I have an interest in "alternative" logic gates, as in not silicon transistors. I have built RAM in a little game called "Powder Game", constructed "one-off" logic gates out of exploding lattices of popsicle sticks, and used electromechanical devices such as motors and relays. Right now, I'm intensely interested in magnetic cores. I would like to somehow experiment with these myself, but short of purchasing the equipment to do so, I haven't found a way. I am hoping that there is, perhaps, a free piece of software that would let me experiment with trying to make logic with magnetic cores. Can anybody help me find one?

Here is a PDF to describe what I am talking about: http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=04051986

  • 1
    \$\begingroup\$ The link doesn't work. You may be able to build some core memory...they don't have to be microscopically tiny (only if you want a lot of them). I saw some made from metal nuts once. He said it worked but I didn't see it powered. \$\endgroup\$ – gbarry Oct 2 '12 at 3:30

You can buy vintage bare 45,000 cores here for $US49.00, with free international shipping from Bulgaria. Find 9 other friends interested in sharing and you have 4,500 each for $5.

One bit core memory project - liable to be of high interest. Use a suitable ferrite bead if needs be. Drive levels etc need looking at.

This vintage core memory board costs $400 on ebay BUT I have bought similar here for about nothing and while they are rarer than way back then they are probably still available for looking as electronic junk.

Almost any small ferrites with a hole in them will do - including RF decoupling beads. Or you could wrap a coil around a stick core etc. You can also just use iron or steel with due attention to characteristics. A steel nail with a suitable amp-turns pulse will be magnetised. Not the most efficint core memory but doable.

How to drive one

Very useful


Old discussion -2010 - may be some use

  • \$\begingroup\$ These are all very good options for working with cores by hand. I was looking for a simulator, but these options seem more attractive now. Since yours is the most complete and helpful answer, with lots of links to good details, I am marking it as my answer. \$\endgroup\$ – Void Star Oct 3 '12 at 0:25
  • \$\begingroup\$ I wonder what would be required to implement something analogous to the Apollo Guidance Computer's rope memory using 1970's technology? Would it have been practical, for example, for a hobbyist with more patience than money to construct e.g. a 320x8 ROM using a one-of-eight high-side driver and four one-of-ten low-side drivers, connected with a total of 320 wires, each of which either passes through or bypasses each of eight cores? Such a thing would seem like it could be more compact than a 320x8 diode ROM, since the contents of each byte would be encoded in the threading of a single wire. \$\endgroup\$ – supercat Sep 26 '18 at 18:40
  • \$\begingroup\$ @supercat I think it would be entirely possible. The small ferrite core memories that I have look like something I could make at home if I had more time than sense or $ :-). I also have a commercial keyboard with an encoder that works by a keypress driving a pulse through a series of ring cores corresponding to its ASCII code. \$\endgroup\$ – Russell McMahon Oct 15 '18 at 12:06

This paper seems to have a fairly complete description of many types of magnetic core logic functions, and it shows how to simultate them in a circuit simulator like Spice.

(I got this link by plugging your question's title verbatim into a well-known search engine.)

  • \$\begingroup\$ I've seen that document before. I have saved on my computer. I will look at spice. \$\endgroup\$ – Void Star Oct 3 '12 at 0:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.