An aerospace engineer explains fireworks | Hacker News Reader

An aerospace engineer explains fireworks

datagenetics.com

105 pointsraybb36 comments

When I was a kid, every 4th I'd buy a bunch of bottle rockets and attempt to make a multi-stage one by taping them together and varying the length of the fuses. What would happen is the one with the shortest fuse would go bang first and blow off the fuse of the other.

I finally resolved this problem by locating stage 2 further down the stick (instead of adjacent), with its fuse up into the exhaust of stage one. Then it would light as it was ascending, and the stage 1 bang was far enough away not to disturb stage 2.

3 stages did not work because stage 1 could not lift the weight. A dual booster bottle rocket did not work because the fuse timings were too erratic. Always one would go first and blow the fuse off the other. (Wasn't that the problem with the Soviet moon rocket?)

Naturally, in college I gravitated towards aerospace and got a minor in Aeronautical Engineering.

Finally, when I had a real job, I could buy all the fireworks I wanted. I bought a huge pile, and set about lighting and throwing them. After an hour or so, I got bored and lit off the rest as strings. It was like someone threw a switch in my brain, I completely lost interest in fireworks and it never returned.

But my interest in jet engines, rocket engines and V8s has never waned.

If compilers paid the way games do you’d be laughing at John Carmack way back there in your rearview mirror!

Carmack is a far better programmer than I am.

I was originally a game programmer (Empire, Mattel Intellivision) but wrestling with the compiler convinced me I could write a better one, and so it went.

I’d almost believe that if I hadn’t been following your work for about 27 years

I was grinning the entire time reading this, we had a very similar childhood. Only I ended up as a sysadmin.

Still have that obsessive love of aviation and making things fly

Decades ago I purchased a printed DIY fireworks guide from skylighter.com, it looks like they've since made a lot of that information freely available on their web site:

https://www.skylighter.com/pages/free-homemade-fireworks-pro...

This is going to be fun…

Thank you for sharing!

Thanks for sharing, this is awesome. When I inevitably lose my eyebrows, I'm blaming you!

When you've got a firework with hundreds of little flammable bits inside making the star pattern, how do they all get ignited? Does the center fuse ignite some main charge, which through heat and expansion ignites the rest of the bits? I mean, the puzzle to me is that the parts are not all connected by fuses, so how do they go off?

Read the article, explains it and shows cross sections of fireworks to aid understanding.

Yeah, I saw that photo, and thanks. But it didn't really explain to me how pieces of disconnected carbon ignite each other without some physical link. Unless somehow the inner particles create a cloud of fire that ignites the larger balls? That's the part I'm not getting.

I agree the article doesn’t really explain it. The short answer is the burst from the shell generates enough heat to ignite the payload (the stars) before they’re blown out into the sky. No fuse is needed because the payload is flammable enough to ignite but not so flammable that nothing interesting is seen from the ground. That sounds like a cop-out but it’s why designing fireworks is a science and art. :)

The photo seems to show a layer of the same material as the exploding core between the layers of spheres. I presume it's gun powder.

The purpose of gunpowder is to burn rapidly. Burning generates a lot of hot gas that tries to expand. The bomb's physical shell, along with the inertia of the stars, resists the expansion for a brief moment, creating pressure that's used to propel the stars away at high speeds, but also ensures that the burning gunpowder charge has time to ignite the stars before they fly away.

Gunpowder is meant to burn at controlled rates, but traditional gunpowder, black powder, used in fireworks and older firearms typically burns very quickly. Smokeless gunpowder for modern firearms have controlled burn rates. Depending on barrel length, etc. you may want the gunpowder to burn faster or slower to affect the ballistics. I imagine they use certain ratios of differing ingredients to control the rate of burn in fireworks as well.

Burn rate for black powder depends on particle size. Fine powder burns faster than coarse powder. So one characteristic of high-quality powder is uniform particle size.

But yes, changing the formulation also affects burn rate. For example, replacing potassium nitrate with potassium perchlorate yields a much faster burn.[0] Also replacing the carbon with aluminum dust gives you flash powder.[1]

0) https://pyrodata.com/compositions/Black-Powder-Perchlorate

1) https://pyrodata.com/composition/flash-powder

Everything is coated with highly flammable powder, so they have no trouble igniting from the burst.

It's a challenge actually to prevent everything from igniting prematurely. Especially with mortars, since they're propelled out of a tube with an explosion that can compromise the shell, igniting its contents immediately rather than up in the air as intended.

If you've ever seen a public fireworks display where the stars exploded awkwardly close to the ground instead of way up with everything else, it's likely because the explosion for thrusting it into the air managed to ignite the contents instead of the delayed fuse.

>> they're propelled out of a tube with an explosion that can compromise the shell

I've always wondered why large fireworks use explosive 'lifting charges' rather than a proper rocket design (slower burning fuel). Is it a legal limitation against creating missiles?

This is partially addressed by the article:

>Simple stick stabilized rockets work out well for small pyrotechnics, but don’t scale well. If you want bigger bangs and larger bursts in the air you need to get more payload up there. A larger rocket, with a few pounds of excitement at the front end, would require a stick 20-30 feet in length; not very practical.

En essence, rocket like designs are less stable, less predictable, and more complicated.

It is hard to have something simpler and cheaper than a ball.

Except even the article acknowledges that the stick's just doing the job that fins normally would do on other rockets. I know from first-hand (well, 1.5 hands; my stepdad was the one building and launching, and I was just there to watch and occasionally help out) that model rockets are capable of putting some decently-heavy payloads quite a ways into the air without a 20-30 foot long stick, and while model rockets are typically designed to be a lot sturdier than bottle rockets, they're also typically designed to be reused whereas bottle rockets are not (if such a requirement didn't exist, then they could be built to disintegrate and explode in the air, and be dirt cheap).

