Random/Pulsed Switch for Driver?

Hey everyone.

So I recently had an idea... I have been toying around with building a neato spirograph for a while now, but then I realized... why not just build one with three colors in it? All it would require would be a dichro and an extra diode (would use a PHR and an LPC).

Except... I would also need a method of having it pulsed between three colors. Now, I thought that the easiest way to do this would be have some sort of 555 timer or something periodically pull a transistor, then two transistors in series (?) then have it set back to 0V output, so that it would cycle between red/magenta/blu-ray, because I think that would be really neat to be able to get a spirograph that changes colors.

However, I have no idea if what I thought of would work, or even if it would work, how it would work.

The idea is that off mode would be red, then one transistor on would turn ON BR (giving magenta), then the second one would switch OFF the red, giving just BR, and then it would cycle like that, probably at .5Hz or something.

But I am not sure how to go about making transistor gates go in SERIES... so that the voltage is added between them.

Anyone have any ideas?

(Of course, these transistors would just turn on/off the normal drivers I would have to be powering these diodes, which would probably be an easy LM1117 build or something. And I would have power knobs for fine tuning, etc.)

I have an idea.

You know how in like gas stations and such there are those little bouncy balls that when they hit the ground two or three different color LEDs start flashing in random patterns? I took a broken one apart once and all it is is a tiny circle board (~1cm) with an IC (the kind not in a package, just a drop of epoxy over it), a "spring switch", and two (or more) LEDs.

You could easily use one of those as your "controller". Simply replace the spring switch with a momentary button, remove the LEDs, and put your transistors in their place. Then you would just flip a switch to start the motors and tap the button to start the pattern sequence. I imagine that all the different quick random pulses would look pretty cool coupled with a spiro.

I have absolutely no idea what you are talking about, those little bouncy balls =p What do you mean?

Umm well let me find some...

Like this:
Soft Plastic Flashing Smile Face Figure Bouncy Balls (2-Pack/3*AG10) - Free Shipping - DealExtreme
Or this:
Motion Activated Auto Wheels Strobe Lights (2-pack) - Free Shipping - DealExtreme
They are both pretty much the same thing, it doesn't really matter what it is as long as it has flashy led's. Surely they must have some of these flashy led toys in your area, they are all over the place here.

Hmm. I will take a look into things like this.

You can also make a square wave generator out of a 555 and a second square wave triggering with a delay on the first. With the right adjustment this would cycle between red, magenta and violet. I don't know if such a ball is a good idea, but you can try.

Hmm. So like the first square wave being generated for 33% duty cycle, the second one is another 33% duty cycle with a 33% duty cycle delay, and then repeat?

As I understand it, the cycle should be:
red on - blue off ==> red
red on - blue on ==> violet
red off - blue on ==> blue
So switching between red, violet and blue would mean a 66% duty cycle because they should overlap to form violet.

Oh, right.

So both switches would be 66% duty cycles, like so:

----__----__----__
__----__----__----

Where the top is red and the bottom is BR.

EDIT: So I think I have drawn up a driver... someone wanna check out my design? It's using 3 555 circuits... so let's see!



EDIT 2: So a bit of an explanation... the top 555 starts out with a duty cycle of 66% at 1Hz (at least that's what my calculators say it should be) and when the 9V is high, it should pull the NPN transistor and let the 18V flow through to the 1085, turning on the LPC. Meanwhile, the second 555 timer is ticking away at .33 seconds, using the delay-on topology of the 555 timer. That then, in turn, pulls the NPN transistor down at the third 555 timer which is set to be in mono mode, which should just give a pulse of .67s, letting the PHR turn on and stay on for the second 2/3 of the cycle. Is the circuit I drew up wired to do that correctly?

Problem solved!

So I figured out a much easier way to do this....

Here is the schematic and the sim from LTSpice.

That looks very good, except for the spikes. A different current source or some filtering would fix this.

Yeah - I actually took care of the spikes in the final product (the picture I printed out wasn't the final development, was just excited to post that I got it working). I just added an input and output cap to both of the 317s and it stabilized.

EDIT:

Alright, so I etched my PCB and just put in an order for parts on Mouser. Now I wait till like Saturday (or Monday) to receive my parts, and then my circuit will be up and running! I will test it... with I have no freakin' clue, until I get the LDs. I suppose I also need to get some motors and some FS mirrors and a dichro. Anyone have any good sources?