Nickel-metal Hydride (NiMH) batteries due to the excellent current delivery capability. The pair consists of two 3.6V cells (300 mAh each) wired in series which deliver a generous current at 7.2V. These cell phone batteries also have excellent cold-weather performance.
Important note: Use of an igniter containing a much greater charge of BP than shown in Figure 1 may result in an initial pressure "spike". This overpressurization of the rocket motor during startup may result in motor failure characterized by shearing of the safety bolts retaining the rocket motor head. The effect of igniter size on the motor chamber pressure curve can be significant for "small" rocket motors (such as the A-100, B-200 & C-400). This is illustrated in Figure 2 for a particular BEM (Ballistic Evaluation Motor).
a) 3 gram igniter
b) 7 gram igniter
c) 15 gram igniter
Ignition PowderThe Ignition Powder that I currently use is a modified Black Powder mixture, and consists of 80% Potassium Nitrate and 20% Charcoal, by weight. Sulphur is not added to this mixture. The function of sulphur in conventional Black Powder is mainly to facilitate easy ignition. This is great for flintlock guns, but for safety reasons, it is not such a good feature for rocket motor igniters, nor is it really necessary.
Chamber Pressure due to Igniter Combustion
The combustion of the igniter charge serves two important purposes:
The pressure that is generated by combustion of the igniter charge may be determined by the following equation:
P = pressure in combustion chamber at time, t lbf/in2
r= density of charge material lbm/in3
D = loading density = C / V lbm/in3 C = original mass of charge lbm V = free volume of combustion chamber at time, t in3
C = original mass of charge lbm
V = free volume of combustion chamber at time, t in3
R = universal gas constant in-lbf /° R-lbm
M = effective molecular weight of combustion products (system mass divided by number of moles of gas) lbm / mole
T = adiabatic flame temperature ° R
G = fraction of original charge mass consumed by time, t dimensionless
Pa = atmospheric pressure lbf/in2
The effective force, l ("lambda"), may be calculated from the values of M and T determined from a combustion analysis (e.g. using PEP program) for any charge material. Alternatively, it can be readily determined if the impetus for a particular charge material is known. The impetus for Black Powder (commercial grade) is approximately 100 000 ft-lbf/lbm. Thus, the effective force for Black Powder is 100 000 * 12 = 1 200 000 in-lbf/lbm.
Derivation of this equation
For convenience, an Excel spreadsheet is available which determines combustion pressure using this method:
Pyrogen IgnitionA pyrotechnic igniter, as described above, works very well for starting smaller sized rocket motors. However, for larger motors (i.e. K-class & larger), a pyrogen ignition system provide superior motor starts. A pyrogen is essentially a small rocket motor mounted at the bulkhead. Nearly instantaneous ignition of the motor grain is assured by the high velocity, particle-laden flame that emanates from the pyrogen. The pyrogen used for the Kappa rocket motor is shown in Figure 3.
The grain used for this particular pyrogen is cast KNSU propellant, chosen for its ease of ignition and rapid burn rate. The pyrogen grain is ignited by a black powder charge, initially contained within the pyrogen canister by a burst diaphragm. This charge additionally aids pressurization of the motor.
An easy to make pyrotechnic igniter can be made in a manner similar to the igniter shown in Figure 1, except that instead of utilizing a nichrome bridgewire, a Xmas-tree mini-bulb can be used in its place. This particular light bulb has the advantages of minimal cost and very small size, which conveniently fits within a soda straw, and requires only a small electrical current to fire the charge. The tiny tungsten filament is instantly vaporized when current (such as from a 9V battery) is applied, and as such, is referred to as an "exploding bridgewire" igniter. This igniter design is based on a concept pioneered by rocketry experimenter Rob Furtak, who has used the Mini-bulb igniter with great success in his rocketry work.
The igniter described here may be used for either motor ignition or for firing a parachute ejection charge.
To make this igniter (shown in Figure 4), the plastic base of the mini-bulb is first removed (by pulling) and discarded. This exposes the two copper wire leads, which are then scraped clean of oxide, The glass bulb is then carefully broken open. The simplest means is to slowly squeeze the upper half of the bulb in a bench vise. The bulb is first wrapped in a tissue to catch the tiny shards of glass that erupt once the vacuum seal is broken. Safely glasses must be worn during this operation as a redundant safety measure. Care must be taken to prevent damage to the filament bridgewire or to break the lower portion of the bulb.
