Firework control board conundrum.

[Slickgsxr5]

If anyone can help me figure this out I might be able to move on with my life!

Ok so I decided as my first project I wanted to replicate and tweak (click here) This design of board and make it my own.

While the finished product "worked" I ran into issues, one in particular i can't seem to rectify. I do fully realize some of the issues that came up were my fault by adding components and changing the original design. Again the original design is 10+ years old and I realize there are readily available professional built systems to buy.

I first only wire up 3 of the 6 Db25 connections as I only possibly needed 3 remote boxes at this time. It also keeps the mess of wires to a minimum.

Instead of bundling a mess of wires for the 24 cue buttons, I daisy chained them together. These buttons connect to only the 12v key switch.

Not part of the original design, I decided to add red LEDs as part of the 12v circuit. Both the red and green LEDs came wired with these resistors...

I connected the negative wires from the green LEDs, red LEDs and the cue button to one side of the terminal strip. This decision introduced the first issue I ran into after full assembly.

Only when the 12v key switch was turned on and I jumped the ground pin with each cue to complete the circuit, not only did the red led light up, but also very faintly lit up the green led on each cue when tested. Figuring that the 12v was enough to backfeed the green led with enough power to dimly light it. (Which to me is strange...LEDs only work one way from +to- right? The positive wires from the green LEDs are ONLY connected to the 3v key switch.) I jumped the negative wires from the green LEDs to the other side of the terminal strip. Which for the most part worked. It also worked on each channel as well.

When I finished one of the remote boxes I connected, jumped each cue terminal and tested for good continuity and that every cue matched with how the box was wired as to what cue LEDs lit up on the control board AND tested an actual igniter, I was satisfied everything was in order. It wasn't until I was out in the field connecting the fireworks to the igniters to the remote boxes till I noticed a HUGE problem.

I DON'T KNOW HOW TO FIX PROBLEM 2....

Problem 2: If ONE remote box is connected say channel 1,2 or 3....everything works exactly how it should. Press cue 1 button - cue 1 igniter fires and lights the firework. Just like it should. The problem arises when MORE than ONE remote box is connected to the system.

If say remote box on channel 1 has all 24 cues filled and I connect the empty remote box to channel 2 then switch to channel 2, naturally the circuit is still open as there is nothing connected. As soon as ONE cue is connected and closes the circuit, all 24 cue light up as if they were ALL connected to igniters.

Here's a video showing the problem.

 

[Visitor]

A schematic would be really helpful to understand how you've wired things up.

 

[Slickgsxr5]

I have not a clue how to draw a schematic and I know for sure there is not one for this designed board. I can probably answer any question you have.

 

[Visitor]

Not of the whole system, just of you additions. It's hard to tell for certain the connections you've made, and any assumptions made are likely to be wrong.

 

[Slickgsxr5]

In the original design a 2 pole 6 position selector switch was used, but the designer said a singe pole switch could be substituted. I went with the single pole. In the original design 22g braided wire was used to connect the green LEDs to the 2x 12 position terminal strip same with the cue buttons. I used 22g solid core for the buttons and I believe the prewired LEDs used 26g braided wire. No resistors in the original design. I added the red LEDs but they are set up just the same as the green LEDs in the original design. My green LEDs negative wire had to be moved to the other side of the 2x 12 position terminal to eliminate (what I believe) was voltage crossover illuminating the green LEDs that were only supposed to be lit by the 3v power supply. The original design initially set up all 6 db25 connectors, I only set up 3. I daisy chained the 12v power across all 24 cue buttons, the original design used extended positive leads with wire nuts with one primary lead connecting to the key switch....same goes for the LEDs. I used studs and jumper wires to connect the LEDs to their respective power sources. Lastly in the remote boxes.... I directly connected each pinned wire from the db25 connection straight to the positive posts on the speaker terminals, the original design used wire nuts to bridge the pinned db25 connections to the + & -. The negative speaker terminal connections were wire nutted together with one primary lead connecting the negative db25 pin....I used 3 ring connectors (2 sets of 4 negative wires per ring) connected to a primary stud in which the 1 negative pinned db25 wire was also connected to. Also I'm the original design 8 d cell batteries for armed side and 4 aa batteries for continuity. I use a 12v sealed lead acid battery and a single 3.7v lipo battery for continuity.

