Basic random generator

[MaxMouseDLL]

Hi all,

I was wondering if it was possible to create a (very) basic random number generator without the use of any IC (Display could be something as simple as some LEDs, an IC controller for the display is fine), and if anyone has a link to a circuit diagram for one?

Regards
Max

 

[BobK]

Are you talking about a true random number generator or a pseudo random number generator, they are two entirely different things.

Bob

 

[Raven Luni]

If you only need a low number you can use an astable multivibrator or transistor chaser circuit that you can interrupt.

For more than that you would need more complex counters (ICs) or a way to quantise a thermal noise source.

 

[MaxMouseDLL]

Are you talking about a true random number generator or a pseudo random number generator, they are two entirely different things.

Bob

Doesn't have to be "true" random (Is there such a thing? ... no, just "very good random" I don't want that, I want pretty lights that look random lol.) just an extremely basic random "number" generator, let's say... 50/50 (on/off)... can it be done and display it in some way (LEDs?) without any kind of IC.

I know I could easily build one with an IC, I just want to build a small veroboard example with NO ic, that has a red and green light, with no discernible pattern, 50/50 is ideal, but, anywhere near that (40/60, 30/70) is acceptable.

EG: a red led and a green led, 50/50... pseudo random is fine, as long as there isn't a predefined set (Although I doubt a defined set would be possible without an IC anyway, and if it is... a very long set would do too.. say a predefined set of reds/greens 500 long [but again.. I can't see that without an IC being involved]).

If you only need a low number you can use an astable multivibrator or transistor chaser circuit that you can interrupt.

For more than that you would need more complex counters (ICs) or a way to quantise a thermal noise source.

Low number... 1 and 0... do you have an example?

Edit: Raven, thermodynamics kinda sucks, although I've been reading about oddities in various experiments which seem to at least in part, defy the second law...
Edit2: It's an interesting one, it's to win a (non monetary) bet, I will post results if successful, including a shot of the build and a gutted engineer....
Edit3: something like this: http://www.qrp.pops.net/LEDs.asp (Sequentially Off LED Pulser) 3 leds, somehow keeping the power on (so it runs) and showing ONE led when a button is pressed..... lol I don't know, I'm clutching at straws...

 

[Raven Luni]

So its a coin toss thingy - heres a decent example: http://talkingelectronics.com/projects/200TrCcts/200TrCcts.html#30

 

[(*steve*)]

For a heads/tail type of circuit, the easiest way is to toggle your output at high frequency and manually interrupt the signal which causes the toggling.

If the signal to generate the random state itself comes from a process which generates it automatically, then your source of randomness almost certainly needs to be better than a simple high frequency oscillator (which will provide regular timing, so a regular sampling may yield predictable results). In this case a noise source (e.g. a reverse biased diode junction) may be preferable.