1

I would like to make some kind of primitive system in the direction of a computer, from scratch.

Someone showed me you can use ferromagnets to do some elementary computation; the “states” are the magnetic poles which can get flipped when you run a current through them (or something similar).

I still needed to buy a kit for that and while it interests me I think I would still need modern computer equipment like a screen and keyboard to interact with it.

I want to know what I can make completely myself - anything like some raw metal which I can pass a current through for some kind of interesting project, to learn more about the possibilities.

I am most interested in long-distance communication using radio waves or any type of basic computation.

In the case of computation, maybe the information output could either be visible somehow - like using sand on top of magnets or something? - or a very elementary binary printout mechanism, perhaps?

I really appreciate if anyone can help me discover some avenues along these lines to get started.

I am open to using bioelectricity or electricity generated in any other way.

Thanks very much.

hmltn
  • 149
  • 3
  • 4
    "long-distance communication using radio waves" is not something to experiment with outside a very well shielded Faraday cage as a beginner and without professional equipment. The risk of interfering with others infrastructure is way too high and dangerous. – asdfex Apr 03 '22 at 16:05
  • 4
    The crystal radio is how some of us engineers got started. Some wire, a cardboard tube, a chunk of galena and a safety pin... Google "crystal radio" for ideas. You can receive radio signals from miles away. – Mark Leavitt Apr 03 '22 at 16:09
  • 2
    If you're interested in long-range communications, try to find an amateur radio club in your area. They'll have all kinds of ideas and resources to get you started. At it stands, this question is both too broad and too vague to begin to attempt to answer here. – Dave Tweed Apr 03 '22 at 16:10
  • Peter, you are off in at least two orthogonal directions in this question. More, you seem to be open, literally, to almost any direction at all that even vaguely fits those broad categories. It could be that you want to start where a solid state physicist might, exploring basic features of solid state matter and forces and then discovering behaviors that might be useful. Or you might not care about that, at all. I can't tell. You need to narrow this down. A lot. – jonk Apr 03 '22 at 18:49
  • 1
    Think about the earliest analog multi-channel signaling devices. Old time mansions had strings throughout the house that converged to a servant's area. The string would ring a bell and also drop a flap with the name of the room. Even if the servant did not see the bell ring, they would see the flap, reset it, and go the the room where needed. Also, at airports and stock exchanges, rotating shafts could cycle and display stock prices or plane arrival times. – John Canon Apr 03 '22 at 19:43
  • 1
    Look up "foxhole radios" - soldiers in wars managed to receive some AM radio signals with basic materials only. – user253751 Apr 04 '22 at 16:46
  • 1
    If you want to make a computer, I suggest that you start from transistors or relays instead of from scratch. You can either make relays from scratch, or transistors from scratch (very difficult), or a computer from relays or transistors, but not both. – user253751 Apr 04 '22 at 16:47
  • @jonk Can you list off the directions you think I could narrow it down in, so I can discover more possibilities? One of my ideal projects would be some kind of mechanical computer, inspired by Babbage's Analytical Engine but much smaller. – hmltn Apr 04 '22 at 17:46
  • @PeterElbert You could refer to [mechanical computers](https://arstechnica.com/information-technology/2020/05/gears-of-war-when-mechanical-analog-computers-ruled-the-waves/) using cams, rods, gears, and so on. These are still quite large. As someone else already mentioned, you could build a [relay computer](http://web.cecs.pdx.edu/~harry/Relay/). Those are smaller. But not small. All are inspired by Babbage's device and Lovelace's visions for it. Or you go towards discrete BJTs -- still smaller. – jonk Apr 04 '22 at 17:53

1 Answers1

3

The only thing you can make from scratch would be relays. But to build a reliable relay from scratch would require a lot of work and several different materials.

If you buy cheap relays to build some kind of primitive system in the direction of a computer you would need hundreds if not thousands of them.

A counter able to count from 0 to 15 would require about 16 to 24 relays.

Here is an example https://www.konstruktionspraxis.vogel.de/zuse-ur-computer-mit-finder-relais-nachgebaut-a-308058/ requiring 600 + 1800 + 100 relays.

Uwe
  • 2,783
  • 1
  • 11
  • 18