First of all, there is a fundamental different between deflagration and detonation; in the former, propagation is limited by the speed of combusion along the exposed surface of the explosive cast or particles. (This is regardless of whether the explosive compound contains its own oxidizer, as with black powder, or uses air, like vaporized gasoline.) A compound that detonates–say, nitroglycerin, does so by propagating the energy required for detonation via a supersonic shock wave; technically, this doesn’t require any “flame” at all, and in fact these explosives are often used to put out well-head fires by detonating a shaped charge which drives away all combustable air from the base.
Note that just because something deflagrates instead of detonates doesn’t make it weak; fuel-air bombs are deflagration events, but because the fuel is dispersed and burning occurs through the cloud rapidly, the momentum of the shock front is continually accelerated, resulting in a massive explosion. However, deflagration doesn’t produce the explosive impulse (called brisance) which allows high explosives the capability of shearing metal and shattering concrete and therefore makes them appropriate for demolition and penetrating work. Low order explosives like black powder and the ANFO (ammonium nitrate-fuel oil) mixture used by the Oklahoma City and WTC-1993 bombers can do considerable damage in large amounts but it’s not what you’d select for cutting down a bridge or steel structure. You can actually get a pretty good blast from a paint-can of flour or charcoal powder on top of an explosive, espeically if it is confined, and if you want a big flaming explosion, you use gasoline or kerosene; that’s how the Hollywood people get those flowery explosions that impress the kids so much.
High explosives word best if they are compactified into a solid or gel without any voids or bubbles, so as to allow unobstructed propagation of the wavefront. Low order explosives that deflagrate require surface area and so are generally more energetic in powdered, uncompacted but confined (as in a pipe bomb) form. Bottle rockets and the like bind up the powder tightly–essentially solidly–in a thin path through a cardboard “case”. And indeed, larger solid rocket motors, like the Minuteman or Space Shuttle SRBs cast their propellent (which is a combination of combustible elastomer, an oxidizer like aluminum perchlorate, and a catalyst) as a solid with a star-shaped chamber (formed by a mandrel in the mold) running down the middle so as to control how fast the propellent burns. Voids or case delaminations which can cause unintended shock waves (possibly resulting in catastrophic grain fracture, case overpressure, or in extreme cases, detonation) are a big no-no.
There are two reasons why this can happen; one is that an underpacked case may not immediately ignite from the primer–this can happen with small pistol primers which generate less impulse–resulting in a delayed firing with higher than rated pressures, owing to the slower but higher ultimate impulse. The other, more significant issue is that many smokeless powders are what is called “double base”, i.e. comprised of nitrocellulose and nitroglycerin. Because of the latter, under certain peak pressure conditions (as described above) the powder can detonate rather than deflagrate. Hercules, Bullseye, 2400, and many other popular smokeless powders are double base. (Du Pont got out of the double base manufacturing business a few years ago and sold those powders to Alliant Technologies, now ATK–the same company that builds Minuteman motors and Shuttle SRBs. :eek: ) This is why it is not a good idea to deviate from reloading manuals when dealing with double base powders; a slight change in propellent weight can result in dispropotionately and potentially dangerous high pressures.
And from the “Don’t Try This At Home” department, I used to make a faux gunpowder from sugar, sulphur, and potassium nitrate; without going into details, you heat up the mixture in an open pan until the sugar carmelizes, then grinding it into a powder. Although I never had an accident, this is quite dangerous, and another would-be chemist from a nearby school managed to remove parts of a hand and some of his face making this. After that, I just stuck with the old nitrogen tri-iodide that my grandfather showed me as a wee lad, and was never tempted to make nitroglycerin or mercury-based primary explosives even when I learned the procedures. It’s just not worth losing vital parts of my anatomy.
Stranger