Copper Foil Roughness: Preparation, Placement and Process Control

Copper foil is not smooth. The side that bonds to the laminate is treated to give it a rough, dendritic surface that grips the resin, and the scale of that roughness is measured in microns. At low frequencies the roughness is irrelevant, but at microwave frequencies and at high data rates it begins to matter a great deal.

The reason is the skin effect. As frequency rises, current concentrates in a thin layer at the surface of the conductor, and that layer becomes thinner than the roughness of the foil. The current then has to follow the profile of the surface, which lengthens the path and increases the loss.

What Foil Roughness Is

Foil is produced by electrodeposition onto a rotating drum, and the drum side is smooth while the deposited side is rough. The rough side is the one bonded to the laminate, so the copper surface facing the dielectric carries the profile that matters for high frequency loss.

Standard foil has a roughness in the region of two to three microns, while a low profile foil can be below one micron. The difference sounds small, but it can change the insertion loss of a long trace by a measurable percentage.

Skin Effect And Frequency

Skin depth falls with the square root of frequency, so the current crowding becomes more severe as the data rate rises. At one gigahertz the skin depth in copper is a little over two microns, which is comparable to the roughness of a standard foil.

Above that frequency the current is effectively travelling over a rough surface, and the extra path length shows up as additional conductor loss. The effect is modest at five gigahertz and significant at twenty five.

How Roughness Is Measured

Roughness is reported as an arithmetic mean deviation of the profile, usually written as Rz or Ra depending on the method. Rz describes the average peak to valley height over the sampled length and is the figure most often quoted for foil.

The measurement is made with a profilometer on a sample of foil or on a cross section of the finished laminate. Because the foil is bonded and pressed, the profile after lamination differs from the profile of the raw foil, and the finished value is the one that matters.

Standard Versus Low Profile Foil

Standard electrodeposited foil is the cheapest and the most widely available. It is used for the great majority of boards, and for anything below a few gigahertz it is perfectly adequate.

Low profile and very low profile foils are produced with a finer dendritic treatment or by a different deposition process. They cost more and they are supplied in a narrower range of weights, so they should be specified only where the loss budget requires them.

Effect On Insertion Loss

The roughness contribution is usually modelled as a correction factor applied to the smooth conductor loss, and the factor rises with both frequency and roughness. Design tools normally include a roughness parameter in the transmission line model.

Leaving the parameter at its default value is a common source of error. If the model assumes a smooth conductor and the board is built with standard foil, the measured loss will be higher than the simulation predicted, and the link margin disappears.

Adhesion And The Trade Off

Roughness exists for a reason. The dendritic surface is what gives the foil its peel strength against the resin, and a smoother foil bonds less well. Low profile foils compensate with improved chemical treatment, but the adhesion margin is still smaller.

That matters for a board that will be thermally cycled or that has large copper areas. Where a very smooth foil is used, the lamination process and the surface treatment become more critical, and the fabricator should be consulted before the stackup is released.

Choosing Foil For A Design

The decision should be driven by the loss budget rather than by a general preference for better material. If the budget can be met with standard foil, there is no reason to pay for a low profile product.

Where the budget is tight, the choice is usually to move to a lower loss laminate first, because the dielectric loss is normally the larger contributor. Foil roughness is the second lever and it is applied when the dielectric has already been improved.

Laminate And Foil Combinations

Low loss laminates are often supplied with a recommended foil, and the two are qualified together. Mixing a low loss laminate with a rough foil wastes much of the benefit of the laminate, because the conductor loss then dominates at high frequency.

The combination should be treated as a single material system and qualified as such. The dielectric properties of the materials are described under low loss laminate selection.

Verification

Verification is by a measurement on a test coupon rather than by a certificate. A coupon with a known trace geometry, measured with a vector network analyser, gives the insertion loss of the actual stackup including the foil effect.

Comparing that measurement with the simulation closes the loop and confirms that the roughness parameter was set correctly. The coupon practice involved is described under impedance coupon design.

Process Control and Verification

Documenting the assumption is part of the design work, and a short note on the drawing prevents a question that would otherwise arrive a day later and cost a day of schedule. 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.

Related reading: our fabrication notes, board quality and design release notes cover the same ground.

A first article check confirms that the process and the drawing agree on the points listed above, and that the coupon data supports the values used in the design.

Where a measurement falls outside the expected window, the sample is retained so that the cause can be established before the balance of the batch is released.

FAQ

Does foil roughness affect impedance? Only slightly, and mainly through the effective conductor thickness. Its dominant effect is on loss rather than on the characteristic impedance of the trace.

Can roughness be reduced after lamination? No. The profile is created when the foil is made and it is locked in by the lamination. The choice is made at the material selection stage.

Is low profile foil worth it for a short trace? Usually not. The loss contribution scales with length, so a short interconnect rarely justifies the extra material cost.