erratic power light controller circuit

[4pyros]

I want to make some erratic power light controller circuits.
The off the shelf units are mosted made to simulate a flickering flame that never goes out entirely.
I would like one that stays mostly on and randomly goes completely out.
I am thinking maybe a clever 555 timer circuit, or using 2 or 3 timers combined somehow to make it more random.
Any ideas?
This would control mostly 12o VAC lamps, but would like to control 12VDC LEDs as well.

 

[dr pepper]

I think you'll need a micro.

 

[4pyros]

I think you'll need a micro.

I was thinking that two, some day I need to get into them.
For now I wanted to explore how hard it would be to do with analog.

 

[ronv]

I've not tried this, but it seems it would work.
http://world.std.com/~reinhold/waynesrngcomp.gif

 

[4pyros]

I've not tried this, but it seems it would work.
http://world.std.com/~reinhold/waynesrngcomp.gif

This seems to be for making audio noise. how do you propose using it i a lighting controller?

 

[4pyros]

I was thinking of something like this using three 555 timers but I can not find a schematic for it.

 

[heydonms]

I would like one that stays mostly on and randomly goes completely out.
I am thinking maybe a clever 555 timer circuit, or using 2 or 3 timers combined somehow to make it more random.

Define "random".

You can't really get truly random behaviour from a 555, but it sounds like maybe you just want a long sequence so that it is hard for someone to visually detect the repetition. What sort of timing are you after? 2-10 seconds on then 1-4 seconds off? or 1-2 seconds on and 0.1-0.2 seconds off?

EDIT: The more I think about it, the more the noise generator sounds like a good approach. Feed the noise into a comparator, the other leg comes from a variable voltage source then you can adjust the voltage to alter the on/off ratio. Add a latch and a 555 to adjust the timing if necessary.

 

[4pyros]

Define "random".

Yes just seemingly random, I know you cant do truly random, I am trying to get something that looks good.

The more I think about it, the more the noise generator sounds like a good approach.

Maybe thats what ronv was getting on about with the circuit he posted.

 

[heydonms]

What's it actually for? A light that occasionally turns off doesn't sound that impressive (no offence )

 

[alec_t]

I doubt the 120V AC lights would survive long if repeatedly switched like that .
Somewhere I have a simple pseudo-random generator schematic using CMOS logic. I'll see if I can dig it out.

Edit: Here you are. Might be the basis for your switcher.

 

[4pyros]

What's it actually for? A light that occasionally turns off doesn't sound that impressive (no offence )

Its for Halloween.

 

[4pyros]

I doubt the 120V AC lights would survive long if repeatedly switched like that .
Somewhere I have a simple pseudo-random generator schematic using CMOS logic. I'll see if I can dig it out.

Edit: Here you are. Might be the basis for your switcher.
View attachment 85841

Now thats interesting, looks like 4 oscillators in a ring and you could tap off the ring anywhere.
I guess this could be done with 555s as well?

 

[dr pepper]

I think I have a similar circuit, a 40106 schmitt inverter wired to make several oscillators, and one of the schmitt oscilators running at a lower freq triggering a data latch.
This was for more than 1 light, I was thinking however you could use 1 o/p to drive your lamp, and another o/p to randomly switch the o/p off altogether to produce your 'off' cycle you wanted.

If you use solid state relays that implement zero cross switching, your lamps will last longer and the system will create less rf noise.

 

[4pyros]

I think I have a similar circuit, a 40106 schmitt inverter wired to make several oscillators, and one of the schmitt oscilators running at a lower freq triggering a data latch.
This was for more than 1 light, I was thinking however you could use 1 o/p to drive your lamp, and another o/p to randomly switch the o/p off altogether to produce your 'off' cycle you wanted.

If you use solid state relays that implement zero cross switching, your lamps will last longer and the system will create less rf noise.

Can you find your circuit?

 

[alec_t]

I guess this could be done with 555s as well?

Reckon so. Each oscillator could apply a voltage to the CV input of another oscillator to change its frequency. But a CD40106 gives you six oscillators in one IC.

 

[4pyros]

I could not find any circuits online for a 555 ring oscillator but here is a vid of one, it doesn't quit do it for me.

 

[4pyros]

I don't think a ring oscillator is going to work.
Has anyone seen a schematic of several 555 oscillators combined?

 

[alec_t]

How about this?