How my Etch-A-Sketch circuit works
Vložit
- čas přidán 13. 09. 2024
- Last time I showed off an Etch-A-Sketch I built using a bunch of old digital electronics chips. This time I'll try to explain how it works!
-- Stuff
Schematic: bigzaphod.com/...
KiCad: www.kicad.org
My Wife's Books: www.skaeth.com
(Maybe you and/or someone you know will like them!)
Nice! Anyone making reasonably advanced projects/designs without a microcontroller get a whole bunch of gold stars from me. :-)
I really enjoyed your previous video on this topic and this one was great too! I hope you plan to do more electronics projects and present them in a similar style. Your ability to present these projects and ideas clearly is really impressive! I subscribed!
Hi, I was inspired by this circuit, so i built my own. Some of the parts i couldn't get hold of in the UK like the encoder dubounce IC's so i had to use some schmitt trigger inverters and xor gates to get the correct signals. But got this working well. For the address generator circuit i used the same 74HC191 counters also for the blink clock. The SRAM was a pain in the ass. I used a AS6C62256, Its got more control signals than the chip you used. And apparently writes on the rising edge following a falling edge of a clock pulse. So i had fun trying to get the timing and everything right. But eventually got it sorted after a couple of days tearing my hair out.
One feature i added was on the left hand rotary encoder, i added a toggle using a D flip-flop to enable a delete function, by selectively inverting some of the signals using some xor gates and allowing pixels to either be lit or unlit.
My kids love playing with it. Thanks. Keep up the great work 👍
Wow that's awesome! Thanks for letting me know! Love it!
Thanks for making this video, it's really appreciated. I've been following along with this build at home, and have built out the four matrix displays, and ended up using 74LS138 decoders to the matrix, with one side being inverted. I've also found that because there are so many other data lines (As you say in this example you are only using 1/4 of the space of the memory IC) you can happily save other images as you go.
My version currently also allows you to choose if you are writing a 'dark' or 'light' spot on the screen for when mistakes are made.
I'm also using 74LS123's to clean up the encoder data because I am cheap!
Wow that sounds cool! I've thought about adapting mine to store more images in the RAM - glad to see someone else go for it! I was considering a more advanced version that could do animation, maybe - draw and save several frames in the RAM and then hit a button to play them all back in a looping sequence. :)
@@bigzaphod that's a nice idea! I've used a 32k for mine because it's the smallest ram I had here. I suspect I might end up adjusting into making a two phase clock one for reading then one phase for writing so I can add some othe devices for dumping an image on from an arduino ettc
Many special-purpose chips like the "rotary encoder debounce" chips are in fact pre-programmed microcontrollers. I don't know if you saw it, but in the other video I describe how a reasonable amount of logic can convert the rotary-encoder outputs into direction and count-enable signals if the same clock is used to latch the inputs as is used to operate a synchronous up/down counter, which would seem a much more authentic "1980s"-style design.
I actually did experiment with handling the rotary encoder the "old fashioned way" at first, but I got excited by that custom chip when I found it and since it reduced the number of components it seemed like a fun idea to try. I didn't realize these kinds of custom chips are basically tiny microcontrollers these days, but I suppose that makes some sense - seems like everything else is. 😛
I wasn't meaning to demean your efforts, especially since a variety of application-specific chips existed in the 1980s to handle tasks like this, sometimes using external resistors and capacitors to control debounce timing. Still, if you can't get one of those the logic to convert encoder inputs to counts shouldn't be hard if you make certain to sample the encoders with the same clock used to control the counter, before they drive logic. If you try to drive logic with encoder inputs before they're latched, race conditions may make things unreliable.