 

[KeepItSimpleStupid]

Your biggest problem is "All red wires". I remember somethng done at work with "all yellow wires" and ..... I also remember a chemical Phd ask me why the wiring had to be neat?
It's so you can troubleshoot it.

Some leds can be bi-color with reverse polarity,

What does a green and red LED mean? Does it check the continuity of the ignitors?

Can you try isolating power to each connector with a diode, so it appears more independent?

Draw one circuit.

Do you use the 3.7V battery and a green LED for continuity and the 12 V battery and a LED to indicate power to one side of the firing button?

ithout knowing anything about the circuit design, what comes to mind is "isolating" the connector power supplies with diodes. Bothe the 3.7 and 12V,

For automation panels I built, I used 18 AWG wires in 10 colors with reduced stranding.

 

[Slickgsxr5]

Ok so I specifically purchased non bi color leds( I did look into going that route to be honest.) The green LEDs are on a separate connected to the 3.7 volt supply for continuity only. The red LEDs were added as a redundant continuity test AND under the assumption that when the ignitor was fired the nichrome wire would break - breaking the circuit which would turn the red led off. I can use diodes (which was mention by friends to isolate possible backfeed) but that to me is the last resort if nothing else works.

I'm in the process of drawing up a schematic.... please tell me if this makes any sense so far...

**broken link removed**

 

[Slickgsxr5]

Ok so I specifically purchased non bi color leds( I did look into going that route to be honest.) The green LEDs are on a separate connection to the 3.7 volt supply for continuity only. The red LEDs were added as a redundant continuity test AND under the assumption that when the ignitor was fired the nichrome wire would break - breaking the circuit which would turn the red led off. I can use diodes (which was mention by friends to isolate possible backfeed) but that to me is the last resort if nothing else works.

I'm in the process of drawing up a schematic.... please tell me if this makes any sense so far...

 

[Musicmanager]

Hi Slick

Quick question .. .. .. have you done anything with the wiring to the rotary switch ??

MM

 

[KeepItSimpleStupid]

It's hard to follow anything, but I think I get the idea.

Basically, you run the LED through a resistor and the ignitor. 12V and 5.6K is not enough to turn on the ignior. The switch should short out the LED/RESISTOR and then the ignitor would fire.

The selector switch makes 6 blocks. Key power keeps the good stuff (12 V power OFF).

The ignitor can probably be considered a piece of wire.

That could leave you with a common side and possibly a feedback path through the LED in one direction.

A push-button switch that's double pole even would get interesting, when you try to push a button for more than one ignitor.

Does you rotary switch cut both positive and negative?

The 3V battery can't set off the firework fuse.

One seems fine. Multiple ones seem problematic especially at the same time.

Buttons can have two NC contacts and two NO contacts too which could act differently.

 

[rjenkinsgb]

I suspect you need a diode in series with each output to each external circuit, at each D connector (so 144 diodes to add).
1N4000 series are OK for up to an amp continuous.

That will allow the LED current or ignition power through to the igniter, but not allow the igniters to "cross-connect" different circuits.

It's a similar effect to a matrix keyboard with more than one key pressed - wires get cross-connected via the switches, unless you add a diode at each key.
The switched banks of igniters are the rows and the outputs the columns of the matrix in your case, with everything totally shorted once you have all the external stuff plugged in...

 

[Nigel Goodwin]

Your biggest problem is "All red wires". I remember somethng done at work with "all yellow wires" and ..... I also remember a chemical Phd ask me why the wiring had to be neat?
It's so you can troubleshoot it.

As kids we rebuild a motorbike, and rewired it entirely with black wires

Simple reason, we had a drum of black wire, and zero money.

But it was a nightmare even doing it, never mind fault finding later.

 

[Musicmanager]

I think there may be an issue with the rotary switch ?

