Don't blow your fingers off with this PIC idea

[bartman]

I thought of this for a project. I don't know if I would ever bother with it, but how would you guys attack this one.

A system to safely set off consumer grade fireworks.

Possible operation:

1) Properly place the various shells and roman candles so they are secure
2) Use model rocket igniters to light the fuses attached with small alligator clips (same as a model rocket launch system works like from Estes)
3) Arm the system
4) Move back a safe distance and use a remote control to start the sequence or push a button to start a 30 second countdown while you move back
5) PIC, depending on program, either randomly ignites the fuses, ignities the fuses in a pre-programmed sequence or ignites the fuses one at a time with each push of the remote

Issues:

1) Limit on the number of firework "events" - while perhaps two or three fireworks could be ignited with the same ignitor the number of igniters is limited to the outputs on the PIC (assuming a system that tries to remain as simple as possible)
2) Safety mechanism - There could be NO chance of power going to the igniters at any time other than when commanded. No output spike when first turned on or anything like that.
3) Will an igniter light a fuse? It lights the powder in a rocket engine working on just a couple volts from a 9v battery.

Possibilities:

In theory it would be possible to program the fireworks to a short piece of music. It would be a "poor man's" system, but since the consumer grade garden variety fireworks usually are roman candles, shooting stars or shells that only go up about 150 feet timing issues are not as critical. Assuming an igniter was placed close to the firework tube one could assume there would be only a very brief delay from ignition to firing so timing the music to the dramatic parts is mostly an issue of PAUSE statements and possibly staring the program .5 seconds before the music starts to allow for burn and launching.

Not something I would want to market, but only something I might create for myself.

Even with my limited knowledge the program itself wouldn't be difficult. I could see using transistors to route a higher current to the igniters. Not sure how I would handle the safety issue with an accidental firing though.

Just throwing the idea out to see what you guys think, but if someone starts to sell it I want a percentage!!! :-)

Bart

[mister_e]

The core of it have to work with some midi sequence and a PC to MIDI adapter. Some of them are directly plugged in the PC Joystick connector. I still have some here. With a google search i found this one
http://www.borg.com/~jglatt/hardware/pc_intfc.htm

Use a PC software capable to generate midi sequence or note like Vegas, Cubase Audio or many other. In there, set your MIDI notes with your favourite audio track

Interface: The PIC grab the MIDI note, for x note it activate x igniter... and you're in business.

Sounds easy hey!

you just have to do a search for the MIDI protocol. There's a ton of example using a PIC. Probably you'll also find some PicBasic code. I remind some thread here so far.

Now your unit may fit in a small box... route few meters of wire for your safety and keep a FireTruck close

Personnaly, without knowing the current requirement of an igniter, i think that some nifty 5V reed-relay would do the job. Some of them need less 25mA and have internal back emf diode, so can be directly connected to the PIC.

Just throwing the idea out to see what you guys think, but if someone starts to sell it I want a percentage!!! :-)

it's still a public forum, anything posted in here could be done with or without your agreement by xyz person...

[mister_e]

Simpler than ever but need a little sacrifice about the Stereo effect... but the whole thing can be done within an hour.

Record your audio track on the Left channel, place some special tones frequency on the right channel... your pic is attach to the Right channel of your CD player/PC soundcard and he just decode the tones. For a x tone it activate 1 or multiple igniter.

KISS engineering!

-------------------------------------------------------------------------------------------------
Depending how many OUTs you need... PIC16F628 is more than enough (16F627 would also do). Use the internal comparator to convert the dinky audio signal to TTL compatible, feed the comparator output to the T0CKI input, use the internal timer/counter to count the frequency (or use count) and use an array, Lookup or whatever else method to convert frequency to outs.

Some other see something much simple?

[bartman]

"...it's still a public forum, anything posted in here could be done with or without your agreement by xyz person..."

It was a joke. Except if you built it Steve, then I want my money!! :-)

I used to launch the rockets using a 6v lantern battery. I guess it wouldn't take much to figure out what is needed. The little relays would be a good idea.

As for timing. I was thinking more along the line of pushing PLAY on the stereo at the same time as pushing "START" on the firing system and doing all the timing strictly on the PIC. So, each "Show" would be its own program. Your idea may be easier to "program" though using some sound editing software and offsetting the right channel for the launch signals to allow for timing of the launch to the explosion.

What about safety measures when hooking up the igniters to the fuses and starting the PIC hardware? I've been thinking of a separate physical switch that would apply the igniter voltage to the system once the software was running and stable.

Bart

[mister_e]

Except if you built it Steve, then I want my money!! :-)

LMAO! you know what else i could build... and no way i will

Back to the topic, It could be something like... when you press 'play', the right channel send another tone to apply power on the relay contacts. on your device you may add a 'Fire fire fire' panic button to shut down the power...

But it could be useless too assuming you'll use the N.O. contact of the relay... You may decide to monitor the target igniter line in case there's a short on one line... a fuse in serie will be enough and more simple for you.

Safer is better.. sometimes.

Hook your igniter, then apply power on your device. In case of a defect relay... it could save your... kinda beauty ROFL!

[bartman]

Okay, this really has me thinking now.

I looked at the dinky sound editing program I have and it can generate any tone of any length. It also can generate touch tones (which might be helpful too).