Most Xmas bulbs have a very fine wire or piece of foil that is wrapped around the base of the two leads that the filament is attached to. The purpose of this shunt is to maintain continuity in case one bulb in a string of bulbs burns out. This shunt must be removed with a toothpick or pair of tweezers. An ohmmeter should subsequently be used to check continuity to ensure the filament has remained intact and that the leads are not shorting. "Infinite" resistance indicates an open circuit, meaning the filament or solder joint has broken. Zero resistance indicates a shorting condition, likely caused by the two leads touching each other. The actual measured resistance value, based on my experience, could be anywhere between a couple of ohms and more than a hundred ohms, and as such does not provide any useful information other than to indicate an "OK" condition. Electrical lead wires are then soldered to the leads of the bulb. It is recommended to use acid flux when soldering the bulb leads to ensure a good solder joint. The other ends of the lead wires should then be stripped, and shunted (bare ends twisted together) for safety.
The mini-bulb is then placed within a 2 inch (5 cm) length of polyethylene soda straw, and the end nearest the leads sealed with hot-melt polyethylene glue.
Ignition Powder (approximately 1 gram) is then carefully loaded, and tamped every so often to eliminate voids. The final step in preparation of the igniter is to tamp in a small ball of glass wool (fibre glass), and seal the end with hot-melt glue.
Despite the fragile appearance of the mini-bulb filament, testing has demonstrated that it is very robust. In one experiment, 20 Xmas-tree bulbs were mounted (radially outward) on the circumference of a 8 inch (20 cm) disc, which was subsequently spun at high rotational speed on a lathe, subjecting the bulbs to over 120 g's. Afterward, the bulbs were tested and all were found to function normally.
Ultra-low Current IgniterThe following is a description of an igniter that requires very low electrical power, requiring only 20 mA at 1.2V (= 25mW) to fire. As such, this design is exceptionally reliable and especially useful in cold weather operation, which greatly reduces a typical battery's available power.
This igniter may be used for either motor ignition or for firing a parachute ejection charge.
The "Ultra-low Current Igniter" was developed for EARco (Experimental Aerospace Research) by Ken Tucker to increase the safety of Rocketry.
Safety is paramount in rocket experimentation, and so for reliable
operation of parachute ejection systems, 2nd stage ignition, and for any
system that may create a hazardous condition in the event of an igniter
malfunction, a very reliable, redundant igniter may be desirable.
Based on Radio Shacks Part # 272-1139 Mini Lamp (or equivalent), a 1.5
Volt, 25 Milliamp electrical igniter is described. These lamps come
equipped with pre-attached wire leads to facilitate hook-up (see Figure 5). In practice,
the lamp will only require 1.2 Volts and 20 Milliamps to function.
Construction: Record the resistance of the bulb using an ohm meter, a
reading of, 20 -30 ohms is usual. A hole is ground through the lamp bulb by
carefully grinding the bulb on a fine file or polishing stone. A magnifying
glass is a handy tool to check the progress of the grind. This requires
some patience but after a little practice requires a total of 5 minutes. As
the glass is ground, a hole will occur, at that moment the vacuum within
the bulb draws in the air and some of the glass shards. To prevent the
glass shards from impacting the bulb filament, the glass should be ground
with the filament perpendicular to the hole.
Once a hole is observed, recheck the ohms reading of the bulb. It should
be the same as previous.
Carefully place some black powder into the bulb through the hole. Just enough
to be loose is fine, packing in too tightly could damage the bridgewire. The
bridgewire will explode when heated and will land on the powder, so the
powder does not need to be in direct contact with the bridgewire. Take an
ohms reading again to make sure no damage was done to the filament. A
little Scotch tape over the hole keeps the powder in. When fired, the igniter will produce the equivalent of a match igniting,
exhausting through the hole. This may be insufficient, so a secondary burn may
be desirable. Place the igniter in a tube an inch or so long, about the
same diameter as the igniter. Fill this with black powder, and seal the
ends. This amplifies the potency of the igniter (see mini-bulb igniter, described above).
