Estimated reading time: 5 minutes

Target Condition: Distribution of Power—Denounce the Ounce

Have you ever wondered why the PCB design segment uses ounces to describe copper thickness? There’s a story behind all of this—a story that’s old, dusty, and more than a little absurd. (Note that I didn’t add “Like many of us.”)

Legend has it that back in the days of copper tinkers and roofing tradesmen, the standard was set when a craftsman hammered out a sheet of copper until it weighed one ounce, when its area conveniently matched the square of the king’s foot.

Fast-forward a few hundred years, and somehow, this medieval measure wormed its way into modern electronics. For too long, the PCB design world has clung to this outdated convention. Even today, it’s not uncommon to see PCB stackups simply calling out “1-ounce copper” or “2-ounce copper,” as though the royal foot still governs our trace widths and current carrying capacities for our power distribution networks.

Thankfully, the manufacturing industry has moved on. Now it’s time for designers to do the same.

The Legacy Lives On, But Should It?
There’s a debilitating loyalty lingering in the design community: Designers are still specifying copper using ounces per square foot (oz/ft²), a unit of mass per area, not thickness. That may be fine for ordering roofing material, but it introduces real problems when trying to define or verify the finished copper thickness of a circuit board.

Modern PCB manufacturing, led by standards like IPC-6012, has shifted to dimensional copper specifications—micrometers (µm) or micro-inches (µin)—because these units reflect actual thickness, not inferred mass. The ounce method doesn’t account for processing variation, doesn’t include tolerances, and isn’t directly measurable. In short, it’s an engineering liability.

To be fair, IPC-6012 doesn’t completely ignore the copper weight convention. It provides reference charts that correlate copper weight to nominal thickness. But then it does something far more important: It defines the actual minimum finished copper thicknesses required, in µm or µin, by class and by layer type (external, internal, plated, etc.). That’s the standard, and it's not just more precise; it’s measurable, enforceable, and documented for the way PCBs are manufactured today.

Why This Matters: Tolerance, Testing, and Traceability
Take a look at the title block tolerance on your company’s PCB drawing format. Do you see a tolerance for “1-ounce” or “2-ounce” copper? Probably not, and that’s exactly the problem. Specifying copper weight in ounces disregards any thickness tolerance. You wouldn’t spec a critical mechanical feature without a tolerance, so why would you do it for a critical electrical one, which could directly affect your power distribution solution?

Plainly speaking, when a PCB designer simply indicates "1-ounce copper" in a stackup, it causes an instant phone call or engineering query (EQ) because a PCB CAM engineer must know whether the designer made important calculations using base or finished copper thickness values. One-ounce thickness might mean 35 µm, or it might not. Depending on plating, etching, and process variation, your finished copper could be thinner—or thicker—than you expected. If your design depends on controlled impedance, thermal management, or precise voltage drops, those variations can cause real-world failures.

When you refer to IPC-6012 and specify copper thickness in micrometers or micro-inches, you're not just using more tangible units, you’re enabling the entire manufacturing and QA process to work more effectively:

Designers can accurately calculate trace width for current carrying and impedance

• Fabricators can measure actual copper thickness via micro-section
• Inspectors can assess quality to a defined spec with measurable criteria

By referring to the IPC-6012 charts and declaring base copper in your stackup, a PCB designer can calculate signal integrity and power distribution based upon minimum finished copper thickness: an absolute. The charts take the guesswork out of the equation.

Once you know the base and finish expectations, you can help your PCB supplier by specifying manufacturable conductor widths. Working with a supplier or an industry standard is the best way to avoid manufacturing pain points while choosing copper width and spacing values. I’ve found this NCAB PCB design guideline to be very helpful in indicating trace and space (gap) capability. As in IPC-6012, this chart also leads designers in referring to copper base material in ounces, but then shows inner layer (IL) and outer layer (OL) print and etch capabilities.

A Call to Designers
The industry has done its part by adopting dimensional copper specs. It’s time for the design community to fully step into the present. Let’s finally lay the king’s foot to rest and stop measuring copper thickness with an ancient unit of roofing material.

Instead, use an accurate, measurable strategy. First, determine your power distribution requirements and how to solve them using minimum foil thickness after processing for inner layers, and minimum surface conductor thickness after processing for outer layers per the IPC-6012 charts.

Second, determine optimal conductor routing widths based on supplier recommendations for copper thickness. Our suppliers are tired of PCB designers routing 4-mil tracks on outer base copper layers, which are too thick to process. For the same reason, they are tired of asking if they can trim back copper topology that only leaves a 4-mil gap between conductors.

Denounce the Ounce
Consider renouncing PCB thickness values expressed in weight. Who needs a measurement based on the length of a king’s foot squared? Consider expecting IPC-6012 minimum processed or finished measurement units for your power distribution values. Your manufacturers will thank you. Your QA team will thank you, andyour boards will perform better because of it.

This column originally appeared in the October 2025 issue of Design007 Magazine.