If I'm understanding you correctly, it wouldn't really matter what the tone was as long as the program was designed to recognize a specific tone on the right channel and ignore any other possible noise. The program would just open the valve to supply power to each igniter with each tone it encounters on the track. Just a simple counter - lite 1, lite 2, etc.. The software itself would be only a few lines of code with the most complicated part (for me) recognizing the tone.

I'd want some other feedback, but would need to do it without using any other outputs. Perhaps a RED led when it is firing turning to GREEN when complete. Either a single one or maybe a row, one for each igniter. A tricoloured led on the output... I need to think on that. Maybe a multiplexer to have more "channels"?

So, in practical terms a music selection of suitable length is loaded into the sound editor, strip away the right track. Study the music and add tones on the right side corresponding to the musical peaks then offset the right track with whatever delay is required to account for fuse burn and timing for the firework bursts.

Burn the CD and put in a discman or laptop. Modify the headphone plug so the left goes into an external amp and speaker set up and the right to the PIC control box.

Start up PIC box then arm igniter power.

Push play on the discman and watch the fun.

I just need to play with the igiters and fuses to see if they will ignite. I can't see any reason they wouldn't.

Bart

P.S. I'd have to totally wrap my head around different safety methods before building that portion. Mentioning shorts is a perfect example as I didn't consider that type of malfunction. I'd need something simple, but reliable. I think that a tone that would trigger the main power to turn on is a good measure. I'd still want to add a physical switch to apply power, but not have the final switch until I'm out of the way.

Another thing is the line from the discman to the control box. I'd say it would need to be 50 feet at least. 100 even better. What kind of signal would you get out the end of a 100 foot line from the earphone plug of a discman? Enough to even see it at the PIC?

[keithdoxey]

A system to safely set off consumer grade fireworks.

Issues:

2) Safety mechanism - There could be NO chance of power going to the igniters at any time other than when commanded. No output spike when first turned on or anything like that.

Given that you are having long cables connected to your PIC both firing the igniters and supplying the audio to sequence the show, how are you going to adequately protect the system from RFI pickup.

A signal from a mobile phone or taxi radio etc could cause the system to behave erratically and fire an igniter at the wrong time.

Whilst I agree that it could be a really cool thing to develop, the potential for serious injury is something that should be considered VERY SERIOUSLY. Personally I believe that this is something that should remain in the domain of the Professional events organisers.

By their very nature "consumer" firework parties tend to be relaxed affairs often with the presence of alcohol and an automated display would have to have an "Emergency Stop" function in case someone strayed into the area of a firework or if a rocket fell over just as it was about to be fired.

I have been to organised displays where things havent gone according to plan and had a rocket miss me by a couple of feet

[bartman]

Stop being a party pooper! :-)

I'm not planning on putting on any shows outside of my own yard in a rural area. This is just a fun little device to think about.

The only long wire is the one from the CD player to the controller. The controller would only be a couple feet away from the fireworks. Emergency off is stopping the CD I think. No tone going to the controller.

I actually think it is safer this way than someone standing within 10 feet lighting fuses one by one or getting bold and trying to light 3 or 4 at once and having one go off in their face. This system has everyone 100 feet away or more.

Bart

[Madscientist]

If you are looking for a way of decoding touch tones, I suggest using the 8870 chip available from BG micro in Texas. It is a very simple to use 18 pin DIP chip and converts touch tones from a telephone or receiving radio into a 4 bit binary code. I use it for remote control projects. It works flawlessly every time! It only needs about 5 extra components such as resistors, a crystal oscillator and a few capacitors. Speaking of which, I need help on another project. Please see my posted question I left on 7-15-2006 under the user name "Madscientist". I am brand new to this site.

Darren

[bartman]

Testing of remote igniton was a success. A model rocket igniter easily sparks a fuse using 4 AA batteries and 20 feet of wire.

The next step in fuse regard is making repeat and reliable connections between the igniter and the fuse itself.

A little Google search finds that I am not so crazy after all. I've found sites where people are making fuses and gadgets to both hold and lauch various consumer fireworks. I haven't found anyone using a computer and PIC controller to make it all automated, but many manual controllers are out there.

Anyone else have any ideas on possible components? I'm thinking now that 16 lines is not enough. Thirty-two would be more reasonable for a good show. Also, one line may need to ignite 5 or 6 fuses at once so the current needs to be enough on those lines not to fry the contoller, but send enough juice down the wire to spark all those fuses at once.

An ability to link 16 line system together may be a better solution. In this case using different tones to talk to each 16 line system might work well.

Bart

[mister_e]

Relays or MOSFET could be a good choice to drive the igniter. Look at your local part supplier if you could find some overstock relay. I found some in here 10Amp rated, 5V coils for less than 2$.

32 outputs... why not. Many 40Pins device gives you 33 i/o. 1 left for the audio input... just perfect. OR, using a smaller one with an i/o expander... in this case you don't just add extra part... but also a little bit of delay... but few ms will never ruine the show. In another hand, if one day you decide to add other outputs, you just need to add extra i/o expander.

Let's say a PCF8574, give you 8 i/o and you can use 8 of them on the same I2C bus... 64 outputs.. not bad.

A pic 12F would do the job, leaving few i/o for audio and stuff.