PCB Copper Thickness Selection: 1oz, 2oz and Heavy Copper

Copper thickness is one of the few board parameters that changes both the electrical and the mechanical behaviour of a design at the same time, and it is chosen far earlier than most engineers realise. Once the laminate is ordered, the finished copper weight is fixed, and every decision about conductor width follows from it.

What One Ounce of Copper Actually Means

Copper weight is quoted as the mass of foil spread over one square foot, so one ounce per square foot corresponds to a layer about thirty-five micrometres thick. Two ounces is seventy micrometres, and the relationship stays linear until the weight reaches the point where the foil has to be laminated and then plated up rather than supplied as a single sheet.

The number that matters for design is the finished thickness after plating, not the starting foil. An outer layer that begins as one-ounce foil gains several micrometres during plating, while an inner layer keeps very close to its starting value. Quoting the outer layer at its starting weight is a common source of disagreement between a drawing and a shop.

Current Capacity and Self Heating

The current a trace can carry is limited by the temperature rise it is allowed to reach, not by a single melting figure. The classical charts relate current, conductor cross-section and permissible temperature rise for a trace on an outer layer in still air, and they show that a doubling of cross-section does not double the current, because the wider trace also has a larger surface from which to lose heat.

Designing to a temperature rise rather than to a current figure is the safer habit, since the same trace behaves differently on an inner layer where the laminate on both sides holds the heat in. Our notes on current capacity set out how the derating is usually applied for inner layers and for parallel conductors.

Copper thickness comparison on PCB laminate samples

Where Heavy Copper Changes the Process

Above about three ounces the fabrication process changes rather than merely scaling. Etching a thick copper layer removes material sideways as well as downwards, so the minimum space between conductors grows with the weight, and the achieved trace width can differ from the drawn width by an amount that is a significant fraction of a fine line.

The response is to design wider minimum features and to expect a trapezoidal cross-section. Plating thick copper also takes time proportional to the thickness, which is why heavy copper panels occupy the plating line for hours and why the lead time is longer than for a standard weight.

Thermal Mass and Soldering

A thick copper layer conducts heat away from a solder joint faster than a thin one, which changes the soldering recipe rather than the joint itself. Pads connected to a heavy plane need thermal relief or a longer dwell at temperature, and a board that solders easily at one ounce can be difficult at four ounces with the same profile.

The same property is an advantage in service, where heavy copper spreads heat from a power device into the surrounding plane. That is why heavy copper and metal-backed constructions are often considered together, and our metal core notes compare the two approaches.

The Effect on Impedance and Loss

Conductor thickness enters the impedance calculation because current crowds toward the surface at high frequency, so a thicker conductor presents a slightly lower impedance than a thin one of the same width. The shift is small compared with a change in dielectric thickness or width, which is why impedance controlled designs are usually specified by geometry and the copper weight is treated as a secondary variable.

Loss behaves differently. Conductor loss falls as the copper thickens, and on a long high-speed channel that reduction can matter. Our high speed design notes describe where the trade between thickness, width and loss belongs.

Heavy copper layer plated on a PCB panel

Copper Balance and Warpage

Uneven copper distribution across a layer is the usual cause of a board that bows after lamination, because the resin shrinks differently where it is reinforced by copper and where it is not. Adding a thieving pattern to the sparse areas equalises the resin content, and the benefit is largest on thin boards and on heavy copper layers.

Balance is judged per layer and across the stack, not just per board. A stack in which every layer has roughly the same copper distribution behaves predictably, while one that alternates between dense and sparse layers will warp regardless of how well each individual layer is balanced. Our copper balance notes describe the pattern rules.

Choosing the Weight

The starting point is the current, derated for the layer and the ambient temperature the product will see. Where the resulting trace width is inconveniently large, the alternatives are a heavier weight, a second conductor on another layer, or a change of topology such as moving the converter closer to its load.

Below that electrical requirement sits a floor set by handling and by the plating process, which is usually one ounce for an outer layer because thinner foil is easily damaged. Above about six ounces the design becomes a specialised heavy copper job with its own design rules, and the sensible step is to talk to the shop before the layout is finished rather than after.

Cost and Lead Time

Copper weight affects price in three ways: the foil itself, the additional plating time, and the larger minimum features that reduce the number of boards that fit on a panel. The third effect is often the largest, because a design that needs a wider spacing cannot be nested as tightly as a fine-line design of the same outline.

Lead time moves for the same reasons. Heavy copper is a smaller share of any shop’s capacity, so a four-ounce job waits for a slot in a line that is reserved for it. Our plating thickness notes explain how the deposit is controlled and measured.

Specifying Copper on the Drawing

A drawing should state the copper weight per layer, the finished thickness where it matters, and the minimum conductor width and space that the weight implies. Stating only the weight leaves the shop to assume a minimum feature size, and the assumption is usually optimistic.

Where a design mixes weights in one stack, the notes should say which layer is which, because a stack-up with a two-ounce inner layer and one-ounce outer layers is perfectly manufacturable but easy to misread from a layer table that lists weights without mapping them to layers.

FAQ

Is two-ounce copper twice as good as one-ounce? It carries roughly twice the current for the same temperature rise, because resistance is inversely proportional to cross-section, but the benefit is not proportional in every respect. The etching tolerance gets worse, the minimum space grows, and the board becomes harder to solder, so the gain is real but it is not free.

Can a heavy copper board still be impedance controlled? Yes, and the calculation simply uses the thicker conductor. The practical difficulty is that the wider minimum features limit how finely the width can be adjusted, so the impedance is usually tuned by changing the dielectric thickness or the layer stack rather than by trimming the trace width.

What does gopcb need to quote a heavy copper board? We need the copper weight for each layer, the minimum conductor width and space, the finished board thickness, the expected current per net and the maximum ambient temperature. With those five items we can confirm the design rules before the layout is released rather than after the panel is built.

Leave A Comment