Etch Uniformity Control Across the PCB Panel

Etch uniformity decides whether the traces on a panel finish at the width the design asked for, and it is measured across the panel rather than at a single point. A line that etches the centre correctly and the edges lightly will produce boards that pass a line width check in the middle and fail it in the corners.

The effect is not small. A difference of 10 percent in etch rate across a panel can move a 0.1 mm trace by several micrometres on each side, which is enough to change the impedance of a controlled line and enough to fail a fine line inspection.

What Uniformity Means

Uniformity is the spread of the etched result across the panel, usually expressed as a percentage of the average. It is not the same as the absolute etch rate, because a process can be uniform and still be set too fast or too slow for the design.

Both numbers should be reported together. Etch rate tells the operator whether the line is running as qualified, while uniformity tells the process owner whether the spray system, the conveyor or the panel layout has changed. A spread of five percent across the working area is a practical target for a conveyorised etcher, while a spread above ten percent should be treated as an out-of-control process rather than as normal variation.

Etch Rate Measurement

Etch rate is measured by etching a known thickness of copper for a known time and dividing the result, and it is normally tracked through the conveyor speed and the measured line width rather than by weighing panels.

A test coupon placed at the leading edge, the centre and the trailing edge of a panel gives the spread in one pass. Doing that once per shift produces a record that shows a worn nozzle or a partial blockage long before the line width reaches its etch bath control limit. The coupons should carry the same copper thickness as the production panels and be etched in the same pass, because a coupon run on its own describes a machine that is not the machine that runs the work.

Spray Pressure and Nozzle Condition

Etching happens at the surface through a boundary layer of solution, and the spray exists to break that layer and replace it with fresh chemistry. Pressure that is too low leaves a thick boundary layer and slows the etch, while pressure that is too high damages fine features.

Nozzles wear and block, and the pattern they produce changes long before the pressure gauge moves. Spray patterns should be checked visually at the start of each shift, and the nozzle set should be replaced as a group rather than one at a time. Oscillating spray banks even out the pattern, and the amplitude and frequency of that oscillation are part of the set-up that has to be recorded with the run.

Conveyor Speed and Dwell

Conveyor speed sets the dwell time in each chamber, and it is the parameter most often adjusted to compensate for another problem. Speeding up the line to reduce over-etch on a dense panel also reduces the etch on the opposite side of the same panel.

Where the chemistry, the etching chemistry and the conveyor speed are all changed in one shift, the record no longer explains the result. One variable should be adjusted at a time, with a coupon measurement between each change. A reduction in conveyor speed increases the etch on both sides at once, while a difference between the top and the bottom points to a spray or drainage problem rather than to dwell.

Copper Balance and Pattern Effects

Copper consumes the etchant locally, so a panel with dense copper area etches differently from one with large open areas. Where a design has an unbalanced layout, the etchant in the dense region becomes depleted and the etch slows there.

Thieving and balanced copper distribution reduce the difference, and the copper balance of the panel should be reviewed whenever a new design is released. A layout that etches uniformly on one panel can behave differently when the copper area changes. Panels that remove less than twenty percent of their copper are the hardest to etch evenly, which is why thieving and a balanced copper distribution are worth adding at the design stage.

Sidewall Profile and Undercut

The sidewall is where the etch result becomes visible as a shape. An ideal profile is close to vertical, while an over-etched trace shows undercut beneath the resist and a trapezoidal cross section that reduces the effective width of the conductor.

Undercut is measured in a microsection, and it grows with etch time, with spray pressure and with the age of the resist. A trace that is within its width tolerance but heavily undercut will behave differently at high frequency, and the etch factor and line width record should carry the profile as well as the number. A vertical sidewall is the aim, and a measurable taper is acceptable only while the resulting line width still falls inside the tolerance band.

Corrective Action

When uniformity falls, the order of investigation is fixed: check the spray pattern, then the pressure, then the chemistry, then the conveyor. The spray system is the most common cause and the cheapest to correct, while chemistry changes are the most disruptive and should come last.

A single nozzle that is blocked produces a localised stripe of under-etching that follows the panel as it travels. That pattern is diagnostic, and it is one of the few defects in a wet process that can be identified from the appearance of the panel alone.

Acceptance Criteria

Acceptance should state the average line width, the tolerance, and the allowed variation across the panel, because a board can be within tolerance everywhere and still show a spread that indicates a process problem.

The criterion should also state where the measurements are taken. Reading a width at the panel centre only is a common shortcut, and it is the reason a corner-related defect reaches the customer rather than being caught in the etch line.

Records and Trend Monitoring

Records should carry the bath chemistry, the specific gravity and the copper concentration, the spray pressure, the conveyor speed and the coupon results with their positions. With those fields, an excursion can be tied to a bath addition or to a mechanical fault. Bath chemistry, specific gravity and copper concentration belong in the daily record, tied to the line width data so that the two can be compared directly.

Trending the spread rather than the average is the most useful view. A stable average with a widening spread points to the spray system, while a moving average with a stable spread points to the chemistry, and the two call for different corrective actions.

FAQ

Why does the etch differ between the centre and the edges of a panel? Spray coverage, panel support and copper distribution all vary across the panel, and the nozzle pattern is the first item to check when the difference appears.

How is etch uniformity measured? By etching coupons or line width patterns at several positions across the panel and comparing the results as a percentage of the average, once per shift and after any change to the line.

Does higher spray pressure improve uniformity? Up to a point. Beyond the pressure the nozzles were designed for, fine features are damaged and the profile becomes trapezoidal without any gain in uniformity.