randomized mystery box

Status
Not open for further replies.

rasterburn

Member
hello, I am trying to work out the circuit for a useless machine, I am sure many of you have seen these on youtube (mystery box) the ones they show are simple, and my wife would like one but only more advanced, first off she would like to have 2 flashing leds (easily done with a 555 timer and a few resistors and a capacitor or 2) second she would like not 1 but 4 arms so that a random arm pops out and turns the the box off, I know this would require a push button to do, the LED circuit has already been designed but I am not sure what ICs I need to randomize which arm pops out to turn things off can someone point me in the direction of a line of ICs that will work for this, bear in mind I dont have any programming skills or any programming boards available to use so if at all possible it will have to be purely un-programmed.
 
Here is a schematic of a 4 bit led chaser. It repeatedly cycles through all the LEDs, but you can use the outputs to enable one of the arms when a button is pressed. The result will be random (if the chaser clock is fast enough), because the button-push event is random.

**broken link removed**

Here is the original project page (the origin of the picture): http://metku.net/index.html?path=mods/vilkkuvalot1/index_eng
 
Last edited:
hhhmmm... this seems to still be a puzzle, I don't think using a 555 timer to trigger the 1 of 4 servos would be the right method as it doesn't seem truly random based on the fact that the capacitor would charge and discharge at the same rate every time and if the trigger was time based from the 555 then it would end up triggering the same one every time.

I was kind of thinking that a decade counter would be the correct route for producing a random output across 4 of the pins but not sure how to get the random output from the input (if possible).

Right now I am thinking that investing in a programming board and relevent IC would be best and try and write some randomizing code. I am also thinking if i was to use a programmable system then i would do away with the 555 timer chip
 
You can use a shift-register to generate a pseudo-random sequence that should suffice for your application such as this.
 

If the 4 bit cycle keeps running constantly the result will be random. If you need to prove this mathematically, you can correlate the 4 bit sequence with a random event (the button-push).. and you will see that the correlation is very small (the events are independent -> button-push can occur evenly at any time during the 4 bit sequence).

My suggestion will be more random than any pseudo-random solution because it is based on truly random events (the button-push event).

Of course this (my suggestion) doesn't work if you start the 4 bit cycle from scratch every time the button is pushed. I assumed that some part of the machine is always running because you mentioned the randomly flashing LEDs. Aren't the random LEDs always on (randomly flashing)?
 
Last edited:
ok so the initial design of the "toy" was that 2 leds would flash back and forth at a set frequency say something like 5hz, that part of the design was designed around a 555 timer, just a simple square wave oscilator. The power system was switched with a regular togle switch and then the push button switch would turn on the leds and then an arm pops out, hits the push button and turns off the leds. The random 1 of 4 arms that pops out was originally to be controlled by a second IC, I am not sure how I would generate a random event if I go down that path.

If I change the design so the LEDs are running constantly and use the push button to cause the random event, what chip would be best to decipher which servo is the active one and how would it decipher which one based on the output from the 555 timer?

I just talked the design out with my wife and she does not want the design to change any so it needs to operate as follows:

turn master power on, press button, LEDs start flashing, arm pops out and presses button, LEDs stop flashing

of course the 4 arms pop out from 1 of 4 sides of the box...

Just thought of a work around, not sure how well it would work, but of course utilize 2 555 timer chips, 1 to operate the LEDs while the switch is on the other as a clock for the random event
 
Last edited:
ok so the schematic has been completed and is working properly using multisim, I am now waiting for half the components needed, just need to get a hold of most of the physical hardware and it will be ready to be built, can anyone tell me at what frequency range will be good so that the servos dont decide to kick in when the counter passes across the pin? I am thinking 100Hz should be sufficient
 
Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…