Models can reveal what design rules "forgot."

A picture is worth a thousand words, and that is especially true when evaluating the impact of changing trace dimensions in your printed circuit board.

The capabilities today of visualizing, in a matter of seconds, the complexity of the electromagnetic fields in one's printed circuit board are astounding. Twenty years ago, software tools to help designers make informed decisions were limited to numerical outputs or very rudimentary images. Often one required a significant level of understanding of the physics involved as well as a solid imagination to properly visualize the electromagnetic field lines and their significance. Even then, most visualizations were representations, not actually calculated fields displayed over the actual board and components. Many tools today provide unprecedented insights through such visualizations, equipping and enabling quick and intelligent decision making by designers on an unprecedented level.

I recall – and I hope you had a similar experience – a science class that used iron shavings and a magnet to make visible the once invisible magnetic fields as in FIGURE 1.

Numbers on a data sheet are a good start, but confirm them with your own testing.

"Trust, but verify." While some readers will recall hearing this in the context of nuclear disarmament, I have it most notably drilled into my conscience by my first engineering manager while working in software quality assurance. And for good reason too! It was literally my role to verify the fixes that the engineers claimed they had implemented and to find the issues that were still open. While most of the time their fixes resolved the issue, sometimes it was only partially dispositioned.

So how do we successfully trust, but verify when it comes to information in data sheets for not just copper roughness, but other values significant for modeling signal integrity in printed circuit boards?

In a previous column, I discussed copper roughness and a partial story of the lifecycle of the copper as it moves along in the process of becoming part of a printed circuit board. In that article, I mentioned the dielectric constant/permittivity (Dk) and the dissipation factor/loss tangent (Df) of the resin/filler/glass combination. The data sheets used to be very secretive, with material suppliers concerned about corporate espionage regarding chemistries if the Dk and Df information was made publicly available. Depending on the material supplier and the materials in question, many times one can also find the construction data – glass weave style and quantity, resin content, and shelf thickness.