From Bob's Basement

Status: Feasible, but needs replanning

Contents 1 Introduction 2 Firing Rig 3 Ignition System 3.1 Firing Board 3.2 Ignitor 3.2.1 Coil of wire 3.2.2 Model Rocket engine igniters 3.2.3 Commercial E-matches 4 Computer Interface 4.1 Parallel Port 4.2 Microcontroller 5 Tests 5.1 Version 1.0 5.2 Version 2.0 6 Credits

Introduction

What kind of geek fireworks party doesn't have electronically fired fireworks?

Here's the scenario: we have a bunch of fireworks. We'd like to make them go boom, preferably in some kind of sequence, using the wonders of modern technology, in a flexible, automated and customisable way.

Bob's Basement to the rescue!

The project is broken up into 3 parts, the Firing Rig, the Ignition System and the Controller.

Firing Rig

This is the part which holds the fireworks to be ignited. It also needs to provide a way for the ignitors to be strapped to the fireworks in a sensible way.

Up until now, this has consisted of a bodged together bit of wood which happens to be lying around on the day, but this is hardly ideal.

The rig ideally needs to be:

• Reusable
• Stable
• Able to hold at least 8 rockets, possibly with the ability to hold more
• Have some way of holding the ignitors in place
• Be cleanable (fireworks kick out a lot of nasty smelling, greasy soot we'd rather not have to carry around with us)

If anyone with some good mechanical design skills has some neat ideas on how to solve this problem, then jump in on the list!

Ignition System

This is the part which actually lights the fuses of the fireworks. It takes a digital signal in, one per channel, which switches a high current supply to ignite the fuse.

Firing Board

The first revision looked something like this:

And had a bill of materials as thus:

• 8x 2N3904 (81-0484)
• 8x R330 (64-0062)
• 8x 1N4002 (47-3422)
• 8x 5mm LED (55-0400)
• 8x 16A PCB FORM A POWER RELAY - 12Vs (60-5032)
• 1x D-25 Male connector + lead (19-1130)
• 1x Veriboard (34-0515)

The circuit takes a low-current digital input, which controls the transistor which provides enough current to flip a relay, which essentially shorts the terminals of the high-current supply (lead acid battery works very well) over the ignitor. However, this was somewhat bulky and expensive, especially as it uses relays. Relays are also slow - so no PWM to warm the ignitors first.

A second revision was devised, which uses a power transistor instead of a relay.

The bill of materials for this was:

• 60W Transistor - TIP41A (81-0162)
• Optoisolator - CYN17-3 (58-0886)
• 10A fuse (26-1040) and holder (26-0196)
• 4 WAY 16A PCB TERMINAL BLOCK (21-0116)
• 2 WAY 16A PCB TERMINAL BLOCK (21-0112)
• 1K resister (1/4 watt carbon film (from stock))
• LED (5mm red (from stock))

Ignitor

A few different kinds of pyrogens have been considered.

Coil of wire

Big wire, with a section of small wire. Short with car battery. Heat, and hopefully ignition of the touchpaper ensues.

Disadvantage is that the firework will not immediately ignite, rather, the touchpaper will. The timings of the fuse would need to be taken into account when firing.

Model Rocket engine igniters

These things are much, much more volatile than a piece of wire and should have a much quicker effect. They will ignite when current is passed through them, rather than just getting hot. They could in fact replace the touchpaper on the fireworks entirely.

Advantage is that the firework will go ignite much more predictably than with a simple wire.

Commercial E-matches

These are designed for the job - they're basically exactly what we need (and what is used commercially in professional pyrotechnics for such things). Only thing is they're a very pricey and kinda tricky to get hold of.

Computer Interface

Various approaches have been suggested for the computer interface

Parallel Port

The Parallel port has 8 1-bit channels which you can play with (so, 8 fireworks). Simply hook each channel up to a firing channel and write some code to flip the bits.

There is a PDF with a full circuit diagram for such a device CK1601.pdf

Disadvantages: The main disadvantage of using the parallel port is you are limited by the number of channels you can use. In the basic case, you only have 8 channels. Also, parallel ports are very rare on modern computers these days

Microcontroller

A microcontroller would allow for any number of different computer interfacing methods to be used, including parallel, RS232, Ethernet or even USB.

Code would need to be written for the microcontroller which would interpret commands sent to it by the computer(s) and enable the appropriate firing channel. There is the possibility of having it do more complex things, too, for example report back data about the firing rig (e.g. which fireworks appear to have been launched) by adding a little more code.

Tests

Version 1.0

Bonfire night was celebrated with the test launch of our fireworks system. We knocked together a simple firing rack out of some old wood and the launch tubes that came with the fireworks. The ignition cables were connected via a choc block to the control board and we wired resisters onto the end for the igntition system. It was found that placing the elctronics underneath the fireworks was a pretty bad idea as it gets rather messy! We also need to review the rig design, maybe next time using something disposable as it too gets in rather a mess. The ignition system also takes a long time to reload, so something new there. There also needs to be a redesign of the wiring as the choc block is very fiddly to wire up and Cat5 tends to break rather easilly.

Version 2.0

Naxxtor made up a PCB for the firing board - unfortunately, it didn't work. In the process of trying to make it work, we managed to blow up 3 power transistors rather spectacularly. The firing PCBs didn't work at all in the end - which was somewhat academic in that there was nothing to trigger them anyway.

Credits

• Naxxtor
• Katje
• BOfH