This igniter is quite shock-proof. Hard data is lacking, but this
igniter survives being thrown against a concrete surface, which causes a
few 100g's of deceleration. Its very low current and voltage requirements
allow for numerous redundant igniters in parallel with the ignition power
supply. For example 3 igniters will only require 1.2 volts and 60 ma (3 x
20 ma) to function.
Cold Soak Test
A semi dead "9V" battery (5.1 Volt with
mini x-mas bulb load prior to freezing) was placed in the freezer at -4F for 6 hours.
Using the mini x-mas bulb, with a resistance load of the igniter found the
voltage to be 2.5 volts. It was very dim using the mini x-mas bulb.
Connecting an "Ultra-low Current Igniter" to this
dead frozen battery and it exploded instantly.
Safety is paramount in rocket experimentation, and so for reliable operation of parachute ejection systems, 2nd stage ignition, and for any system that may create a hazardous condition in the event of an igniter malfunction, a very reliable, redundant igniter may be desirable. Based on Radio Shacks Part # 272-1139 Mini Lamp (or equivalent), a 1.5 Volt, 25 Milliamp electrical igniter is described. These lamps come equipped with pre-attached wire leads to facilitate hook-up (see Figure 5). In practice, the lamp will only require 1.2 Volts and 20 Milliamps to function.
Record the resistance of the bulb using an ohm meter, a reading of, 20 -30 ohms is usual. A hole is ground through the lamp bulb by carefully grinding the bulb on a fine file or polishing stone. A magnifying glass is a handy tool to check the progress of the grind. This requires some patience but after a little practice requires a total of 5 minutes. As the glass is ground, a hole will occur, at that moment the vacuum within the bulb draws in the air and some of the glass shards. To prevent the glass shards from impacting the bulb filament, the glass should be ground with the filament perpendicular to the hole.
Once a hole is observed, recheck the ohms reading of the bulb. It should be the same as previous.
Carefully place some black powder into the bulb through the hole. Just enough to be loose is fine, packing in too tightly could damage the bridgewire. The bridgewire will explode when heated and will land on the powder, so the powder does not need to be in direct contact with the bridgewire. Take an ohms reading again to make sure no damage was done to the filament. A little Scotch tape over the hole keeps the powder in.
When fired, the igniter will produce the equivalent of a match igniting, exhausting through the hole. This may be insufficient, so a secondary burn may be desirable. Place the igniter in a tube an inch or so long, about the same diameter as the igniter. Fill this with black powder, and seal the ends. This amplifies the potency of the igniter (see mini-bulb igniter, described above).
This igniter is quite shock-proof. Hard data is lacking, but this igniter survives being thrown against a concrete surface, which causes a few 100g's of deceleration. Its very low current and voltage requirements allow for numerous redundant igniters in parallel with the ignition power supply. For example 3 igniters will only require 1.2 volts and 60 ma (3 x 20 ma) to function.
Cold Soak Test
A semi dead "9V" battery (5.1 Volt with mini x-mas bulb load prior to freezing) was placed in the freezer at -4F for 6 hours. Using the mini x-mas bulb, with a resistance load of the igniter found the voltage to be 2.5 volts. It was very dim using the mini x-mas bulb.
Connecting an "Ultra-low Current Igniter" to this dead frozen battery and it exploded instantly.
"Ferocious" IgniterDuring the development phase of the RNX epoxy composite propellant, it was learned early on that the "BP straw" igniter (described above) was not suitable for igniting RNX propellant grains. As this epoxy-based propellant has a higher decomposition temperature, a hotter and more sustained burn is necessary. The solution that arose was an igniter that used a more energetic oxidizer together with epoxy as a binder. The resulting product, aptly referred to as a Ferocious igniter, burns with a hot flame for a duration of about two seconds, reliably lighting an RNX composite propellant grain.
The developmental formulation utilized potassium chlorate (KC) as the oxidizer, but this was later replaced with potassium perchlorate (KP) for safety reasons. The Ferocious igniter is made as follows:
A nichrome bridgewire is soldered to a pair of electrical leads, similar to that used for the "straw" igniter. For motors that have a small throat diameter, or if the igniter is to be installed at the grain outside surface (rather than in the grain core), fine gauge wire such as 30 AWG "wrapping wire" (RS p/n 278-501) should be used. A pyrolant coating is then prepared, using the following materials:
A small quantity of epoxy mixture is prepared, using the manufacturer recommended resin/hardener ratio. One half gram of epoxy is sufficient for 4 or 5 igniters. A sprinkling of sulphur is then added to the epoxy, just enough to colour the mixture a distinct yellow or light amber. The mixture is then blended well using a wooden craft stick, for a minimum of two minutes. A plastic "pan" cut from a 2 litre soda bottle works well as a disposable mixing pan. The next step is to add the KP, a little at a time, until the resulting mixture resembles a stiff paste.
