Copper Weight: Preparation, Placement and Process Control
Copper spreads heat laterally, and the amount it spreads depends on how thick it is. A design that moves heat away from a component is partly a thermal design and partly a copper weight decision, and the two are usually made in different meetings.
How Copper Spreads Heat
Heat entering a copper plane spreads outward in all directions, and the spreading resistance falls as the copper gets thicker. The plane acts as a fin, so the further the heat travels the less effective each additional area becomes.
That means the first millimetre of spreading does most of the work, which is why a small increase in copper weight near the source can matter more than a large plane far away. Our thermal design notes describe how the paths are usually arranged.
Copper Weight and Its Effect on Temperature Rise
A one ounce plane and a two ounce plane of the same outline do not differ by a factor of two in spreading because the fin behaviour is non-linear. The improvement is real and it is smaller than the ratio of the thicknesses.
The measurement that matters is the temperature rise at the component for a given power, which is taken on a test board rather than calculated from the copper weight alone. Our metal core notes describe the alternative when copper alone is not enough.
Where the Copper Has to Be
Spreading works when the plane is close to the source, and it works poorly when a thin dielectric separates the source from a plane two layers away. The first copper layer under the component is therefore the one that matters most.
That is why a thermal pad is placed on the top layer as well as on the bottom, and why the vias that connect them are placed under the pad rather than at its edge. A via at the edge of a pad does not help the centre.
Vias and Their Contribution
A thermal via array carries heat through the board to the plane below, and its thermal resistance depends on the copper in the barrel. A via that is filled and plated over conducts well and costs more; an open via conducts less and is cheaper.
The array is usually most effective under the centre of the pad, where the temperature is highest. A single ring of vias around the periphery spreads less than a grid because the centre remains the hot spot. Our current capacity notes describe the same geometry from the electrical side.
Etch Limits on Heavy Copper
A heavy copper layer cannot hold a fine line, because the etch removes copper sideways in proportion to the thickness. A design that needs both a heavy plane and a fine pattern uses two layers rather than one.
The minimum line width for a given copper weight is a fabrication figure rather than a design preference, and it is worth stating on the stack-up. Our plating thickness notes describe how the figure is controlled.
Plating Versus Foil Weight
The copper weight on an inner layer is the foil weight, and the weight on an outer layer is the foil plus the plating. That means an outer layer described as two ounce may be one ounce foil with one ounce of plating, and the plating is not uniform.
Where the thermal or current figure is specified on the outer layer, the distinction matters. The specification should say whether it is the finished weight or the base foil, and the measurement should be taken at the finished figure.
Cost and Trade-Offs
Heavy copper costs more in the material, in the etch, and in the drilling because a thick laminate is needed to carry it. The cost rises faster than the weight, so a design that can meet its requirement at one and a half ounces usually should.
The alternative to more copper is a different structure: a metal core, a thicker dielectric, or a thermal interface to a heatsink. Each is cheaper than heavy copper in some cases and more expensive in others, and the comparison is made on the measured temperature rather than on the drawing.
Verifying the Result
The verification is a thermal measurement on the assembled board at the specified power, with the thermocouple placed at the component case or at the pad rather than at the board edge. The measurement is compared with the target rather than with the previous design.
Keeping the test board with its copper weights recorded makes the result reusable. A measurement without the copper weight and the stack is a number that cannot be compared with anything.
Process Control and Verification
The process window is set by the narrowest step in the flow, so an improvement anywhere else shows up as margin rather than as yield until that step is addressed. A short note on the drawing about handling, storage or packaging is often worth more than an extra decimal place on a tolerance.
Where a value sits close to a process limit, the drawing should say so, since the shop can then open the process window rather than working to a nominal figure that carries no tolerance.
Checks Before Release
The checks that matter are the ones performed on the product rather than on a sample kept for the purpose, because a coupon that travels with the panel is the only evidence about that panel. Where a requirement can be measured, it should be measured at the point of manufacture and recorded against the board or the lot it applies to.
A parameter that is set once and never re verified drifts, and the drift is usually discovered by a defect rather than by the record.
FAQ
Does doubling the copper halve the temperature rise? No. Spreading is non-linear, so the improvement is real and considerably smaller than the ratio of thicknesses.
Are filled vias worth the cost for thermal paths? Where the joint underneath must be flat they are, and for thermal conduction alone an open via array is usually adequate.
What does gopcb provide for thermal copper design? We provide copper weight and plane placement recommendations, via array layout under the pad, etch limits stated for the weight, the finished weight specified rather than the foil, thermal measurement on the assembled board, and records that tie the result to the stack.
