Negative Capacitance

What would be the characteristics of Negative Capacitance?

Looks to me like a special capacitor circuit configuration used to cancel out specific other capacitive effects.

http://www.cpes.vt.edu/public/showcase/negativecancellation.php

Could such a thing be simulated using something like a gyrator circuit?

Yea I am sure you could model it and use capacitors with a negative value.

You have to do some magic with op amps...

That's the correct fundamental circuit. However, the problems with nearly all Negative Impedance Circuit (NIC) designs are stability from the positive feedback and complications from all of the non-ideal attributes of real versus ideal operational amplifiers.

But but but....the design I built in the computer modelling program clearly proves........

Thanks to all for their info!

In a network, where is the op amp connected?

Are there op-amps that might work at RF? Say up to 15 MHz?

Is this a low-power-only (milliwatts) approach, or could it be used at power levels up to, say, 10s of watts?

A capacitor between input and output is positive capacitance , a capacitor between input and ground is negative capacitance, as viewed from the output. Correct?

No, the input impedance of the whole circuit (op=amp, capacitor, resistors and op-amp supply) that Rixter posted has an input impedance of -C.

Negative capacitance is a new term for me. Capacitance is not a vector.

Parasitic capacitance cancellation is usually done with inductors and a trim cap.

Is an inductor a negative capacitance?

Negative capacitance is a functional concept not an electrical parameter.

Negative Impedance Converters are atypical circuits in the Gyrator family. (The Wikibooks link gives a very nice derivation of this analog circuit.) An inductor and capacitor can produce a non-reactive load at a particular frequency when properly tuned. A capacitor canceled by a capacitive negative impedance circuit will only have a frequency dependence of the linear bandwidth of the op-amp.

You should be able to find circuits called "negative impedance converter" [NIC] and "general impedance converter" [GIC] on the web. I did some work some years ago in electrically small antenna matching using negative impedances in matching networks. They are pretty simple, if you can build an active filter you can build an NIC/GIC.

With a GIC one can (and I have) also build an impedance of 100 Henry in the size of a small postage stamp. If one desires one can even construct an impedance that otherwise does not exist, like the squared capacitor impedance of 1/(s^2*Xc).

Electrically Small antennas are just what I am interested in. Did you publish any work? Anything you might share offline? --

CR4 ADMIN - email address removed

From the CR4 Rules: Do not post phone numbers or email addresses. The CR4 Admin will delete all phone numbers and email addresses posted in threads or comments. You can share this information via the CR4 internal messaging system.

As a recent college graduate, I worked with Stephen E Sussman Fort [AB2EW]. Steve has published papers relative to this topic. My greatest contribution was the derivation that the required gain for the gain blocks needed to be > 4000 for minimum acceptable performance. Efficiency is horrible to some power but we were working around 10W using cascode MOSFET gain blocks I believe.

This is a completely new concept for me, If I had a set of calculations that said I needed a negative capacitance, negative inductance, or negative resistance. My first thought would be to double-check the math by hand, as the formula may not be right. Then I'd look for every point where roots are taken, because the nth root of Xⁿ is ±X, when n is even, and polarities may get flipped there. Finally I'd conclude that the circuit needs to have some of its values adjusted, as some section is WAY out of bounds to drive another into negative fundamental values to balance things out.

Remember, computers are wonderful tools, they allow humans to make thousands more mistakes per minute than they could on their own. In other words, don't trust computer simulations implicitly, whether an RF circuit design program or a GPS-based navigation system. All it takes is one bit to flop when it should have flipped, and suddenly you're looking for a capacitor with negative farads, or driving down a runway to get to the airport because the navigation map has incorrectly identified it as a drivable street.

Yes.