Squares of polyethylene sheet , 1" x 1" (25x25 mm) are cut out from a plastic sandwich bag (or similar), two per igniter. The igniter electrical leads are placed such that the bridgewire is centred onto one of these poly squares. A dab of pyrolant (approximately the amount equivalent to a pencil eraser) is placed onto the bridgewire, as shown in Figure 7. The second poly square is then placed over the pyrolant and carefully pressed down such that the bridgewire is fully embedded. The resulting thickness should be no greater than 3 mm to prevent fracture of the pyrolant coating upon firing. If the igniter is to be installed at the grain outside surface, the pyrolant should be pressed such that a thin disc is formed of approximately 2 mm thickness maximum. The disc form is recommended for both the Epoch motor and the Paradigm motor.
The igniter is then set aside to cure overnight at room temperature. The poly squares are simply peeled off after curing. The cured igniters can be readily trimmed as necessary. For a more rapid cure, the igniters may be placed inside a dedicated shop oven set at 65oC. (150oF.). Curing at this elevated temperature is complete after approximately 1/2 hour. .
The Ferocious igniter has proven to be highly reliable. In over 30 motor firings, not a single misfire has occurred.
Bottom left: Pressing between poly sheets
Right: Ferocious igniter in action
"Spitfire" IgniterSurely the best igniter I've ever used for igniting composite propellant is the Spitfire. This igniter is the recent outcome of a prolonged research project into developing a hot burning, reliable igniter conducted by Rob Furtak. This igniter is based on the mini-bulb igniter, but instead of using a Black Powder composition as the combustion agent, a pyrolant is used to coat the bridgewire (filament). The pyrolant consists of a neoprene binder combined with the following composition referred to as Thermex Powder:
The Thermex Powder must first be well blended. I place the constituents into a tupperware container which is mounted to a rotating "tumbler", and allow the mixing to occur for one or two hours, depending on the batch size.
The mini-bulb (Xmas bulb) must be broken open to expose the filament. I have found that the most reliable method of doing this is to place the glass ampule of the bulb into the jaws of a workshop vise, then carefully and slowly close the vise. The bulb glass ampule will suddenly fracture and leave the filament totally exposed. The glass tends to shatter so the bulb is covered completely with a suitable cloth to catch the glass fragments. Wear safety glasses when performing this step (in fact, it is a wise practice to wear safety glasses all the time when preparing igniters, especially pyrolant). I've broken open about a hundred bulbs using this method, without damaging a single filament. Note that most Xmas bulbs have a very fine wire or piece of foil that is wrapped around the base of the two leads that the filament is attached to. The purpose of this "shunt" is to maintain continuity in case one bulb in a string of bulbs burns out. This shunt should be removed with a fine pick or pair of tweezers. Otherwise, the ohmmeter test may falsely indicate that the bulb's filament is intact.
The copper-plated electrical leads, which typically are quite oxidized, should then be lightly scraped with a hobby knife or sandpaper to ensure good electrical conduction.
The Thermex Powder is then added to a small amount of neoprene contact cement at a ratio of approximately 2:1 by mass, and using a craft stick, is well blended. The resulting pyrolant blend should be as stiff as toothpaste. Using a toothpick, the pyrolant is then carefully placed onto the filament of the mini-bulb (a twirling motion around the filament seems to work well). The filament should be completely covered, and to help make the resulting igniter more resilient, the pyrolant should also extend down to the base of the mini-bulb. Working time is quite short, only a few minutes due to the high evaporation rate of the solvent, so only a few igniters should be made at one time. The igniter is then set aside to cure. Curing takes place in a few hours or even less in an environment with a slightly elevated temperature. After the cure is complete, the igniter is tested for continuity using an ohmmeter. Prior to storage of the finished igniter, the leads should be shunted (twist the leads together) for safety.
