CAD Project for Single Layer Capacitors

What is a capacitor?

[1] ca·pac·i·tor




A device used to store an electric charge, consisting of one or more pairs of conductors separated by an insulator.


But... “one or more pairs of conductors separated by an insulator” can also be a transmission line.

It turns out that even though a capacitor is not connected to other circuit elements in the same way as one might connect a transmission line, their mathematical models are very similar.

An EM simulation of a capacitor

 slc_wireframe  slc_surface

The figure on the left is a wire-frame model of a capacitor with a fictitious internal charging wire (the red object). The figure on the right is the same model with the wire-frame colored in‒ green represents conductor and brown represents dielectric.

If a current is impressed onto the charging wire, then the capacitor will accumulate negative charge on one plate and positive charge on the opposite plate. An alternating current will result in charges flowing onto and off the plates until the plates are either filled up or emptied. Although we don’t think about it much, electric current flow is speed-limited. This means that as the frequency of the alternating current is increased the capacitor will no longer fill up totally and the alternating current will flow continuously. We also don’t think about the physical size of the capacitor, but when the charging frequency gets high enough, the capacitor begins to exhibit behavior similar to a vibrating string– called normal modes.

Refer to this YouTube video: Simulated Resonances of a Single Layer Capacitor

The behavior of a real physical capacitor is complicated, but repeatable and subject to a set of rules. The behavioral rules depend on the capacitor’s physical dimensions, its constituent parameters and the frequency of operation. Applying the behavioral model is beyond the capabilities of many capacitor users, thus making a software application that computes the behavior given the rules and the independent variables mentioned above.

Evolution of CAD

GUI not just for artwork

When personal computers began to appear on the desks of design engineers in the 1980’s the inputs, and behavioral rules were programmed by the designers themselves. Graphical output was so crude that data produced by CAD was still hand plotted for publication. Slowly, the graphical capability of the hardware improved, and with it the availability of software that created informative graphical presentation.

The first widespread CAD program for circuit design was SPICE. Circuit elements (resistors, capacitors, inductors, transistors, diodes and so on) were entered via the keyboard. To facilitate error checking and documentation, circuit schematic output was soon added. Then, with the advent of the GUI, the schematic evolved to become a method of inputting data, as well as being an output.

Virtual reality not just for gamers

For most of the 1980’s & ‘90’s mechanical CAD was predominantly aimed at automating the drafting process to create two-dimensional projected views (top, front, side, …) of structural elements that were then handed off to the artisans to make prototypes. Sometimes complex shapes were first modeled and then the CAD drawings made from take-offs. The CAD process was not primarily an act of creativity; it served to document creativity.

During the 1990’s 3D CNC equipment became robust and affordable which encouraged the development of solid modeling software used to drive the CNC equipment. Synergies between computer gaming, computer graphics, parametric modeling and CNC further popularized the use of computer workstations as creative mechanical design tools.

First-pass design and ‘fab-less’ manufacturing

As behavioral models for circuit elements and computational capacity improved portions of the electronics industry moved into fab-less manufacturing. Detailed proprietary design rules allowed creating integrated circuits without ever going through a prototype phase. The specialized computer programs were closely guarded trade secrets, and not available commercially.

The possibility for first-pass design success attracted many others in the electronics industry to develop accurate models for the electronic components they used to build circuits. The models themselves became marketing tools for the electronic component manufacturers to attract customers. Clearly, a device model that is inaccurate or difficult to understand and  use doesn’t offer much marketing clout.

Synthesis– top-down design

Electronic CAD has now progressed to the point where circuits can be designed and optimized by envisioning an output and then (semi)automatically creating the necessary inputs. Instead of asking the question, “If I have a set of circuit elements wired together a certain way can I predict the circuit’s performance?” the designers of tomorrow will ask the question, “If I want a certain circuit performance, can I (semi)automatically come up with a set of circuit elements and a method for wiring them together?” In order to move toward that goal, the designer needs modeling tools that allow the inputs to be rapidly and easily adjusted by watching the dynamic response of the outputs. These models must be graphical, intuitive and “virtually real” in their precision.

Instead of entering a bunch of parameters and clicking ‘run’ the model should (ideally) display a flexible, tunable, realistic virtual device that, once the desired output is observed, displays the inputs used to arrive at the desired result. The ideal component CAD model will act as both the prototype lab and the documentation service. It must at the same time be virtually real and really accurate.

Share Button
Comments are now closed for this article.