Firing Sequential Ignitors with Just One Switch?

Hi Tim...

I'm really looking for a way to chain any number of ignitors, not just

1. One button is pressed, and the first ignitor in the chain is lit and blown off the chain (electrically speaking), then the button is pressed and the next ignitor in the chain is lit, and so on.

Thanks!

The easiest way is with a rotary switch. Electronic step circuits are possible but you would need to examine all failure modes for safety reasons. You would have 10 "hot" rockets on the pad. Fly safely, Tom

Here's a thought that occurred to me - it's probably a bad idea, but I'd like to have it shot down professionally.

Construct a number of independant counters - each initialized with the sequence number of the igniter, and set to actually ignite when and only when the counter reaches zero.

Connect all your counters to a shared bus, and make it so that pushing the button signals all counters simultaneously to 'count down one step'. Thus, evey time you press the button, another counter will reach zero, and that counter will then ignite its rocket.

You can, in this setup, even configure several counters to fire simultaneously, simply by setting them to the same sequence number.

You could even then have a completely automatic launching sequence, simply by replacing your button with a 555 or similar.

As everything is nice and modular, you could start out with modules that use a four-bit counter (which will allow for a max of 15 distinct launch events), and later add wider ones if necessary, while the narrower ones still retain their usefulness.

Each counter module would need a counter, a way to pre-set the value, and perhaps a small display to display the current remaining count just so you can check things are as they should be.

Let the shooting-down begin!

Best wishes,

// Christian Brunschen

Sounds a nice idea, I'd say. And don't forget a main switch, perhaps a keyswitch, for the ignition circuit only so you can set the counters without running the risk of igniting anything by writing a 0 in one of them. ;)

**** Stephan ****

This sounds like a trip to the emergency room just waiting to happen.

You really, REALLY, need to know EXACTLY which ignitor will fire when you push the button. Don't skimp on the status readouts if you do it with a PIC. AVOID hairbrained lashed together logic. All it takes is one missed switch bounce to fire the wrong rocket.

The easiest and safest might be a row of 10 pushbuttons. If you need more, bank switch the ten buttons with a rotary switch. mike...I hate emergency rooms...

--
Wanted, Serial cable for Dell Axim X5 PDA.
Return address is VALID but some sites block emails
with links.  Delete this sig when replying.
FS 500MHz Tek DSOscilloscope TDS540 Make Offer
Bunch of stuff For Sale and Wanted at the link below.
MAKE THE OBVIOUS CHANGES TO THE LINK
http://www.geocities.com/SiliconValley/Monitor/4710/

I don't know what your budget is, but a basic PLC - like the pico from A-B - can do that along with inputs from safety observers with kill switches. It has the dry relay contacts to provide the power handling, too. And a nifty little LCD display you can program to show the countdown.....

I'd guess less than $100 for the PLC and a handful of pushbuttons. I think a pico has 12 dry relay contacts. You'd have to chain them together to get more....

For a low-tech alternative, see the nail-and-wire pegboards the fireworks guys use. Basically, a bunch of 12 penny nails with wires; you hit each one in turn to fire the next sequence of fireworks.

Good idea - but you don't need a counter for each igniter. Use a decade counter - like a 4017 - and each press of the switch will increment the count by 1. One chip will count from 0 to 9 with an individual output for each count. He can add as many counters as he needs to count from 1 to N, with each chip controlling 9 igniters. The pushbutton will need a debounce circuit - cross connected nand's can do that. I don't know what current the igniters need, but I expect he'll need a transistor on each output to fire them.

A sample 4017 circuit is shown here:

Look for 18 stage LED sequencer.

Personally, I would use the rotary switch approach to select the rocket, and a pushbutton to fire it. He could add as many rotary switches as needed to select N rockets. That way, it takes two operations to fire a rocket - 1) select the rocket to be fired and 2) push the button. Seems safer that way.

Ed

One way to do this safely would be for each rocket to physically close a switch as it goes, making the next one hot. Perhaps a little pin/string arrangement might work? A nonconductive pin holds two contacts apart. Once the pin is pulled, the contacts close, and the next one is ready. There are a few issues to work out here (like not launching two rockets when you only wanted to do one), but it is probably simple to do, and practically free.

If you can get hold of an old rotary selector from an old electromechanical phone exchange that'd do the trick... (with the one I had you could switch

50 ignitors sequentially) electromech is two generations ago so they may be rare even in the best stocked surplus outlets, the one I had needed 48V to fire the solenoid but was easily dissasembled (all bolts, no rivets) and could have been re-wound for 12 or 6V operation with only hand tools and a bench vice...

if you go with the microcontroller you can matrix the output so that 10 output devices (relays/transistors etc) can control 25 igniters, do 5 rows and 5 columns with a diode in series with each ignitor.

you could also use a decade counter (two for a matrix) instead of the microcontroller that may save you a few bucks.

a ULN2003 driver chip may be cheaper than 8 transistors, (but not in the catalogue I have here)

hmm, if you want cheap driver hardware dissasemmble an old 24-pin dot matrix printer :) 24 pin drivers and 4 drivers each for the two motors...

Another possibility would be using some contacts that close when the rocket departs so that the next ignitor is connected to the circuit, you'd need to set your button up to send only a short pulse so you don't get accidental multiple launches. +V -----------------------+ +--------------->

|nc | || __o | +---||------o-~~ | | || com o---+---/\\/\\/\\--+ to ignitors | no | | | 0V -----+---------------------------------+--->

the optimum component values depend on the specifications of the ignitors.

Bye. Jasen

Seems like timed charges get used in these building demolitions. They really dont want one of the charges to hang and stop the process once started, so I believe there are standard procedures and protocols that are proven effective in this area, I just don't know if a demo expert would tell us, because then he'd have to whack us.

It seems to me that there is a time to buggerise around,and a time to go safe and simple - I reckon firing rockets should be safe, simple, and with minimum error risks

Another poster suggested a rotary switch to select the rocket, and a pb switch to fire - sounds like the best safe and simple way to go safely

David

--- I'd go with a pushbutton for each rocket and a separate momentary spring-loaded "ARM" toggle switch under a switchguard to enable the array of pushbuttons:

ARM

+V>--O---> | | O---> | | | | O---[IGNITER 1]---+ | | | | O---> | | | | | | O---[IGNITER 2]---+ | | . | . | . | | | O---> | | | | O---[IGNITER n]---+ | RETURN>----------------------------+

-- John Fields Professional Circuit Designer

JF: I'd go with a pushbutton for each rocket and a separate momentary spring-loaded "ARM" toggle switch under a switchguard to enable the array of pushbuttons: ============================ Elegant