"Spitfire II" IgniterThe Spitfire II igniter is a hotter buring version of the original Spitfire. Spitfire II was developed as a result of a need for a very hot burning initiator for the hard-to-light A24 AN-based composite propellant. Theoretical flame temperature is 3180 C. The base material is Thermex II powder:
The Thermex II powder is first well-blended using a rotating tumbler. To make Spitfire II pyrolant, Thermex II is blended with 20% (by mass) of neoprene-based contact cement, and allowed to fully dry. The product is then granulated and used in a standard "straw" igniter or may be dissolved in MEK or acetone for dipping.
Hot-Particle Igniter CompoundsHot-burning formulations that produce liquid (and/or solid) products of combustion are useful for igniting hard-to-light formulations such as composite propellants. However, hot-particle formulations are also useful for rapid startup of motors with conventional sugar propellants. The first two formulations presented here are conventional thermites (see Thermite Experiments ) and the third is based on a pyrotechnic igniter formulation described in reference .
Nitrate-based Igniters for Composite PropellantThe reliability and performance of a solid composite rocket motor depends partially on the composition and design of the electrical igniter. Ignition Powder (KNO3/Charcoal) based igniters, referred to as IP igniters, continue to be used successfully in potassium nitrate/sucrose (and other sugar-based) motors because they are easy to make, relatively safe, reliable and provide for fast start-up of the motor, but fall short when used with RNX and other composite propellants that are harder to ignite by their very nature.
Potasium chlorate or ammonium perchlorate is frequently used as an oxidizer in igniter composition because of the low melting point, low ignition point and high caloric value with thermic fuels. However, an igniter composition based on potassium chlorate can be sensitive to impact, friction and heat stimuli, and as such, possess certain inherent hazards in their fabrication and usage. Potassium chlorate and the other energetic oxidizers may also be hard to obtain. Therefore, two igniter compositions based on potassium nitrate and other non exotic chemicals that are also reasonably safe to prepare were developed. Both types, despite their different designs, provide satisfactory results, particularly when used to initiate composite motors...
Read the complete article written by
Richard Nakka & Peter Lee Ericksson
MDP PyrolantThe MDP pyrolant, which is an acronym for "Modified Dan Pollino", is a pyrolant that is well-suited for igniting sugar-propellant. This formulation is based on rocketeer Dan Pollino's standard igniter pyrolant for his motors. Simple to make and uses common materials, it burns with good vigour and is reliable and easy to initiate. I have used this formulation to power my "MDP Pyrogen" igniter.
The above ingredients are blended into 100 ml water, placed in a pot (or deep-fryer), and boiled to drive off water. When product reaches wet paste-like consistency, heating is stopped. Product is spread onto aluminum foil and allowed to fully dry. It is then granulated to a particle size in the range of 10-60 mesh. The product should be stored in a sealed container with a silica-gel pack to keep it dry.
MDP PyrogenThis is a very simple pyrogen unit that I have used regularly to fire my Impulser rocket motor. This pyrogen style igniter brings the motor up to pressure nearly instantly, maximizing motor performance. This unit consists of a plastic drinking straw loaded with MDP pyrolant. To fabricate this igniter, a length of plastic tube is cut and one end closed with a small packed wad of tissue sealed with hot glue. MDP pyrolant is then firmly packed into the tube using a wooden dowel. Loose MDP is then added and a nichrome bridgewire assembly is inserted. The end is then closed off with a dab of hot glue. Two slits are then carefully cut into the side of the tube to act as 'nozzles'. The pyrogen is then inserted into the forward end of the rocket motor. When the igniter is fired, jets of flame emanate from the pair of slits, rapidly igniting the motor grain. The pyrogen burn duration is approximately 1/2 second.