It's hard to be sure from the picture, but I have one here that looks very similar which suggests that ' channel 1 negative wire ' is connected to a pole not a switched connection, which would explain the ' cross-connect' ??

MM

 

[rjenkinsgb]

Your biggest problem is "All red wires".

That's what cable markers are for!

Most industrial machine tools & robotics systems etc. are wired with one colour per voltage range, but every wire is identified by numbering at each end.

eg. All blue - but all identified.



It's well worth getting a set of cable markers to use when doing any kind of system where wiring could be confusing either during assembly or with later faultfinding. It also means you can identify each connection on your schematic & know exactly what wire it refers to.

 

[KeepItSimpleStupid]

Most industrial machine tools & robotics systems etc. are wired with one colour per voltage range, but every wire is identified by numbering at each end.

Interesting.

I used those big ones once and in a very strange way.
Like 1A2 on one end and 2A1 on the other. Cable A from location 1 and 2. The 1 and 2 were in different rooms. They were Teflon insulated cables.

I've used the little pre-printed ones too. I brought in my p-Touch labeler which can print flag labels too.

Motors might have things like T2 actually written on the wires.

Then there was this X-ray set with about 50 black wires on a connector. Each wire had numbers on the insulation.

Yep, wires were expensive, but not that expensive. I found a wire manufacturer that had low minimum orders ($50.00 USD), BUT bad lead times. 100' minimum PU for the wire I wanted too. I kept 10 colors of 18 AWG with the reduced stranding in stock and 4 colors of 8 AWG in stock. A friend ordered wires too.

 

[Nigel Goodwin]

Yep, wires were expensive, but not that expensive.

It's all relative - as kids we had close to zero money, so couldn't buy wires even if we wanted - any money we managed to scrounge up we spent on petrol for the motorbikes we ran in the fields, and it was very difficult to save up to buy a gallon of petrol.

As for industrial wire use, cost isn't usually an issue, and paying big money for wires is the norm.

I'm quite horrified even now, when I'm spending (at work) close to £300 for a reel of cable - or I would be if it wasn't out of stock! (expected 31/7/2020). In the mean time we've had to buy smaller reels, about £80 for 100m, instead of the 500m reels.

And in cable terms of course, that's only cheap cable.

 

[Slickgsxr5]

Hi Slick

Quick question .. .. .. have you done anything with the wiring to the rotary switch ??

MM

No I haven't after further testing I don't know if that is the problem or not. I'm about to upload some pictures and videos.

 

[Slickgsxr5]

I think there may be an issue with the rotary switch ?

It's hard to be sure from the picture, but I have one here that looks very similar which suggests that ' channel 1 negative wire ' is connected to a pole not a switched connection, which would explain the ' cross-connect' ??

MM

 

[Musicmanager]

Nope .. ... looking at that picture, clearly I'm wrong ! Looks OK to me.

MM

 

[Slickgsxr5]

I suspect you need a diode in series with each output to each external circuit, at each D connector (so 144 diodes to add).
1N4000 series are OK for up to an amp continuous.

That will allow the LED current or ignition power through to the igniter, but not allow the igniters to "cross-connect" different circuits.

It's a similar effect to a matrix keyboard with more than one key pressed - wires get cross-connected via the switches, unless you add a diode at each key.
The switched banks of igniters are the rows and the outputs the columns of the matrix in your case, with everything totally shorted once you have all the external stuff plugged in...

It's hard to follow anything, but I think I get the idea.

Basically, you run the LED through a resistor and the ignitor. 12V and 5.6K is not enough to turn on the ignior. The switch should short out the LED/RESISTOR and then the ignitor would fire.

The selector switch makes 6 blocks. Key power keeps the good stuff (12 V power OFF).

The ignitor can probably be considered a piece of wire.

That could leave you with a common side and possibly a feedback path through the LED in one direction.

A push-button switch that's double pole even would get interesting, when you try to push a button for more than one ignitor.

Does you rotary switch cut both positive and negative?

The 3V battery can't set off the firework fuse.

One seems fine. Multiple ones seem problematic especially at the same time.

Buttons can have two NC contacts and two NO contacts too which could act differently.