Power Design

In the early days of feedback and Electrical Engineering analysis, advances were made to define and explain the operation of feedback circuits in terms of pure mathematics. Feedback was sufficiently useful, and stability sufficiently problematic!

When the first theories proved too mathematical for design use, and too weak in the face of nonlinearity, a crisis arose. Engineers who designed intuitively arrived at poor results only slowly, while theoreticians drowned in the impracticality of excess elaboration.

The resolution came when Engineers began “pushing” as much mathematics (and physics!) as possible back into their schematic diagrams. And, as manufacturers slowly idealized their components, designers began to shift more easily between actual circuits, and circuit models featuring idealized voltage and current generators mixed with R’s, L’s, and C’s.

Further, the problem of nonlinearity was dispensed with when such behavior came to be treated linearly, piece by piece, in complex models. The idea was to let a complicated model carry the information defining the nonlinearlity, allowing the engineer to comprehend its complexities visually. He had then only to choose those elements pertinent to his design to avoid irrelevant complication.

More advances came when it became clear that substituting the Thevenin or Norton equivalent model of a part of a circuit made the inevitable math much easier. This maneuver allowed generators and impedances to be combined, simplifying the circuit, until the answer was available by inspection!

Caltech’s Dr. R.D. Middlebrook has perfected this technique, pointing out that most of the necessary circuit analysis for power supply design can be performed with reference to an idealized schematic of the switcher itself [his famous Canonical Model!] Remarkably, the technique not only eliminates much of the mathematical calculation previously done (by hand, or later with SPICE) but allows the engineer to see what he is doing as he designs. Thus, he can make his design do what he wants it to, and guide the project to a successful conclusion, meeting a spec with confidence.

This “Design Oriented” analysis, then, is no longer the work of the math major, but that of an Engineer! Similar ideas make up the growing field of Engineering Theory, which serves the Engineer’s needs, and may even someday be profitably taught to the rank and file of impractical Scientists and Mathematicians!

DC Conditions

In modeling the BJT, account had had to be taken for the fact that such devices operate at a steady state, or quiescent operating point. That is, in order for the device to work at all, and if it is not to pass into saturation on positive or negative excursion, a DC bias is necessary.

Fortunately, the design and analysis problems posed by DC conditions in small and large signal linear circuits are rarely difficult, and solution of “the DC circuit” is usually straightforward.

Power DC Conditions

The situation in Power Electronics is considerably different….

Here, the DC conditions involved are numerous, nonlinear, critical to the design, and even pose something of a safety hazard to the inexperienced engineer;) Those most often encountered are the source voltage, Vg; the output voltage V; the switch voltage and current Vs,Is; the stresses on the L’s and C’s; and the excitation voltage of the transformer.

In order to determine the DC operating point in a switcher, an engineer needs, as in the AC case, Design Oriented Analysis. His design equations contain too many variables, too few variables, and/or clusters of undetermined constants. [Note to self: Is there such a thing as a variable constant?]

The result is that these sets of equations (usually not even linear equations) cannot be solved, but only cast into the most useable form. Miraculously, when one does so, it becomes easy to use them to find the DC current and voltage stresses in your converter. Such things as low line, high load, switch stress, and so on can be found easily, even during design when other factors are changing.

Design Oriented Form of the DC Conditions

To the above schematic of the Boostbuck (Cuk) Converter are appended design oriented equations for the various DC stresses. {Assuming 100% efficiency, i.e. V/Vg = D/D’}

Inventor vs. Engineer

An inventor works intuitively, and is at his best genius when he is at his wits’ end.

An Engineer is a trained professional, has already realized he is not a genius, and is humble enough to work responsibly with the full knowledge of the subject he has been taught.

An Inventor dreams of glory, and of the impossible.

An Engineer has a job to do, and is a steward of the world’s hardware & software.

Who is an Engineer?

[Not an email, just found it browsing!]

Some ideas from the article: “Who’s an engineer” published in “IEEE Spectrum” magazine, July 1999.
Q&A; attributed to Abraham Lincoln:

Q: If you call a horse’s tail a leg, how many legs does a horse have?
A: Four. Calling a tail a leg doesn’t make it a leg.

Some people should not call themselves engineers without proper training in engineering.

Learning to use a computer does not make one a computer or software engineer, any more than driving a car makes one an automotive engineer.

www.powersupplies.net Then click on “Ethics”