Fusion Pyrolant/PyrogenI've recently begun experimenting with AP-based and KP-based propellants and as such, soon realized that a special high-energy igniter was required to ignite some of the more stubborn compositions. Another requirement was a sustained burn that generated copious amounts of hot gases and particles. In other words, not a brief "flash" that generates a pressure pulse. This is particularly important for those propellant compositions that have a rather high pressure exponent, in order to prevent over-pressurization of the motor at the moment the propellant grain ignites. After much experimenting with various pyrolant compositions, I came to the conclusion that chlorate and perchlorate based pyrolants were not satisfactory, as these tend to burn very rapidly when confined. As such, I set out to develop a pyrolant that was based on potassium nitrate. I recalled that Formula 2 propellant is a hot burning potassium nitrate based propellant and, as such, may fit the bill as a suitable pyrolant. I wanted the pyrolant to be in a powder form (to optimize the burn rate) so I substituted the polyurthane with powdered sucrose. The resulting pyrolant ignited readily and burned exceptionally well. Due to the intense flame developed, I decided to name this pyrolant Fusion. Not only is it well suited to igniting AP and KP compositions, I am now using Fusion for igniting my sugar-propellant motors.
The potassium nitrate is milled to a fine powder. Sucrose is finely powdered "icing sugar". To prepare, the ingredients are blended together fully, preferably using a rotating tumbler for an hour or more. A couple of small stones helps break up any clumps.
Harvesting Aluminum from PaintA number of hot-burning igniters featured on this webpage use metallic aluminum as a thermic agent. Flake (or leaf) aluminum is best for igniters, igniting more reliably than atomized (spherical) aluminum powder, such as West Systems 420. Spherical aluminum requires greater thermal energy to start burning due to its relatively thick shell of aluminum oxide. Flake also tends to ignite more readily due to the particular shape of its particles. Only problem is, flake aluminum can be expensive or hard to find.
Fortunately, in the true spirit of experimental rocketry, a source of flake aluminum is as close as your nearest Home Depot or hardware store. Aluminum paint is largely a mixture of linseed oil and metallic aluminum flake. According to various MSDS's, such paint contains 20%-30% aluminum, by mass.
I first gave thought to the possibility of extracting metallic aluminum from paint some years ago. Early attempts were not very successful -- the challenge of efficiently separating the metallic aluminum pigment from the linseed oil base seemed daunting. Recently, I tried a different and rather simple approach, with excellent results -- a nice harvest of flake aluminum.
From my own experience, a typical harvest is around 15%-20% of the paint mass (depending on brand), taking into consideration that some aluminum is lost in the extraction process. I have harvested aluminum from both TremClad and Rust-Oleum 7715 brands, with the former having a slightly higher yield.
To test the product, a batch of HotFlash igniters were prepared and test fired. HotFlash pyrolant is made from 30% aluminum flake, 60% KN, 10% Sulfur and NC lacquer as a bonding medium. As hoped, the pyrolant burned fiercly, as can be seen in the videoclip at the bottom of this page.
The following section describes how to harvest aluminum from a 1/2 pint (273 ml) of paint. Net harvest will typically be 35-40 grams (1.2-1.4 oz.) which is sufficient for a hundred or more igniters. Before beginning the harvesting process, allow the tin of paint to sit undisturbed for several weeks to allow metallic aluminum to settle to the bottom. To begin the process, carefully remove the lid from the container and use a turkey baster (or other similar siphoning device) to carefully draw away most of the linseed oil base. Some aluminum particles will be drawn up, but this is not important - the net loss will not be significant.
Fill the tin with acetone, then stir contents, using a wooden craft stick, for several minutes. Make sure and scrape the settled aluminum off the entire bottom and sides. Replace the lid and allow to stand overnight undisturbed. The acetone is next drawn off using the baster. Refill the container with acetone and repeat the cleansing process. The purpose of the cleansing is to remove any remaining linseed oil which, if not removed, will cause the dried aluminum to form into hard clumps.
After the acetone has been drawn off a second time, tilt the tin slightly and use a piece of paper towel to "wick" away the remaining pool of acetone. The remaining "wet slurry" of aluminum flakes and acetone are then scooped into a suitable polyethylene pan and spread out to allow the acetone to evaporate. Occasional stirring will speed up the process which will take a couple of days at room temperature to fully dry.
Using the convex side of a spoon, break up the clumps to a fine powder. Using a fine nylon kitchen sifter, sift the powder to remove any clumps. Continue until all the aluminum has been harvested as a fine powder. Store in a suitable sealed container.
Photos & Video:
References George P. Sutton, "Rocket Propulsion Elements", 4th ed., John Wiley & Sons, 1976
 Richard B. Morrow, Mitchell S. Pines: "Small Sounding Rockets", Small Rockets Press, 2000
 GUIPEP Propellant Evaluation Program