I find existing explanations of memristors are either overly vague, or an unapproachable mass of academic, mathematical analysis. I want to gain an intuitive understanding of what a memristor is, what it does, and what its applications might be in terms accessible to the ordinary engineer.

If an inductor can be emulated with a capacitor and active components, then I bet it's also possible a sufficiently clever circuit could emulate a memristor, at least within a limited region of operation. Though this circuit may not have the same properties that make memristors potentially useful, it could still be a useful demonstration and platform for experimentation.

Is there something I can build on a breadboard that looks, within some reasonable operating operating parameters, like a memristor? If not, what about something that can be simulated with ideal components? Bonus points for summarizing the theory of operation, and succinctly and intuitively describing what a memristor is without resorting to abstract mathematics or models far removed from even idealized real components.


The answer is: Mutator. Or maybe: M-R mutator.

The gyrator, used to emulate some properties of an inductor using a capacitor, is a subset of a mutator. I am not aware of a specific name for the type of mutator that gives you a memristic behavior (other than M-R mutator), but it seems like you can actually build it with real, existing components:

M-R Mutator

Source: Memristor - The Missing Circuit Element, Leon O. Chua, IEEE Transactions on Circuit Theory, Vol. CT-18, No. 5, September 1971.

The paper also shows how you can build a curve tracer for a memristor, i.e. a device that shows you a diagram for charge vs. flux-linkage. While there's a lot of theory in the paper, I like it because there are actual circuits I am sure one can breadboard and hack around with.

| improve this answer | |
  • 1
    \$\begingroup\$ I was hoping that the circuit would be simpler, but I guess my hopes don't change reality. \$\endgroup\$ – Phil Frost Jul 18 '13 at 18:21
  • \$\begingroup\$ @PhilFrost: Well, usually, Chua seems to be one of those who go for the most elegant solution, where elegant = simple. (Just replace the vintage OpAmps ;-) The whole idea of predicting the Memristor back in 1971 would not have happened without this philosophy, I guess. Same with his research on chaos ("Chua's Circuit"): His paper "The Genesis of Chua's Circuit" is a piece of beauty (www-inst.cs.berkeley.edu/~ee129/sp10/handouts/…). I really like how he rules out everything too complicated and arrives at the easiest circuit to do the job. \$\endgroup\$ – zebonaut Jul 18 '13 at 18:55

People have been working on this. A lot (most) of the stuff you see (as you mention) isn't something that anyone without a huge lab would even think of trying.

Here is a paper by Muthuswamy, who has worked with Chua, I think.

Here is a copy of the circuit schematic:

enter image description here

The circuit looks doable. It does use 2 multipliers (AD633JN) that are probably about $5 each, but it is not too complicated.

The paper has a lot of information about the circuit and its design. If you like Mathematica, it has a model based in that as well.

It looks like to be really up on the Mem_x_technology you will have to read up on Memcapacitors and Meminductors too. (Not kidding, they are out there).

| improve this answer | |

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.