Estimated reading time: 4 minutes

Flex Connections: Four Lessons in Engineering Judgment for PCB Designers

Editor’s note: This begins a new column series by Flexible Circuit Technologies, which will appear monthly in I-Connect007 Magazine. The column will feature five rotating authors who share expertise from their many years of working with flexible circuits.

In my more than four decades in the industry, I’ve watched design tools and manufacturing capabilities evolve in ways we could hardly have imagined when I started, touching everything from conventional rigid boards to today’s highly sophisticated flexible and rigid-flex circuits. But anyone with meaningful experience in the PCB industry can relate to the phrase, “Just because you can doesn’t mean you should.”

Some of the advances I’ve seen have made us better engineers. Others have made it easier to create unnecessary complexity. Our CAD systems let us specify incredibly tight tolerances, route astonishingly dense designs, and call out requirements that were once impossible to manufacture. It’s impressive, but using it wisely requires engineering judgment, something software can’t provide.

Looking back over my career, where I have worked in various PCB manufacturing environments, including production, prototype, and R&D, each setting provided distinct technical and process-related insights, and reinforcing the importance of anticipating downstream challenges and planning several steps ahead to ensure successful execution.

Thousands of PCB designs have crossed my desk over the years. Most have been practical manufacturing designs with requirements aligned to available process capability. Others, however, included specifications and layout decisions that were difficult to justify from a manufacturing standpoint.

Here are a few examples that illustrate the “just because you can doesn’t mean you should” idea.

From Tape to CAD
When I started, PCB artwork was created with tape on enlarged layouts. Every change took time and effort, so designers naturally thought carefully before making one. You didn’t make a design more complicated unless there was a good reason.

Today’s software allows us to make changes almost instantly, which is an incredible advantage. But that same convenience also makes it easy to add complexity without stopping to ask whether it’s necessary.

But just because your software lets you route one more trace through a congested area doesn’t mean that’s the best engineering solution. Sometimes moving a component or changing the stackup creates a board that’s easier to manufacture, easier to inspect, and ultimately more reliable.

Tolerances
One example I’ve seen many times is dimensional tolerances. Designers can specify incredibly tight positional tolerances for connector holes, pads, and other critical features. I’ve seen flex designs where every bend area carried unnecessarily tight fabrication requirements. Sometimes that’s exactly what’s needed.

But just because you can specify a tighter tolerance doesn’t mean the product requires it. Every tighter tolerance limits your manufacturing options. It may increase cost, extend lead times, and reduce the number of suppliers capable of building the board, all without improving how the product actually performs.

That same principle applies to highly dense routing combined with fabrication requirements that only a limited number of suppliers can meet. While modern layout software makes extremely challenging designs possible, those decisions must still be evaluated against program cost targets, schedule constraints, and practical manufacturability.

IPC Class 3
I see the same thinking when IPC Class 3 gets specified. It’s easy to assume that asking for the highest level of workmanship automatically produces the best board.

But just because you can specify Class 3 doesn’t mean your application benefits from it. Class 1 applies to products for which the primary requirement is the basic function of the completed assembly. Class 2 applies to products that require continued performance and extended service life, where uninterrupted operation is desired but not mission-critical.

If the product truly requires a Class 3 high performance/harsh environment, then it’s absolutely the right decision. But if the product is really a Class 2 application, you’ve added inspection, documentation, testing, and cost without adding value to the customer. Class 3 generally carries the highest fabrication and verification cost, and it can extend lead time depending on the qualification and inspection requirements imposed.

This becomes especially important in flex and rigid-flex designs, where manufacturing complexity is already higher than in conventional rigid boards.

Each IPC class establishes distinct acceptance criteria, including annular ring dimensions, copper plating thickness, and allowable feature-level anomalies such as nicks, pits, and other discontinuities.

Working With Flex
Flexible circuits allow us to do some remarkable things today. We can create incredibly dense rigid-flex designs with multiple bend areas, exotic materials, and complex constructions.

But just because a design can be built doesn’t necessarily mean it should be built that way. Sometimes a slightly larger bend radius, a simpler layer count, or a different material selection creates a product that’s easier to manufacture and more reliable in the field.

Final Thoughts
Looking back over 40 years, I’ve learned that good engineering isn’t about pushing every design to the limit of today’s technology. I want to understand the product’s requirements and make decisions that support them. The best designs achieve the right balance of performance, reliability, manufacturability, and cost. Input from a capable fabricator can reduce total program cost by aligning material selection and fabrication methods with the actual functional requirements of the design.

Every time I review a design, I find myself coming back to this simple reminder: Just because you can doesn’t mean you should. It’s a lesson that has served me well throughout my career, and one that’s just as relevant today as it was when I first entered this industry.

Dan Skweres is a senior applications engineer for Flexible Circuit Technologies.