While I can't seem to find any specific rules from the FAA, Tripoli, or NAR specifically prohibiting the launch of a rocket that deliberately explodes in mid-air, my impression from my recollections of talking to expert rocketeers at launches (and asking them the sorts of questions a 10-year-old might reasonably ask, like "can I launch a rocket from an RC plane?" or "Can I launch a rocket from another rocket?") is that this would be thoroughly frowned upon my launch officials and/or regulatory bodies, and if the rocket exceeds the specifications for "Class 1" it'd require prior FAA authorization (and I can't imagine "Hey I'm gonna strap an H motor to this explosive ball and hope for the best" to go well on that front).

Author of article here. Thanks for kinds comments.

Yes, it's possible to stabilize bottle rockets with fins (to adjust the CP), but as the article hints, the stick is multipurpose in helping with launch, as well as stability (plus being easy to manufacture with low precision).

I just published a follow-up article talking about bigger rockets. http://datagenetics.com/blog/july22020/index.html

Aeronautics is a complex science and my articles are intended to pique interest, not oversimplify or belittle. It's a hard balance to explain things in a few hundred words. I hope you enjoy them and, if there is interest, they become the catalyst for you to learn more. It's a fascininating subject.

That is a fantastic follow-up article. Bookmarked :)

Re: the stick...

> the stick is multipurpose in helping with launch, as well as stability (plus being easy to manufacture with low precision)

True. Model rockets achieve mostly the same, except instead of keeping the stick attached to the rocket, the stick is instead attached to the launchpad, and is stuck through two (or more) holes on the side of the rocket (fancier/bigger rockets typically opt for rails, but a simple metal rod is typically good enough for smaller rockets).

Like I mentioned, if such a rocket were to be built to be disposable (i.e. with a "recovery system" that just disintegrates the rocket instead of something like a parachute), they could probably be cost-competitive and performance-competitive with mortars (probably a slightly-higher upfront cost due to the launchpad and electric ignition system instead of a traditional fuse, but not by a whole lot); that they ain't typically used for pyrotechnics seems to imply a regulatory issue with doing so.

I haven’t done the math, but I would assume it’s more efficient. Ignoring drag, if you launch out of a tube, you don’t need to carry any of your fuel with you, and, at least for very subsonic muzzle velocities, you might be able to arrange for the pressure in the barrel behind the shell to be near atmospheric when the shell exits. In other words, most of the launch charge energy could actually go to kinetic energy of the shell.

In contrast, rockets are very inefficient. Except in the special case where the exhaust velocity equals the rocket’s velocity, a large amount of the engine’s energy ends up as kinetic energy in the exhaust.

With a mortar shell most of the lofted mass is burnt or converted into tiny shreds. With a rocket the casing plus stick/other-aerodynamic-guide falls in a few coherent pieces, sometimes burning, which means for large (public display) fireworks would restrict the launch site and direction, ideally out over a body of water. Somewhere like Cape Canaveral in fact...

Thanks for that info!

That video of the world record-setting firework is impressive. I started to think, the amount of lifting charge needed to get a car's worth of mass up that high must be big enough to start worrying about blowing the launch tube (and firework itself) into pieces?

Yes for amateurs. But we have plenty of experience from military applications about how to launch a massive explosive really far. WWI artillery shells could weigh that much and travel much further, though they weren't launched straight up.

The caption says it was a 26 ft launch tube. I wonder, would you put the explosive at the very bottom, or in the middle? Packed right next to the firework, or some distance from it, for the gases to be able to act most effectively?

Years ago I was watching a public firework + music and was a bit disappointed in how out of sync everything looked.

I figured there must be an economically feasible way to trigger the explosions remotely and in sync for such public displays. So essentially shooting the firework the traditional way but igniting the payload on demand. Is someone working on this?

I'd imagine fireworks + music would be difficult because of the difference in the speed of sound vs the speed of light, which would explain the out of sync experience you had.

Apparently there are software simulations for this.

Something like https://www.youtube.com/watch?v=CNjggrxUQ78

Not sure at what distance the speed of sound vs speed of light becomes disrupting

At 343 meters / 1125 feet the sound is going to be about 1 second out of sync with the light, and that's going to sound terrible.

Wind direction and atmospheric pressure is going to make a small difference.

You can have fireworks go off such that the light from the next one reaches the crowd at the same time as the sound from the first. Tempo can be adjusted by changing the location/altitude where they go off. Then you only notice something off when you transition between fireworks that sound different, and even that can be mitigated by having transitions where one of the old type goes off simultaneously with one of the new type. You can't really adjust it on the fly, but you can break the show into discrete blocks that are synced to pre-recorded music, for example each song, and trigger each block on a cue to make what looks like a smooth continuous show.

I've been to several firework displays where the synchronisation is actually pretty good. Specifically, a few by Kimbolton Fireworks in the UK.

I'd imagine one challenge is the area over which the crowd is spread. For large displays you can have the crowd spread over quite an area with quite a variation in arrival time of the sounds. I suppose that's just the normal sound engineering problem of a large gig - lots of speakers with suitably configured delay.

Very interesting — I can’t wait for the opportunity to see this counterintuitive property whereby bottle rockets launched into a wind will steer themselves into the wind.

See also: what happens when you have a helium balloon in your car and you hit the brakes.

does anyone know where you can find bottle rockets without a report or a whistle? i do not want to cut open existing rockets.

Reply on news.ycombinator.com